Pharmaceutical

Process Validation
An International Third Edition, Revised and Expanded

edited by
Robert A. Nash
Stevens Institute of Technology
Hoboken, New Jersey, U.S.A.

Alfred H.Wachter
Wachter P h r m a Projects
Therwil, Switzerland

MARCEL

MARCELDEKKER,
INC. NEWYORK * BASEL
DEKKER
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 2003 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Version Date: 20141006

International Standard Book Number-13: 978-0-203-91211-9 (eBook - PDF)

This book contains information obtained from authentic and highly regarded sources. While all reasonable
efforts have been made to publish reliable data and information, neither the author[s] nor the publisher can
accept any legal responsibility or liability for any errors or omissions that may be made. The publishers wish to
make clear that any views or opinions expressed in this book by individual editors, authors or contributors are
personal to them and do not necessarily reflect the views/opinions of the publishers. The information or guid-
ance contained in this book is intended for use by medical, scientific or health-care professionals and is provided
strictly as a supplement to the medical or other professional’s own judgement, their knowledge of the patient’s
medical history, relevant manufacturer’s instructions and the appropriate best practice guidelines. Because of
the rapid advances in medical science, any information or advice on dosages, procedures or diagnoses should
be independently verified. The reader is strongly urged to consult the relevant national drug formulary and the
drug companies’ printed instructions, and their websites, before administering any of the drugs recommended
in this book. This book does not indicate whether a particular treatment is appropriate or suitable for a particular
individual. Ultimately it is the sole responsibility of the medical professional to make his or her own professional
judgements, so as to advise and treat patients appropriately. The authors and publishers have also attempted to
trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if
permission to publish in this form has not been obtained. If any copyright material has not been acknowledged
please write and let us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted,
or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, includ-
ing photocopying, microfilming, and recording, or in any information storage or retrieval system, without writ-
ten permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com
(http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Dan-
vers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a
variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of
payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only
for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com
 DRUGS AND THE PHARMACEUTICAL SCIENCES

Executive Editor
James Swarbrick
PharmaceuTech, Inc
Pinehurst, North Carolina

Advisory Board

Larry L. Augsburger David E. Nichols

University of Maryland Purdue University
Baltimore, Maryland West Lafayette, Indiana

Douwe D. Breimer Stephen G. Schulman

Gorlaeus Laboratories University of Florida
Leiden, The Netherlands Gamesville, Florida

Trevor M Jones Jerome P. Skelly

The Association of the Alexandria, Virginia
British Pharmaceutical Industry
London, United Kingdom

Hans E. Junginger Felix Theeuwes

Leiden/Amsterdam Center Alza Corporation
for Drug Research Palo Alto, California
Leiden, The Netherlands

Vincent H. L. Lee Geoffrey T Tucker

University of Southern California University of Sheffield
Los Angeles, California Royal Hallamshire Hospital
Sheffield, United Kingdom

Peter G. Welling
Institut de Recherche Jouvemal
Fresnes, France
 DRUGS AND THE PHARMACEUTICAL SCIENCES

A Series of Textbooks and Monographs

1. Pharmacokmetics, Milo Gibaldi and Donald Perrier

2. Good Manufacturing Practices for Pharmaceuticals: A Plan for Total
Quality Control, Sidney H. Willig, Murray M. Tuckerman, and William
S. Hitchings IV
3. Microencapsulation, edited by J. R Nixon
4. Drug Metabolism. Chemical and Biochemical Aspects, Bernard Testa
and Peter Jenner
5. New Drugs: Discovery and Development, edited by Alan A. Rubin
6. Sustained and Controlled Release Drug Delivery Systems, edited by
Joseph R. Robinson
7. Modern Pharmaceutics, edited by Gilbert S. Banker and Christopher
T. Rhodes
8. Prescription Drugs in Short Supply Case Histories, Michael A.
Schwartz
9. Activated Charcoal' Antidotal and Other Medical Uses, David O.
Cooney
10. Concepts in Drug Metabolism (in two parts), edited by Peter Jenner
and Bernard Testa
11. Pharmaceutical Analysis: Modern Methods (in two parts), edited by
James W, Munson
12. Techniques of Solubilization of Drugs, edited by Samuel H Yalkow-
sky
13. Orphan Drugs, edited by Fred E. Karch
14. Novel Drug Delivery Systems: Fundamentals, Developmental Con-
cepts, Biomedical Assessments, Yie W. Chien
15. Pharmacokmetics: Second Edition, Revised and Expanded, Milo
Gibaldi and Donald Perrier
16 Good Manufacturing Practices for Pharmaceuticals' A Plan for Total
Quality Control, Second Edition, Revised and Expanded, Sidney H
Willig, Murray M Tuckerman, and William S. Hitchings IV
17 Formulation of Veterinary Dosage Forms, edited by Jack Blodinger
18 Dermatological Formulations. Percutaneous Absorption, Brian W
Barry
19. The Clinical Research Process in the Pharmaceutical Industry, edited
by Gary M. Matoren
20. Microencapsulation and Related Drug Processes, Patrick B. Deasy
21. Drugs and Nutrients The Interactive Effects, edited by Daphne A.
Roe and T. Colin Campbell
22. Biotechnology of Industrial Antibiotics, Enck J. Vandamme
23 Pharmaceutical Process Validation, edited by Bernard T Loftus and
Robert A Nash
24 Anticancer and Interferon Agents Synthesis and Properties, edited by
Raphael M Ottenbrtte and George B Butler
25 Pharmaceutical Statistics Practical and Clinical Applications, Sanford
Bolton
26 Drug Dynamics for Analytical, Clinical, and Biological Chemists,
Benjamin J Gudzmowicz, Burrows T Younkm, Jr, and Michael J
Gudzmowicz
27 Modern Analysis of Antibiotics, edited by Adjoran Aszalos
28 Solubility and Related Properties, Kenneth C James
29 Controlled Drug Delivery Fundamentals and Applications, Second
Edition, Revised and Expanded, edited by Joseph R Robinson and
Vincent H Lee
30 New Drug Approval Process Clinical and Regulatory Management,
edited by Richard A Guarino
31 Transdermal Controlled Systemic Medications, edited by Yie W Chien
32 Drug Delivery Devices Fundamentals and Applications, edited by
Praveen Tyle
33 Pharmacokinetics Regulatory • Industrial • Academic Perspectives,
edited by Peter G Welling and Francis L S Tse
34 Clinical Drug Trials and Tribulations, edited by Alien E Cato
35 Transdermal Drug Delivery Developmental Issues and Research Ini-
tiatives, edited by Jonathan Hadgraft and Richard H Guy
36 Aqueous Polymeric Coatings for Pharmaceutical Dosage Forms,
edited by James W McGmity
37 Pharmaceutical Pelletization Technology, edited by Isaac Ghebre-
Sellassie
38 Good Laboratory Practice Regulations, edited by Alien F Hirsch
39 Nasal Systemic Drug Delivery, Yie W Chien, Kenneth S E Su, and
Shyi-Feu Chang
40 Modern Pharmaceutics Second Edition, Revised and Expanded,
edited by Gilbert S Banker and Chnstopher T Rhodes
41 Specialized Drug Delivery Systems Manufacturing and Production
Technology, edited by Praveen Tyle
42 Topical Drug Delivery Formulations, edited by David W Osborne and
Anton H Amann
43 Drug Stability Principles and Practices, Jens T Carstensen
44 Pharmaceutical Statistics Practical and Clinical Applications, Second
Edition, Revised and Expanded, Sanford Bolton
45 Biodegradable Polymers as Drug Delivery Systems, edited by Mark
Chasm and Robert Langer
46 Preclmical Drug Disposition A Laboratory Handbook, Francis L S
Tse and James J Jaffe
47 HPLC in the Pharmaceutical Industry, edited by Godwin W Fong and
Stanley K Lam
48 Pharmaceutical Bioequivalence, edited by Peter G Welling, Francis L
S Tse, and Shrikant V Dinghe
49. Pharmaceutical Dissolution Testing, Umesh V. Sana/car
50. Novel Drug Delivery Systems: Second Edition, Revised and
Expanded, Yie W. Chien
51. Managing the Clinical Drug Development Process, David M. Coc-
chetto and Ronald V. Nardi
52. Good Manufacturing Practices for Pharmaceuticals: A Plan for Total
Quality Control, Third Edition, edited by Sidney H. Willig and James
R. Stoker
53. Prodrugs: Topical and Ocular Drug Delivery, edited by Kenneth B.
Sloan
54. Pharmaceutical Inhalation Aerosol Technology, edited by Anthony J.
Mickey
55. Radiopharmaceuticals: Chemistry and Pharmacology, edited by
Adrian D. Nunn
56. New Drug Approval Process: Second Edition, Revised and Expanded,
edited by Richard A. Guarino
57. Pharmaceutical Process Validation: Second Edition, Revised and Ex-
panded, edited by Ira R. Berry and Robert A. Nash
58. Ophthalmic Drug Delivery Systems, edited byAshim K. Mitra
59. Pharmaceutical Skin Penetration Enhancement, edited by Kenneth A.
Walters and Jonathan Hadgraft
60. Colonic Drug Absorption and Metabolism, edited by Peter R. Bieck
61. Pharmaceutical Particulate Carriers1 Therapeutic Applications, edited
by Alain Rolland
62. Drug Permeation Enhancement: Theory and Applications, edited by
Dean S. Hsieh
63. Glycopeptide Antibiotics, edited by Ramakrishnan Nagarajan
64. Achieving Sterility in Medical and Pharmaceutical Products, Nigel A.
Halls
65. Multiparticulate Oral Drug Delivery, edited by Isaac Ghebre-Sellassie
66. Colloidal Drug Delivery Systems, edited byJorg Kreuter
67 Pharmacokinetics: Regulatory • Industrial • Academic Perspectives,
Second Edition, edited by Peter G. Welling and Francis L. S. Tse
68. Drug Stability: Principles and Practices, Second Edition, Revised and
Expanded, Jens T. Carstensen
69. Good Laboratory Practice Regulations: Second Edition, Revised and
Expanded, edited by Sandy Weinberg
70. Physical Characterization of Pharmaceutical Solids, edited by Harry
G. Bnttain
71. Pharmaceutical Powder Compaction Technology, edited by Goran Al-
derborn and Christer Nystrom
72. Modern Pharmaceutics. Third Edition, Revised and Expanded, edited
by Gilbert S. Banker and Christopher J Rhodes
73. Microencapsulation. Methods and Industrial Applications, edited by
Simon Benita
74. Oral Mucosal Drug Delivery, edited by Michael J. Rathbone
75. Clinical Research in Pharmaceutical Development, edited by Barry
Bleidt and Michael Montagne
76 The Drug Development Process Increasing Efficiency and Cost Ef-
fectiveness, edited by Peter G Welling, Louis Lasagna, and Umesh
V Banakar
77 Microparticulate Systems for the Delivery of Proteins and Vaccines,
edited by Smadar Cohen and Howard Bernstein
78 Good Manufacturing Practices for Pharmaceuticals A Plan for Total
Quality Control, Fourth Edition, Revised and Expanded, Sidney H
Willig and James R Stoker
79 Aqueous Polymeric Coatings for Pharmaceutical Dosage Forms
Second Edition, Revised and Expanded, edited by James W
McGmity
80 Pharmaceutical Statistics Practical and Clinical Applications, Third
Edition, Sanford Bolton
81 Handbook of Pharmaceutical Granulation Technology edited by Dilip
M Pankh
82 Biotechnology of Antibiotics Second Edition, Revised and Expanded,
edited by William R Strohl
83 Mechanisms of Transdermal Drug Delivery, edited by Russell O Potts
and Richard H Guy
84 Pharmaceutical Enzymes edited by Albert Lauwers and Simon
Scharpe
85 Development of Biopharmaceutical Parenteral Dosage Forms, edited
by John A Bontempo
86 Pharmaceutical Project Management, edited by Tony Kennedy
87 Drug Products for Clinical Trials An International Guide to Formula-
tion • Production • Quality Control, edited by Donald C Monkhouse
and Christopher T Rhodes
88 Development and Formulation of Veterinary Dosage Forms Second
Edition, Revised and Expanded, edited by Gregory E Hardee and J
Desmond Baggot
89 Receptor-Based Drug Design, edited by Paul Leff
90 Automation and Validation of Information in Pharmaceutical Pro-
cessing, edited by Joseph F deSpautz
91 Dermal Absorption and Toxicity Assessment, edited by Michael S
Roberts and Kenneth A Walters
92 Pharmaceutical Experimental Design, Gareth A Lewis, Didier
Mathieu, and Roger Phan-Tan-Luu
93 Preparing for FDA Pre-Approval Inspections, edited by Martin D
Hynes III
94 Pharmaceutical Excipients Characterization by IR, Raman, and NMR
Spectroscopy, David E Bugay and W Paul Fmdlay
95 Polymorphism in Pharmaceutical Solids, edited by Harry G Brittam
96 Freeze-Drymg/Lyophihzation of Pharmaceutical and Biological Prod-
ucts, edited by Louis Rey and Joan C May
97 Percutaneous Absorption Drugs-Cosmetics-Mechanisms-Metho-
dology, Third Edition, Revised and Expanded, edited by Robert L
Bronaugh and Howard I Maibach
 98. Bioadhesive Drug Delivery Systems: Fundamentals, Novel Ap-
proaches, and Development, edited by Edith Mathiowitz, Donald E.
Chtckering III, and Claus-Michael Lehr
99. Protein Formulation and Delivery, edited by Eugene J. McNally
100. New Drug Approval Process: Third Edition, The Global Challenge,
edited by Richard A. Guarino
101. Peptide and Protein Drug Analysis, edited by Ronald E. Reid
102 Transport Processes in Pharmaceutical Systems, edited by Gordon L
Amidon, Ping I. Lee, and Elizabeth M. Topp
103. Excipient Toxicity and Safety, edited by Myra L. Weiner and Lois A.
Kotkoskie
104 The Clinical Audit in Pharmaceutical Development, edited by Michael
R. Hamrell
105. Pharmaceutical Emulsions and Suspensions, edited by Francoise
Nielloud and Gilberte Marti-Mestres
106. Oral Drug Absorption: Prediction and Assessment, edited by Jennifer B.
Dressman and Hans Lennernas
107. Drug Stability: Principles and Practices, Third Edition, Revised and
Expanded, edited by Jens T. Carstensen and C. T. Rhodes
108. Containment in the Pharmaceutical Industry, edited by James P.
Wood
109. Good Manufacturing Practices for Pharmaceuticals: A Plan for Total
Quality Control from Manufacturer to Consumer, Fifth Edition, Revised
and Expanded, Sidney H Willig
110. Advanced Pharmaceutical Solids, Jens T Carstensen
111. Endotoxins: Pyrogens, LAL Testing, and Depyrogenation, Second
Edition, Revised and Expanded, Kevin L. Williams
112 Pharmaceutical Process Engineering, Anthony J. Hickey and David
Ganderton
113. Pharmacogenomics, edited by Werner Kalow, Urs A. Meyer, and Ra-
chel F. Tyndale
114. Handbook of Drug Screening, edited by Ramaknshna Seethala and
Prabhavathi B. Fernandas
115. Drug Targeting Technology: Physical • Chemical • Biological Methods,
edited by Hans Schreier
116. Drug-Drug Interactions, edited by A. David Rodngues
117. Handbook of Pharmaceutical Analysis, edited by Lena Ohannesian
and Anthony J. Streeter
118. Pharmaceutical Process Scale-Up, edited by Michael Levin
119. Dermatological and Transdermal Formulations, edited by Kenneth A.
Walters
120. Clinical Drug Trials and Tribulations: Second Edition, Revised and
Expanded, edited by Alien Cato, Lynda Sutton, and Alien Cato III
121. Modern Pharmaceutics: Fourth Edition, Revised and Expanded, edi-
ted by Gilbert S. Banker and Chnstopher T. Rhodes
122. Surfactants and Polymers in Drug Delivery, Martin Malmsten
123. Transdermal Drug Delivery: Second Edition, Revised and Expanded,
edited by Richard H. Guy and Jonathan Hadgraft
124. Good Laboratory Practice Regulations: Second Edition, Revised and
Expanded, edited by Sandy Weinberg
125. Parenteral Quality Control: Sterility, Pyrogen, Particulate, and Pack-
age Integrity Testing. Third Edition, Revised and Expanded, Michael
J. Akers, Daniel S. Larnmore, and Dana Morion Guazzo
126. Modified-Release Drug Delivery Technology, edited by Michael J.
Rathbone, Jonathan Hadgraft, and Michael S. Roberts
127. Simulation for Designing Clinical Trials' A Pharmacokinetic-Pharma-
codynamic Modeling Perspective, edited by Hui C Kimko and Ste-
phen B Duffull
128. Affinity Capillary Electrophoresis in Pharmaceutics and Biopharma-
ceutics, edited by Remhard H. H. Neubert and Hans-Hermann Rut-
tinger
129. Pharmaceutical Process Validation: An International Third Edition, Re-
vised and Expanded, edited by Robert A Nash and Alfred H. Wachter
130. Ophthalmic Drug Delivery Systems: Second Edition, Revised and Ex-
panded, edited byAshim K. Mitra
131 Pharmaceutical Gene Delivery Systems, edited by Alam Rolland and
Sean M. Sullivan

