How does cumulative evidence from older adult-specific trials influence clinical practice? A difference-in-differences analysis in early-stage breast cancer

Background: Despite increasing focus on conducting cancer clinical trials in older adults, it is unclear whether such evidence influences practice patterns. We aimed to estimate the impact of cumulative evidence from older adult-specific trial results from the CALGB 9343 and PRIME II trials that found post-lumpectomy irradiation has little benefit among older adults with early-stage breast cancer (ESBC). Methods: Patients diagnosed with ESBC between 2000 and 2018 were identified from the SEER registry data. We examined the incremental immediate effect, incremental average yearly effect, and cumulative effect of a series of CALGB 9343 and PRIME II results on the utilization level of post-lumpectomy irradiation. We conducted difference-in-differences analyses, comparing those aged 70 or older vs. <65 years old. Results: The initial 5-year CALGB 9343 results in 2004 led to a significant immediate (-0.038, 95% CI:-0.064,-0.012) and average yearly decrease (-0.008, 95% CI:-0.013,-0.003) in the probability of irradiation use among those aged 70 or older compared to those below 65 years of age. 11-year CALGB 9343 results in 2010 significantly accelerated the average yearly effect by 1.7 percentage points (95% CI:-0.030,-0.004). The other later results did not significantly change the time trend. The cumulative effect of all results between 2004 and 2018 was-26.3 percentage points (95% CI:-0.29,-0.24). Conclusion: Cumulative evidence from older adult-specific trials in ESBC led to decreasing use of irradiation over time among elderly patients. The rate of decrease after the initial results was accelerated by long-term follow-up results.