WO2002046751A2

WO2002046751A2 - Methods of predicting chemotherapy response

Info

Publication number: WO2002046751A2
Application number: PCT/US2001/051196
Authority: WO
Inventors: Vladimir D. Kravtsov; Mark J. Koury
Original assignee: Vanderbilt University
Priority date: 2000-11-13
Filing date: 2001-11-13
Publication date: 2002-06-13
Also published as: WO2002046750A3; WO2002046751A9; WO2002046751A3; WO2002046750A2; AU2002241791A1; AU2002239771A1

Abstract

A method of predicting the effect of chemotherapy on a patient is disclosed. In this method, the patient is suffering form a disease or condition that can be treated or prevented by the administration to the patient of a drug which acts by inducing apoptosis. A method of predicting the effect of chemotherapy on the survival rate of a patient population suffering from a disease or condition that can be treated by chemotherapy is also disclosed.

Description

METHODS OF PREDICTING CHEMOTHERAPY RESPONSE

Priority is claimed to U.S. provisional application no. 60/247,053, filed November 13, 2000, the entirety of which is incoxporated herein by reference.

1. FIELD OF THE INVENTION

The invention relates to methods of predicting the effect of a drug on a patient .. suffering from a disease, and to methods of determining and/or optimizing chemotherapy regimens.

2. BACKGROUND

Despite the use of aggressive chemotherapeutic protocols, about 80 percent of patients suffering from acute non-lymphocytic leukemia ("ANLL") have resistant disease or relapse after achieving a remission. Bennett, J. M., et al, "Long-term survival in acute myeloid leukemia: the Eastern Cooperative Oncology Group experience," Cancer 1997; 80 Suppl 11:2205-9. Specific clinical characteristics combined with laboratory analyses of the leukemic cells are routinely used to predict clinical response to chemotherapy in ANLL. Among the clinical characteristics that have demonstrated prognostic significance are age, WBC at the time of diagnosis, and disease status (de novo leukemia vs. relapsed or myelodysplasia-associated leukemia). Buchner, T. and Heinecke, A., "The role of prognostic factors in acute myeloid leukemia," Leukemia 10 Suppl 1.-S28-29 (1996); Estey, E. H., "Treatment of acute meylogenous leukemia and myelodysplastic syndromes," Semin. Hematol. 32:132-51 (1995). Laboratory analyses for cytogenetic abnormalities or expression of the P-glycoprotein product of the multiple drug resistance- 1 gene are also used as prognostic indicators of clinical outcome following chemotherapy. Slovak, M. L., et al, "Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study," Blood 96:4075-83 (2000); Leith, C. P., et al., "Frequency and clinical significance of the expression of the multidrug resistance proteins MDR1 /P-glycoprotein, MRPl, and LRP in acute meyloie leukemia: A Southwest Oncology Group Study," Blood 94:1086-99 (1999); Ross, D. D., "Novel mechanisms of drug resistance in leukemia," Leukemia 14:467-73 (2000); Ollivier, L., et al, "The immunophenotype of 177 adults with acute myeloid leukemia: proposal of a prognostic score," Blood 96:870-7 (2000). Several types of in vitro assays that measure leukemia cell survival following exposure to cytotoxic agents have been used in attempts to predict clinical outcomes in patients treated for ANLL. See, e.g., Lihou, M. G., et al., "Quantitation of chemosensitivity in acute myelocytic leukaemia," Br. J. Cancer 48:559-67 (1983); Nara, N., et al, "Relationship between the in vitro sensitivity to cytosine arabinoside of blast progenitors and the outcome of treatment in acute myeloblastic leukaemia patients," Br. J. Haematology 70:187-91 (1988); Park, C. H., et al, "Clinical correlations of leukemic clonogenic cell chemosensitivity assessed by in vitro continuous exposure to drugs," Cancer Res. 43:2346-9 (1983); Dow, L. W., et al, "Correlation of drug sensitivity in vitro with clinical responses in childhood acute myeloid leukemia," Blood 68:400-5 (1986); Sargent, J. M., and Taylor, C. G., "Appraisal of the ^" MTT assay as a rapid test of chemosensitivity in acute myeloid leukaemia," Br. J. Cancer 60:206-10 (1988); Santini, V., et al, "In vitro chemosensitivity testing of leukemic cells: prediction of response to chemotherapy in patients with acute non-lympocytic leukemia," Hematological Oncology 7:287-3 (1989); Barzi, A., et al, "A 4-day chemosensitivity assay in vitro reliably predicts clinical response of patients with acute leukemia," Haematologica 74:449-54 (1989); Hwang, W-S., etal, "Prediction of chemotherapy response inhuman leukemia using in vitro chemosensitivity test," Leuk. Res. 17:685-8 (1993); Larsson, R., et al. , "In vitro testing of chemotherapeutic drug combinations in acute myelocytic leukaemia using the fluorometric microculture cytotoxicity assay (FMCA)," Brit. J. Cancer 67:969-74 (1993); Klumper, E., et al, "In vitro chemosensitivity assessed with the MTT assay in childhood acute non-lymphoblastic leukemia," Leukemia 9:1864-9 (1995); Norgaard, J. M., et al, "Relation of blast cell survival and proliferation to chemotherapy resistance in AML," Brit. J. Haematol. 93:888-97 (1996). Despite varied success in predicting remissions, these assays have never been reported to be independent predictors of survival of ANLL patients. Apoptosis is the mechanism by which chemotherapeutic agents exert their antitumor effect. Sachs, L. and Lotem, J., "Control of progrmmed cell death in normal and leukemic cells: new implications for therapy," Blood 82:15-21 (1993); Kerr, J. F. R., "Apoptosis: Its significance in cancer and cancer therapy," Cancer 73:2013 (1994); Hannun, Y. A., "Apoptosis and the dilemma of cancer chemotherapy," Blood 89:1845-53 (1997); Kamesaki, H., "Mechanisms involved in chemotherapy-induced apoptosis and their implications in cancer chemotherapy," Int. J. Hematol 68:29-43 (1998). Recent advances in the measurement of apoptosis have allowed the automated study of this phenomenon. See, e.g., U.S. Patent Nos. 6,077,684 and 6,258,553. However, the extent of drug-induced apoptosis in a population of diseased (e.g., cancerous) cells has never been shown to be an independent predictor of chemotherapy outcome.

A method is needed for predicting the outcome of chemotherapy. Methods are also needed for optimizing the drugs and administration regimes used to treat specific patients or populations of patients suffering from diseases such as, but not limited to, cancer. 3. SUMMARY

This invention encompasses a method of predicting the effect of chemotherapy on a patient (e.g., a fish, bird, or mammal) suffering from a disease or condition that can be treated or prevented by the administration to the patient of a drug which acts by inducing apoptosis. Also encompassed by the invention is a method of predicting the effect of chemotherapy on the survival rate of a patient population suffering from a disease or condition that can be treated by chemotherapy. The invention fiαrther encompasses a method of optimizing and/or developing a chemotherapy regimen for use in an individual patient or in a patient population suffering from a disease or condition that can be treated by chemotherapy.

The invention further encompasses a method of treating or prevention a disease or condition which comprises removing one or more cells that cause the disease or condition from a patient in need of such treatment or prevention, determining the effect of one or more drugs on the cells in vitro, and administering to the patient a drug that causes apoptosis of the cells in vitro.

3.1. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows the abilities of various drugs to induce apoptosis of tumor cells isolated from a patient suffering from acute myelogeneous leukemia. FIG. 2 shows Kaplan-Meier plots of probability of overall survival of evaluable patients with ANLL who had the MiCK assay performed prior to treatment. Top panel: MiCK assay sensitive group vs. resistant group for cytarabine; Middle panel: MiCK assay sensitive group vs. resistant group for idarubicin. Bottom panel: Three groups based on MiCK assay results for cytarabine and idarubicin: cytarabine sensitive and idarubicin sensitive vs. cytarabine resistant and idarubicin sensitive vs. cytarabine resistant and idarubicin resistant. Data for the one patient who was cytarabine sensitive and idarubicin resistant, and who died at day 325, is not shown. All p- values are by log-rank test.