ADDITIONAL VOLUMES IN PREPARATION

Biomarkers in Clinical Drug Development, edited by John C Bloom

and Robert A. Dean

Pharmaceutical Inhalation Aerosol Technology: Second Edition, Re-

vised and Expanded, edited by Anthony J Mickey

Pharmaceutical Extrusion Technology, edited by Isaac Ghebre-Sellas-

sie and Charles Martin

Pharmaceutical Compliance, edited by Carmen Medina

 Dedicated to Theodore E. Byers, formerly of the U.S. Food and Drug
Administration, and Heinz Sucker, Professor at the University of Berne,
Switzerland, for their pioneering contributions with respect to
the pharmaceutical process validation concept. We also acknowledge
the past contributions of Bernard T. Loftus and Ira R. Berry toward the
success of Pharmaceutical Process Validation.
Preface

The third edition of Pharmaceutical Process Validation represents a new ap-

proach to the topic in several important respects.
Many of us in the field had made the assumption that pharmaceutical
process validation was an American invention, based on the pioneering work of
Theodore E. Byers and Bernard T. Loftus, both formerly with the U.S. Food &
Drug Administration. The truth is that many of our fundamental concepts of
pharmaceutical process validation came to us from “Validation of Manufactur-
ing Processes,” Fourth European Seminar on Quality Control, September 25,
1980, Geneva, Switzerland, and Validation in Practice, edited by H. Sucker,
Wissenschaftliche Verlagsegesellschaft, GmbH, Stuttgard, Germany, 1983.
There are new chapters in this edition that will add to the book’s impact.
They include “Validation for Medical Devices” by Nishihata, “Validation of
Biotechnology Processes” by Sofer, “Transdermal Process Validation” by Neal,
“Integrated Packaging Validation” by Frederick, “Statistical Methods for Uni-
formity and Dissolution Testing” by Bergum and Utter, “Change Control and
SUPAC” by Waterland and Kowtna, “Validation in Contract Manufacturing”
by Parikh, and “Harmonization, GMPs, and Validation” by Wachter.
I am pleased to have Dr. Alfred Wachter join me as coeditor of this edi-
tion. He was formerly head of Pharmaceutical Product Development for the
CIBA Pharmaceutical Company in Basel, Switzerland, and also spent a number
of years on assignment in Asia for CIBA. Fred brings a very strong international
perspective to the subject matter.

Robert A. Nash

v
Contents

Preface v
Contributors ix
Introduction xiii

1. Regulatory Basis for Process Validation 1

John M. Dietrick and Bernard T. Loftus

2. Prospective Process Validation 7

Allen Y. Chao, F. St. John Forbes, Reginald F. Johnson,
and Paul Von Doehren

3. Retrospective Validation 31
Chester J. Trubinski

4. Sterilization Validation 83
Michael J. Akers and Neil R. Anderson

5. Validation of Solid Dosage Forms 159

Jeffrey S. Rudolph and Robert J. Sepelyak

6. Validation for Medical Devices 191

Toshiaki Nishihata

7. Validation of Biotechnology Processes 213

Gail Sofer

8. Transdermal Process Validation 237

Charlie Neal, Jr.

9. Validation of Lyophilization 289

Edward H. Trappler
vii
viii Contents

10. Validation of Inhalation Aerosols 329

Christopher J. Sciarra and John J. Sciarra
11. Process Validation of Pharmaceutical Ingredients 363
Robert A. Nash
12. Qualification of Water and Air Handling Systems 401
Kunio Kawamura
13. Equipment and Facility Qualification 443
Thomas L. Peither
14. Validation and Verification of Cleaning Processes 465
William E. Hall
15. Validation of Analytical Methods and Processes 507
Ludwig Huber
16. Computer System Validation:
Controlling the Manufacturing Process 525
Tony de Claire
17. Integrated Packaging Validation 603
Mervyn J. Frederick
18. Analysis of Retrospective Production Data Using
Quality Control Charts 647
Peter H. Cheng and John E. Dutt
19. Statistical Methods for Uniformity and Dissolution Testing 667
James S. Bergum and Merlin L. Utter
20. Change Control and SUPAC 699
Nellie Helen Waterland and Christopher C. Kowtna
21. Process Validation and Quality Assurance 749
Carl B. Rifino
22. Validation in Contract Manufacturing 793
Dilip M. Parikh
23. Terminology of Nonaseptic Process Validation 805
Kenneth G. Chapman
24. Harmonization, GMPs, and Validation 823
Alfred H. Wachter
Index 857
Contributors

Michael J. Akers Baxter Pharmaceutical Solutions, Bloomington, Indiana,

U.S.A.

Neil R. Anderson Eli Lilly and Company, Indianapolis, Indiana, U.S.A.

James S. Bergum Bristol-Myers Squibb Company, New Brunswick, New Jer-

sey, U.S.A.

Kenneth G. Chapman Drumbeat Dimensions, Inc., Mystic, Connecticut,

U.S.A.

Allen Y. Chao Watson Labs, Carona, California, U.S.A.

Peter H. Cheng New York State Research Foundation for Mental Hygiene,
New York, New York, U.S.A.

Tony de Claire APDC Consulting, West Sussex, England

John M. Dietrick Center for Drug Evaluation and Research, U.S. Food and
Drug Administration, Rockville, Maryland, U.S.A.

John E. Dutt EM Industries, Inc., Hawthorne, New York, U.S.A.

Mervyn J. Frederick NV Organon–Akzo Nobel, Oss, The Netherlands

William E. Hall Hall & Pharmaceutical Associates, Inc., Kure Beach, North
Carolina, U.S.A.

Ludwig Huber Agilent Technologies GmbH, Waldbronn, Germany

ix
x Contributors

F. St. John Forbes Wyeth Labs, Pearl River, New York, U.S.A.

*Reginald F. Johnson Searle & Co., Inc., Skokie, Illinois, U.S.A.

Kunio Kawamura Otsuka Pharmaceutical Co., Ltd., Tokushima, Japan

Christopher C. Kowtna DuPont Pharmaceuticals Co., Wilmington, Dela-

ware, U.S.A.

*Bernard T. Loftus Bureau of Drugs, U.S. Food and Drug Administration,

Washington, D.C., U.S.A.

Robert A. Nash Stevens Institute of Technology, Hoboken, New Jersey,

U.S.A.

Charlie Neal, Jr. Diosynth-RTP, Research Triangle Park, North Carolina,

U.S.A.

Toshiaki Nishihata Santen Pharmaceutical Co., Ltd., Osaka, Japan

Dilip M. Parikh APACE PHARMA Inc., Westminster, Maryland, U.S.A.

Thomas L. Peither PECON—Peither Consulting, Schopfheim, Germany

Carl B. Rifino AstraZeneca Pharmaceuticals LP, Newark, Delaware, U.S.A.

Jeffrey S. Rudolph Pharmaceutical Consultant, St. Augustine, Florida, U.S.A.

Christopher J. Sciarra Sciarra Laboratories Inc., Hicksville, New York, U.S.A.

John J. Sciarra Sciarra Laboratories Inc., Hicksville, New York, U.S.A.

Robert J. Sepelyak AstraZeneca Pharmaceuticals LP, Wilmington, Delaware,

U.S.A.

Gail Sofer BioReliance, Rockville, Maryland, U.S.A.

Edward H. Trappler Lyophilization Technology, Inc., Warwick, Pennsylva-

nia, U.S.A.

*Retired
Contributors xi

Chester J. Trubinski Church & Dwight Co., Inc., Princeton, New Jersey,
U.S.A.

Merlin L. Utter Wyeth Pharmaceuticals, Pearl River, New York, U.S.A.

Paul Von Doehren Searle & Co., Inc., Skokie, Illinois, U.S.A.

Alfred H. Wachter Wachter Pharma Projects, Therwil, Switzerland

Nellie Helen Waterland DuPont Pharmaceuticals Co., Wilmington, Dela-

ware, U.S.A.
Introduction

Robert A. Nash
Stevens Institute of Technology, Hoboken, New Jersey, U.S.A.

I. FDA GUIDELINES

The U.S. Food and Drug Administration (FDA) has proposed guidelines with
the following definition for process validation [1]:
Process validation is establishing documented evidence which provides a
high degree of assurance that a specific process (such as the manufacture
of pharmaceutical dosage forms) will consistently produce a product meet-
ing its predetermined specifications and quality characteristics.

According to the FDA, assurance of product quality is derived from care-

ful and systemic attention to a number of important factors, including: selection
of quality components and materials, adequate product and process design, and
(statistical) control of the process through in-process and end-product testing.
Thus, it is through careful design (qualification) and validation of both the
process and its control systems that a high degree of confidence can be estab-
lished that all individual manufactured units of a given batch or succession of
batches that meet specifications will be acceptable.
According to the FDA’s Current Good Manufacturing Practices (CGMPs)
21CFR 211.110 a:
Control procedures shall be established to monitor output and to validate
performance of the manufacturing processes that may be responsible for
causing variability in the characteristics of in-process material and the drug
product. Such control procedures shall include, but are not limited to the
following, where appropriate [2]:
1. Tablet or capsule weight variation
2. Disintegration time
xiii
xiv Nash

3. Adequacy of mixing to assure uniformity and homogeneity

4. Dissolution time and rate
5. Clarity, completeness, or pH of solutions

The first four items listed above are directly related to the manufacture
and validation of solid dosage forms. Items 1 and 3 are normally associated
with variability in the manufacturing process, while items 2 and 4 are usually
influenced by the selection of the ingredients in the product formulation. With
respect to content uniformity and unit potency control (item 3), adequacy of
mixing to assure uniformity and homogeneity is considered a high-priority con-
cern.
Conventional quality control procedures for finished product testing en-
compass three basic steps:
1. Establishment of specifications and performance characteristics
2. Selection of appropriate methodology, equipment, and instrumenta-
tion to ensure that testing of the product meets specifications
3. Testing of the final product, using validated analytical and testing
methods to ensure that finished product meets specifications.
With the emergence of the pharmaceutical process validation concept, the fol-
lowing four additional steps have been added:
4. Qualification of the processing facility and its equipment
5. Qualification and validation of the manufacturing process through ap-
propriate means
6. Auditing, monitoring, sampling, or challenging the key steps in the
process for conformance to in-process and final product specifications
7. Revalidation when there is a significant change in either the product
or its manufacturing process [3].

II. TOTAL APPROACH TO PHARMACEUTICAL

PROCESS VALIDATION

It has been said that there is no specific basis for requiring a separate set of
process validation guidelines, since the essentials of process validation are em-
bodied within the purpose and scope of the present CGMP regulations [2]. With
this in mind, the entire CGMP document, from subpart B through subpart K,
may be viewed as being a set of principles applicable to the overall process of
manufacturing, i.e., medical devices (21 CFR–Part 820) as well as drug prod-
ucts, and thus may be subjected, subpart by subpart, to the application of the
principles of qualification, validation, verification and control, in addition to
change control and revalidation, where applicable. Although not a specific re-
Introduction xv

quirement of current regulations, such a comprehensive approach with respect

to each subpart of the CGMP document has been adopted by many drug firms.
A checklist of qualification and control documentation with respect to
CGMPs is provided in Table 1. A number of these topics are discussed sepa-
rately in other chapters of this book.

III. WHY ENFORCE PROCESS VALIDATION?

The FDA, under the authority of existing CGMP regulations, guidelines [1], and
directives [3], considers process validation necessary because it makes good
engineering sense. The basic concept, according to Mead [5], has long been

Table 1 Checklist of Qualification and Control Documentation

Qualification and
Subpart Section of CGMPs control documentation

A General provisions
B Organization and personnel Responsibilities of the quality con-
trol unit
C Buildings and facilities Plant and facility installation and
qualification
Maintenance and sanitation
Microbial and pest control
D Equipment Installation and qualification of
equipment and cleaning methods
E Control of components, containers Incoming component testing proce-
and closures dures
F Production and process controls Process control systems, reprocess-
ing control of microbial contami-
nation
G Packaging and labeling controls Depyrogenation, sterile packaging,
filling and closing, expire dating
H Holding and distribution Warehousing and distribution pro-
cedures
I Laboratory controls Analytical methods, testing for re-
lease component testing and sta-
bility testing
J Records and reports Computer systems and information
systems
K Return and salvaged drug products Batch reprocessing

Sterilization procedures, Air and water quality are covered in appropriate subparts of Table 1.
xvi Nash

applied in other industries, often without formal recognition that such a concept
was being used. For example, the terms reliability engineering and qualification
have been used in the past by the automotive and aerospace industries to repre-
sent the process validation concept.
The application of process validation should result in fewer product re-
calls and troubleshooting assignments in manufacturing operations and more
technically and economically sound products and their manufacturing processes.
In the old days R & D “gurus” would literally hand down the “go” sometimes
overformulated product and accompanying obtuse manufacturing procedure,
usually with little or no justification or rationale provided. Today, under FDA’s
Preapproval Inspection (PAI) program [4] such actions are no longer accept-
able. The watchword is to provide scientifically sound justifications (including
qualification and validation documentation) for everything that comes out of the
pharmaceutical R & D function.

IV. WHAT IS PROCESS VALIDATION?

Unfortunately, there is still much confusion as to what process validation is

and what constitutes process validation documentation. At the beginning of this
introduction several different definitions for process validation were provided,
which were taken from FDA guidelines and the CGMPs. Chapman calls process
validation simply “organized, documented common sense” [6]. Others have said
that “it is more than three good manufactured batches” and should represent a
lifetime commitment as long as the product is in production, which is pretty
much analogous to the retrospective process validation concept.
The big problem is that we use the term validation generically to cover
the entire spectrum of CGMP concerns, most of which are essentially people,
equipment, component, facility, methods, and procedural qualification. The spe-
cific term process validation should be reserved for the final stage(s) of the
product/process development sequence. The essential or key steps or stages of
a successfully completed product/process development program are presented
in Table 2 [7].
The end of the sequence that has been assigned to process validation is
derived from the fact that the specific exercise of process validation should
never be designed to fail. Failure in carrying out the process validation assign-
ment is often the result of incomplete or faulty understanding of the process’s
capability, in other words, what the process can and cannot do under a given
set of operational circumstances. In a well-designed, well-run overall validation
program, most of the budget dollars should be spent on equipment, component,
facility, methods qualification, and process demonstration, formerly called pro-
cess qualification. In such a program, the formalized final process validation
Introduction xvii

Table 2 The Key Stages in the Product/Process

Development Sequence

Development stage Pilot scale-up phase

Product design 1 × batch size

Product characterization
Product selection (“go” formula)
Process design
Product optimization 10 × batch size
Process characterization
Process optimization
Process demonstration 100 × batch size
Process validation program
Product/process certification

With the exception of solution products, the bulk of the work is nor-
mally carried out at 10 × batch size, which is usually the first scale-up
batches in production-type equipment.

sequence provides only the necessary process validation documentation required

by the regulatory authorities—in other words, the “Good Housekeeping Seal of
Approval,” which shows that the manufacturing process is in a state of control.
Such a strategy is consistent with the U.S. FDA’s preapproval inspection
program [4], wherein the applicant firm under either a New Drug Application
(NDA) or an Abbreviated New Drug Application (ANDA) submission must
show the necessary CGMP information and qualification data (including appro-
priate development reports), together with the formal protocol for the forthcom-
ing full-scale, formal process validation runs required prior to product launch.
Again, the term validation has both a specific meaning and a general one,
depending on whether the word “process” is used. Determine during the course
of your reading whether the entire concept is discussed in connection with the
topic—i.e., design, characterization, optimization, qualification, validation, and/
or revalidation—or whether the author has concentrated on the specifics of the
validation of a given product and/or its manufacturing process. In this way the
text will take on greater meaning and clarity.

