Uchospitals.edu

Relationship of Type II Diabetes and
Metformin Use to Ovarian Cancer
Progression, Survival, and Chemosensitivity
Iris L. Romero, MD, MS, Anna McCormick, MD, DO, Kelsey A. McEwen, BS, SeoYoung Park, PhD,Theodore Karrison, PhD, S. Diane Yamada, MD, Silvana Pannain, MD, and Ernst Lengyel, MD, PhD OBJECTIVE: To estimate whether metformin use by
and the diabetic patients who did not use metformin,
ovarian cancer patients with type II diabetes was associ-
respectively (P؍.03). The patients with diabetes received
ated with improved survival.
the same treatment for ovarian cancer as the patients
METHODS: We reviewed the effect of diabetes and
without diabetes. The association of metformin use and
diabetes medications on ovarian cancer treatment and
increased progression-free survival, but not overall sur-
outcomes in a single-institution retrospective cohort.
vival, remained significant after controlling for standard
Inclusion criteria were International Federation of Gy-
clinicopathologic parameters.
necology and Obstetrics stage I–IV epithelial ovarian,
CONCLUSION: In this ovarian cancer cohort, the pa-
fallopian, or peritoneal cancer. Exclusion criteria were
tients with type II diabetes who used metformin had
noninvasive pathology or nonepithelial malignancies.
longer progression-free survival, despite receiving similar
The primary exposures analyzed were history of type II
treatment for ovarian cancer.
diabetes and diabetes medications. The primary out-
comes were progression-free and overall ovarian can-
cer survival.
LEVEL OF EVIDENCE: II
RESULTS: Of the 341 ovarian cancer patients included in
the study, 297 did not have diabetes, 28 were type II
diabetic patients who did not use metformin, and 16
It is estimated that two in five women born in the were type II diabetic patients who used metformin. The
United States in 2000 will have type II diabetes progression-free survival at 5 years was 51% for diabetic
diagnosed during their lifetime.1 Diabetes is associ- patients who used metformin compared with 23% for the
ated with an increased incidence of most cancers and nondiabetic patients and 8% for the diabetic patients
decreased cancer survival.2 Understanding how dia- who did not use metformin (P؍.03). The overall survival
betes influences cancer treatment and prognosis is of at 5 years was 63%, 37%, and 23% for the diabetic
particular importance in cancers that have high mor- patients who used metformin, the nondiabetic patients,
tality rates, such as ovarian cancer, which is the mostlethal gynecologic cancer.3 Although limited, the From the Departments of Obstetrics and Gynecology/Section of Gynecologic available data suggest that ovarian cancer patients Oncology–Center for Integrative Science, Health Studies, and Medicine/Sectionof Endocrinology, University of Chicago, Chicago, Illinois. with type II diabetes have decreased survival.4,5 Supported by grants from the Reproductive Scientist Development Program It is biologically plausible that the hyperinsulin- (National Institutes of Health 2K12HD00849-22) and the Gynecologic Cancer emia and hyperglycemia induced by type II diabetes Foundation/St. Louis Ovarian Cancer Awareness (to Dr. Romero). Dr. Lengyel promotes tumorigenesis. Insulin stimulates the holds a Clinical Scientist Award in Translational Research from the BurroughsWellcome Fund and is supported by grants from the National Cancer Institute growth of cancer cells by activating insulin-like growth factor I and decreasing insulin-like growth Corresponding author: Iris Romero, MD, University of Chicago, Department of factor binding protein.6 Hyperglycemia provides a Obstetrics and Gynecology, 5841 South Maryland Avenue, MC 2050, Chicago, nutrient-rich microenvironment for rapidly dividing IL 60637; e-mail: iromero@uchicago.edu. cancer cells, which have elevated metabolic demands Financial Disclosure
and consume glucose at a higher rate than normal The authors did not report any potential conflicts of interest. cells.7 Consistent with this concept, elevated plasma 2012 by The American College of Obstetricians and Gynecologists. Publishedby Lippincott Williams & Wilkins. glucose levels at the time of ovarian cancer cytore- ductive surgery are predictive of decreased survival.8 OBSTETRICS & GYNECOLOGY
Interestingly, metformin, a diabetic treatment unaware of the individual’s cancer survival status. All that reduces both insulin and glucose levels, may have records of diabetic patients were reviewed again by a anticancer effects.9 Epidemiologic studies indicate second blinded investigator (I.R.). For secondary re- that patients who use metformin have decreased view of records for nondiabetic patients, the entries in cancer incidence10,11 and increased cancer surviv- the dataset were consecutively numbered and every al.11–14 Preclinical studies also corroborate the antitu- tenth medical record was reviewed again.