4. DETAILED DESCRIPTION The invention is based on a discovery that a kinetic assay (referred to herein as

"MiCK") can be used to determine the effects of drugs and/or administration protocols (e.g., timing of administration, amounts of administration) on the apoptosis of particular types of cells. The invention is further based on the discovery that results of this in vitro assay can be directly correlated with the in vivo effect of a particular drug or chemotherapy regime. Indeed, it has been found that results of the MiCK assay can, in many instances, predict the effect of a drug in vivo with a degree of accuracy that far exceeds prior predictive methods. A first embodiment of the invention encompasses a method of predicting the effect of a drug on a patient suffering from a disease that can be treated by killing certain cells within the patient, which comprises: forming first and second cultures of some of the certain cells isolated from the patient; contacting the first culture with the drug; measuring the optical density of the first culture at more than one time point; measuring the optical density of the second culture at more than one time point, wherein the second culture was not contacted with the drug; and determining a net slope, which is the difference between the optical density change over time of the first culture and the optical density change over time of the second culture; wherein a positive net slope indicates that the drug can be effective in the treatment of the disease.

In a preferred method of this embodiment, the cells are cancer cells. In another method, the cells are bacteria or fungi. In still another method, the cells are infected with a virus.

As used herein in connection with a disease or condition, the term "treatment" means the reduction or elimination of one or more symptoms of the disease or condition.

As used herein in connection with the treatment of a disease caused or aggravated by the existence and/or proliferation of certain cells, the term "predicting the effect of a drug" means determining whether the administration of the drug to a patient suffering from the disease will kill within the patient a sufficient number of the certain cells to effect treatment of the disease. Typically, such a sufficient number is a significant number. Examples of significant numbers include, but are not limited to, greater than about 50 percent, greater than about 60 percent, greater than about 70 percent, greater than about 80 percent, greater than about 90 percent, greater than about 95 percent, and greater than about 99 percent. In particular methods of this invention, the net slope is determined by a method which comprises subtracting at each time point the optical density measurement of the second culture from the corresponding optical density measurement of the first culture. In other methods of the invention, the net slope is determined by a method which comprises: calculating the rate at which the optical density of the first culture changes over time to provide a first rate of change; calculating the rate at which the optical density of the second culture changes over time to provide a second rate of change; and subtracting the second rate of change from the first rate of change.

