For years, Sara worked in a hospital, making care possible for others. Now, after being diagnosed with breast cancer, she was the one in need of care.
A team of researchers has implemented a groundbreaking method for modeling survival rates in patients with metastatic cancer — an approach that could expedite patient access to novel, much-needed therapies. The team, which includes six Parexel health economics statisticians, details its findings in a paper published in Medical Decision Making.
Thanks to widespread and prolonged responses to immuno-oncology (IO) therapies, patient survival rates are on the rise, with many patients living long past the follow-up period of the clinical trials in which they were enrolled. Until now, models for survival extrapolation have largely been based on study follow-up data, which may account for only the first two or three years of treatment. This compromises the accuracy of standard parametric models (SPMs). As a result, payers often lack confidence in long-term projections derived from such methods.
But our statisticians and their collaborators offer a different kind of model in their paper “Use of Advanced Flexible Modeling Approaches for Survival Extrapolation from Early Follow-up Data in two Nivolumab Trials in Advanced NSCLC with Extended Follow-up.” They evaluate a flexible parametric model (FPM) that employs trial data while also making use of longer-term external data sources. By including data from several national cancer registries (including the U.S. National Cancer Institute’s Surveillance, Epidemiology, and End Results program) as well as general population mortality data, the team’s approach improves on simpler models, returning projections that are more accurate, more reliable, and more valuable to drug developers, payers, and health technology assessment agencies.
Through their study of FPMs, the authors demonstrate that such models can incorporate realistic effects not captured by naïve approaches, including age-related mortality and treatment waning, when the impact of treatment-specific effects decrease over time. These advanced methods can also represent complex treatment effects common to IO therapies, such as heterogeneous responses, in which some metastases respond to treatment while others remain stable or progress. As a result, FPMs may provide more accurate short- and long-term extrapolations, even with limited study follow-up available.
Simply put, FPMs allow for better predictions based on early trial data cuts, projecting long-term clinical outcomes and cost-effectiveness more accurately than standard models. The rigorous approach produces consistent projections using successive database locks (DBLs), including for the earliest DBL at two years. This stability makes FPM estimates less susceptible to criticism than results from simpler models not based on additional evidence. Equipped with these reliable extrapolations, researchers and regulators can move medicines forward faster, giving patients accelerated access to novel IO therapies — one of the most promising ways to help people facing cancer.
Expedite payer coverage with surrogate endpoints
Sep 6, 2023
What's the Future of FSP?
Jan 6, 2022
China's Market Approval Policy and Medical Insurance Payment System for Rare Disease
Jul 21, 2022
Parexel names first Patient Ambassador
Jun 17, 2022
Mitigating FSP talent shortfalls through development
Jun 14, 2022
Overview of China's Market Approval Policy Med Insurance Payment System
Apr 7, 2022
How biotechs can strengthen their value story with advanced analytics
Feb 15, 2022
Pragmatic Trials: Targeting evidence generation to inform market access and meet payers’ needs
Feb 8, 2022
New endpoints for early-stage cancer are gaining regulatory traction
Jan 28, 2022
Three strategies for articulating a coherent product value story
Dec 18, 2021
What Do the Next 10 Years Hold for CRISPR?
Jul 27, 2022
Rethinking FSP outsourcing
Oct 4, 2021