On 2015 Sep 28, Gauthier Bouche commented:

The placebo-controlled randomized trial reported in this article explores the effects of anti-inflammatory drugs (steroidal and non-steroidal) on the inflammation induced by surgery of rectal cancer. Knowing the critical role that the peri-operative period may have on long-term cancer outcomes (Horowitz M, 2015), results of this trial are of major importance for clinicians and researchers interested in this topic. Since the full article is only available in Chinese, it is important to report the number of recurrences in each of the group as only the final p-value is reported in the abstract. The number of recurrence was 1, 3 and 3 in group A (celecoxib), group B (methylpredinosolone) and group C (placebo) respectively. The statistical test used was a Chi-square test. In the method section, the authors do not specify any statistical hypothesis (power calculation) for the recurrence endpoint. Therefore the conclusion of the author that celecoxib does not lower postoperative recurrence rate of rectal cancer is not supported by the results. The trial was not sufficiently powered to detect a difference (even a large one).

On 2015 Sep 28, Gauthier Bouche commented:

The placebo-controlled randomized trial reported in this article explores the effects of anti-inflammatory drugs (steroidal and non-steroidal) on the inflammation induced by surgery of rectal cancer. Knowing the critical role that the peri-operative period may have on long-term cancer outcomes (Horowitz M, 2015), results of this trial are of major importance for clinicians and researchers interested in this topic. Since the full article is only available in Chinese, it is important to report the number of recurrences in each of the group as only the final p-value is reported in the abstract. The number of recurrence was 1, 3 and 3 in group A (celecoxib), group B (methylpredinosolone) and group C (placebo) respectively. The statistical test used was a Chi-square test. In the method section, the authors do not specify any statistical hypothesis (power calculation) for the recurrence endpoint. Therefore the conclusion of the author that celecoxib does not lower postoperative recurrence rate of rectal cancer is not supported by the results. The trial was not sufficiently powered to detect a difference (even a large one).