A second embodiment of the invention encompasses a method of determining the probability that a patient suffering from a disease that can be treated with a particular chemotherapy protocol, which comprises: isolating some of the certain cells from the patient; characterizing the cells; and correlating the cells with treatment information obtained for patient populations suffering from the disease; wherein the treatment information is obtained using the methods described herein (e.g., the method of the first embodiment). In a preferred method, the probability is determined with a p-value of less than about 0.2, less than about 0.1, less than about 0.05, less than about 0.01, less than about 0.005, less than about 0.001, less than about 0.0005, or less than about 0.0001. Methods of the invention can be used to predict the effect of chemotherapy on patients and patient populations suffering from diseases and conditions such as, but not limited to: bacterial, viral and fungal infections; and cancer. Examples of cancer include, but are not limited to, primary and metastatic cancer of the head, neck, eye, mouth, throat, esophagus, chest, bone, lung, colon, rectum, stomach, prostate, breast, ovaries, kidney, liver, pancreas, and brain. Specific types of cancers include, but are not limited to: AIDS associated leukemia and adult T-cell leukemia lymphoma; anal carcinoma; astrocytoma; biliary tract cancer; cancer of the bladder, including bladder carcinoma; brain cancer, including glioblastomas and medulloblastomas; breast cancer, including breast carcinoma; cervical cancer; choriocarcinoma; colon cancer including colorectal carcinoma; endometrial cancer; esophageal cancer; Ewing's sarcoma; gastric cancer; gestational trophoblastic carcinoma; glioma; hairy cell leukemia; head and neck carcinoma; hematological neoplasms, including acute and chronic lymphocytic and myelogeneous leukemia; hepatocellular carcinoma; Kaposi's sarcoma; kidney cancer; multiple myeloma; intraepithelial neoplasms, including Bowen's disease and Paget's disease; liver cancer; lung cancer including small cell carcinoma; lymphomas, including Hodgkin's disease, lymphocytic lymphomas, non-Hodgkin's lymphoma, Burkitt's lymphoma, diffuse large cell lymphoma, follicular mixed lymphoma, and lymphoblastic lymphoma; lymphocytic leukemia; neuroblastomas; oral cancer, including squamous cell carcinoma; ovarian cancer, including those arising from epithelial cells, stromal cells, germ cells and mesenchymal cells; pancreatic cancer; prostate cancer; rectal cancer; sarcomas, including soft tissue sarcomas, leiomyosarcoma, rhabdomyosarcoma, liposcarcoma, fibrosarcoma, and osteosarcoma; skin cancer, including melanoma, Kaposi's sarcoma, basal cell cancer and squamous cell cancer; testicular cancer, including testicular carcinoma and germinal tumors (e.g., semicoma, non-seminoma[teratomas, choriocarcinomas]), stromal tumors and germ cell tumors; thyroid cancer, including thyroid adenocarcinoma and medullar carcinoma; and renal cancer including adenocarcinoma and Wilm's tumor. A specific type of cancer is acute non-lymphocytic leukemia.

The kinetic assay used in the methods of the invention is disclosed by U.S. Patent Nos. 6,077,684 and 6,258,553, both of which are incorporated herein by reference. It has been discovered that this assay (referred to herein as the "MiCK assay") can be used to predict the effects of drugs cells isolated from individual patients and groups of patients. Methods of the invention can be used to deteπnine whether or not chemotherapy is an appropriate method of treatment. If it is, methods of the invention can provide information useful in the design and or optimization of chemotherapy regimens, that are particularly safe and effective for specific individuals or groups of individuals. For example, because methods of the invention can be used to determine the minimum dosage amounts and times suitable for the treatment of a disease, they can be used to miiώnize and/or avoid adverse effects. Examples of adverse effects typical of chemotherapy drugs include, but are not^" limited to, early and late-forming diarrhea, nausea, vomiting, anorexia, constipation, flatulence, leukopenia, anemia, neutropenia, asthenia, abdominal cramping, fever, pain, loss of body weight, dehydration, alopecia, dyspnea, insomnia, and dizziness.

Without being limited by theory, the ability of methods of this invention to accurately predict the effect of chemotherapy is likely due to at least two characteristics of the MiCK assay. First, the assay measures the extent of drug-induced apoptotic cell death by directly detecting the sensitivity of a cell population to a cytotoxic agent. Prior laboratory tests used to study cytotoxicity in vitro in ANLL did not measure apoptosis, but rather they evaluated the fraction of cells surviving exposure to a cytotoxic agent. Therefore, as opposed to the MiCK assay which detects drug sensitivity, these other laboratory tests detected resistance of a cell population to a cytotoxic agent. Second, considering that apoptosis in, for example, leukemic cells from different patients can occur at varying times after drug exposure, the continuous monitoring of cell-drug interactions by the MiCK assay provides for an accurate evaluation of drug-induced apoptosis, whenever it occurs in the culture.

Unlike established clinical predictive factors, which do not influence chemotherapy outcomes, the predictive ability of the MiCK assay is expected to improve clinical responses of individual patients by guiding adjustments in regimens used for the treatment of a wide range of diseases and condition. For example, patients with relapsed ANLL but no appropriate transplantation donors have several possible reinduction regimens. For these patients, the MiCK assay may help in choosing a chemotherapy regimen containing drugs to which the patient's ANLL cells were sensitive in vitro. On the other hand, for patients with relapsed leukemia and a suitable transplantation donor, methods of this invention may help in determining whether further standard chemotherapy is likely to be effective. 5. EXAMPLES

5.1. EXAMPLE 1

The MiCK assay was applied to study responses of tumor cells isolated from 74 AML patients to cytarabine (C), idarubicin (T), mitoxantrone (M), etoposide (E) and daunorubicin (D). Concentrations of tested chemotherapeutic agents ranged from 0.01 to 160 μ . Apoptotic responses were monitored for 24 hours and measured in kinetic units (KU) which are directly related to the percentages of apoptotic cells in the cultures. The distribution of the peak amount of apoptosis was analyzed for each drug and these peak values were separated into two groups of responsive or non-responsive for the induction of apoptosis. Patients' median age was 43 years (range, 3-77); male: female ration was 30:44. Clinical features [cytogenetics (cyto), preceding myelodysplasia (MDS), de novo verus relapse, and prior chemotherapy] as well as the MiCK assay results were correlated with survival and clinical response. The 1 year survival was 35% ± 7% with a median of 257 days. The MiCK assay based on sensitivity to I, C, and M, but not to D or E, was statistically more significant in predicting survival than any of the clinical features.

TABLE 1

Feature N 1 Yr (%) P

Cyto-Good 6 44

Intermediate 27 43 .42

Poor 24 18

Prior MDS 22 18 .69

No MDS 52 34

Relapses 21 14

De Novo 53 38 -07

Prior Chem 1* 10 22 .09*

> 2* 11 14

* Courses of prior chemotherapy compared with de novo (no chemotherapy). TABLE 2

There were 29 complete remission ("CR"), 8 partial remission ("PR"), and 12 no responses ("NR") in 49 patients evaluable for clinical response; 25 patients were not evaluable due to either bone marrow transplant (n=17) or early death/no therapy (n=8). The 1 year survival for CR was 58% vs 10% for PR+NR (p=.0001). There was a significant different in predicting response (CR vs PR+NR) for I (p<.001), C (p=.046), and M (p= .034); neither E (ρ=.l 11) nor D (p=.136) predicted response. Preliminary data indicates the MiCK assay is a better predictor of response and survival in AML than clinical features.

5.2. EXAMPLE 2

The extent of apoptosis induced in leukemic blasts by chemotherapeutic agents was compared with clinical parameters for the ability to predict CR and survival after chemotherapy in ANLL patients. The MiCK assay was employed to measure in vitro apoptotic responses of leukemic blasts to five chemotherapeutic agents.

5.2.1. METHODS Patients. Between 1996 and 2000, bone marrow aspirates or venous blood samples were obtained before initiation of therapy from 108 patients with newly diagnosed or relapsed ANLL at the Vanderbilt and Nashville VA Medical Centers. These samples were obtained with informed consent as approved by the Vanderbilt University Institution Review Board. Of these 108 samples, 104 had sufficient numbers and purity of leukemic cells for testing in the MiCK assay. Of these 104 patients, 80 received a full course of chemotherapy and were evaluable for complete remission (CR; Table 3). The patients' median age was 51.5 years. The chemotherapeutic regimen for each patient was chosen by that patient's attending physician without knowledge of MiCK assay results. Of the 24 patients who were considered to be not evaluable for response, eleven underwent early stem cell transplantation, ten received no therapy, and three died during induction chemotherapy. Among the 35 patients with relapse or an associated MDS, 24 had MDS and eleven were relapsed patients, including five with preceding MDS. CR was defined as less than 5% blasts, and at least 20% cellularity in representative bone marrow on day 28 after induction therapy, and an absolute neutrophil count greater than 1.5 x 10⁹/L and a platelet count of 1.0 x 10ⁿ/L or more for at least four weeks or until consolidation therapy. Cheson, B. D., et al, "Report of the National Cancer Institute-sponsored workshop on definitions of diagnosis and response in acute myeloid leukemia," J. Clin. Oncol. 8:813-9 (1990). For patients with preceding myelodysplasia, CR was defined as a reduction in bone marrow blasts to less than 5%, a return to a marrow with myelodysplasia and transfusion independence. Cheson, B. D., et al, "Report of an international working group to standardize response criteria for myelodysplastic syndromes," Blood 96:3671-4 (2000).