V. PILOT SCALE-UP AND PROCESS VALIDATION

The following operations are normally carried out by the development function
prior to the preparation of the first pilot-production batch. The development
activities are listed as follows:
xviii Nash

1. Formulation design, selection, and optimization

2. Preparation of the first pilot-laboratory batch
3. Conduct initial accelerated stability testing
4. If the formulation is deemed stable, preparation of additional pilot-
laboratory batches of the drug product for expanded nonclinical and/
or clinical use.
The pilot program is defined as the scale-up operations conducted subse-
quent to the product and its process leaving the development laboratory and
prior to its acceptance by the full scale manufacturing unit. For the pilot program
to be successful, elements of process validation must be included and completed
during the developmental or pilot laboratory phase of the work.
Thus, product and process scale-up should proceed in graduated steps with
elements of process validation (such as qualifications) incorporated at each stage
of the piloting program [9,10].

A. Laboratory Batch
The first step in the scale-up process is the selection of a suitable preliminary
formula for more critical study and testing based on certain agreed-upon initial
design criteria, requirements, and/or specifications. The work is performed in
the development laboratory. The formula selected is designated as the (1 × )
laboratory batch. The size of the (1 × ) laboratory batch is usually 3–10 kg of a
solid or semisolid, 3–10 liters of a liquid, or 3000 to 10,000 units of a tablet or
capsule.

B. Laboratory Pilot Batch

After the (1 × ) laboratory batch is determined to be both physically and chemi-
cally stable based on accelerated, elevated temperature testing (e.g., 1 month at
45°C or 3 months at 40°C or 40°C/80% RH), the next step in the scale-up
process is the preparation of the (10 × ) laboratory pilot batch. The (10 × )
laboratory pilot batch represents the first replicated scale-up of the designated
formula. The size of the laboratory pilot batch is usually 30–100 kg, 30–100
liters, or 30,000 to 100,000 units.
It is usually prepared in small pilot equipment within a designated CGMP-
approved area of the development laboratory. The number and actual size of the
laboratory pilot batches may vary in response to one or more of the following
factors:
1. Equipment availability
2. Active pharmaceutical ingredient (API)
Introduction xix

3. Cost of raw materials

4. Inventory requirements for clinical and nonclinical studies
Process demonstration or process capability studies are usually started in this
important second stage of the pilot program. Such capability studies consist of
process ranging, process characterization, and process optimization as a prereq-
uisite to the more formal validation program that follows later in the piloting
sequence.

C. Pilot Production
The pilot-production phase may be carried out either as a shared responsibility
between the development laboratories and its appropriate manufacturing coun-
terpart or as a process demonstration by a separate, designated pilot-plant or
process-development function. The two organization piloting options are pre-
sented separately in Figure 1. The creation of a separate pilot-plant or process-
development unit has been favored in recent years because it is ideally suited to
carry out process scale-up and/or validation assignments in a timely manner. On
the other hand, the joint pilot-operation option provides direct communication
between the development laboratory and pharmaceutical production.

Figure 1 Main piloting options. (Top) Separate pilot plant functions—engineering

concept. (Bottom) Joint pilot operation.
xx Nash

The object of the pilot-production batch is to scale the product and process
by another order of magnitude (100 × ) to, for example, 300–1,000 kg, 300–
1,000 liters, or 300,000–1,000,000 dosage form units (tablets or capsules) in
size. For most drug products this represents a full production batch in standard
production equipment. If required, pharmaceutical production is capable of scal-
ing the product/process to even larger batch sizes should the product require
expanded production output. If the batch size changes significantly, additional
validation studies would be required. The term product/process is used, since
one can’t describe a product with discussing its process of manufacture and,
conversely, one can’t talk about a process without describing the product being
manufactured.
Usually large production batch scale-up is undertaken only after product
introduction. Again, the actual size of the pilot-production (100 × ) batch may
vary due to equipment and raw material availability. The need for additional
pilot-production batches ultimately depends on the successful completion of a
first pilot batch and its process validation program. Usually three successfully
completed pilot-production batches are required for validation purposes.
In summary, process capability studies start in the development labora-
tories and/or during product and process development, and continue in well-
defined stages until the process is validated in the pilot plant and/or pharmaceu-
tical production.
An approximate timetable for new product development and its pilot
scale-up program is suggested in Table 3.

VI. PROCESS VALIDATION: ORDER OF PRIORITY

Because of resource limitation, it is not always possible to validate an entire

company’s product line at once. With the obvious exception that a company’s
most profitable products should be given a higher priority, it is advisable to
draw up a list of product categories to be validated.
The following order of importance or priority with respect to validation is
suggested:

A. Sterile Products and Their Processes

1. Large-volume parenterals (LVPs)

2. Small-volume parenterals (SVPs)
3. Ophthalmics, other sterile products, and medical devices
Introduction xxi

Table 3 Approximate Timetable for New Product Development and Pilot

Scale-Up Trials

Calendar
Event months

Formula selection and development 2–4

Assay methods development and formula optimization 2–4
Stability in standard packaging 3-month readout (1 × size) 3–4
Pilot-laboratory batches (10 × size) 1–3
Preparation and release of clinical supplies (10 × size) and
establishment of process demonstration 1–4
Additional stability testing in approved packaging 3–4
6–8-month readout (1 × size)
3-month readout (10 × size)
Validation protocols and pilot batch request 1–3
Pilot-production batches (100 × size) 1–3
Additional stability testing in approved packaging 3–4
9–12-month readout (1 × size)
6–8-month readout (10 × size)
3-month readout (100 × size)
Interim approved technical product development report with
approximately 12 months stability (1 × size) 1–3
Totals 18–36

B. Nonsterile Products and Their Processes

1. Low-dose/high-potency tablets and capsules/transdermal delivery sys-
tems (TDDs)
2. Drugs with stability problems
3. Other tablets and capsules
4. Oral liquids, topicals, and diagnostic aids

VII. WHO DOES PROCESS VALIDATION?

Process validation is done by individuals with the necessary training and experi-
ence to carry out the assignment.
The specifics of how a dedicated group, team, or committee is organized
to conduct process validation assignments is beyond the scope of this introduc-
tory chapter. The responsibilities that must be carried out and the organizational
structures best equipped to handle each assignment are outlined in Table 4. The
xxii Nash

Table 4 Specific Responsibilities of Each Organizational Structure within the Scope

of Process Validation

Engineering Install, qualify, and certify plant, facilities, equipment, and sup-
port system.
Development Design and optimize manufacturing process within design limits,
specifications, and/or requirements—in other words, the estab-
lishment of process capability information.
Manufacturing Operate and maintain plant, facilities, equipment, support sys-
tems, and the specific manufacturing process within its design
limits, specifications, and/or requirements.
Quality assurance Establish approvable validation protocols and conduct process
validation by monitoring, sampling, testing, challenging, and/
or auditing the specific manufacturing process for compliance
with design limits, specifications, and/or requirements.

Source: Ref. 8.

best approach in carrying out the process validation assignment is to establish a

Chemistry, Manufacturing and Control (CMC) Coordination Committee at the
specific manufacturing plant site [10]. Representation on such an important lo-
gistical committee should come from the following technical operations:
• Formulation development (usually a laboratory function)
• Process development (usually a pilot plant function)
• Pharmaceutical manufacturing (including packaging operations)
• Engineering (including automation and computer system responsibili-
ties)
• Quality assurance
• Analytical methods development and/or Quality Control
• API Operations (representation from internal operations or contract
manufacturer)
• Regulatory Affairs (technical operations representative)
• IT (information technology) operations
The chairperson or secretary of such an important site CMC Coordination Com-
mittee should include the manager of process validation operations. Typical
meeting agendas may include the following subjects in the following recom-
mended order of priority:
• Specific CGMP issues for discussion and action to be taken
• Qualification and validation issues with respect to a new product/pro-
cess
Introduction xxiii

• Technology transfer issues within or between plant sites.

• Pre-approval inspection (PAI) issues of a forthcoming product/process
• Change control and scale-up, post approval changes (SUPAC) with
respect to current approved product/process [11].

VIII. PROCESS DESIGN AND CHARACTERIZATION

Process capability is defined as the studies used to determine the critical process
parameters or operating variables that influence process output and the range of
numerical data for critical process parameters that result in acceptable process
output. If the capability of a process is properly delineated, the process should
consistently stay within the defined limits of its critical process parameters and
product characteristics [12].
Process demonstration formerly called process qualification, represents
the actual studies or trials conducted to show that all systems, subsystems, or
unit operations of a manufacturing process perform as intended; that all critical
process parameters operate within their assigned control limits; and that such
studies and trials, which form the basis of process capability design and testing,
are verifiable and certifiable through appropriate documentation.
The manufacturing process is briefly defined as the ways and means used
to convert raw materials into a finished product. The ways and means also
include people, equipment, facilities, and support systems required to operate
the process in a planned and effectively managed way. All the latter functions
must be qualified individually. The master plan or protocol for process capabil-
ity design and testing is presented in Table 5.
A simple flow chart should be provided to show the logistical sequence
of unit operations during product/process manufacture. A typical flow chart used
in the manufacture of a tablet dosage form by the wet granulation method is
presented in Figure 2.

IX. STREAMLINING VALIDATION OPERATIONS

The best approach to avoiding needless and expensive technical delays is to

work in parallel. The key elements at this important stage of the overall process
are the API, analytical test methods, and the drug product (pharmaceutical dos-
age form). An integrated and parallel way of getting these three vitally important
functions to work together is depicted in Figure 3.
Figure 3 shows that the use of a single analytical methods testing function
is an important technical bridge between the API and the drug product develop-
ment functions as the latter two move through the various stages of develop-
xxiv Nash

Table 5 Master Plan or Protocol for Process Capability Design and Testing

Objective Process capability design and testing

Types of process Batch, intermittent, continuous
Typical processes Chemical, pharmaceutical, biochemical
Process definition Flow diagram, in-process, finished product
Definition of process output Potency, yield, physical parameters
Definition of test methods Instrumentation, procedures, precision, and
accuracy
Process analysis Process variables, matrix design, factorial design
analysis
Pilot batch trials Define sampling and testing, stable, extended runs
Pilot batch replication Different days, different materials, different equip-
ment
Process redefinition Reclassification of process variables
Process capability evaluation Stability and variability of process output, eco-
nomic limits
Final report Recommended SOP, specifications, and process
limits

Figure 2 Process flow diagram for the manufacture of a tablet dosage form by wet
granulation method. The arrows show the transfer of material into and out of each of the
various unit operations. The information in parentheses indicates additions of material to
specific unit operations. A list of useful pharmaceutical unit operations is presented in
Table 6.
Introduction xxv

Table 6 A List of Useful Pharmaceutical Unit Operations According to Categories

Heat transfer processes: Cooking, cooling, evaporating, freezing, heating, irradiating,

sterilizing, freeze-drying
Change in state: Crystallizing, dispersing, dissolving, immersing, freeze-drying, neutral-
izing
Change in size: Agglomerating, blending, coating, compacting, crushing, crystallizing,
densifying, emulsifying, extruding, flaking, flocculating, grinding, homogenizing,
milling, mixing, pelletizing, pressing, pulverizing, precipitating, sieving
Moisture transfer processes: Dehydrating, desiccating, evaporating, fluidizing, humidify-
ing, freeze-drying, washing, wetting
Separation processes: Centrifuging, clarifying, deareating, degassing, deodorizing, dia-
lyzing, exhausting, extracting, filtering, ion exchanging, pressing, sieving, sorting,
washing
Transfer processes: Conveying, filling, inspecting, pumping, sampling, storing, trans-
porting, weighing

Source: Ref. 13.

ment, clinical study, process development, and process validation and into pro-
duction. Working individually with separate analytical testing functions and
with little or no appropriate communication among these three vital functions is
a prescription for expensive delays. It is important to remember that the concept
illustrated in Figure 3 can still be followed even when the API is sourced from
outside the plant site or company. In this particular situation there will probably
be two separate analytical methods development functions: one for the API
manufacturer and one for the drug product manufacturer [14].

X. STATISTICAL PROCESS CONTROL AND

PROCESS VALIDATION

Statistical process control (SPC), also called statistical quality control and pro-
cess validation (PV), represents two sides of the same coin. SPC comprises the
various mathematical tools (histogram, scatter diagram run chart, and control
chart) used to monitor a manufacturing process and to keep it within in-process
and final product specification limits. Lord Kelvin once said, “When you can
measure what you are speaking about, and express it in numbers, then you know
something about it.” Such a thought provides the necessary link between the
two concepts. Thus, SPC represents the tools to be used, while PV represents
the procedural environment in which those tools are used.
xxvi Nash

Figure 3 Working in parallel. (Courtesy of Austin Chemical Co., Inc.)

There are three ways of establishing quality products and their manufac-
turing processes:
1. In-process and final product testing, which normally depends on sam-
pling size (the larger the better). In some instances, nothing short of
excessive sampling can ensure reaching the desired goal, i.e., sterility
testing.
2. Establishment of tighter (so called “in-house”) control limits that hold
the product and the manufacturing process to a more demanding stan-
Introduction xxvii

dard will often reduce the need for more extensive sampling require-
ments.
3. The modern approach, based on Japanese quality engineering [15], is
the pursuit of “zero defects” by applying tighter control over process
variability (meeting a so-called 6 sigma standard). Most pharmaceuti-
cal products and their manufacturing processes in the United States
today, with the exception of sterile processes are designed to meet a
4 sigma limit (which would permit as many as eight defects per 1000
units). The new approach is to center the process (in which the grand
average is roughly equal to 100% of label potency or the target value
of a given specification) and to reduce the process variability or noise
around the mean or to achieve minimum variability by holding both
to the new standard, batch after batch. In so doing, a 6 sigma limit
may be possible (which is equivalent to not more than three to four
defects per 1 million units), also called “zero defects.” The goal of 6
sigma, “zero defects” is easier to achieve for liquid than for solid
pharmaceutical dosage forms [16].
Process characterization represents the methods used to determine the
critical unit operations or processing steps and their process variables, that usu-
ally affect the quality and consistency of the product outcomes or product attri-
butes. Process ranging represents studies that are used to identify critical process
or test parameters and their respective control limits, which normally affect the
quality and consistency of the product outcomes of their attributes. The follow-
ing process characterization techniques may be used to designate critical unit
operations in a given manufacturing process.

A. Constraint Analysis
One procedure that makes subsystem evaluations and performance qualification
trials manageable is the application of constraint analysis. Boundary limits of
any technology and restrictions as to what constitutes acceptable output from
unit operations or process steps should in most situations constrain the number
of process variables and product attributes that require analysis. The application
of the constraint analysis principle should also limit and restrict the operational
range of each process variable and/or specification limit of each product attri-
bute. Information about constraining process variables usually comes from the
following sources:
• Previous successful experience with related products/processes
• Technical and engineering support functions and outside suppliers
• Published literatures concerning the specific technology under investi-
gation
xxviii Nash

A practical guide to constraint analysis comes to us from the application

of the Pareto Principle (named after an Italian sociologist) and is also known
as the 80–20 rule, which simply states that about 80% of the process output is
governed by about 20% of the input variables and that our primary job is to
find those key variables that drive the process.
The FDA in their proposed amendments to the CGMPs [17] have desig-
nated that the following unit operations are considered critical and therefore
their processing variables must be controlled and not disregarded:
• Cleaning
• Weighing/measuring
• Mixing/blending
• Compression/encapsulation
• Filling/packaging/labeling

B. Fractional Factorial Design

An experimental design is a series of statistically sufficient qualification trials
that are planned in a specific arrangement and include all processing variables
that can possibly affect the expected outcome of the process under investigation.
In the case of a full factorial design, n equals the number of factors or process
variables, each at two levels, i.e., the upper (+) and lower (−) control limits.
Such a design is known as a 2n factorial. Using a large number of process
variables (say, 9) we could, for example, have to run 29, or 512, qualification
trials in order to complete the full factorial design.
The fractional factorial is designed to reduce the number of qualification
trials to a more reasonable number, say, 10, while holding the number of ran-
domly assigned processing variables to a reasonable number as well, say, 9. The
technique was developed as a nonparametric test for process evaluation by Box
and Hunter [18] and reviewed by Hendrix [19]. Ten is a reasonable number of
trials in terms of resource and time commitments and should be considered an
upper limit in a practical testing program. This particular design as presented in
Table 7 does not include interaction effects.

XI. OPTIMIZATION TECHNIQUES

Optimization techniques are used to find either the best possible quantitative
formula for a product or the best possible set of experimental conditions (input
values) needed to run the process. Optimization techniques may be employed in
the laboratory stage to develop the most stable, least sensitive formula, or in the
qualification and validation stages of scale-up in order to develop the most sta-
Introduction xxix

Table 7 Fractional Factorial Design (9 Variables in 10 Experiments)

Trial no. X1 X2 X3 X4 X5 X6 X7 X8 X9

1 − − − − − − − − −
2 + − − − − − − − −
3 − − − + − − − − +
4 + − + − − − + − −
5 − + − + − + − + −
6 + − + − + − + − +
7 − + − + + + − + +
8 + + + − + + + + −
9 − + + + + + + + +
10 + + + + + + + + +

Worst-case conditions: Trial 1 (lower control limit). Trial 10 (upper control limit). X variables
randomly assigned. Best values to use are RSD of data set for each trial. When adding up the data
by columns, + and − are now numerical values and the sum is divided by 5 (number of +s or −s).
If the variable is not significant, the sum will approach zero.

ble, least variable, robust process within its proven acceptable range(s) of opera-
tion, Chapman’s so-called proven acceptable range (PAR) principle [20].
Optimization techniques may be classified as parametric statistical meth-
ods and nonparametric search methods. Parametric statistical methods, usually
employed for optimization, are full factorial designs, half factorial designs, sim-
plex designs, and Lagrangian multiple regression analysis [21]. Parametric
methods are best suited for formula optimization in the early stages of product
development. Constraint analysis, described previously, is used to simplify the
testing protocol and the analysis of experimental results.
The steps involved in the parametric optimization procedure for pharma-
ceutical systems have been fully described by Schwartz [22]. Optimization tech-
niques consist of the following essential operations:
1. Selection of a suitable experimental design
2. Selection of variables (independent Xs and dependent Ys) to be tested
3. Performance of a set of statistically designed experiments (e.g., 23 or
32 factorials)
4. Measurement of responses (dependent variables)
5. Development of a predictor, polynomial equation based on statistical
and regression analysis of the generated experimental data
6. Development of a set of optimized requirements for the formula based
on mathematical and graphical analysis of the data generated
xxx Nash

XII. WHAT ARE THE PROCESS VALIDATION OPTIONS?

The guidelines on general principles of process validation [1] mention three

options: (1) prospective process validation (also called premarket validation),
(2) retrospective process validation, and (3) revalidation. In actuality there are
four possible options.

A. Prospective Process Validation

In prospective process validation, an experimental plan called the validation
protocol is executed (following completion of the qualification trials) before the
process is put into commercial use. Most validation efforts require some degree
of prospective experimentation to generate validation support data. This particu-
lar type of process validation is normally carried out in connection with the
introduction of new drug products and their manufacturing processes. The for-
malized process validation program should never be undertaken unless and until
the following operations and procedures have been completed satisfactorily:
1. The facilities and equipment in which the process validation is to
be conducted meet CGMP requirements (completion of installation
qualification)
2. The operators and supervising personnel who will be “running” the
validation batch(es) have an understanding of the process and its re-
quirements
3. The design, selection, and optimization of the formula have been
completed
4. The qualification trials using (10 × size) pilot-laboratory batches have
been completed, in which the critical processing steps and process
variables have been identified, and the provisional operational control
limits for each critical test parameter have been provided
5. Detailed technical information on the product and the manufacturing
process have been provided, including documented evidence of prod-
uct stability
6. Finally, at least one qualification trial of a pilot-production (100 × size)
batch has been made and shows, upon scale-up, that there were no
significant deviations from the expected performance of the process
The steps and sequence of events required to carry out a process validation
assignment are outlined in Table 8. The objective of prospective validation is to
prove or demonstrate that the process will work in accordance with a validation
master plan or protocol prepared for pilot-product (100 × size) trials.
In practice, usually two or three pilot-production (100 × ) batches are pre-
pared for validation purposes. The first batch to be included in the sequence
Introduction xxxi

Table 8 Master Plan or Outline of a Process Validation Program

Objective Proving or demonstrating that the process works

Type of validation Prospective, concurrent, retrospective, revalidation
Type of process Chemical, pharmaceutical, automation, cleaning
Definition of process Flow diagram, equipment/components, in-process, fin-
ished product
Definition of process output Potency, yield, physical parameters
Definition of test methods Method, instrumentation, calibration, traceability, preci-
sion, accuracy
Analysis of process Critical modules and variables defined by process capa-
bility design and testing program
Control limits of critical vari- Defined by process capability design and testing pro-
ables gram
Preparation of validation pro- Facilities, equipment, process, number of validation tri-
tocol als, sampling frequency, size, type, tests to perform,
methods used, criteria for success
Organizing for validation Responsibility and authority
Planning validation trials Timetable and PERT charting, material availability,
and disposal
Validation trials Supervision, administration, documentation
Validation finding Data summary, analysis, and conclusions
Final report and recommenda- Process validated, further trials, more process design,
tions and testing

may be the already successfully concluded first pilot batch at 100 × size, which
is usually prepared under the direction of the organizational function directly
responsible for pilot scale-up activities. Later, replicate batch manufacture may
be performed by the pharmaceutical production function.
The strategy selected for process validation should be simple and straight-
forward. The following factors are presented for the reader’s consideration:
1. The use of different lots of components should be included, i.e., APIs
and major excipients.
2. Batches should be run in succession and on different days and shifts
(the latter condition, if appropriate).
3. Batches should be manufactured in equipment and facilities desig-
nated for eventual commercial production.
4. Critical process variables should be set within their operating ranges
and should not exceed their upper and lower control limits during
process operation. Output responses should be well within finished
product specifications.
xxxii Nash

5. Failure to meet the requirements of the validation protocol with re-

spect to process inputs and output control should be subjected to re-
qualification following a thorough analysis of process data and formal
review by the CMC Coordination Committee.

B. Retrospective Validation
The retrospective validation option is chosen for established products whose
manufacturing processes are considered stable and when on the basis of eco-
nomic considerations alone and resource limitations, prospective validation pro-
grams cannot be justified. Prior to undertaking retrospective validation, wherein
the numerical in-process and/or end-product test data of historic production
batches are subjected to statistical analysis, the equipment, facilities and subsys-
tems used in connection with the manufacturing process must be qualified in
conformance with CGMP requirements. The basis for retrospective validation
is stated in 21CFR 211.110(b): “Valid in-process specifications for such charac-
teristics shall be consistent with drug product final specifications and shall be
derived from previous acceptable process average and process variability esti-
mates where possible and determined by the application of suitable statistical
procedures where appropriate.”
The concept of using accumulated final product as well as in-process nu-
merical test data and batch records to provide documented evidence of product/
process validation was originally advanced by Meyers [26] and Simms [27] of
Eli Lilly and Company in 1980. The concept is also recognized in the FDA’s
Guidelines on General Principles of Process Validation [1].
Using either data-based computer systems [28,29] or manual methods,
retrospective validation may be conducted in the following manner:
1. Gather the numerical data from the completed batch record and in-
clude assay values, end-product test results, and in-process data.
2. Organize these data in a chronological sequence according to batch
manufacturing data, using a spreadsheet format.
3. Include data from at least the last 20–30 manufactured batches for
analysis. If the number of batches is less than 20, then include all
manufactured batches and commit to obtain the required number for
analysis.
4. Trim the data by eliminating test results from noncritical processing
steps and delete all gratuitous numerical information.
5. Subject the resultant data to statistical analysis and evaluation.
6. Draw conclusions as to the state of control of the manufacturing pro-
cess based on the analysis of retrospective validation data.
7. Issue a report of your findings (documented evidence).
Introduction xxxiii

One or more of the following output values (measured responses), which

have been shown to be critical in terms of the specific manufacturing process
being evaluated, are usually selected for statistical analysis.

1. Solid Dosage Forms

1. Individual assay results from content uniformity testing
2. Individual tablet hardness values
3. Individual tablet thickness values
4. Tablet or capsule weight variation
5. Individual tablet or capsule dissolution time (usually at t50%) or disinte-
gration time
6. Individual tablet or capsule moisture content

2. Semisolid and Liquid Dosage Forms

1. pH value (aqueous system)
2. Viscosity
3. Density
4. Color or clarity values
5. Average particle size or distribution
6. Unit weight variation and/or potency values
The statistical methods that may be employed to analyze numerical output
data from the manufacturing process are listed as follows:
1 Basic statistics (mean, standard deviation, and tolerance limits) [21]
2. Analysis of variance (ANOVA and related techniques) [21]
3. Regression analysis [22]
4. Cumulative sum analysis (CUSUM) [23]
5. Cumulative difference analysis [23]
6. Control charting (averages and range) [24,25]
Control charting, with the exception of basic statistical analysis, is proba-
bly the most useful statistical technique to analyze retrospective and concurrent
process data. Control charting forms the basis of modern statistical process con-
trol.

C. Concurrent Validation
In-process monitoring of critical processing steps and end-product testing of
current production can provide documented evidence to show that the manufac-
turing process is in a state of control. Such validation documentation can be
provided from the test parameter and data sources disclosed in the section on
retrospective validation.
xxxiv Nash

Test parameter Data source

Average unit potency End-product testing

Content uniformity End-product testing
Dissolution time End-product testing
Weight variation End-product testing
Powder-blend uniformity In-process testing
Moisture content In-process testing
Particle or granule size distribution In-process testing
Weight variation In-process testing
Tablet hardness In-process testing
pH value In-process testing
Color or clarity In-process testing
Viscosity or density In-process testing

Not all of the in-process tests enumerated above are required to demon-
strate that the process is in a state of control. Selections of test parameters
should be made on the basis of the critical processing variables to be evaluated.

D. Revalidation
Conditions requiring revalidation study and documentation are listed as follows:
1. Change in a critical component (usually refers to raw materials)
2. Change or replacement in a critical piece of modular (capital) equip-
ment
3. Change in a facility and/or plant (usually location or site)
4. Significant (usually order of magnitude) increase or decrease in batch
size
5. Sequential batches that fail to meet product and process specifications
In some situations performance requalification studies may be required
prior to undertaking specific revalidation assignments.
The FDA process validation guidelines [1] refer to a quality assurance
system in place that requires revalidation whenever there are changes in packag-
ing (assumed to be the primary container-closure system), formulation, equip-
ment or processes (meaning not clear) which could impact on product effective-
ness or product characteristics and whenever there are changes in product
characteristics.
Approved packaging is normally selected after completing package perfor-
mance qualification testing as well as product compatibility and stability studies.
Since in most cases (exceptions: transdermal delivery systems, diagnostic tests,
and medical devices) packaging is not intimately involved in the manufacturing
process of the product itself, it differs from other factors, such as raw materials.
Introduction xxxv

The reader should realize that there is no one way to establish proof or
evidence of process validation (i.e., a product and process in control). If the
manufacturer is certain that its products and processes are under statistical con-
trol and in compliance with CGMP regulations, it should be a relatively simple
matter to establish documented evidence of process validation through the use
of prospective, concurrent, or retrospective pilot and/or product quality informa-
tion and data. The choice of procedures and methods to be used to establish
validation documentation is left with the manufacturer.
This introduction was written to aid scientists and technicians in the phar-
maceutical and allied industries in the selection of procedures and approaches
that may be employed to achieve a successful outcome with respect to product
performance and process validation. The authors of the following chapters ex-
plore the same topics from their own perspectives and experience. It is hoped
that the reader will gain much from the diversity and richness of these varied
approaches.

REFERENCES

1. Guidelines on General Principles of Process Validation, Division of Manufacturing

and Product Quality, CDER, FDA, Rockville, Maryland (May 1987).
2. Current Good Manufacturing Practices in Manufacture, Processing, Packing and
Holding of Human and Veterinary Drugs, Federal Register 43(190), 45085 and
45086, September 1978.
3. Good Manufacturing Practices for Pharmaceuticals, Willig, S. H. and Stoker, J.
R., Marcel Dekker, New York (1997).
4. Commentary, Pre-approval Inspections/Investigations, FDA, J. Parent. Sci. & Tech.
45:56–63 (1991).
5. Mead, W. J., Process validation in cosmetic manufacture, Drug Cosmet. Ind., (Sep-
tember 1981).
6. Chapman, K. G., A history of validation in the United States, Part I, Pharm. Tech.,
(November 1991).
7. Nash, R. A., The essentials of pharmaceutical validation in Pharmaceutical Dosage
Forms: Tablets, Vol. 3, 2nd ed., Lieberman, H. A., Lachman, L. and Schwartz, J.
B., eds., Marcel Dekker, New York (1990).
8. Nash, R. A., Product formulation, CHEMTECH, (April 1976).
9. Pharmaceutical Process Validation, Berry, I. R. and Nash, R. A., eds., Marcel
Dekker, New York (1993).
10. Nash, R. A., Making the Paper Match the Work, Pharmaceutical Formulation &
Quality (Oct/Nov 2000).
11. Guidance for Industry, Scale-Up & Postapproval Changes, CDER, FDA (Nov
1995).
12. Bala, G., An integrated approach to process validation, Pharm. Eng. 14(3) (1994).
13. Farkas, D. F., Unit operations optimization operations, CHEMTECH, July 1977.
xxxvi Nash

14. Nash, R. A., Streamlining Process Validation, Amer. Pharm. Outsourcing May
2001.
15. Ishikawa, K., What is Total Quality Control? The Japanese Way, Prentice-Hall,
Englewood Cliffs, NJ (1985).
16. Nash, R. A., Practicality of Achieving Six Sigma or Zero-Defects in Pharmaceutical
Systems, Pharmaceutical Formulation & Quality, Oct./Nov. 2001.
17. CGMP: Amendment of Certain Requirements, FDA Federal Register, May 3,
1996.
18. Box, G. E. and Hunter, J. S., Statistics for Experimenters, John Wiley, N.Y. (1978).
19. Hendrix, C. D., What every technologist should know about experimental design,
CHEMTECH (March 1979).
20. Chapman, K. G., The PAR approach to process validation, Pharm. Tech., Dec.
1984.
21. Bolton, S., Pharmaceutical Statistics: Practical and Clinical Applications, 3rd ed.,
Marcel Dekker, New York (1997).
22. Schwartz, J. B., Optimization techniques in product formulation. J. Soc. Cosmet.
Chem. 32:287–301 (1981).
23. Butler, J. J., Statistical quality control, Chem. Eng. (Aug. 1983).
24. Deming, S. N., Quality by Design, CHEMTECH, (Sept. 1988).
25. Contino, AV., Improved plant performance with statistical process control, Chem.
Eng. (July 1987).
26. Meyer, R. J., Validation of Products and Processes, PMA Seminar on Validation
of Solid Dosage Form Processes, Atlanta, GA, May 1980.
27. Simms, L., Validation of Existing Products by Statistical Evaluation, Atlanta, GA,
May 1980.
28. Agalloco, J. P., Practical considerations in retrospective validation, Pharm. Tech.
(June 1983).
29. Kahan, J. S., Validating computer systems, MD&DI (March 1987).
1
Regulatory Basis for
Process Validation

John M. Dietrick
U.S. Food and Drug Administration, Rockville, Maryland, U.S.A.

