Abstract 3376: Real-world cancer treatment patterns and their effects in older adults with lung cancer: a SEER-MHOS analysis

Abstract Introduction: Lung cancer primarily affects older adults, with a median diagnosis age of 71. Due to high competing death in older adults, the feasibility and efficacy of treatment may differ from clinical trials. This study examined treatment patterns and mortality outcomes in older adults with lung cancer using real-world analytic methods. Methods: We used SEER-Medicare Health Outcomes Survey (MHOS) to identify adults aged ≥65 years with non-small cell lung cancer (2007-2019). The study had 2 primary outcomes: (1) uptake for first-course treatments, and (2) lung cancer survival. Treatment uptake was calculated by age, frailty at diagnosis (assessed with a cancer-specific frailty index from MHOS), and neighborhood socioeconomic status (nSES) across disease stages. We used inverse probability of treatment weighting (IPTW) and clone-censoring methods in multivariable Cox models to mitigate selection and immortal time biases. Results: In 11, 443 patients (mean age, 77.3 years; 72% White; 47% distant-stage), treatment uptake varied by age, frailty, and nSES. Surgery rates for localized-stage patients declined with age (67% vs. 16% for ages 65-69 vs. ≥85, P<10-3), as did chemotherapy rates for distant-stage patients (58% vs. 21% for ages 65-69 vs. ≥85, P<10-3). Patients who received surgery or chemotherapy were less frail at diagnosis (P<10-3) compared to those who did not, while frailty score was comparable by radiotherapy. Survival benefits for all three treatments were not observed in patients aged ≥85 using both IPTW and clone-censoring models, though IPTW-only models suggested survival benefits, indicating residual bias. Conclusion: Age at diagnosis, frailty, and nSES influence treatment uptake in older adults with lung cancer. Rigorous real-world analytic methods indicate limited survival benefits from intensive treatments for patients≥85 years, underscoring the need for tailored treatment decision-making in this vulnerable population. Table. Association between treatment and lung cancer survival Real-world Analytics Age at LC Surgery (vs. no surgery) Chemotherapy (vs. no chemotherapy) Radiation (vs. no radiotherapy) aHR (95% CI) aHR (95% CI) aHR (95% CI) IPTW 65-74 0.34 (0.31-0.37)* 0.74 (0.69-0.80)* 0.86 (0.81-0.92)* 75-84 0.40 (0.38-0.43)* 0.88 (0.84-0.93)* 0.86 (0.82-0.90)* 85+ 0.62 (0.56-0.68)* 0.75 (0.67-0.83)* 0.73 (0.67-0.80)* IPTW + Clone-censoring 65-74 0.56 (0.52-0.59)* 1.08 (1.02-1.14)* 0.84 (0.80-0.89)* 75-84 0.61 (0.59-0.63)* 1.12 (1.07-1.16)* 0.88 (0.85-0.92)* 85+ 0.97 (0.90-1.04) 1.08 (1.00-1.16)* 1.00 (0.93-1.08) Abbreviations: IPTW, inverse probability of treatment weighting; aHR, adjusted hazard ratio; CI (confidence interval); LC (lung cancer); * denotes P<0.05; Hazard ratios were adjusted for key prognostic factors. Citation Format: Shuyan Qiu, Yueyi Xu, Laura Pinheiro, Jialin Mao, Yushu Shi, Yiwey Shieh, Christine Garcia, Eunji Choi. Real-world cancer treatment patterns and their effects in older adults with lung cancer: a SEER-MHOS analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3376.