The diagnoses of ANLL were made by the Vanderbilt hematopathology service, using criteria that included: a) microscopic examinations of blood or marrow smears stained with Wright's, Sudan black, periodic acid-Schiff, chloracetate esterase, and α-naphthylacetate esterase stains; b) cytogenetic analyses; c) flow cytometry immunophenotypic analysis with a panel of monoclonal antibodies (Mabs) which included Mabs pairs for dual staining of CD2/CD4, CD13/CD19, CD20/CD10, CD7/HLA-DR, CD33/CD34, and CD14/CD45. Each patient was classified according to the French- American-British (FAB) classification. Bennett, J. M., et al, "Proposed revised criteria for the classification of acute myeloid leukemia. A report of the French- American-British Cooperative Group," Ann. Intern. Med. 103:620-5 (1985). Samples for cytogenetic analysis were obtained simultaneously with samples for the MiCK assay. Cytogenetic profiles were considered to have a good prognosis if they showed inv(16), t(15;17), or t(8;21), an intermediate prognosis if they were normal, and a poor prognosis if they included any other abnormality. MiCK assay for apoptosis. The MiCK assay was performed as described. See, e.g., Kravtsov, V. D., et al, "Use of the microculture kinetic (MiCK) assay of apoptosis to determine chemosensitivities of leukemias," Blood 92:968-80 (1998). Briefly, leukemic blasts were purified from the heparinized bone marrow or blood by Ficoll-Hipaque gradient centrifugation and depleted of T lymphocytes and monocytes using magnetic beads conjugated with monoclonal antibodies to CD2 and CD14, respectively (Dynal, Inc., Lake Success, NY). All cell populations studied with the MiCK assay contained more than 95% blast cells and more than 90% viable cells. The purified blasts were suspended in RPMI- 1640 medium with 10% fetal bovine serum and plated in 240 μL aliquots in 96-well microtiter plates (Corning-Costar, Cambridge, MA). Concentrations of seeded cells varied from 0.8xl0⁶ to 1.2xl0⁶ cells/mL depending upon the cellular size. Id. The five chemotherapeutic agents and their dose ranges tested were: cytarabine (0.1 - 160 μM), idarubicin (0.1 - 20.0 μM), daunorubicin (0.1 - 20.0 μM), mitoxantrone (0.1 - 5.0 μM), and etoposide (0.1 - 80 μM). Appropriate dilutions of the agents were added to wells in 10 μL aliquots. After 30 minutes incubation at 37°C in an humidified atmosphere of 5% CO₂ in air, 50 μL of sterile mineral oil was layered on each microculture. The microtiter plate was placed in the incubated chamber of a spectrophotometer (SPECTRAmax 340, Molecular Devices Corp., Sunnyvale, CA), incubated at 37°C and the OD at 600 nm was read every 5 minutes for a period of 24 hours. Data processing and quantitation of drug-induced apoptosis in leukemic blasts were performed as described. Id. In brief, the OD readings were plotted against time providing a kinetic representation of the drug responses. The extent of apoptosis was determined from the slope of the steepest increase of OD over time and expressed as kinetic units ("KU"). Id.

Statistical Analyses. Apoptotic responses in KU were computed for each dose of each drug. The maximal KU value induced by a drug was considered the "best apoptotic response" to the drug. To discriminate between the drug sensitive and drug resistant leukemia cell populations, the best responses to each agent were compared in a receiver operator characteristic ("ROC") analysis as well as logistic regression analysis with attainment of CR after induction therapy (CR vs No CR) in the 80 evaluable patients. Analyses were made at "cut-off points" of 1, 2, 3, 4, or 5 KU for the best responses. Based on the best combination of positive predictive value ("PPV"), negative predictive value ("NPV") and p- value, thresholds demarcating sensitivity and resistance were established for each agent. The 3 KU cut-off point was found optimal for demarcating sensitivity and resistance with idarubicin and mitoxantrone, while the 2 KU cut-off point was optimal for cytarabine, daunorubicin, and etoposide (Table 4). For each chemotherapeutic agent, a patient's leukemic blasts were considered drug-resistant, if the best apoptotic response was less than that drug's cut-off point, or sensitive, if the best apoptotic response was equal to or greater than that drug' s cut-off point.

The Fisher's exact test, chi-square test and analysis of variance method were employed to test the correlation between the clinical response results and other clinical variables. For lifetime data analyses, the possible risk factors were compared for survival with Kaplan-Meier estimates and log-rank tests. The Cox proportional hazards model was used for adjusting the tests of significance and estimating the adjusted odds ratios.

Significance was based on two-sided tests. The statistical analyses was completed using SAS 8.2 statistical program or S-Plus 2000 (SAS Institute, Inc., Cary, NC). TABLE 3: Characteristics of ANLL Patients Tested

* Evaluable pts are 80 patients eligible for CR evaluation. **p-value is by chi-square analysis or Fisher's exact test.

TABLE 4. Cut-off Points, Positive Predictive Value (PPV) And Negative Predictive Value (NPV) of the MiCK Assay for 80 Evaluable Patients

p-value is by chi-square analysis.

TABLE 5. Comparisons between the in vitro drug responses and attainment of CR.

* p-value is by chi-square analysis or Fisher's exact test. TABLE 6. The relation between in vitro drug responses and probability of survival after therapy.

*p-value is determined by log-rank test.

** range of confidence interval not yet reached.

TABLE 7 Multivariate Analysis

*p-value is by Cox proportion model. 5.2.2. RESULTS Complete remission ("CR")was achieved by 43 (54%) of evaluable patients. The median survival time for all patients was 9.3 months with an estimated two-year survival of 30%. The median event-free survival ("EFS") was 7.3 months with an estimated two-year EFS of 22%). The clinical parameters that were statistically significant predictors of CR were cytogenetics and disease status at the time of entry into the study (Table 3). By univariate analysis, age (p=0.0007), disease status (p=0.005), and cytogenetics (p=0.029) were statistically significant clinical predictors of survival.

Of the 80 patients evaluable for CR and survival, responses in the MiCK assay were completed in 79 patients for cytarabine, in 77 patients for idarubicin, daunorubicin and mitoxantrone and in 75 patients for etoposide. For each chemotherapeutic agent, responses in the MiCK assay were significantly associated with attainment of CR (Table 5), however, only the responses to idarubicin in the MiCK assay were significant for the prediction of survival following chemotherapy (Table 6). Since the combination of cytarabine and idarubicin was the most commonly used induction protocol (52 patients, 65%), Kaplan- Meier plots of survival are shown for patients with leukemias that were sensitive or resistant in the MiCK assay to cytarabine and or idarubicin (FIG. 2). In the top panel of FIG. 2, in vitro sensitivity to cytarabine was associated with increased survival at a level of confidence that approached statistical significance, with most of the effect occurring in the first two years after the treatment. Patients whose leukemic cells were sensitive to idarubicin in vitro had a markedly increased rate of survival over the entire period of observation (FIG. 2, middle panel). Survival of patients based on their in vitro responses to both drugs depended on their sensitivity or resistance to idarubicin, irrespective of the response to cytarabine (FIG. 2, bottom panel). Multivariate analysis was performed to compare the prognostic power of the drug responses in the MiCK assay with those clinical characteristics known to have prognostic significance for survival (Table 7). The results show that idarubicin sensitivity, age less than 40 years and WBC less than 10⁹ per liter predicted survival. The independent prognostic significance of the in vitro sensitivity to idarubicin was superior to all analyzed factors (Table 7).

