BBA Clinical 6 (2016) 45–48

Contents lists available at ScienceDirect

BBA Clinical

journal homepage: www.elsevier.com/locate/bbaclin

Relationship between level of HbA1C and breast cancer

Fariba Jousheghany a, Joshua Phelps a, Tina Crook a, Reza Hakkak a,b,c,⁎
a
Departments of Dietetics and Nutrition, University of Arkansas for Medical Sciences, 4301 W. Markham St., Little Rock, AR 72205, USA
b
Departments of Pediatrics, University of Arkansas for Medical Sciences, 4301 W. Markham St., Little Rock, AR 72205, USA
c
Arkansas Children's Hospital Research Institute, 13 Children's Way., Little Rock, AR 72202, USA

a r t i c l e i n f o a b s t r a c t

Article history: Background: Diabetes and cancer are public health issues worldwide; studies have shown that diabetes is related
Received 23 March 2016 to increased breast cancer mortality. The purpose of this study was to examine associations between HbA1C and
Accepted 18 April 2016 obesity with tumor stage and mortality among breast cancer patients.
Available online 23 April 2016 Methods: Data for 82 patients with breast cancer (36–89 years of age, diagnosed /treated 1999–2009) were
provided by the University of Arkansas for Medical Sciences (UAMS) Data Trust Warehouse. Survival time was
Keywords:
estimated from start date of service to date of last follow-up or date of death. The Kaplan–Meier method provided
Breast cancer
HbA1C
analysis of survival curves for two groups of HbA1C (HbA1C b 6.5% vs HbA1C ≥ 6.5%) and two groups of BMI
Diabetes (BMI b 30 vs BMI ≥ 30 kg/m2); survival curves were compared using log-rank tests. Associations between
Tumor stage HbA1C and BMI, and between HbA1C and tumor stage were determined by chi-square.
Obesity Results: The relationship between tumor stages and HbA1C was not statistically significant (X2 = 0.093, p = 0.47,
df = 1). The relationship between obesity and HbA1C was statistically significant (X2 = 6.13, p = 0.013, df = 1).
Log-rank tests did not show statistically significant differences between survival curves (HbA1C curves, p = 0.4;
Obesity curves, p = 0.09).
Conclusion: While there was a statistically significant association between HbA1C and obesity, there were no
significant associations found with this analysis. However, there are clinically meaningful relationships based
on observed trends. Future directions for research may involve exploring a larger sample of patients and
the role of therapeutic regimens on blood sugar control and BMI of breast cancer patients and influence on
cancer prognosis.
© 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction will contribute toward improving cancer screening and assessment

practices. The aim of this study was to examine associations between
Breast cancer is the most common malignant tumor among women, HbA1C and obesity with tumor stage and mortality among breast
and of all cancers, it is the second leading cause of mortality in women in cancer patients.
the US. Estimates for 2016 predicted that 246,660 women were likely to
be diagnosed with invasive breast cancer and that 40,450 women were 2. Methods
likely to die from breast cancer [1]. The rate of mortality from breast
cancer has declined due to improvements in early detection and treat- 2.1. Data source
ment [1]. Early detection of the cancer and evaluation of comorbidities
result in prompt and relevant treatment that improves breast cancer Data for this project came from i2b2 (informatics for integrating
survival and ultimately saves lives. biology and the bedside), a large computerized data base at UAMS
Type 2 diabetes may increase the risk of cancer [2–5], is considered a [13].The dataset consisted of de-identified secondary data and was
common comorbidity present in a considerable number of breast cancer exempt from review by the UAMS Institutional Review Board (IRB).
patients and correlates with poor clinical outcomes [6–9]. Furthermore,
diabetes may alter the efficacy of treatment in breast cancer patients 2.2. .Study population
affecting drug response and survival endpoints [8,10–12]. Therefore un-
derstanding the associations between diabetes and cancer development A cohort was formed including subjects diagnosed with breast dur-
ing the years of 2001–2009. African-American (AA) and Caucasian
⁎ Corresponding author at: Department of Dietetics and Nutrition, 4301 W. Markham
women who were ≥ 18 years of age at diagnosis, and who had at
St., Mail Slot 627, Little Rock, AR 72205, USA. least one HbA1C on file were included. Subjects with documented
E-mail address: RHakkak@uams.edu (R. Hakkak). pathological tumor stage, BMI and dates of first and last clinic visits

http://dx.doi.org/10.1016/j.bbacli.2016.04.005
2214-6474/© 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
46 F. Jousheghany et al. / BBA Clinical 6 (2016) 45–48