morigenic effect of the drug in breast, prostate, and The primary exposures of interest for the study colon cancer.15–17 In ovarian cancer, two preclinical were a history of type II diabetes and the type of studies have shown that metformin inhibits the pro- diabetic medications used. The primary outcome liferation of cancer cell lines in a dose-dependent and measures included progression-free and overall sur- vival. Recurrence was defined using previously pub- Based on preclinical evidence of a strong antican- lished clinical criteria22 and included evidence of cer effect, we hypothesized that the use of metformin reappearance of the cancer by clinical examination may be associated with improved ovarian cancer (eg, tumor, ascites), new tumor findings on computed outcomes. To test this hypothesis, we evaluated tomography or ultrasound scan, or an increase in CA whether ovarian cancer patients with type II diabetes 125 two times or more the upper limit of normal (70 who used metformin had increased progression-free units/mL). Progression-free survival was calculated from the date of diagnosis until the date of ovariancancer recurrence or death. Patients without recur- MATERIALS AND METHODS
rence or death were censored at last follow-up. Over- This is a retrospective, single-institution, cohort study all survival was calculated from the date of diagnosis utilizing an established dataset of women treated for until death from ovarian cancer or the observation ovarian cancer at the University of Chicago from 1992 was censored as of the date of last follow-up. The six to 2010. All women with International Federation of patients out of 341 who died from causes other than Gynecology and Obstetrics stage I–IV epithelial ovar- ovarian cancer were censored at the time of death.
ian, fallopian, or peritoneal cancer were included in the All statistical analyses were performed using R study. We identify, and refer to, these three cancers as 2.11.0 (The R Project for Statistical Computing, ovarian cancer because of their common origin in the http://www.r-project.org/). For comparison, the co- Mu¨llerian epithelium. Patients were excluded from the hort was stratified into three groups: patients without study if they had noninvasive pathology, nonepithelial diabetes, patients with diabetes who did not use malignancies, or nonovarian primary cancer that had metformin, and patients with diabetes who used met- metastasized to the ovary. They were also excluded if formin. F tests were used for comparing continuous they did not receive their primary cancer treatment at variables and the Fisher exact tests were used for the University of Chicago but were only treated for categorical variables. Kaplan-Meier survival curves recurrences. The study was approved by the institu- were plotted for the three groups and compared with tional review board at the University of Chicago.
log-rank tests. Of note, in the metformin group, the As previously reported, the ovarian cancer data- event rates and the duration of follow-up were not set contains information on clinicopathologic param- sufficient to estimate the upper limit of the confidence eters, treatment, and outcomes.20,21 All pathologic interval (CI) of progression-free and overall survival.
diagnoses had been confirmed by a subspecialty- These CIs are reported as not estimable. A Cox trained gynecologic pathologist. Follow-up data were proportional hazards model was used to estimate obtained from medical records at the University of hazard ratios (HRs) for progression-free and overall Chicago, the Illinois Cancer Registry, the United survival while adjusting for confounders. For model States Social Security Index, and by communicating selection, a univariable Cox regression was per- with physicians involved in the patient’s care. For this formed with each of the potential confounders, and study, an additional chart abstraction was performed those found to be significant in predicting recurrence or to extract data pertaining to diagnoses of diabetes, survival were included in the final model. The resulting diabetic medications, body mass index, glycosylated Cox regression calculated the HRs for diabetic patients hemoglobin A1C, fasting glucose, and renal function.
who used metformin and diabetic patients who did not Fasting blood glucose and glycosylated hemoglobin use metformin, with patients without diabetes as the A1C values were missing for a large portion of the reference group. The HR of diabetic patients who used cohort, prohibiting analysis of these variables. The metformin relative to diabetic patients not using met- person abstracting the data regarding diabetes was Romero et al
Metformin Use and Ovarian Cancer Survival OBSTETRICS & GYNECOLOGY
patients who used metformin (nϭ16). Among the met- The study included 341 women. The cancer types formin users, five used only metformin, four used met- included epithelial ovarian (nϭ273), fallopian (nϭ34), formin and insulin, and seven used metformin plus and peritoneal (nϭ34) cancer. The median duration of another oral antidiabetic agent. The baseline character- follow-up was 63 months (range 1–245 months). For istics of the three groups are reported in Table 1. The comparison, the cohort was stratified into three groups patients with diabetes were more likely to have higher as follows: nondiabetic patients (nϭ297), diabetic pa- body mass indexes, and to be African American than tients who did not use metformin (nϭ28), and diabetic Table 1. Baseline Characteristics and Ovarian Cancer Treatment of Study Cohort
Diabetic Patients Not
Diabetic Patients
Nondiabetic Patients
Using Metformin
Using Metformin
P
American Society of Anesthesiologists class BMI, body mass index.
Data are n (%) or meanϮstandard deviation unless otherwise specified.
American Society of Anesthesiologists class was determined at time of surgery.
* Patients were excluded from the analysis if they declined chemotherapy or if chemotherapy was not indicated (eg, International Federation of Gynecology and Obstetrics 1A).
† Among the patients who did not receive a taxane, all received cytoxan except five patients in the nondiabetic group.
‡ Platinum-sensitiveϭno recurrence of disease for 6 months or more after the end of chemotherapy. Patients who did not receive platinum agent or did not complete six cycles of chemotherapy were excluded from platinum sensitivity analysis.
Romero et al
Metformin Use and Ovarian Cancer Survival The patients with diabetes received the same treatment for ovarian cancer as the patients without diabetes. The rate of primary cytoreductive surgery with residual disease smaller than 1 cm after surgery,the type of chemotherapy agent used, and the meannumber of chemotherapy cycles were comparable among the groups. Despite the long study interval (18years), 95% of patients received both platinum-based and taxane-based chemotherapy; the most commonagents were carboplatin (72%) and taxol (94%). Theresponse to chemotherapy was assessed using the clinical parameter of platinum sensitivity, defined as 6months or more without recurrence of disease afterthe end of chemotherapy. The group using metformin had the highest percentage of patients sensitive toplatinum chemotherapy, although the difference did not reach statistical significance (Pϭ.18) (Table 1).
Despite similar ovarian cancer treatment, the type II diabetic patients who used metformin had longer progression-free and overall survival. The pro- gression-free survival at 5 years for diabetic patientswho used metformin was 51% (median 72 months, 95% CI 13.3–not estimable) compared with 23%(median 16 months, 95% CI 13.9 –19.5 months) forpatients without diabetes and 8% (median 10 months, 95% CI 13.3–37.2 months) for the diabetic patientswho did not use metformin (log rank test Pϭ.03; Fig.
1A). The overall survival at 5 years for diabetic patients who used metformin was 63% (median 138months, 95% CI 31.1–not estimable) compared with37% (median 42 months, 95% CI 35.1– 49.6 months) for patients without diabetes and 23% (median 35 months, 95% CI 24.6 –54.3 months) for the diabeticpatients who did not use metformin (log rank test Pϭ.03; Fig. 1B). The difference in survival could notbe explained by stage, grade, or histologic subtype, because the tumor characteristics of the three groupswere similar (Table 2).
Fig. 1. Kaplan-Meier estimates of survival outcomes. The
In a survival analysis adjusted for confounders, three groups are ovarian cancer patients with type II when comparing diabetic patients who used met- diabetes using metformin (nϭ16), ovarian cancer patientswithout type II diabetes (nϭ297), and ovarian cancer formin to diabetic patients who did not use met- patients with type II diabetes not using metformin (nϭ28). P formin, the metformin group had a significantly de- values are from the log-rank test. A. Progression-free sur-
creased hazard for disease recurrence (HR 0.38, 95% vival. B. Overall survival.
CI 0.16 – 0.90). The metformin group also had a Romero. Metformin Use and Ovarian Cancer Survival. Obstet decreased hazard of dying (HR 0.43, 95% CI 0.16 – 1.19), but this difference was not statistically signifi-cant. Variables significantly predictive of progression- excluded from the models were American Society of free or overall survival (or both) and included in the Anesthesiologists class, ethnicity, and history of car- models were age, body mass index, creatinine, Inter- diovascular disease. The hazards for both disease national Federation of Gynecology and Obstetrics recurrence and dying were also lower in patients with stage, tumor grade, residual implants larger than 1cm diabetes who used metformin when compared with after surgery, and histologic subtype. The variables the group without diabetes, but this reduction was not that were not significant predictors of survival and statistically significant. In contrast, patients with dia- Romero et al
Metformin Use and Ovarian Cancer Survival OBSTETRICS & GYNECOLOGY
Table 2. Tumor Characteristics
Diabetic Patients Not
Diabetic Patients
Nondiabetic Patients
Using Metformin
Using Metformin
P
FIGO, International Federation of Gynecology and Obstetrics.
Data are n (%) unless otherwise specified.
betes who did not use metformin had an increased On activation, adenosine monophosphate-activated hazard of ovarian cancer recurrence (HR 1.42, 95% protein kinase contributes to energy conservation by CI 0.87–2.33) and an increased hazard of dying from decreasing cancer cell proliferation.23 Second, insulin ovarian cancer (HR 1.33, 95% CI 0.77–2.28) when and glucose promote tumorigenesis6,8 and, through compared with patients without diabetes (Table 3).
inhibiting hepatic gluconeogenesis and increasing in-sulin sensitivity, metformin decreases both insulin DISCUSSION
and glucose levels.24 Finally, clinical and laboratory In a single-institution retrospective cohort, we found that studies indicate that metformin may improve re- ovarian cancer patients with type II diabetes who used sponse to chemotherapy.25 In a study of neoadjuvant metformin had increased progression-free survival, but chemotherapy for breast cancer, Jiralerspong et al26 not overall survival, when compared with type II dia- reported that diabetic patients who used metformin betic patients who did not use metformin. These find- had a pathologic complete response rate of 24% ings are consistent with those of Landman et al13 who compared with an 8% pathologic complete response analyzed a prospective diabetic cohort and found that rate among diabetic patients who did not use met- diabetic patients who used metformin had an adjusted formin. In our ovarian cancer cohort, we also noted overall cancer mortality hazards ratio of 0.43 (95% CI that the type II diabetic patients who used metformin 0.23–0.80) when compared with diabetic patients not had the best response to chemotherapy.
using metformin. In fact, we report a similar magnitude The findings reported here are provocative, but of risk reduction in ovarian cancer patients.
given the retrospective study design, they can only be The suggestion that metformin use is associated considered hypothesis-generating and should not be with improved ovarian cancer outcomes may not be generalized to clinical practice at this time. That being intuitive to most gynecologists because patients with said, the study has some notable strengths. We in- type II diabetes have more comorbidities. However, clude an analysis of the association of metformin use our findings are congruent with the translational and ovarian cancer survival, whereas previous studies research that has demonstrated a distinct anticancer have only examined the association of diabetes and effect of the drug in several cancers (eg, breast, ovarian cancer survival without consideration of dia- prostate, colon), including ovarian cancer.15–19 Three betic medications.5 Also, by using a large ovarian possible mechanisms have been proposed. First, the cancer dataset that contained detailed chemotherapy anticancer effect of metformin may be a result of the and platinum sensitivity data, we were able to esti- activation of a critical energy sensor in cancer cells, mate the relationship between metformin use and adenosine monophosphate-activated protein kinase.
response to chemotherapy. Finally, the homogeneity Romero et al
Metformin Use and Ovarian Cancer Survival Table 3. Cox Proportional Model Hazard Ratios for Progression-Free and Overall Survival
Progression-Free Survival
P
Overall Survival
P
FIGO, International Federation of Gynecology and Obstetrics; BMI, body mass index.
Data are hazard ratio for recurrence (95% confidence interval) unless otherwise specified.
* Alternative parameterization with diabetic patients not using metformin as a reference group is presented to show the direct comparison between diabetic patients not using metformin and diabetic patients using metformin.
† Modeled as continuous variables.
‡ Creatinine level at the time of ovarian cancer diagnosis. Creatinine predicted overall survival but not progression-free survival.
of ovarian cancer treatment among diabetic patients antitumorigenic effects. The idea that specific diabetic and nondiabetic patients in the cohort allowed us to treatments affect cancer survival is clinically relevant make relatively strong inferences about the effects of given the increasing prevalence of diabetes. According to the World Health Organization, 171 million people An important limitation of the study was the worldwide have type II diabetes, and this number is small sample size. The sample size limited our ability expected to double by 2030.27 If future studies continue to detect a difference in survival between diabetic to support the protective effect of metformin in cancer, patients using metformin and patients without diabe- then this will be an important consideration when tes. Based on retrospective power calculations, assum- managing diabetic patients with cancer.
ing a true HR of the magnitude reported here andassuming that 5% of ovarian cancer patients have type REFERENCES
II diabetes and use metformin, for progression-free 1. Narayan KM, Boyle JP, Thompson TJ, Sorensen SW, William- survival a total sample size of 870 patients would be son DF. Lifetime risk for diabetes mellitus in the United States.
required to have 80% statistical power. For overall survival, 1,570 patients would be required. The sam- 2. Giovannucci E, Harlan DM, Archer MC, Bergenstal RM, ple size also prohibited an analysis of the effects of Gapstur SM, Habel LA, et al. Diabetes and cancer: A consen- diabetic medications other than metformin. A final sus report. CA Cancer J Clin 2010l;60:207–21.
limitation was that because of lack of available data, 3. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, 2009. CA Cancer J Clin 2009;59:225– 49.
we could not control for diabetes severity, which may 4. van de Poll-Franse LV, Houterman S, Janssen-Heijnen ML, represent an independent risk factors for death.
Dercksen MW, Coebergh JW, Haak HR. Less aggressive In summary, the findings in this ovarian cancer treatment and worse overall survival in cancer patients with cohort of improved progression-free survival among diabetes: A large population based analysis. Int J Cancer2007;120:1986 –92.
type II diabetic patients who used metformin add to a 5. Bakhru A, Buckanovich RJ, Griggs JJ. The impact of diabetes growing body of evidence from epidemiologic and on survival in women with ovarian cancer. Gynecol Oncol preclinical studies indicating that metformin may have Romero et al
Metformin Use and Ovarian Cancer Survival OBSTETRICS & GYNECOLOGY
6. Pollak M. Insulin and insulin-like growth factor signalling in exerts an antitumoral effect in vitro and in vivo through a neoplasia. Nat Rev Cancer 2008;8:915–28.
decrease of cyclin D1 level. Oncogene 2008;27:3576 – 86.
7. Warburg O, Wind F, Negelein E. The metabolism of tumors in 18. Gotlieb WH, Saumet J, Beauchamp MC, Gu J, Lau S, Pollak the body. J Gen Physiol 1927;8:519 –30.
MN, et al. In vitro metformin anti-neoplastic activity in epi- 8. Lamkin DM, Spitz DR, Shahzad MM, Zimmerman B, Lenihan thelial ovarian cancer. Gynecol Oncol 2008;110:246 –50.
