Background: Insulin-sensitizing drugs (i.e. metformin) and lipid-lowering agents (i.e. statins) demonstrate pleiotropic effects against cancer cell lines and show potential as therapeutic adjuvants for cancer patients. Our objective was to examine the association of specific classes of diabetes drugs and statins on survival time among pancreatic cancer patients with metabolic syndrome. Methods: Linked Surveillance, Epidemiology and End Results (SEER) cancer registry and Medicare claims data were used identify pancreatic cancer patients diagnosed in 2008-2009 with the following: resection of the pancreas, metabolic syndrome prior to cancer diagnosis, continual enrollment in Medicare Parts A, B, and D. Metabolic syndrome was defined as having 3 out of the 4 conditions: hyperglycemia, dyslipidemia, obesity, or hypertension. Exposure to diabetes drugs and statins 12 months prior to resection of cancer was assessed in the prescription drug event database and categorized as insulin/sulfonylureas vs. metformin/thiazolidinediones (TZD) vs. none; and as no statin vs. lipophilic statin vs. hydrophilic statin. We employed Cox regression adjusting for age, sex, race, stage at diagnosis and chemotherapy, to estimate the impact of individual drug classes on survival with pancreatic cancer. Results: We analyzed data on 263 pancreatic cancer cases with metabolic syndrome. Users of insulin/sulfonylureas demonstrated shorter survival by 6 months and had higher hazard of death (HR=1.43, 95%CI (1.05, 1.95)) compared to non-users. In contrast, users of hydrophilic statins experienced longer survival by 9 months and showed lower hazard of death compared to those who did not use statin (HR = 0.67, 95%CI (0.46, 0.96)) (Table). Conclusion : Insulin/sulfonylureas use is associated with shorter survival, while hydrophilic statin use was associated with longer survival in pancreatic patients with metabolic syndrome.

Table. Relative hazard of death in pancreatic cancer patients by diabetes drugs and statin

Variables Category n Median survival (months, 95%) unadjusted HR (95%) adjusted HR* (95%) P-value 
Diabetes medications administered prior to cancer diagnosis      
 No insulin/sulfonylurea/metformin/TZD 131 20.2 (15.6, 25.0) Ref Ref  
 Metformin/TZD 27 21.7 (10.5, NA) 0.73 (0.43, 1.24) 0.80 (0.50, 1.39) 0.42 
 Insulin/sulfonylurea 105 13.9 (9.4, 21.2) 1.29 (0.96, 1.73) 1.43 (1.05, 1.95) 0.02 
Statins administered prior to cancer diagnosis      
 No statin 106 14.6 (10.7, 18.3) Ref Ref  
 Lipophilic statins 83 20.8 (14.9, 28.6) 0.76 (0.55, 1.06) 0.72 (0.51, 1.02) 0.07 
 Hydrophilic statins 74 23.5 (17.8, 31.0) 0.63 (0.44, 0.90) 0.67 (0.46, 0.96) 0.03 
Variables Category n Median survival (months, 95%) unadjusted HR (95%) adjusted HR* (95%) P-value 
Diabetes medications administered prior to cancer diagnosis      
 No insulin/sulfonylurea/metformin/TZD 131 20.2 (15.6, 25.0) Ref Ref  
 Metformin/TZD 27 21.7 (10.5, NA) 0.73 (0.43, 1.24) 0.80 (0.50, 1.39) 0.42 
 Insulin/sulfonylurea 105 13.9 (9.4, 21.2) 1.29 (0.96, 1.73) 1.43 (1.05, 1.95) 0.02 
Statins administered prior to cancer diagnosis      
 No statin 106 14.6 (10.7, 18.3) Ref Ref  
 Lipophilic statins 83 20.8 (14.9, 28.6) 0.76 (0.55, 1.06) 0.72 (0.51, 1.02) 0.07 
 Hydrophilic statins 74 23.5 (17.8, 31.0) 0.63 (0.44, 0.90) 0.67 (0.46, 0.96) 0.03 

*Adjusted for age, sex, race, stage and chemotherapy

View Large

Citation Format: Christie Y. Jeon, Stephen J. Pandol, Marc T. Goodman. Survival time in pancreatic cancer patients with metabolic syndrome varies by use of insulin and statins. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2173. doi:10.1158/1538-7445.AM2014-2173

©2014 American Association for Cancer Research.
2014