were included in the cohort. Subjects were registered at the oncology HbA1C category (b 6.5%) and 36 (45.1%) were grouped into the high
clinic. The first visit was defined as the date the patient started HbA1C category (≥6.5%).
service at the clinic and the last follow-up was defined as the last visit. Thirty-eight subjects were categorized as having low stage tumors
Life /death status was defined according to the Social Security Death (46.3%) and 44 were categorized with high stage tumors (53.7%).
Index updated in 2013. A Chi-square test indicated no significant association between HbA1C
category and tumor stage (x2 = 0.093, p = 0.47, df = 1).
2.3. HbA1C, BMI and tumor stage classification We further examined the association of high HbA1C on tumor
survival. The Kaplan–Meier method provided descriptive analysis of
HbA1C was classified in two groups (normal versus high) with a survival curves for the two HbA1C groups (Fig. 1) and for the two BMI
cut-point value of 6.5% [14,15]. BMI was classified into 2 groups; non- groups (Fig. 2). Survival curves were compared using log-rank tests,
obese was defined as BMI b 30 kg/m2 and obese was defined as which determined there were no statistically significant differences
BMI ≥ 30 kg/m2. Tumor stage was classified as low or high based on between survival curves (HbA1C curves, p = 0.4; Obesity curves,
the TNM classification system (low stages 0–I; high stages II–III) [16]. p = 0.09).
Result of chi-square analysis (Table 2) testing the association be-
tween obesity and HbA1C was statistically significant (X2 = 6.126,
2.4. Data analysis and statistics p = 0.013, df = 1). The percentage of obese patients is clearly larger
in the high HbA1C group compared to normal HbA1C group; the per-
Descriptive statistics including frequencies were used to describe centage of non-obese patients is lower in high HbA1Cgroup compared
study participant characteristics including age, BMI, HbA1C levels, and to normal HbA1C group (Fig. 3).
tumor stage.
The Kaplan–Meier method provided descriptive analysis of survival 4. Discussion
curves for the two HbA1C groups and for two BMI groups (BMI b 30
versus BMI ≥ 30 kg/m2); survival curves were compared using log- Diabetes and cancer are two major public health issues worldwide
rank tests. Associations between HbA1C and other variables including and many studies have shown that diabetes is related to increased
BMI and tumor stage were determined by chi-square test. All statistical cancer mortality [8,17,18]. According to a review completed by Long
analyses were performed using Excel (Microsoft Corporation, Seattle, and colleagues [19], many studies have shown dysregulation in cellular
WA), Prism 5 (GraphPad Software, Inc., LaJolla, CA), and IBM SPSS metabolism may associate with resistance to cancer treatment drugs.
Statistics Version 22 (IBM Corp., Armonk, NY) software. Various factors, including insulin, insulin-like growth factor (IGF), and
inflammatory cytokines, may link diabetes to breast cancer mortality.
3. Results In the study reported here, we examined the association of HbA1C as
an indicator of blood glucose control with tumor stage and patient
The characterization of the studied population is shown in Table 1. mortality among 82 patients diagnosed with breast cancer. We used
The age of the study population ranged from 36 to 89 years (mean = tumor stage as an indicator of aggressiveness and did not find a statisti-
60). Sixty-two subjects were ≥ 50 years of age (categorized as cally significant association between HbA1C levels and tumor stage.
post-menopausal) and 20 were b 50 years of age (categorized as These data suggest that blood glucose control as measured by HbA1C
pre-menopausal). The study population consisted of 56% Caucasian is not associated with cancer stage in this population. Our results
and 41.5% AA women. contradict other studies showing diabetes correlates with late stage
BMI of the study population ranged from 18.73 kg/m2 to 53.81 kg/m2 diagnosis [9,20]. Peairs et al. [9] performed a meta-analysis examining
(mean = 31.72 ± 7.5 kg/m2). Forty-two subjects (51.2%) were the effect of pre-existing diabetes on outcomes of breast cancer.
classified as obese (BMI ≥ 30 kg/m2), and 40 (48.8%) were non-obese Among 4 datasets included in their analysis, 3 showed an association
(BMI b 30 kg/m2). The levels of HbA1C varied from 4.8% to 17% with a between diabetes and diagnosis at late stage. One study did not find
mean of 6.9%. Forty-six subjects (54.9%) were grouped into the normal an association between diabetes and breast cancer stage. Peairs et al.
[9] suggest that a lack of association between diabetes and breast cancer
stage could be due to large number of samples with missing stage in
Table 1 the study. However, a study investigating associations between type
Characterization of the studied population. 2 diabetes and colon cancer found that, despite a negative association
Number (%) Mean (±SD) between diabetes and survival rate, there was not an association
Age
between diabetes and tumor stage [21]. Together, the data suggest
b 50 20 (24) 60 that involvement of diabetes with mortality in breast cancer patients
≥ 50 62 (76) may be due to non-breast cancer specific causes.
Total 82 (100) According to results reported here, there was not a statistically sig-
Race
nificant difference between survival curves for the different HbA1C
White 46 (56)
African American 36 (44)
BMI
b 30 40 (49) 25.6 (±2.72)
≥ 30 42 (51) 37.3 (±5.7)
Total 82 (100) 31.7 (±7.5)
Hb1Ac
b 6.5 46 (55) 5.7 (±0.5)
≥ 6.5 36 (45) 8.3 (±2.6)
Total 82 (100) 6.9 (±2.2)
Tumor stage
Stage 0 13 (16)
Stage I 28 (34)
Stage II 27 (33)
Stage III 14 (17)
Low stage (Stages 0 & I) 38 (46.3)
Fig. 1. Survival analysis was performed using Kaplan–Meier method. Log rank test was
High stage (Stages II & III) 44 (53.7)
used to statistically compare the curves and p value is shown.
 F. Jousheghany et al. / BBA Clinical 6 (2016) 45–48 47

Fig. 2. Survival analysis was performed using Kaplan–Meier method. Log rank test was
Fig. 3. Visual representation chi-square analysis demonstrating the distribution of obese
used to statistically compare the curves and p value is shown.
and non-obese patients across high and normal HbA1C groups.

groups. Our results are in contrast with Erickson et al., who reported
elevated HbA1C levels in plasma was independently associated Funding
with higher mortality in breast cancer survivors [20]. Lack of statistical
significance in the current study is most likely due to small sample This research was supported by Department of Dietetics and
size that results in a low-power study. Despite our small sample size, Nutrition at University of Arkansas for Medical Sciences.
we observed a potentially clinically relevant decrease in survival of
patients with HbA1C ≥ 6.5%. However, such a decrease is not accompa- Transparency document
nied with high tumor stage, suggesting that deaths could be unrelated
to cancer. The Transparency document associated with this article can be
As expected, we observed a positive and statistically significant found, in online version.
association between obesity and HbA1C, which may suggest the
HbA1C groups (≥ 6.5% vs b6.5%) actually represent individuals with Acknowledgements
differing mortality risk, and is worth additional investigation involving
a larger sample size. Obesity is a major factor associated with type 2 We thank the UAMS Data Warehouse staff for enabling access to
diabetes [22] that may also affect breast cancer survival [23]. Result of data used for this study.
the current study leans toward poorer survival in the obese group,
which is in harmony with published data [23]. References

