The Association of Metformin, Other Antidiabetic Medications and Statins on the Prognosis of Rectal Cancer in Patients with Type 2 Diabetes: A Retrospective Cohort Study

Abstract

Metformin and statin use have been associated with an improved prognosis for colorectal cancer in persons with type 2 diabetes (T2D). Data regarding rectal cancer (RC) have been inconclusive; therefore, we investigated the issue with high-quality data and a robust study design. We identified 1271 eligible patients with T2D and incident RC between 1998 and 2011 from the Diabetes in Finland (FinDM) database. Cox models were fitted for cause-specific mortality rates to obtain adjusted estimates of the hazard ratios (HR) with 95% confidence intervals (CI) in relation to use of antidiabetic medication (ADM) and statins before the RC diagnosis and for post-diagnostic use in a time-dependent exposure manner. No sufficient evidence was found for either pre- or post-diagnostic metformin use and RC mortality (HR 0.96, 95% CI 0.67–1.38, and 0.70, 95% CI 0.45–1.10, respectively) when compared to other oral ADMs. Both pre- and post-diagnostic statin use appeared to be inversely associated with mortality from RC (HR 0.77 95% CI 0.63–0.94, and 0.57, 95% CI 0.42–0.78, respectively). Our study was inconclusive as to the association of metformin use with the prognosis of RC, but statin use was found to predict reduced mortality, both from RC and from other causes of death in persons with T2D.

Metformin Does not Affect Outcomes in Patients With Locally Advanced Rectal Cancer Treated With Neoadjuvant Therapy and Resection

“…32 On the other hand, Erkinantti did not find any different in OS or DFS between metformin users, patients on other anti-diabetics, or patients on no anti-diabetic medications. 31 Hence, we evaluated the role of metformin in Locally Advanced Rectal Cancer (LARC) treated with neoadjuvant therapy and resection at our institution.…”

“…The current literature on the effect of DM on the prognosis of rectal cancer remains debatable. 14,15,31,33 The conflicting results of the effect of metformin in population-based studies can be explained at least in part, due to the heterogeneity in study designs (observational/ trials with variable regimens/systematic reviews and meta-analyses) and comparison groups (diabetic metformin users/diabetic non-metformin users/non-diabetics). Further, observational studies studying drug effects tend to be susceptible to time-related biases such as immortal-time bias (misclassification or ignorance of variation in the timing of treatment initiation from cohort entry), time-window bias (cases and controls not matched on the duration of exposure opportunity time within a fixed observation period), and time-lagging bias (when a first-line therapy is compared to the second-or third-line of therapy given at a different disease stage) resulting in potential overestimation of a drug effect when there is none.…”

“…Experiments suggest that metformin exerts its anti-cancer action both by activating AMP-activated protein kinase (AMPK interacting with LKB1, a tumor suppressor gene) and in an AMPK-independent manner by inhibiting hexokinase-2 activity, leading to a state of calorie restriction, apoptosis, reduced protein synthesis, and reduced proliferation of cancer cells. 29 While the promising results from the in-vitro analyses encouraged the prospect of metformin as an adjunct treatment in CRC, the evidence from population-based studies (including trials) is conflicting; from a protective effect, 18-21,30 no association, 6,22,31 as well as to a worse risk of developing CRC. 23,24 Specifically for rectal cancer, we could only find three studies in the literature, national studies from Taiwan and Finland, and a single-institution study from the United States.…”

“…The current literature on the effect of DM on the prognosis of rectal cancer remains debatable. 14,15,31,33…”