DJ, Degeest K, et al. Glucose as a prognostic factor in ovarian 19. Rattan R, Giri S, Hartmann L, Shridhar V. Metformin atten- carcinoma. Cancer 2009;115:1021–7.
uates ovarian cancer cell growth in an AMP-kinase dispensable 9. Ben Sahra I, Le Marchand-Brustel Y, Tanti JF, Bost F. Met- manner. J Cell Mol Med 2009;15:166 –78.
formin in cancer therapy: A new perspective for an oldantidiabetic drug? Mol Cancer Ther 2010;9:1092–9.
20. Terplan M, Temkin S, Tergas A, Lengyel E. Does equal treatment yield equal outcomes? The impact of race on 10. Evans JM, Donnelly LA, Emslie-Smith AM, Alessi DR, Morris survival in epithelial ovarian cancer. Gynecol Oncol 2008;111: AD. Metformin and reduced risk of cancer in diabetic patients.
11. Libby G, Donnelly LA, Donnan PT, Alessi DR, Morris AD, 21. Kenny HA, Leonhardt P, Ladanyi A, Yamada SD, Montag A, Evans JM. New users of metformin are at low risk of incident Im HK, et al. Targeting the urokinase plasminogen activator cancer: A cohort study among people with type 2 diabetes.
receptor inhibits ovarian cancer metastasis. Clin Cancer Res 12. Baur DM, Klotsche J, Hamnvik OP, Sievers C, Pieper L, 22. Rustin GJ, Quinn M, Thigpen T, du Bois A, Pujade-Lauraine Wittchen HU, et al. Type 2 diabetes mellitus and medications E, Jakobsen A, et al. Re: New guidelines to evaluate the for type 2 diabetes mellitus are associated with risk for and response to treatment in solid tumors (ovarian cancer). J Natl mortality from cancer in a German primary care cohort.
23. Shackelford DB, Shaw RJ. The LKB1-AMPK pathway: Metab- 13. Landman GW, Kleefstra N, van Hateren KJ, Groenier KH, olism and growth control in tumour suppression. Nat Rev Gans RO, Bilo HJ. Metformin associated with lower cancer mortality in type 2 diabetes: ZODIAC-16. Diabetes Care2010;33:322– 6.
24. DeFronzo RA, Goodman AM. Efficacy of metformin in patients with non-insulin-dependent diabetes mellitus. The 14. Bowker SL, Yasui Y, Veugelers P, Johnson JA. Glucose- Multicenter Metformin Study Group. N Engl J Med 1995;333: lowering agents and cancer mortality rates in type 2 diabetes: Assessing effects of time-varying exposure. Diabetologia 2010;53:1631–7.
25. Iliopoulos D, Hirsch HA, Struhl K. Metformin decreases the 15. Buzzai M, Jones RG, Amaravadi RK, Lum JJ, DeBerardinis RJ, dose of chemotherapy for prolonging tumor remission in Zhao F, et al. Systemic treatment with the antidiabetic drug mouse xenografts involving multiple cancer cell types. Cancer metformin selectively impairs p53-deficient tumor cell growth.
26. Jiralerspong S, Palla SL, Giordano SH, Meric-Bernstam F, 16. Liu B, Fan Z, Edgerton SM, Deng XS, Alimova IN, Lind SE, Liedtke C, Barnett CM, et al. Metformin and pathologic et al. Metformin induces unique biological and molecular complete responses to neoadjuvant chemotherapy in diabetic responses in triple negative breast cancer cells. Cell Cycle patients with breast cancer. J Clin Oncol 2009;27:3297–302.
27. Wild S, Roglic G, Green A, Sicree R, King H. Global 17. Ben Sahra I, Laurent K, Loubat A, Giorgetti-Peraldi S, Colos- prevalence of diabetes: Estimates for the year 2000 and etti P, Auberger P, et al. The antidiabetic drug metformin projections for 2030. Diabetes Care 2004;27:1047–53.
Romero et al
Metformin Use and Ovarian Cancer Survival

Source: http://www.uchospitals.edu/pdf/uch_033093.pdf