[1] American Cancer Society, Breast Cancer Facts & Figs. 2015–2016, American cancer
society, Inc., Atlanta, 2016.
5. Study limitations [2] H.L. Xu, H. Fang, W.H. Xu, G.Y. Qin, Y.J. Yan, B.D. Yao, et al., Cancer incidence in
patients with type 2 diabetes mellitus: a population-based cohort study in
Limitations of the current study included a small sample size, and Shanghai, BMC Cancer 15 (2015) 852.
[3] L. Vona-Davis, D.P. Rose, Type 2 diabetes and obesity metabolic interactions:
unavailability of prognostic factors and cancer-specific survival data. common factors for breast cancer risk and novel approaches to prevention and
therapy, Curr. Diabetes Rev. 8 (2) (2012) 116–130.
[4] S.H. Saydah, C.M. Loria, M.S. Eberhardt, F.L. Brancati, Abnormal glucose tolerance
and the risk of cancer death in the United States, Am. J. Epidemiol. 157 (12)
6. Conclusions (2003) 1092–1100.
[5] P. Ferroni, S. Riondino, O. Buonomo, R. Palmirotta, F. Guadagni, M. Roselli, Type 2
Breast cancer has a low mortality rate and survivors are living diabetes and breast cancer: the interplay between impaired glucose metabolism
and oxidant stress, Oxidative Med. Cell. Longev. 2015 (2015) 183928.
longer; therefore, there is need for studies with a longer follow-up
[6] C.C. Cowie, K.F. Rust, E.S. Ford, M.S. Eberhardt, D.D. Byrd-Holt, C. Li, et al., Full
period. Future directions for research may involve exploring a larger accounting of diabetes and pre-diabetes in the U.S. population in 1988–1994 and
sample of patients and the role of therapeutic regimens on blood 2005–2006, Diabetes Care 32 (2) (2009) 287–294.
[7] M. Cazzaniga, B. Bonanni, A. Guerrieri-Gonzaga, A. Decensi, Is it time to test
sugar control and BMI of breast cancer patients and influence on
metformin in breast cancer clinical trials? Cancer Epidemiol. Biomark. Prev. 18 (3)
cancer prognosis. (2009) 701–705.
[8] L.L. Lipscombe, P.J. Goodwin, B. Zinman, J.R. McLaughlin, J.E. Hux, The impact of
diabetes on survival following breast cancer, Breast Cancer Res. Treat. 109 (2)
(2008) 389–395.
Authors' contributions [9] K.S. Peairs, B.B. Barone, C.F. Snyder, H.C. Yeh, K.B. Stein, R.L. Derr, et al., Diabetes
mellitus and breast cancer outcomes: a systematic review and meta-analysis, J.
Fariba Jousheghany developed the hypothesis and designed the Clin. Oncol. 29 (1) (2011) 40–46.
[10] S.S. Coughlin, E.E. Calle, L.R. Teras, J. Petrelli, M.J. Thun, Diabetes mellitus as a predic-
study, organized, analyzed and interpreted data and wrote the manu- tor of cancer mortality in a large cohort of US adults, Am. J. Epidemiol. 159 (12)
script. Josh Phelps contributed to development of the hypothesis and (2004) 1160–1167.
study design, data organization and analysis, and manuscript develop- [11] C.J. Currie, M. Peyrot, C.L. Morgan, C.D. Poole, S. Jenkins-Jones, R.R. Rubin, et al., The
impact of treatment noncompliance on mortality in people with type 2 diabetes,
ment. Tina Crook contributed to development of the hypothesis and Diabetes Care 35 (6) (2012) 1279–1284.
study design, and manuscript development. Reza Hakkak contributed [12] K. Ranc, M.E. Jorgensen, S. Friis, B. Carstensen, Mortality after cancer among patients
to development of the hypothesis and study design, and manuscript with diabetes mellitus: effect of diabetes duration and treatment, Diabetologia 57
(5) (2014) 927–934.
development.
[13] http://tri.uams.edu/research-resources-services-directory/biomedical-informatics-
data-services/electronic-health-care-data-for-research/.
[14] M.J. Gillett, International expert committee report on the role of the A1c assay in the
Table 2 diagnosis of diabetes: diabetes care 2009; 32(7): 1327–334, Clin. Biochem. Rev. 30
The results of Chi-square testing association of HbA1C and BMI. (4) (2009) 197–200.
[15] R.M. Cohen, S. Haggerty, W.H. Herman, HbA1c for the diagnosis of diabetes and
Number of patients Pearson chi-square p value prediabetes: is it time for a mid-course correction? J. Clin. Endocrinol. Metab. 95
82 6.129 0.013⁎ (12) (2010) 5203–5206.
[16] http://www.breastcancer.org.
HbA1c with cut-point 6.5%, BMI with cut-point 30 kg/m2. [17] S.C. Larsson, C.S. Mantzoros, A. Wolk, Diabetes mellitus and risk of breast cancer:
⁎ The association of HbA1c and BMI is statistically significant (p b 0.05). a meta-analysis, Int. J. Cancer 121 (4) (2007) 856–862.
48 F. Jousheghany et al. / BBA Clinical 6 (2016) 45–48

[18] P. Minicozzi, F. Berrino, F. Sebastiani, F. Falcini, R. Vattiato, F. Cioccoloni, et al., High [21] J.A. Meyerhardt, P.J. Catalano, D.G. Haller, R.J. Mayer, J.S. Macdonald, A.B. Benson 3rd,
fasting blood glucose and obesity significantly and independently increase risk of et al., Impact of diabetes mellitus on outcomes in patients with colon cancer, J. Clin.
breast cancer death in hormone receptor-positive disease, Eur. J. Cancer 49 (18) Oncol. 21 (3) (2003) 433–440.
(2013) 3881–3888. [22] S.E. Kahn, R.L. Hull, K.M. Utzschneider, Mechanisms linking obesity to insulin
[19] J.P. Long, X.N. Li, F. Zhang, Targeting metabolism in breast cancer: how far we can resistance and type 2 diabetes, Nature 444 (7121) (2006) 840–846.
go? World J. Clin. Oncol. 7 (1) (2016) 122–130. [23] J. Ligibel, Obesity and breast cancer, Oncology (Williston Park) 25 (11) (2011)
[20] K. Erickson, R.E. Patterson, S.W. Flatt, L. Natarajan, B.A. Parker, D.D. Heath, et al., 994–1000.
Clinically defined type 2 diabetes mellitus and prognosis in early-stage breast
cancer, J. Clin. Oncol. 29 (1) (2011) 54–60.