Abstract 4792: Nationwide patterns of incidence and surveillance of second primary lung cancer among long-term lung cancer survivors using SEER Medicare data

Abstract Background: With advances in screening and therapy, the number of lung cancer (LC) survivors is rapidly increasing, with over half a million as of 2019. Prior studies demonstrated that the incidence of second primary lung cancer (SPLC) among LC survivors is 4-6 times that of initial primary lung cancer (IPLC) in the general population. However, the incidence of SPLC and surveillance strategies using computed tomography (CT) have not been examined in long-term LC survivors with over 5-year survival. We aim to estimate SPLC incidence among long-term survivors and the effectiveness of annual surveillance CT exams. Methods: We used data from SEER-Medicare with a cohort of 1,232,338 LC patients from the SEER cancer registry followed till death or last follow-up, who were also enrolled in Medicare in 1999-2020. The study cohort included 12,948 patients who were long-term survivors (≥ 5-year survival from IPLC) by age 65, received curative therapy for IPLC, and were continuously enrolled in Medicare 1 year prior to 5 years after IPLC diagnosis. The primary outcome was the 10-year cumulative SPLC incidence accounting for competing risk of death in the full study cohort and by receipt of surveillance CTs post-5-year survival, defined using detailed billing and LC-related ICD codes. To qualify as surveillance CTs, two annual CTs 9-18 months apart is required. Cox regression was used as exploratory analysis to examine the association between LC-specific survival (LC as cause of death) and the receipt of two CTs post-5-year survival as a binary variable, adjusting for SPLC diagnosis, age, sex, marital status, race, IPLC stage, income, education, and poverty. Results: Of 12,948 long-term LC survivors, 46% (n=6,015) were male, 90% (n=11,476) had early-stage disease, 77% (n=9,945) had undergone surgery for initial curative treatment, and 3.8% (n = 486) received two surveillance CTs initiated within 18 months post-5-year survival. The 10-year cumulative incidence of SPLC was 5.06% [95% CI: 5.00-5.12%] in the SEER cohort but was substantially higher among long-term LC survivors at 13.6% [13.0-14.2%]. Those who received two annual surveillance CTs after 5-year survival had a significantly higher 10-year SPLC incidence (23.3% [19.4 - 27.2%]) than those who received one or no CT (13.2% [12.6 - 13.8%]). The adjusted hazard ratio of the association between LC-specific survival and receipt of two CTs was 0.67 [0.59-0.80, p < 0.0001]. Conclusions: The high cumulative risk of SPLC among long-term LC survivors warrants tailored surveillance strategies. Our exploratory analysis showed a high SPLC detection rate among those who received two annual surveillance CTs vs. those who did not, with significantly higher LC-specific survival in the former. Future work will include extensive analysis accounting for time-varying exposure of multiple surveillance CTs using rigorous causal inference methods. Citation Format: Chloe C. Su, