In response. Adjuvant chemotherapy in colorectal cancer patients with microsatellite instability.

To the Editor: In a recent issue of Clinical Cancer Research (1), we read Benatti et al.'s article with interest, therein they showed significantly better survival in microsatellite instability high (MSI-H) patients than microsatellite stable patients, and concluded that the type of genomic instability could influence the prognosis of colorectal cancer, in particular, in stages II and III. Furthermore, in relation to adjuvant treatment, they concluded that fluorouracil-based chemotherapy does not seem to improve survival among MSI-H patients. Although Benatti et al. introduced our results, our research on the similar topic came to a different conclusion (2). There are major issues to be discussed as to how Benatti et al. drew their conclusions. The first point is the total number of MSI-H patients who received chemotherapy. In Benatti et al.'s study, only 65 MSI-H patients received fluorouracil-based chemotherapy; among these 65 patients, 31 patients had stage IV disease. Considering that most of the patients who had stage IV disease would have received postoperative chemotherapy, the total number of MSI-H patients, in stages II and III, who received chemotherapy seems to be much fewer than 65. The second point is how they examined the survival rate. Benatti et al. examined the survival in stage II and stage III patients separately. Therefore, the number of MSI-H patients who received chemotherapy in either stage II or III should be 20 at most. This number of patients seems to be too small to find any difference in survival. In fact, MSI-H patients who underwent chemotherapy showed higher survival rates than those who did not, although it did not reach statistical significance. Furthermore, according to the survival curve of stage II cancers, there seems to be no cancer-related death in MSI-H patients who had undergone chemotherapy. Considering all these points, the small number of MSI-H patients who underwent chemotherapy seems to be the main reason why they could not find significant differences in survival between MSI-H patients who did or did not undergo chemotherapy. All of these points are also true in an analysis of MSI-H patients in stage III. In Benatti et al.'s study, the number of MSI-H patients with chemotherapy in stage II or III seems to be too small to draw any conclusion. In our previous study, we examined 73 MSI-H patients who received chemotherapy and showed a significant difference in survival between MSI-H patients and microsatellite stable patients (2). Taken together, Benatti et al.'s study may potentially bias the significance of MSI-H as a predictor of survival in patients with adjuvant-treated colon cancer.