The manuscript is a report on randomised, phase 3, OlympiAD trial, where the authors claimed that ,among patients with HER 2 negative metastatic breast cancer and a germline BRCA mutation, median progression- free survival was significantly longer with oral olaparib monotherapy than with standard single agent chemotherapy ( capecitabine, eribulin or vinorelbine). Methodologically, the study was conducted appropriately, as true clinical trial. The study was randomized, controlled, open-label, multicenter, international, and adequate outcome measures were employed. They also analyzed data according to intention-to-treat principle, that is appropriate. The main problem is, clinical significance of the difference in duration of progression-free survival. This is actually the most important outcome, since Hazard ratio (which is presented in detail, with confidence intervals) is only saying that patients with olaparib have less probability in any moment to progress or die, but not how big is that change in real life. The only clinically significant measure is duration of progression-free survival, and here median is only 2.8 months longer with olaparib. However, if the patients are widely distributed around the median, it may happen that difference is not statistically significant.

In this study, the authors made a claim.They missed important information in the Results: they claimed median progression-free survival was significantly longer with olaparib (7 months), then with standard therapy (4.2 months). However, they neither showed range of the survival times (what is essential measure of variability if one gives median as measure of central tendency), nor showed results of statistical test that proved that the difference was statistically significant! The question is why? Maybe because the statistical test is actually NOT significant. We do not know, and we have to ask from the authors to present this data. I think we should ask for additional data about survival duration. The title is not in line with CONSORT guideline (2010), though it reflects the content of the manuscript. The language, style of the articles is of high quality. In terms of novelty of findings, it can be considered as widening our understanding in a new way.