Bernard T. Loftus
U.S. Food and Drug Administration, Washington, D.C., U.S.A.

I. INTRODUCTION

Bernard T. Loftus was director of drug manufacturing in the Food and Drug
Administration (FDA) in the 1970s, when the concept of process validation was
first applied to the pharmaceutical industry and became an important part of
current good manufacturing practices (CGMPs). His comments on the develop-
ment and implementation of these regulations and policies as presented in the
first and second editions of this volume are summarized below [1].

II. WHAT IS PROCESS VALIDATION?

The term process validation is not defined in the Food, Drug, and Cosmetic Act
(FD&C) Act or in FDA’s CGMP regulations. Many definitions have been of-
fered that in general express the same idea—that a process will do what it
purports to do, or that the process works and the proof is documented. A June
1978 FDA compliance program on drug process inspections [2] contained the
following definition:

This chapter was written by John M. Dietrick in his private capacity. No official support or endorse-
ment by the Food and Drug Administration is intended or should be inferred.

1
2 Dietrick and Loftus

A validated manufacturing process is one which has been proved to do what

it purports or is represented to do. The proof of validation is obtained
through the collection and evaluation of data, preferably, beginning from
the process development phase and continuing through the production
phase. Validation necessarily includes process qualification (the qualifica-
tion of materials, equipment, systems, buildings, personnel), but it also in-
cludes the control on the entire process for repeated batches or runs.
The first drafts of the May 1987 Guideline on General Principles of Process
Validation [3] contained a similar definition, which has frequently been used in
FDA speeches since 1978, and is still used today: “A documented program which
provides a high degree of assurance that a specific process will consistently pro-
duce a product meeting its pre-determined specifications and quality attributes.”

III. THE REGULATORY BASIS FOR

PROCESS VALIDATION

Once the concept of being able to predict process performance to meet user
requirements evolved, FDA regulatory officials established that there was a le-
gal basis for requiring process validation. The ultimate legal authority is Section
501(a)(2)(B) of the FD&C Act [4], which states that a drug is deemed to be
adulterated if the methods used in, or the facilities or controls used for, its
manufacture, processing, packing, or holding do not conform to or were not
operated or administrated in conformity with CGMP. Assurance must be given
that the drug would meet the requirements of the act as to safety and would
have the identity and strength and meet the quality and purity characteristics
that it purported or was represented to possess. That section of the act sets the
premise for process validation requirements for both finished pharmaceuticals
and active pharmaceutical ingredients, because active pharmaceutical ingredi-
ents are also deemed to be drugs under the act.
The CGMP regulations for finished pharmaceuticals, 21 CFR 210 and
211, were promulgated to enforce the requirements of the act. Although these
regulations do not include a definition for process validation, the requirement is
implicit in the language of 21 CFR 211.100 [5], which states: “There shall be
written procedures for production and process control designed to assure that
the drug products have the identity, strength, quality, and purity they purport or
are represented to possess.”

IV. THE REGULATORY HISTORY OF

PROCESS VALIDATION

Although the emphasis on validation began in the late 1970s, the requirement
has been around since at least the 1963 CGMP regulations for finished pharma-
ceuticals. The Kefauver-Harris Amendments to the FD&C Act were approved
Regulatory Basis for Process Validation 3

in 1962 with Section 501(a)(2)(B) as an amendment. Prior to then, CGMP and

process validation were not required by law. The FDA had the burden of prov-
ing that a drug was adulterated by collecting and analyzing samples. This was
a significant regulatory burden and restricted the value of factory inspections of
pharmaceutical manufacturers. It took injuries and deaths, mostly involving
cross-contamination problems, to convince Congress and the FDA that a revi-
sion of the law was needed. The result was the Kefauver–Harris drug amend-
ments, which provided the additional powerful regulatory tool that FDA re-
quired to deem a drug product adulterated if the manufacturing process was not
acceptable. The first CGMP regulations, based largely on the Pharmaceutical
Manufacturers Association’s manufacturing control guidelines, were then pub-
lished and became effective in 1963. This change allowed FDA to expect a
preventative approach rather than a reactive approach to quality control. Section
505(d)(3) is also important in the implementation of process validation require-
ments because it gives the agency the authority to withhold approval of a new
drug application if the “methods used in, and the facilities and controls used
for, the manufacture, processing, and packing of such drug are inadequate to
preserve its identity, strength, quality, and purity.”
Another requirement of the same amendments was the requirement that
FDA must inspect every drug manufacturing establishment at least once every
2 years [6]. At first, FDA did this with great diligence, but after the worst
CGMP manufacturing situations had been dealt with and violations of the law
became less obvious, FDA eased up its pharmaceutical plant inspection activi-
ties and turned its resources to more important problems.
The Drug Product Quality Assurance Program of the 1960s and 1970s
involved first conducting a massive sampling and testing program of finished
batches of particularly important drugs in terms of clinical significance and
dollar volume, then taking legal action against violative batches and inspecting
the manufacturers until they were proven to be in compliance. This approach
was not entirely satisfactory because samples are not necessarily representative
of all batches. Finished product testing for sterility, for example, does not assure
that the lot is sterile. Several incidents refocused FDA’s attention to process
inspections. The investigation of complaints of clinical failures of several prod-
ucts (including digoxin, digitoxin, prednisolone, and prednisone) by FDA found
significant content uniformity problems that were the result of poorly controlled
manufacturing processes. Also, two large-volume parenteral manufacturers ex-
perienced complaints despite quality control programs and negative sterility test-
ing. Although the cause of the microbiological contamination was never proven,
FDA inspections did find deficiencies in the manufacturing process and it be-
came evident that there was no real proof that the products were sterile.
What became evident in these cases was that FDA had not looked at the
process itself—certainly not the entire process—in its regulatory activities; it
was quality control- rather than quality assurance-oriented. The compliance offi-
4 Dietrick and Loftus

cials were not thinking in terms of process validation. One of the first entries
into process validation was a 1974 paper presented by Ted Byers, entitled “De-
sign for Quality” [7]. The term validation was not used, but the paper described
an increased attention to adequacy of processes for the production of pharma-
ceuticals. Another paper—by Bernard Loftus before the Parenteral Drug Associ-
ation in 1978 entitled “Validation and Stability” [8]—discussed the legal basis
for the requirement that processes be validated.
The May 1987 Guideline on General Principles of Process Validation [3]
was written for the pharmaceutical, device, and veterinary medicine industries.
It has been effective in standardizing the approach by the different parts of the
agency and in communicating that approach to manufacturers in each industry.

V. UPDATE

As discussed in the preceding sections, process validation has been a legal re-
quirement since at least 1963. Implementation of the requirement was a slow
and deliberate process, beginning with the development and dissemination of an
agency policy by Loftus, Byers, and others, and leading to the May 1987 guide-
line. The guideline quickly became an important source of information to phar-
maceutical manufacturers interested in establishing a process validation pro-
gram. Many industry organizations and officials promoted the requirements as
well as the benefits of validation. Many publications, such as Pharmaceutical
Process Validation [1] and various pharmaceutical industry journal articles,
cited and often expanded on the principals in the guideline. During the same
period, computer validation—or validation of computer controlled processes—
also became a widely discussed topic in both seminars and industry publications.
The regulatory implementation of the validation requirement was also a
deliberate process by FDA. During the 1980s, FDA investigators often reported
processes that had not been validated or had been inadequately validated. Batch
failures were often associated with unvalidated manufacturing processes. The
FDA issued a number of regulatory letters to deficient manufacturers citing the
lack of adequate process validation as a deviation from CGMP regulations
(21CFR 211.100), which causes the drug product to be adulterated within the
meaning of Section 501(a)(2)(B) of the federal FD&C Act. Process validation
was seldom the only deficiency listed in these regulatory letters. The failure of
some manufacturers to respond to these early warnings resulted in FDA filing
several injunction cases that included this charge in the early 1990s. Most of
these cases resulted in consent decrees, and ultimately the adoption of satisfac-
tory process validation programs by the subject manufacturers. One injunction
case filed in 1992, however, was contested in court and led to a lengthy written
order and opinion by the U.S. District Court in February of 1993 [9]. The court
Regulatory Basis for Process Validation 5

affirmed the requirement for process validation in the current good manufactur-
ing regulations, and ordered the defendants to perform process validation studies
on certain drug products, as well as equipment cleaning validation studies. This
case and the court’s ruling were widely circulated in the pharmaceutical industry
and became the subject of numerous FDA and industry seminars.
The court also criticized the CGMP regulations for their lack of specific-
ity, along with their ambiguity and vagueness. Responding to this criticism,
FDA drafted revisions to several parts of these regulations. The proposed revi-
sions were published in the Federal Register on May 3, 1996 [10]. One of the
main proposed changes was intended to emphasize and clarify the process vali-
dation requirements. The proposal included a definition of process validation
(the same definition used in the 1987 guideline), a specific requirement to vali-
date manufacturing processes, and minimum requirements for performing and
documenting a validation study. These were all implied but not specific in the
1978 regulation. In proposing these changes, FDA stated that it was codifying
current expectations and current industry practice and did not intend to add new
validation requirements. Comments from all interested parties were requested
under the agency’s rule-making policies, and approximately 1500 comments
were received. Most of the responses to the changes regarding process validation
supported the agency’s proposals, but there were many comments regarding the
definitions and terminology proposed about which processes and steps in a pro-
cess should or should not require validation, the number of batches required for
process validation, maintenance of validation records, and the assignment of
responsibility for final approval of a validation study and change control deci-
sions. Because of other high-priority obligations, the agency has not yet com-
pleted the evaluation of these responses and has not been able to publish the
final rule. In addition to the official comments, the proposed changes prompted
numerous industry and FDA seminars on the subject.
Process validation is not just an FDA or a U.S. requirement. Similar re-
quirements are included in the World Health Organization (WHO), the Pharma-
ceutical Inspection Co-operation Scheme (PIC/S), and the European Union (EU)
requirements, along with those of Australia, Canada, Japan, and other interna-
tional authorities.
Most pharmaceutical manufacturers now put substantial resources into
process validation for both regulatory and economic reasons, but despite contin-
ued educational efforts by both the agency and the pharmaceutical industry,
FDA inspections (both domestically and internationally) continue to find some
firms manufacturing drug products using unvalidated or inadequately validated
processes. Evidently there is still room for improvement, and continued discus-
sion, education, and occasional regulatory action appears warranted.
The future of process validation is also of great interest, especially with
the worldwide expansion of pharmaceutical manufacturing and the desire for
6 Dietrick and Loftus

harmonized international standards and requirements. Many manufacturers are

also working on strategies to reduce the cost of process validation and incorpo-
rate validation consideration during product design and development. New tech-
nologies under development for 100% analysis of drug products and other inno-
vations in the pharmaceutical industry may also have a significant effect on
process validation concepts and how they can be implemented and regulated.

REFERENCES

1. Loftus, B. T., Nash, R. A., ed. Pharmaceutical Process Validation. vol. 57. New
York: Marcel Dekker (1993).
2. U.S. Food and Drug Administration. Compliance Program no. 7356.002.
3. U.S. Food and Drug Administration. Guideline on General Principles of Process
Validation. Rockville, MD: FDA, 1987.
4. Federal Food Drug and Cosmetic Act, Title 21 U.S. Code, Section 501 (a)(2)(B).
5. Code of Federal Regulations, Title 21, Parts 210 & 211. Fed Reg 43, 1978.
6. U.S. Code, Federal Food Drug and Cosmetic Act, Title 21, Section 510 (h).
7. Byers, T. E. Design for quality, Manufacturing Controls Seminar, Proprietary Asso-
ciation, Cherry Hill, NJ, Oct. 11, 1974.
8. Loftus, B. T. Validation and stability, meeting of Parenteral Drug Association,
1978.
9. U.S. v. Barr Laboratories, Inc., et al., Civil Action No. 92-1744, U.S. District Court
for the District of New Jersey, 1973.
10. Code of Federal Regulations, Title 21, Parts 21 & 211, Proposed Revisions, Fed
Reg (May 3, 1996).
2
Prospective Process Validation

Allen Y. Chao
Watson Labs, Carona, California, U.S.A.

F. St. John Forbes

Wyeth Labs, Pearl River, New York, U.S.A.

Reginald F. Johnson and Paul Von Doehren

Searle & Co., Inc., Skokie, Illinois, U.S.A.

I. INTRODUCTION

Validation is an essential procedure that demonstrates that a manufacturing pro-

cess operating under defined standard conditions is capable of consistently pro-
ducing a product that meets the established product specifications. In its
proposed guidelines, the U.S. Food and Drug Administration (FDA) has offered
the following definition for process validation [1].
Process validation is establishing documented evidence that provides a
high degree of assurance that a specific process (such as the manufacture of
pharmaceutical dosage forms) will consistently produce a product meeting its
predetermined specifications and quality characteristics.
Many individuals tend to think of validation as a stand-alone item or an
afterthought at the end of the entire product/process development sequence.
Some believe that the process can be considered validated if the first two or
three batches of product satisfy specifications.
Prospective validation is a requirement (Part 211), and therefore it makes
validation an integral part of a carefully planned, logical product/process devel-
opmental program. An outline of the development sequence and requirements
relevant to process validation is presented in Figure 1. After briefly discussing
organizational aspects and documentation, the integration of validation into the
product development sequence is discussed. At the end of the chapter there is a
7
8 Chao et al.

brief discussion of specific ways in which experimental programs can be defined

to ensure that critical process development and validation objectives are met.

II. ORGANIZATION

Prospective validation requires a planned program and organization to carry it

to successful completion. The organization must have clearly defined areas of
responsibility and authority for each of the groups involved in the program so
that the objective of validating the process can be met. The structure must be
tailored to meet the requirements in the specific organization, and these will
vary from company to company. The important point is that a defined structure
exists, is accepted, and is in operation. An effective project management struc-
ture will have to be established in order to plan, execute, and control the pro-
gram. Without clearly defined responsibilities and authority, the outcome of
process validation efforts may not be adequate and may not comply with CGMP
requirements.

III. MASTER DOCUMENTATION

An effective prospective validation program must be supported by documenta-

tion extending from product initiation to full-scale production. The complete
documentation package can be referred to as the master documentation file.
It will accumulate as a product concept progresses to the point of being
placed in full-scale production, providing as complete a product history as possi-
ble. The final package will be the work of many individual groups within the
organization. It will consist of reports, procedures, protocols, specifications, ana-
lytical methods, and any other critical documents pertaining to the formulation,
process, and analytical method development. The package may contain the ac-
tual reports, or it may utilize cross-references to formal documentation, both
internal and external to the organization.
The ideal documentation package will contain a complete history of the
final product that is being manufactured. In retrospect, it would be possible to
trace the justification or rationale behind all aspects of the final product, process,
and testing.
The complete master documentation file not only provides appropriate
rationale for the product, process, and testing, but also becomes the reference
source for all questions relating to the manufacture of a product at any plant
location. This master documentation file, however, should not be confused with
the concept of the master product document, which is essential for routine manu-
facturing of the product and is described later in the chapter. The master docu-
 Prospective Process Validation
9

Figure 1 Prospective process validation.

10 Chao et al.

mentation file should contain all information that was generated during the en-
tire product development sequence to a validation process.

IV. PRODUCT DEVELOPMENT

Product development usually begins when an active chemical entity has been
shown to possess the necessary attributes for a commercial product. The product
development activities for the active chemical entity, formulation, and process
form the foundation upon which the subsequent validation data are built.
Generally, product development activities can be subdivided into formula-
tion and process development, along with scale-up development.

A. Formulation Development
Formulation development provides the basic information on the active chemical,
the formula, and the impact of raw materials or excipients on the product. Typi-
cal supportive data generated during these activities may include the following:
1. Preformulation profile or characterization of the components of the
formula, which includes all the basic physical or chemical information
about the active pharmaceutical ingredients (API, or the chemical en-
tity) and excipients
2. Formulation profile, which consists of physical and chemical charac-
teristics required for the products, drug-excipient compatibility stud-
ies, and the effect of formulation on in vitro dissolution
3. Effect of formulation variables on the bioavailability of the product
4. Specific test methods
5. Key product attributes and/or specifications
6. Optimum formulation
7. Development of cleaning procedures and test methods
Formulation development should not be considered complete until all those fac-
tors that could significantly alter the formulation have been studied. Subsequent
minor changes to the formulation, however, may be acceptable, provided they
are thoroughly tested and are shown to have no adverse effect on product.

B. Process Development
Even though the process development activities typically begin after the formu-
lation has been developed, they may also occur simultaneously. The majority of
the process development activities occur either in the pilot plant or in the pro-
Prospective Process Validation 11

posed manufacturing plant. The process development program should meet the
following objectives:
1. Develop a suitable process to produce a product that meets all
a. Product specifications
b. Economic constraints
c. Current good manufacturing practices (CGMPs)
2. Identify the key process parameters that affect the product attributes
3. Identify in-process specifications and test methods
4. Identify generic and/or specific equipment that may be required
It is important to remember that cleaning procedures should at least be in the
final stages of development, as equipment and facilities in the pilot or proposed
manufacturing plant are involved, and the development of the cleaning verifica-
tion test methods must be complete.
Process development can be divided into several stages.
Design
Challenging of critical process parameters
Verification of the developed process
Typical activities in these areas are illustrated in Figure 2.

1. Design
This is the initial planning stage of process development. The design of the
process should start during or at the end of the formulation development to
define the process to a large extent. One aspect of the process development
to remember is end user (manufacturing site) capabilities. In other words, the
practicality and the reality of the manufacturing operation should be kept in
perspective. Process must be developed in such a manner that it can easily be
transferred to the manufacturing site with minimal issues. During this stage,
technical operations in both the manufacturing and quality control departments
should be consulted.
Key documents for the technical definition of the process are the flow
diagram, the cause-and-effect diagram, and the influence matrix. The details of
the cause-and-effect diagram and the influence matrix will be discussed under
experimental approach in a later section.
The flow diagram identifies all the unit operations, the equipment used,
and the stages at which the various raw materials are added. The flow diagram
in Figure 3 outlines the sequence of process steps and specific equipment to be
used during development for a typical granulation solid dosage from product.
The flow diagram provides a convenient basis on which to develop a detailed
list of variables and responses.
 12

Figure 2 Product development flow.

Chao et al.
Prospective Process Validation 13

Figure 3 Typical process flow—granulated product.

14 Chao et al.

Preliminary working documents are critical, but they should never be cast
in stone, since new experimental data may drastically alter them. The final ver-
sion will eventually be an essential part of the process characterization and
technical transfer documents.
Regardless of the stage of formulation/process development being consid-
ered, a detailed identification of variables and responses is necessary for early
program planning. Typical variables and responses that could be expected in a
granulated solid dosage form are listed in Table 1. This list is by no means
complete and is intended only as an example.

Table 1 Typical Variables and Responses: Granulated Product

Process step Control variables Measured responses

Preblending Blending time Blend uniformity

rpm
Load size
Order of addition
Granulating Load size Density
Amount of granulating agent Yield
Solvent addition rate
rpm
Granulation time
Drying Initial temperature Density
Load size Moisture content
Drying temperature program Yield
Air flow program
Drying time
Cooling time
Sizing Screen type Granule size distribution
Screen size Loose density
Feed rate Packed density
Blending Load size Blend uniformity
rpm Flow characteristics
Blending time Particle size distribution
Tableting Compression rate Weight variation
Granule feed rate Friability
Precompression force Hardness
Compression force Thickness
Disintegration time
Dissolution
Dosage form uniformity
Prospective Process Validation 15

As the developmental program progresses, new discoveries will provide

an update of the variables and responses. It is important that current knowledge
be adequately summarized for the particular process being considered. It should
be pointed out, however, that common sense and experience must be used in
evaluating the variables during process design and development. An early trans-
fer of the preliminary documentation to the manufacturing and quality control
departments is essential, so that they can begin to prepare for any new equip-
ment or facilities that may be required.

2. Challenging of Process Parameters

Challenging of process parameters (also called process ranging) will test
whether or not all of the identified process parameters are critical to the product
and process being developed. These studies determine:
The feasibility of the designed process
The criticality of the parameters
This is usually a transition stage between the laboratory and the projected
final process. Figure 4 also shows typical responses that may have to be evalu-
ated during the ranging studies on the tableted product.

3. Challenging of Critical Process Parameters or

Characterization of the Process
Process characterization provides a systematic examination of critical variables
found during process ranging. The objectives of these studies are
Confirm critical process parameters and determine their effects on product
quality attributes.
Establish process conditions for each unit operation.
Determine in-process operating limits to guarantee acceptable finished
product and yield.
Confirm the validity of the test methods.
A carefully planned and coordinated experimental program is essential
in order to achieve each of these objectives. Techniques to assist in defining
experimental programs are mentioned later in the chapter.
The information summarized in the process characterization report pro-
vides a basis for defining the full-scale process.

4. Verification
Verification is required before a process is scaled up and transferred to produc-
tion. The timing of this verification may be critical from a regulatory point of
view, as the there is little or no room for modifying the parameter values and
16 Chao et al.

specifications, particularly shifting or expanding after the regulatory submission

is made. This ensures that it behaves as designed under simulated production
conditions and determines its reproducibility. Key elements of the process verifi-
cation runs should be evaluated using a well-designed in-process sampling pro-
cedure. These should be focused on potentially critical unit operations. Vali-
dated in-process and final-product analytical procedures should always be used.
Sufficient replicate batches should be produced to determine between- and
within-batch variations.
Testing during these verification runs will be more frequent and cover
more variables than would be typical during routine production. Typically the
testing requirements at the verification stage should be the same or more than
the proposed testing for process validation runs. The typical process verification
analysis of tabulated product includes the following:

Unit operation Analysis

Preblending Potency (if required)

Granulation Potency (if required)
Sizing Particle size distribution
Loss on drying (LOD)
Blending Uniformity
Particle size distribution
Tableting Weight
Hardness
Thickness
Disintegration and/or dissolution
Friability
Potency
Dosage uniformity
Degradants

For maximum information, the process should not be altered during the verifica-
tion trials.
5. Development Documentation
The developmental documentation to support the validation of the process may
contain the following:
Process challenging and characterization reports that contain a full de-
scription of the studies performed
Development batch record
Raw material test methods and specifications
Prospective Process Validation 17

Equipment list and qualification and calibration status

Process flow diagram
Process variable tolerances
Operating instructions for equipment (where necessary)
In-process quality control program, including:
Sampling intervals
Test methods
Finished Product
Stability
Critical unit operation
Final product specifications
Safety evaluation
Chemical
Process
Special production facility requirements
Cleaning
Procedure for equipment and facilities
Test methods
Stability profile of the product
Produced during process development
Primary packaging specification

V. DEVELOPMENT OF MANUFACTURING CAPABILITY

There must be a suitable production facility for every manufacturing process

that is developed. This facility includes buildings, equipment, staff, and support-
ing functions.
As development activities progress and the process becomes more clearly
defined, there must be a parallel assessment of the capability to manufacture the
product. The scope and timing of the development of manufacturing capability
will be dependent on the process and the need to utilize or modify existing
facilities or establish new ones.

VI. FULL-SCALE PRODUCT/PROCESS DEVELOPMENT

The development of the final full-scale production process proceeds through the
following steps:
Process scale-up studies
Qualification trials
Process validation runs
18 Chao et al.

A. Scale-Up Studies
The transition from a successful pilot-scale process or research scale to a full-
scale process requires careful planning and implementation. Although a large
amount of information has been gathered during the development of the process
(i.e., process characterization and process verification studies), it does not neces-
sarily follow that the full-scale process can be completely predicted.
Many scale-up parameters are nonlinear. In fact, scale-up factors can be
quite complex and difficult to predict, based only on experience with smaller-
scale equipment. In general, the more complex the process, the more complex
the scale-up effect.
For some processes, the transition from pilot scale or research scale to full
scale is relatively easy and orderly. For others the transition is less predictable.
More often than not there will be no serious surprises, but this cannot be guaran-
teed. Individuals conducting the transfer into production should be thoroughly
qualified on both small- and large-scale equipment.
The planning for scale-up should follow the same general outline followed
for process characterization and verification. It usually begins when process
development studies in the laboratory have successfully shown that a product
can be produced within specification limits for defined ranges of process param-
eters.
Frequently, because of economic constraints, a carefully selected excipient
may be used as a substitute for the expensive active chemical in conducting
initial scale-up studies. Eventually, the active chemical will have to be used to
complete the scale-up studies, however.
It is common sense that every effort will be made to conduct the final
scale-up studies under CGMP conditions, thus any product produced with speci-
fications can be considered for release as a finished salable product (for over-
the-counter products only).

B. Qualification Trials
Once the scale-up studies have been completed, it may be necessary to manufac-
ture one or more batches at full scale to confirm that the entire manufacturing
process, comprising several different unit operations, can be carried out
smoothly. This may occur prior to or after the regulatory submission, depending
on the strategy used in filing.

C. Process Validation Runs

After the qualification trials have been completed, the protocol for the full-scale
process validation runs can be written. Current industry standard for the valida-
tion batches is to attempt to manufacture them at target values for both process
Prospective Process Validation 19

parameters and specifications. The validation protocol is usually the joint effort
of the following groups:

Research and development

Pharmaceutical technology or technical services
Quality control (quality assurance)
Manufacturing
Engineering
One of these groups usually coordinates the activities.
A complete qualification protocol will contain specific sections; however,
there can be considerable variation in individual protocol. Section content typi-
cal validation protocol may consist of the following:

Safety instructions
Environmental restrictions
Gas or liquid discharge limitations
Solid or scrap disposal instructions
Equipment
Description
Operation
Cleaning
Raw materials
Pertinent characteristics
Acceptance limits
Analytical methods
Packaging and storage
Handling precautions
Process flow chart
Critical parameters and related means of controls
Responsibilities of each of the groups participating
Cleaning validation/verification requirements
Master batch components (percentage by weight)
Production batch component (by weight)
Process batch record
Process sequence
Process instructions
Material usage
Product testing
In-process testing and acceptance criteria
Finished product testing and acceptance criteria
Test method references
Formulation
20 Chao et al.

Validation sampling and testing

In-process
Finished product
Definition of validation criteria
Lower and upper acceptance limits
Acceptable variation
Cleaning sampling plan (locations, type, and number of samples)
It is expected that acceptable, salable products will be produced, since all quali-
fication batches will be produced using a defined process under CGMP condi-
tions with production personnel.
A question that always arises is how many replicate batches or lots must
be produced for a validation protocol to be valid or correct. There is no absolute
answer. Obviously, a single batch will provide the minimum amount of data.
As the number of replicated batches increases, the information increases. The
FDA, however, has determined that the minimum number of validation batches
should be three.

D. Master Product Document

An extensive quantity of documents is generated at each stage of the develop-
ment and validation of the final production process. Some of these documents
will be directly related to the manufacture of the final products. Others may
provide the basis for decisions that ultimately result in the final process.
The documents that are required for manufacturing the product then be-
come the master product document. This document must be capable of provid-
ing all of the information necessary to set up the process to produce a product
consistently and one that meets specifications in any location.
Items that will normally be included in the master product document are
Batch manufacturing record
Master formulation
Process flow diagram
Master manufacturing instructions
Master packaging instructions
Specifications
Sampling (location and frequency)
Test methods
Process validation data
Each of the above items must contain sufficient detailed information to permit
the complete master product document to become an independent, single pack-
age that will provide all information necessary to set up and produce a product.
Prospective Process Validation 21

VII. DEFINING EXPERIMENTAL PROGRAMS

The objective in this section is to examine experiments or combinations of re-

lated experiments that make up development programs so that adequate justifi-
cation can be developed for the formulation, process, and specifications. The
emphasis will be on techniques to increase developmental program effective-
ness.
A logical and systematic approach to each experimental situation is essen-
tial. Any experiment that is performed without first defining a logical approach
is certain to waste resources. The right balance between overplanning and under-
planning should always be sought.
It is usually impossible to define a substantial experimental effort at the
beginning and then execute it in every detail without modification. To overcome
this, it is convenient to split the program into a number of stages.
Each stage will normally consist of several specific experiments. The ear-
lier experiments tend to supply initial data concerning the process and define
preliminary operating ranges for important variables. As results become avail-
able from each stage, they can be used to assist in defining subsequent stages
in the experimental program. In some cases it may be necessary to redefine
completely the remainder of the experimental program on the basis of earlier
results.
The following discussion describes some techniques to help improve ex-
perimental program effectiveness. A logical and systematic approach coupled
with effective communication among individuals associated with the program is
emphasized. Topics to be discussed include
Defining program scope
Process summary
Experimental design and analysis
Experiment documentation
Program organization

A. Program Scope
Defining a clear and detailed set of objectives is a necessary first step in any
experimental program. Some similarity exists between objectives for different
products and processes using similar existing technology. For products and pro-
cesses at the forefront of technology, the definition of specific experimental
objectives can be a continuing activity throughout product development.
Constraints on planning experimental programs can be classified accord-
ing to their impact on time, resources, and budget. The effect and impact of
these should be incorporated into the experimental program early to avoid com-
promising critical program objectives.
22 Chao et al.

B. Process Summary
An initial clear understanding of the formulation and/or process is important.
The following techniques can assist in summarizing current process knowledge.

1. Flow Diagram
A process flow diagram (Fig. 3) can often provide a focal point of early program
planning activities. This diagram outlines the sequence of process steps and
specific equipment to be used during development for a typical granulated prod-
uct. Flow diagram complexity will depend on the particular product and process.
The flow diagram provides a convenient basis on which to develop a detailed
list of variables and responses.

2. Variables and Responses

For process using existing technology, many of the potential variables and re-
sponses may have already been identified in previous product-development
studies or in the pharmaceutical literature. Once properly identified, the list of
variables and responses for the process is not likely to change appreciably. Typi-
cal variables and responses that could be expected in a granulated solid dosage
form are listed in Table 1.
In addition, the relative importance of variables and responses already
identified will likely shift during development activities.

3. Cause-and-Effect Diagram
An efficient representation of complex relationships between many process and
formulation variables (causes), and a single response (effect) can be shown by
using a cause-and-effect diagram [1]. Figure 4 is a simple example.
A central arrow in Figure 4 points to a particular single effect. Branches
off the central arrow lead to boxes representing specific process steps. Next,
principle factors of each process step that can cause or influence the effect are
drawn as subbranches of each branch, until a complete cause-and-effect diagram
is developed. This should be as detailed a summary as possible. An example of
a more complex cause-and-effect diagram is illustrated in Figure 5. A separate
summary for each critical product characteristic (e.g., weight variation, dissolu-
tion, friability) should be made.

4. Influence Matrix
Once the variables and responses have been identified, it is useful to summarize
their relationships in an influence matrix format, as shown in Figure 6. Based
on the available knowledge, each process variable is evaluated for its potential
 Prospective Process Validation

Figure 4 Simple cause-and-effect diagram.

23
 24

Figure 5 Cause-and-effect diagram (granulated product).

Chao et al.
 Prospective Process Validation
25

Figure 6 Influence matrix for variables and responses (simplified).

26 Chao et al.

effects on each of the process responses or product characteristics. The strength

of the relationship between variables and responses can be indicated by some
appropriate notation, such as strong (S), moderate (M), weak (W), or none (N),
together with special classifications such as unknown (?).
Construction of the influence matrix assists in identifying those variables
with the greatest influence on key process or product characteristics. These vari-
ables are potentially the most critical for maintaining process control and should
be included in the earliest experiments. Some may continue to be investigated
during development and scale-up.

VIII. EXPERIMENTAL DESIGN AND ANALYSIS

Many different experimental designs and analysis methods can be used in devel-
opment activities (Fig. 7). Indeed, the possibilities could fill several books. For-
tunately, in any given situation, it is not necessary to search for that single
design or analysis method that absolutely must be used; there are usually many
possibilities. In general, designs that are usable offer different levels of effi-
ciency, complexity, and effectiveness in achieving experimental objectives.

A. Types of Design
It is not possible to list specific designs that will always be appropriate for
general occasions. Any attempt to do so would be sure to be ineffective, and
the uniqueness of individual experimental situation carefully, including
Specific objectives
Available resources
Availability of previous theoretical results
Relevant variables and responses
Qualifications and experience of research team members
Cost of experimentation
It should also be determined which design is appropriate. A statistician who is
experienced in development applications can assist in suggesting and evaluating
candidate designs. In some cases, the statistician should be a full-time member
of the research team.

B. Data Analysis
The appropriate analysis of the experimental results will depend on the experi-
mental objectives, the design used, and the characteristics of the data collected
during the experiment. In many cases, a simple examination of a tabular or
 Prospective Process Validation

Figure 7 Experimental design example.

27
28 Chao et al.

graphical presentation of the data will be sufficient. In other cases, a formal

statistical analysis may be required in order to draw any conclusions at all. It
depends on the particular experimental situation. No rules of thumb are avail-
able. In general, the simplest analysis consistent with experimental objectives
and conditions is the most appropriate.

C. Experiment Documentation
Documentation is essential to program planning and coordination, in addition to
the obvious use for the summary of activities and results. Written communica-
tion becomes important for larger complex programs, especially when con-
ducted under severe constraints on time and resources. Documentation can con-
sist of some or all of the following items:
1. Objectives; an exact statement of quantifiable results expected from
the experiment
2. Experimental design; a detailed list of the experimental conditions to
be studied and the order of investigation
3. Proposed/alternate test methods
a. A list of test methods consistent with the type of experiment be-
ing performed
b. A detailed description of the steps necessary to obtain a valid
measurement
c. Documentation supporting the accuracy, precision, sensitivity,
and so on of the test methods
4. Equipment procedures; documentation of safety precautions and step-
by-step methods for equipment setup, operation, and cleanup
5. Sampling plans; the type, number, location, and purpose of samples
to be taken during the experiment; in addition, the type and number
of all measurements to be performed on each sample
6. Protocol; a formal written experimental plan that presents the afore-
mentioned experimental documentation in a manner suitable for re-
view
7. Data records
a. Experiment log; details of events in the experiment noting process
adjustments and any unusual occurrences
b. In-process measurements; records of the magnitude of critical
process parameters during the experimental sequence
Sample measurements; recorded values of particular measure-
ments on each sample
8. Report; documentation of experiment implementation, exceptions/
modifications to the protocol, results, and conclusion
Prospective Process Validation 29

D. Program Organization
Throughout the experimental phases of the development program, it is essential
to maintain effective communication among various team members. This is fa-
cilitated by having one individual with the necessary technical and managerial
skills assume responsibility for the experimental program, including procuring
resources and informing management of progress.
In a large experimental program, the responsible individual may serve as
a project leader or manager with little or no technical involvement.

IX. SUMMARY

Prospective validation of a production process utilizes information generated

during the entire development sequence that produced the final process.
Validation is supported by all phases of development from the product
concept.
As a potential product moves through the various developmental stages,
information is continually generated and incorporated into a master documenta-
tion file. When the validation runs are planned for the final process, they will
be based on the master documentation file contents. The information generated
during the validation runs is usually the last major item to go into the master
documentation file.
An abstract of the master documentation file is the master product docu-
ment, which is the source of all information required to set up the process at
any location.
Though validation may seem to be a stand-alone item, it actually is an
integral portion of the entire product/process development sequence.

REFERENCES

1. FDA. Guidelines on General Principles of Process Validation. Rockville, MD: Divi-

sion of Manufacturing and Product Quality (HFN-320) Center for Drugs and Biolog-
ics (May 1987).
2. Box, G. E. P., Gunter, W. G., and Hunter, J. S. Statistics for Experimenters: An
Introduction to Design, Data Analysis, and Model Building. New York: Wiley
(1978).
3. Box, G. E. P., and Draper, N. R. Evolutionary Operation: A Statistical Method for
Process Improvement. New York: Wiley (1969).
4. Cornell, J. A. Experiments with Mixtures: Design, Models and the Analysis of Mix-
ture Data. New York: Wiley (1981).
5. Daniel, C. Applications of Statistics to Industrial Experiments. New York: Wiley
(1976).
30 Chao et al.

6. Davies, O. L., ed. The Design and Analysis of Industrial Experiments. New York:
Longman Group (1978).
7. Diamond, W. J. Practical Experiment Designs for Engineers and Scientists. Bel-
mont, CA: Lifetime Learning Publications (1981).
8. Ott, E. R. Process Quality Control: Troubleshooting and Interpretation of Data.
New York: McGraw-Hill (1975).
9. Anderson, N. R., Banker, G. S., and Peck, G. E. Pharmaceutical Dosage Forms:
Tablets. vol. III. New York: Marcel Dekker (1981).
3
Retrospective Validation

Chester J. Trubinski
Church & Dwight Co., Inc., Princeton, New Jersey, U.S.A.

I. INTRODUCTION

In the present-day pharmaceutical industry the Food and Drug Administration

(FDA) expects firms to have validated manufacturing processes. Process valida-
tion has been defined as a documented program that provides a high degree of
assurance that a specific process will consistently produce a product meeting
predetermined specifications [1]. For new products or existing products that
have recently undergone reformulation, validation is usually an integral part of
the process development effort. No such opportunity exists for older established
products, however. Of the brands recognized as medical or scientific break-
throughs of the 20th century that continue to be marketed, 21 were introduced
before 1980 [2]. This suggests product lines are likely to contain a product for
which the manufacturing processes have not been validated, at least not to the
extent that is now expected.

II. PROCESS VALIDATION STRATEGIES

The FDA has published a guideline for use by industry that outlines general
principles considered acceptable parts of process validation [1]. Pharmaceutical
firms have been inspected against this standard and those found wanting have
been cited or had approval to manufacture product denied. Indeed, statistics
compiled by the FDA for fiscal year 1997 show inadequate process validation
as one of the top 10 reasons for withholding approval [3]. One way for a firm
to satisfy the requirement for validated processes is to identify those products
that have been on the market for some time and use the wealth of production,
31
32 Trubinski

testing, and control data to demonstrate that the process is reliable. This strate-
gy is commonly referred to as retrospective validation. Historical data also may
be used to augment an earlier validation in cases in which the product has
changed.

A. Product Selection Criteria for Retrospective Validation

For a product to be considered for retrospective validation, it must have a stable
process; that is, one in which the method of manufacture has remained essen-
tially unchanged for a period of time.
The first step in the product selection process is therefore to obtain a
summary of changes in the method of manufacture. In most companies such
information is part of the master batch record file. Then a time interval is se-
lected that represents the last 20 to 30 batches. Products for which there is no
record of a change in the method of manufacture or control during this period
can be regarded as candidates for validation. The 20-to-30-batch rule originates
from control chart principals, which consider 20 to 30 points that plot within
the limits as evidence of a stable process [4]. Once this criterion is met, the
number selected is actually somewhat arbitrary, as there is no one number that
is correct for every product. The ideal number of batches required to study a
product is theoretically the number that permits all process variables to come
into play. By process variables, we mean raw materials from different but ap-
proved vendors, introduction of similar but different pieces of equipment, per-
sonnel and seasonal changes, and the like. This academic approach may present
a rather unwieldy situation, especially for a high-volume product, for which
change in process variables occurs infrequently. The influence of seasonal
changes is such an example. In such instances, compromise will need to be
reached between process variables included for study and the number of batches
that can be examined for data. This decision making is best handled by a valida-
tion committee, the organization and makeup of which is covered in detail later
in this chapter.
The second step in the product selection process addresses the situation in
which a change in the method of manufacture or control was implemented dur-
ing the last 20 or so production batches. The fact that a change has occurred
does not automatically disqualify the product for retrospective validation. One
must first know whether the particular modification has caused an expected
result to be different to the extent that it is no longer comparable to previous
batches. An example may be helpful. Suppose the method of granulating was
changed midway through the series of 20 batches selected for the validation
study. The number of batches representing the new process would be signifi-
cantly reduced and could be insufficient to capture some of the interactions that
can affect process reproducibility. In general, a history of any one of the follow-
Retrospective Validation 33

ing changes to the method of manufacture and control should be fully investi-
gated before any decision is made to validate retrospectively:
1. Formulation changes involving one or more of the active ingredients
or key excipients
2. Introduction of new equipment not equivalent in every respect to that
previously in use
3. Changes in the method of manufacture that may affect the product’s
characteristics
4. Changes to the manufacturing facility
A product found to be unsuitable for retrospective validation because of a
revised manufacturing process is a likely candidate for prospective validation,
which is beyond the scope of this chapter [1]. Such a discovery, however, should
be brought to the attention of the appropriate authority. In today’s regulatory
environment ignoring the matter would be imprudent.
The third and last step in our selection process is to identify which prod-
ucts are likely to be discontinued because of a lack of marketing interest or
regulatory consideration, to be sold, or to be reformulated. The timing of these
events will dictate whether the product in question remains a viable candidate
for retrospective validation.
The foremost discussion on developing a list of suitable products for study
is summarized in Figure 1.

B. Organizing for Retrospective Validation

To this point we have produced a list of products that may be validated retro-
spectively; that is, their manufacturing processes are relatively stable, and so
adequate historical data exist on which to base an opinion. The next consider-
ation is the formal mechanism for validating the individual products. Appro-
priate organizational structures for effectively validating processes have been
put forth, but mostly in conjunction with the validation of new product introduc-
tions. Still, these recommendations can serve as models. Because the products
being studied are marketed products, the quality assurance and production de-
partments can be expected to make major contributions. In fact, as far as retro-
spective validation is concerned, it may be more appropriate for one of these
departments to coordinate the project. The research and engineering depart-
ments, of course, will be needed, especially where recent process changes have
been encountered or equipment design is at issue.
Operating as a team, the previously discussed disciplines will determine
which data should be collected for each product and from how many batches;
subsequently, they will evaluate the information and report their findings. Per-
sonnel resources beyond this committee are necessary to accomplish the tasks
34 Trubinski

Figure 1 Selection of candidates for retrospective validation.

of data collection and analysis. The time requirements dictate that such work be
assigned to a function with discretionary time, possibly a technical services
group or a quality engineer. Management commitment is especially crucial if
disruptive influences are to be minimized. The loss of a committee member to
another project is such an example.

C. Written Operating Procedures

The various activities and responsibilities associated with retrospectively vali-
dating a product must be put in writing. All too often this simple but crucial
step is omitted for the sake of expediency only to find at a later date that the
Retrospective Validation 35

initial assumptions cannot be recalled. Aside from maintaining consistency, a

written procedure to describe the work being performed satisfies the intent of
the current good manufacturing practice (CGMP) regulations.
In general, the written operating procedure should delineate in reasonable
detail how the validation organization will function. Not every situation can be
anticipated, and this should not be the goal. There should be sufficient detail,
however, to ensure consistency of performance in an undertaking that may con-
tinue for several months. In the preparation of such a document, the following
questions should be answered:
1. Which organizational functions will be represented on the validation
committee?
2. What mechanism exists for validation protocol preparation and ap-
proval?
3. What criteria are used to select critical process steps and quality con-
trol tests for which data will be collected?
4. How often will the committee meet to ensure prompt evaluation of
study data?
5. Who has responsibility for documenting committee decisions? For
report preparation?
6. Is there a provision for follow-up in the event of unexpected findings?
7. Where will the original study data and reports be archived?
In the preceding discussion of areas of interest to the validation organiza-
tion, two concepts were introduced that deserve further clarification: (1) critical
process steps and quality control tests that characterize the operation, and (2)
validation protocol.

1. Critical Process Steps and Control Tests

Critical process steps are operations performed during dosage-form manufacture
that can contribute to variability of the end product if not controlled. Since each
type of dosage form requires different machinery and unit operations to produce
the end product, the critical process steps will also differ. For each product
considered suitable for retrospective validation, a list of these steps must be
compiled following careful analysis of the process by technically competent
persons. In a similar manner, in-process and finished-product tests should be
screened to identify those that may be of some value. As a rule, tests in that the
outcome is quantitative will be of greatest interest.
A flow diagram of the entire operation, but particularly of the manufactur-
ing process, may be helpful in identifying critical steps, especially where the
process involves many steps. Such a diagram is also a useful addition to the
validation report prepared at the conclusion of the study.
36 Trubinski

2. Validation Protocol
A written protocol that describes what is to be accomplished should be prepared
[5]. It should specify the data to be collected, the number of batches to be
included in the study, and how the data, once assembled, will be treated for
relevance. The criteria for acceptable results should be described. The date of
approval of the protocol by the validation organization should also be noted.
The value of a protocol is to control the direction of the study, as well as provide
a baseline in the event unanticipated developments necessitate a change in strat-
egy. A written protocol is also an FDA recommendation [1].

D. Other Considerations
Comprehensive records of complaints received either directly from the customer
or through a drug problem reporting program should be reviewed. Furthermore,
a record of any follow-up investigation of such complaints is mandatory [6] and
should be part of this file. Review of customer complaint records can furnish a
useful overview of process performance and possibly hint at product problems.
Complaint analysis should therefore be viewed as a meaningful adjunct to the
critical process step and control test selection process.
Batch yield reflects efficiency of the operation. Because yield figures are
the sum of numerous interactions, they fail in most cases to provide specific
information about process performance and therefore must be used with caution
in retrospective validation. In any event, this information should be collected,
as it can contribute to further refinement of the yield limits that appear in the
batch record.
Lot-to-lot differences in the purity of the therapeutic agent must be consid-
ered when evaluating in-process and finished-product test results. In addition to
potency such qualities as particle size distribution, bulk density, and source of
the material will be of interest. Such information should be available from the
raw material test reports prepared by the quality control laboratory for each lot
of material received. The physical characteristics of the excipients should not
be overlooked, especially for those materials with inherent variability. Metallic
stearates is a classic example. In such instances, the source of supply is desirable
information to have available.
There is value in examining logs of equipment and physical plant mainte-
nance. These documents can provide a chronological profile of the operating
environment and reveal recent alterations to the process equipment that may
have enough impact to disqualify the product from retrospective validation con-
sideration. For this reason, it is always prudent to contemplate equipment status
early in the information-gathering stage. The availability of such information
should be ascertained for yet another reason: rarely is equipment dedicated to
Retrospective Validation 37

one product. More often than not, each blender, comminutor, tablet press, and
so forth is used for several operations. Information gathered initially can there-
fore be incorporated into subsequent studies.
Retrospective validation is directed primarily toward examining the rec-
ords of past performance, but what if one of these documents is not a true
reflection of the operation performed? Suppose that changes have crept into the
processing operation over time and have gone unreported. This condition would
result in the validation of a process that in reality does not exist. It is therefore
essential to audit the existing operation against the written instructions. There is
obvious advantage to undertaking this audit before commencing data acquisi-
tion. Ideally, the manufacture of more than one batch should be witnessed, espe-
cially where multiple-shift operations are involved. The same logic would apply
to the testing performed in process and at the finished stage. If any deviation
from the written directions is noted, an effort must be made to measure its
impact. In this regard, the previously described validation organization is a logi-
cal forum for discussion and evaluation.
As a rule, batches that are rejected or reworked are not suitable for inclu-
sion in a retrospective validation study [7]. Indeed, a processing failure that is
not fully explainable should be cause to rethink the application of retrospective
validation. Nonconformance to specification that is attributable to a unique
event–operator error, for example, may be justifiably disregarded. In such cases,
the batch is not considered when the historical data are assembled.
Raw materials, both actives and excipients, can be a source of product
variability. To limit this risk, there should be meaningful acceptance specifica-
tions and periodic confirmation of test results reported on the supplier’s certifi-
cate of analysis. Also, purchases must be limited to previously qualified suppli-
ers. A determination that such controls are in place should be part of any
retrospective validation effort.

