Résumé
<p><b>OBJECTIVE</b>To evaluate the colorectal cancer (CRC) prevention effect of metformin in comparison with that of other T2DM medications from a Markov model perspective.</p><p><b>METHODS</b>Literature concerning CRC morbidity of T2DM patients with metformin or other diabetes medications treatment was reviewed in PubMed and Cochrane Library database from September 2010 to December 2016.</p><p><b>INCLUSION CRITERIA</b>(1)enrolled population was adult patients with T2DM but without CRC; (2) any of the parameters applied in our model was reported; (3) randomized clinical trials (RCTs), quasi-randomized trials, prospective or retrospective cohort studies. With CRC morbidity as endpoint, parameters were extracted to construct Markov model to assess CRC morbidity and cumulative tumor-free survival in each group over 11 years' follow-up period. Finally, Monte Carlo analysis was performed to evaluate the influence of parameter instability on the model.</p><p><b>RESULTS</b>Seven literatures were recruited and 10 000 patients were virtually allocated for each arm. In contrast with non-metformin group, T2DM patients treated with metformin had a lower rate of CRC(1.670% vs. 2.146%, P=0.016). Moreover, cumulative tumor-free survival of metformin group was, slightly but significantly, better than that of non-metformin group (10.908 years vs. 10.882 years, P=0.000).</p><p><b>CONCLUSIONS</b>T2DM patients treated with metformin have a lower morbidity of CRC and a better cumulative tumor-free survival than those of non-metformin group. Large scale RCTs are needed to illustrate the role of metformin in the prevention of CRC.</p>