Adjuvant chemotherapy in colorectal cancer patients with microsatellite instability - Response

Inresponse: The major criticism raised by Watanabe in his letter about our recently published article (1) concerns the number of patients with stage II and III MSI-Hcolorectal cancer (CRC) who received chemotherapy, that ‘‘should be around 20 at most’’ and thus ‘‘seems too small to draw any conclusions.’’ The assumption of Watanabe on the small number of patients in our study is based on his interpretation that among those 65 MSI-Hwho received chemotherapy, 31 were in stage IV. This assumption is not correct. In fact, we considered as patients receiving adjuvant chemotherapy only those who underwent surgical intervention with curative intent (i.e., radical surgery), thus excluding most of stage IV patients who received only palliative chemotherapy. We apologize for the fact that this explanation was not present in the text leading to the misinterpretation of the data. On this basis, among the 65 MSI-Hpatients considered as treated with adjuvant chemotherapy, only 2 were in stage IV whereas 25 were in stage II and the remaining 38 were in stage III, which is almost double of what was hypothesized. Furthermore, Watanabe states that the results of our study ‘‘could potentially bias the significance of MSI-Has predictor of survival in adjuvant-treated colon cancer patients’’ as shown by the results of his study (2), which compared the survival of MSI-Hand MSS patients in stage II and III who received chemotherapy. We think that our results are not completely comparable with those of Watanabe, mainly for the reason that our study had a different design. In his work, Watanabe compared, among patients who received chemotherapy, those affected by MSI-HCRC versus those with stable tumors and found a survival advantage in the first group; this study design could not clearly establish whether this prognostic advantage is conferred by the better sensitivity of MSI-HCRC to chemotherapeutic agents or is due to the presence of instability by itself. On the contrary, our study was designed to investigate the sensitivity of MSI-HCRC to chemotherapy. For this reason, we compared among MSI-HCRC patients the outcome of those who underwent 5-fluorouracil-based chemotherapy versus those who were not treated. The results showed that chemotherapy did not confer any survival advantage in MSI-HCRC patients, confirming the report of another study with larger sample size (3). In addition, the results of multivariate analysis in stage II and III CRC cases showed the presence of MSI-H as an independent prognostic factor: this means that even in our study, in accordance with Watanabe’s study, among patients who received chemotherapy, MSI-Hpatients have a better outcome than MSS patients. Finally, another difference between the two studies involved the definition of MSI. To define MSI, in our study, we used the reference marker panel, as established in Bethesda guidelines. Watanabe used eight dinucleotide and two polyadenine markers in most cases, or two mononucleotide markers in those cases without normal DNA available. This could have important implications leading to heterogeneity in the population defined as MSI in the two investigations.