III. SELECTION AND EVALUATION

OF PROCESSING DATA

The following discussion will focus on how to apply the previously discussed
concepts to the validation of marketed products. To provide a fuller understand-
ing of this procedure, the manufacture of several dosage forms designed for
different routes of administration will be examined. For each dosage form, criti-
cal process steps and quality control tests will be identified. Useful statistical
techniques for examining the assembled data will be illustrated. It is also impor-
tant to note that not all of the collected information for a product lends itself to
this type of analysis. This will become more apparent as we proceed with the
evaluation of the five drugs under consideration.
38 Trubinski

A. Compressed Tablet (Drug A)

Drug A is a compressed tablet containing a single active ingredient. Inspection
of the batch record reveals that the following operations are involved in the
manufacture of the dosage unit. The active ingredient is combined with several
excipients in a twin-shell blender. The premix just prepared is granulated using
a purified water-binder solution. The resulting wet mix is milled using a speci-
fied screen and machine setting, then dried using either an oven tray dryer or a
fluid bed dryer. When dry, the blend is oscillated, combined with previously
sized lubricant, and blended. The granulation is then compressed. See Figure 2
for a flow diagram of the manufacturing process.
At the premix blending step, the batch record provides two pieces of infor-

Figure 2 Drug A: flow diagram of manufacturing process.

Retrospective Validation 39

mation: recommended blending time and blender load. The latter will be of little
interest, as only one size batch is produced for this product. Blender speed is
not specified in the batch record because it is fixed. Because mixing time has
been recognized as influencing blend uniformity, this operation will become the
first of the critical process steps for which we will want to collect historical
information [8].
The second major step is granulation. The process is controlled by the
operator, whose judgment is relied on for the appropriate end point. As no
information useful for process validation is available, we will move on to the
next step, comminution.
The batch record calls for passing the wet mix through a comminutor
using a no. 5 or 7 drilled stainless steel screen. Knife position and rotational
speed are two other factors that influence particle size; however, the step instruc-
tion is quite specific about machine setup. Therefore, only screen size is a source
of variability for this step. We will want to know the frequency of use of each
screen.
Next, the granulation is dried to a target moisture of 1%. Either a tray or
fluid bed dryer may be used, at the discretion of area supervision. Regardless
of the method, drying time will be of interest. In addition, the final moisture
content should be ascertained for each batch. The dried granulation and lubricant
are then oscillated using a no. 10 or 12 wire screen. This is the last sizing
operation of the process; it will determine the particle size distribution of the
final blend. Knowing the history of use of each screen size is thus important.
The lubricant and granulation are blended for several minutes. The elapsed
mixing time is of interest because of its impact on drug distribution and the
generally deleterious effect of the lubricant on dissolution.
Because excess moisture is thought to have a negative effect on the dosage
form, loss on drying (LOD) is determined on the final blend.
Blending is followed by tableting. During compression, online measure-
ments such as tablet weight, hardness, and disintegration are made by the pro-
cess operator in order to ensure uniformity of the tablets. The weight of the
tablets is not measured individually; rather, the average weight of 10 tablets is
recorded. Although these data are good indicators of operation and machine
performance, we would prefer to have the more precise picture provided by
individual tablet weight.
Disintegration time and tablet hardness data could be collected from the
manufacturing batch records; however, for ease of administration these figures
will be obtained from the quality control test results, which also contain individ-
ual tablet weighings.
Disintegration time was selected as a critical variable because for a drug
substance to be absorbed it must first disintegrate and then dissolve. The resis-
tance of a tablet to breakage, chipping, and so forth depends on its hardness.
40 Trubinski

Disintegration, too, can be influenced by hardness of the tablet. For these rea-
sons, hardness testing results also will be examined.
Specifications used by quality control to release drug A are found in a
laboratory procedure. In addition to the previously discussed hardness and disin-
tegration time requirements, the procedure calls for determining the average
tablet weight by the United States Pharmacopeia (USP) procedure; that is, 20
individual tablets are weighed.
The control procedure also requires assay of individual tablets. Of all the
information available, these data will be the most useful in reaching an opinion
of the adequacy of the process to distribute the therapeutic agent uniformly.
In addition, the laboratory checks the moisture content of the bulk tablets.
It will be interesting to compare these results to the LOD of the final blend to
measure the contribution of material handling.
Critical manufacturing steps and quality control tests for drug A, identified
as a result of the review, are summarized in Table 1.

1. Evaluation of Historical Data

Earlier in the discussion of process validation strategies, 20 production batches
were suggested as a minimum number upon which to draw conclusions about
the validity of the process. In this particular example, however, two distinct
methods of drying are provided. In order to have sufficient history on each
operation, the number of batches examined was increased to 30.
The batches were selected so that the same number was dried by each
process. For the other critical manufacturing steps and release tests listed in
Table 1, data were collected for all 30 batches.
The first manufacturing step, premix blending time, was consistently re-
ported as 10 min, but with one exception. In this instance, the powders were
tumbled for 20 min, which is still within the limits (10 to 20 min) prescribed
by the batch record. It would be interesting to know if this source of variability

Table 1 Drug A: Selected Critical Process Steps and Quality

Control Tests

Process steps Quality control tests

Premix blending time Disintegration time

Comminutor screen size Hardness
Drying time and method Average tablet weight (ATW)
Loss on drying (LOD)—granulation Assay
Oscillator screen size Water content-bulk tablet
Final mix blending time
LOD—final blend
Retrospective Validation 41

can materially affect attributes of the final product. Unfortunately, having only
one batch produced by the 20-min process does not permit statistically valid
comparisons. At best, test results for the single 20-min batch can be screened
using summary data from the remainder of the study. Under different circum-
stances, batches would have been grouped by mixing time and compared by
dosage form attributes. More than likely, subsequent manipulation of the blend
would have negated any contribution, allowing us to conclude that a mixing
time of 10 to 20 min is not unreasonable.
At the wet milling step we encounter a situation similar to preblending;
that is, only two of the 30 study batches are prepared using the no. 5 drilled
screen. The no. 7 is obviously the screen of choice. The purpose of this step is
to produce particles of reasonably uniform size, which in turn will improve
drying. From the records, we also know that the no. 5 screen was used
only with batches that were tray dried. Elapsed drying time and residual
moisture were compared for the two batches from the no. 5 screen process
and the other 13 batches that were tray dried. No important differences were
detected. Still, in light of the limited use of the no. 5 screen, it would not
be inappropriate to recommend this option be eliminated from the processing
instructions.
Mean drying time for the oven tray process is 19.2 hr. All 15 batches
were dried within the specified time of 16 to 20 hr. No seasonal influence was
apparent. The average moisture content of these batches is 1.2%; the standard
deviation is 0.3%. The 15 batches dried using the fluid bed dryer had a residual
moisture of 0.8% (SD = 0.1%). Drying time is mechanically controlled and not
recorded. The statistics favor the fluid bed process; it is more efficient and
uniform. There is nothing in these data to disqualify the oven tray dryer from
further use, however.
Oscillation of the dried granulation and lubricant was accomplished in
every instance using a no. 10 wire screen. Reference to the no. 12 screen, the
alternative method for pulverizing the batch, must be deleted from the manufac-
turing instructions for the process to be validated retrospectively.
The final mix blending time was reported as either 10 or 15 min. Twenty-
one of the 30 batches were tumbled for 10 min and the remainder were mixed
for 15 min. The mixing time is not mechanically controlled or automatically
recorded; it is left to the operator to interpret elapsed time. Because of the
importance of the step to distribution of the therapeutic agent, a comparison was
made between the distribution of the percentage of relative tablet potency [(tab-
let assay/tablet weight) × 100] for the two mixing times. The frequency distribu-
tions of the two populations are shown in Figure 3.
The two histograms are visually different, with the 15-min process exhibit-
ing more dispersion. Despite this difference both populations are tightly grouped,
which is a reflection of the uniformity of the blend.
42 Trubinski

Figure 3 Histogram of drug A granulation uniformity resulting from different blending

times. Percentage of relative potency = (tablet assay/tablet weight) × 100.
Retrospective Validation 43

The processes may be studied quantitatively by comparing the means and

standard deviations of the two populations. The effect of final blend time on
lubricant distribution was examined by comparing disintegration time statistics
for the grouped data. None was noted.
The moisture content of the 15 tray-dried batches following final mix
remained essentially unchanged from the drying step. The batches from the fluid
bed process gained moisture. This is probably attributable to handling very dry
material in a relatively humid environment. Both groups are still below the
target for this step of 1.5 %, however.
Table 2 gives a comparison of the moisture contents following the drying
and tumbling steps. The sizable increase in mean moisture content of the fluid
bed-dried batches deserves further study. To determine whether or not all
batches were uniformly affected, the mean moisture content was plotted in the
order in which the batches were produced. Whereas the plot for the tray-dried
batches is unremarkable, the fluid bed process chart (Fig. 4) depicts an unnatural
pattern. Further investigation discloses that heating, ventilation, and air condi-
tion (HVAC) problems were experienced by the area in which a number of
these batches were blended.
During compression, 1000 tablets were randomly selected for use by qual-
ity control. Inspection of the batch records revealed that all 30 batches were
compressed on the same model press operating at approximately the same speed.
All presses were fed by overhead delivery systems of the same design, thus
tableting equipment will not be a source of variability from batch to batch.
The test for disintegration is performed as described in the USP, and the
results are rounded to the nearest half-min. Disintegration time varied over a
narrow range for all batches studied. The 15-batch average for the tray dryer
process (2.7 min) is well below the specification (10 min) for this test. Hardness
of tablets from the tray dryer process averaged 15 Strong–Cobb units (SCU).
All batches exceeded the minimum specification (9 SCU); there is no upper

Table 2 Drug A: Comparison of Oven Tray Dryer and Fluid

Bed Dryer Processes

Oven tray Fluid bed

Test process (x̄) process (x̄)

Moisture dried granulation (%) 1.20 0.80

Moisture final mix (%) 1.10 1.30
Moisture bulk tablet (%) 1.26 1.50
Hardness, Strong–Cobb units (SCU) 15.00 16.70
Disintegration (min) 2.70 3.00
44 Trubinski

Figure 4 x̄-control chart for drug A percentage moisture at final blend step (fluid bed
process).

limit. Hardness and disintegration time are not well correlated, probably due to
rounding of test results and the need to compare averages.
On average, tablets from the fluid bed process were slightly harder. Also,
the individual batches had a greater range of hardness than batches from the
alternative drying process. Disintegration time for the fluid bed process averaged
3.0 min. Individual batches ranged from 2.0 to 4.5 min. As with the tray process,
no correlation was found between hardness and disintegration time. In summary,
tablets from the fluid bed dryer process were somewhat harder and took slightly
longer to disintegrate. (See Table 2.) These differences are considered insignifi-
cant, however. If any recommendations were made, it would be to lower the
disintegration time specification or establish an internal action limit closer to
the historical upper range of the process.
Control charts were plotted for hardness and average tablet weight (ATW)
to evaluate process performance over time. Separate charts were prepared for
the tray dryer and fluid bed processes. Hardness values are an average of 10
individual measurements. The ATW subgroups are the result of weighing 20
tablets individually. The control charts were inspected for trends and evidence
of instability using well-established methods [9]. Only the control chart for hard-
ness of tablets from the fluid bed process responded to one of the tests for
pattern instability (Fig. 5); that is, two of three consecutive points exceeded the
2-sigma limit. From the chart it is obvious the general trend toward greater
tablet hardness (from 11 to 25 SCU) is the underlying cause of the instability.
The trend to greater hardness was subsequently arrested and may have to do
Retrospective Validation 45

Figure 5 x̄-control chart for drug A tablet hardness (fluid bed process).

with attempts to regulate another tablet variable—thickness, for example—

although the records are vague in this regard.
Water content of the bulk tablets irrespective of the drying process was
higher than at the final mix stage (Table 2). This is probably due to the compres-
sion room environment and the low initial moisture of the powder. Still, the
specification limit of 2% is easily met.
The FDA has recently issued draft guidelines that recommend blend uni-
formity analysis for all products for which USP requires content uniformity
analysis [10]. The USP requires this test when the product contains less than 50
mg of the active ingredient per dosage form or when the active ingredient is
less than 50% of the dosage form by weight. The concern FDA has is that if
blend uniformity is not achieved with mixing of the final granulation, then some
dosage units are likely not to be uniform [11]. Blend uniformity is not routinely
determined for drug A, nor is there a requirement because the dosage form is
over 50% active ingredient. In the absence of historical information about uni-
formity of the blend, the relationship between tablet weight and potency should
be carefully examined.
Tablet weight should bear a direct relationship to milligrams of active
ingredient available where the final blend is homogeneous. This conclusion as-
sumes that demixing does not occur as the compound is transferred to intermedi-
ate storage containers or to a tablet press hopper [12]. To measure the likelihood
that controlling tablet weight assures dosage uniformity, 50 tablet assays se-
lected at random (from 300 tablet assays) were compared to tablet weight using
regression analysis. Because the same model tablet press and blender were em-
ployed for every batch, assay results from all 30 batches were pooled. The mean
purity of the 25 receipts of active ingredients used to manufacture the 30 batches
in the validation study was 99.7%, or 0.3% below target. Individual lots ranged
46 Trubinski

from 98.8–102%. Because of these lot-to-lot differences, active ingredient raw

material potency was also included in the regression analysis.
The general model from the regression analysis is [13]
y = bo + b1X1 + b2Y2
where
y = tablet potency
bo = constant
X1 = raw material purity
X2 = tablet weight
Tablet potency was found to be related to raw material purity and tablet
weight as follows:
y = −414.6 + 6.605OX1 + 0.4303X2
We would expect the regression plane to have a significant positive slope;
that is, as purity of the active ingredient and tablet weight increase, so will
tablet potency, and this was found to be the case. Both slopes are statistically
significantly different from 0 at α = 0.025. When the above equation is used to
predict tablet potency given the ideal tablet weight (600 mg) for the product
and mean raw material purity of 99.7%, the resulting value is only 2.1 mg
different from the theoretical value of 500 mg.
In conclusion, drug A production was shown to be within established
specifications, and there is no reason to believe this will not be the case for
future production as long as all practices are continued in their present form.
Furthermore, there is no significant difference between batches produced by the
tray dryer process and the fluid bed process. A validation report should memori-
alize these findings. The report should also recommend eliminating the option
to use a no. 5 screen for the wet milling step and a no. 12 screen to pulverize
the dried granulation. There is no experience or only limited experience with
this equipment that supports its continued availability. In the same vein, the
final blend time should be standardized at 10 min and automatically controlled
by means of a timer.

B. Coated Tablet (Drug B)

Let’s now turn our attention to a different dosage form, applying some of the
strategies developed during the examination of drug A. Again we want to iden-
tify the process steps that are responsible for distributing the active ingredient
as well as the tests that measure the effectiveness of those actions. Drug B is a
sugar-coated tablet prepared in the traditional manner; that is, layers are slowly
built up around a core by applying a coat of shellac and then subcoating, gross-