These findings indicate that in vitro responses to each of the five chemotherapeutic agents were significantly associated with attainment of CR (Table 6). Multivariate analysis performed with four clinical variables (age, disease status, cytogenetics and WBC) and in vitro sensitivities to cytarabine or idarubicin confirmed that in vitro sensitivity to idarubicin was the best independent predictor of survival in these patients (Table 7). It is possible that the prediction of survival by in vitro responses to etoposide, daunorubicin and mitoxantrone did not reach statistical significance due to low number of patients who received these drugs in the course of their therapy. The predictive power of the response to idarubicin in the MiCK assay is consistent with empirically derived evidence from clinical studies suggesting that idarubicin is a more effective agent than daunorubicin in the therapy of ANLL. See Berman, E., et al, "Results of a randomized trial comparing idarubicin and cytosine arabinoside with daunorubicin and cytosine arabinoside in adult patients with newly diagnosed acute myelogenous leukemia," Blood 77:1666-74 (1991); Wiernik, P. H., et al, "Cytarabine plus idarubicin or daunorubicin as induction and consolidation therapy for previously untreated adult patients with acute myeloid leukemia," Blood 79:313-9 (1992); AML Collaborative Group, "A systematic collaborative overview of randomized trials comparing idarubicin with daunorubicin (or other anthracyclines) as induction therapy for acute myeloid leukaemia," Br. J. Haematology 103:100-9 (1998).

In Table 5, varying numbers of patients with in vitro resistance to drugs achieved CR. The highest proportions of patients who achieved CR despite in vitro resistance to a chemotherapeutic agent were those with cytarabine or etoposide resistance (Table 5). Considering that all 80 patients were induced with a combination of two or more agents, those demonstrating in vitro resistance to cytarabine or etoposide were analyzed for their sensitivity to other drugs. Of 28 cytarabine-resistant patients who achieved CR, 26 [93%; 95% CI — (76.50%) , 99.12%)] were sensitive to an anthracycline used as the second component of the combination therapy. Similarly, of 24 etoposide-resistant patients who achieved CR, 23 [96%; 95% CI = (78.88%, 99.89%)] were sensitive to another chemotherapeutic agent used in the induction protocols. Therefore, for both cytarabine- and etoposide-resistant patients who achieved CR, there was a significant association between their sensitivity to another drug and their CR rate.

It should be understood that variations and modifications within the spirit and scope of the invention may occur to those skilled in the art to which the invention pertains. Accordingly, all expedient modifications readily attainable by one versed in the art from the disclosure set forth herein that are within the scope and spirit of the present invention are to be included as further embodiments of the present invention. The scope of the present invention accordingly is to be defined as set forth in the appended claims.

Claims

CLAIMSWhat is claimed is:

1. A method of predicting the effect of a drug on a patient suffering from a disease that can be treated by killing certain cells within the patient, which comprises: foπning first and second cultures of some of the certain cells isolated from the patient; contacting the first culture with the drug; measuring the optical density of the first culture at more than one time point; measuring the optical density of the second culture at more than one time point, wherein the second culture was not contacted with the drug; and determining a net slope, which is the difference between the optical density change over time of the first culture and the optical density change over time of the second culture; wherein a positive net slope indicates that the drug can be effective in the treatment of the disease.

2. The method of claim 1 wherein the cells are cancer cells.

3. The method of claim 1 wherein the cells are bacteria or fungi.

4. The method of claim 1 wherein the cells are infected with a virus.

5. The method of claim 1 wherein the prediction is made with a p-value of less than about 0.2.

6. The method of claim 5 wherein the p-value is less than about 0.1.

7. The method of claim 6 wherein the p-value is less than about 0.05.