RISW 2024: An opportunity to learn, be inspired, and network
Godwin Yung (Genentech/Roche)
RISW, in my humble opinion, is the conference that every statistician working in drug development should attend each year. Indeed, I encourage young career statisticians whenever possible to go to Maryland in September to stay up-to-date on the latest statistical topics/thinking, to find inspiration/direction for their own work, and to network and foster collaboration with others. The funny thing is, before 2024, the last time I attended RISW was 7 years ago (due to Covid and having young children). But this year, I was finally able to take my own advice, and RISW 2024 did not disappoint. Here are some highlights from my experience.
Continued learning and exchange of perspectives
There were many great sessions at RISW 2024. One that I enjoyed in particular was PS11 “Bayesian shrinkage estimation for subgroups: Is it ready for prime time?” In this session, Mark Rothmann (FDA) shared about FDA’s use of shrinkage estimation to inform US patients of their potential benefit-risk based on their demography (FDA CDER, 2023). David Ohlssen (Novartis) talked about some important methodological issues, including how priors that influence variability between subgroups and model selection can impact results, as well as how to accommodate for overlapping subgroups (Sun et al., 2024). Properly analyzing and interpreting subgroups is a frequent challenge that clinical trialists face, and I enjoyed bringing what I learned from this session back to my colleagues, some of whom are also active in this area of research (Wolbers et al., arXiv).
Finding inspiration and direction
Everyone wants to innovate. However, bringing about innovation is no easy task. It requires invention and commercialization, which in turn requires finding the right problems, coming up with the most appropriate solutions, and having the business acumen to drive adoption (Rufibach et al., 2024).
At RISW 2024, there were ample opportunities to find inspiration and direction for statistical innovation. During PS22 “Causality in Trials”, Daniel Rubin (FDA) shared his thoughts on 5 important applications (hence opportunities) for causal inference in drug development: missing data, covariate adjustment, time-varying treatment, generalizability, and surrogate endpoints. Likewise, in a separate session, Gregory Levin declared FDA’s recently released guidance on covariate adjustment to be his “favorite” guidance, because he saw covariate adjustment as a “free lunch” that was underutilized. And as aptly remarked by Margaret Gamalo (Pfizer) in PS11 “Influencing regulatory policy and guidance development to drive statistical innovation”, navigating different interests is critical for successful collaboration.
Fostering connections and collaborations
During meals and breaks, I enjoyed catching up with old acquaintances, making new connections, and meeting with collaborators in-person, some for the first time. One such group of collaborators consisted of Emmanuel Zuber (Cogitamen), Lisa Hampson (Novartis), and Arunuva Chakravartty (Novartis), statisticians who recently published on the topic of overall survival safety monitoring and from whom Mark Yan (Roche), myself, and several other Genentech/Roche statisticians had been seeking feedback from for our own manuscript (Yung et al., arXiv). The following picture was taken after our evening meal of good food, drinks, and conversation. It captures the fruits of a positive collaboration and outreach effort between two pharma companies.
Some parting words, if I may. Sitting at the cross-section between regulatory and industry statistics, RISW really is a fantastic conference. If you missed RISW 2024, I highly encourage you to attend next year’s workshop so that you continue to learn, to find inspiration/direction in your own work, and to grow your network of collaborators. Finally, I would like to thank Meijing Wu for her invitation to contribute to this report. May her kindness, passion for furthering the field of biopharmaceutical statistics, and other fond memories live on.
References
FDA Center for Drug Evaluation and Research (2023) `Drug Trials Snapshots Summary Report 2023’. Available at: https://www.fda.gov/media/178602/download?attachment
Sun S, Sechidis K, Chen Y, et al. (2024) `Comparing algorithms for characterizing treatment effect heterogeneity in randomized trials’, Biometrical Journal 66(1):e2100337.
Wolbers M, Rabunal MV, Li K, et al. (submitted) `Using shrinkage methods to estimate treatment effects in overlapping subgroups in randomized clinical trials with a time-to-event endpoint’, arXiv. Available at: https://doi.org/10.48550/arXiv.2407.11729
Rufibach K, Wolber M, Davenport J, et al. (2024) `Implementation of statistical innovation in a pharmaceutical company’, Statistics in Biopharmaceutical Research. Available at: https://doi.org/10.1080/19466315.2024.2327291
FDA Center for Drug Evaluation and Research (2023) `Guidance for covariates in randomized clinical trials for drugs and biological products’. Available at: https://www.fda.gov/media/148910/download
Fleming TR, Hampson LV, Bharani-Dharan B, et al. (2024) `Monitoring overall survival in pivotal trials in indolent cancers’, Statistics in Biopharmaceutical Research. Available at: https://www.tandfonline.com/doi/abs/10.1080/19466315.2024.2365648
Yung G, Rufibach K, Wolbers M, et al. (submitted) `”6 choose 4”: A framework to understand and facilitate discussion of strategies for overall survival safety monitoring’, arXiv. Available at: