Abstract PS11-23: Current landscape in phase 3 trials in breast cancer at major oncology conferences

Abstract

Background: The oral presentation of studies at conferences may impact treatment practice even before the publication or regulatory approval. Methodological aspects and reporting patterns are evolving, and an assessment of these features may help understand the current landscape and the way forward. Methods: We analyzed the characteristics of primary analyses of phase 3 trials from oral sessions presented at ASCO, ESMO and SABCS 2017-19 and the timing of their subsequent publications. We excluded non-inferiority trials and duplicate presentations within the period. Results: Of 36 unique trials, 19/10/7 had a first author from Europe/United States/Asia or elsewhere, and 21/15 were in the (neo)adjuvant/palliative settings. All but one trial had 2 arms. The number of patients enrolled ranged from 226 to 4884 (median, 646). All but 2 trials were on systemic therapy, with 16 dedicated exclusively to hormone-receptor+, 8 to HER2+ and 6 to triple-negative disease. According to the authors, 23 (63.9%, 95% CI, 46.2 to 79.2%) trials were positive, and 13 were negative. For only 2 of the negative trials was the conclusion not entirely consistent with results. When sponsor information was available, 18/21 industry vs 4/14 academic trials were positive (P=0.002). Paradoxically, negative trials were larger than positive trials (medians of 2639 and 585 patients; P=0.002), perhaps because all but one of 13 negative trials were in the (neo)adjuvant setting, vs 9/23 among positive trials (P=0.004). The time to publication was significantly shorter for positive than negative trials (medians of 215 vs 785 days; HR=0.32; P=0.013). Of 30 trials with time-to-event endpoints, the ratio between observed:expected HR could be computed for 24: the mean was 0.88 for 15 positive and 1.43 for 9 negative trials (P=0.004; 1.09 overall). Conclusion: In recent phase 3 trials in breast cancer, positive trials are more likely than negative trials to be industry-sponsored, to assess the metastatic setting, and to be published earlier. These results also suggest that the treatment effect has been underestimated in positive trials. Research is ongoing to elucidate potential determinants of the latter hypothesis. Citation Format: Gustavo Oliveira Bretas, Jose Bines, Everardo Delforge Saad. Current landscape in phase 3 trials in breast cancer at major oncology conferences [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS11-23.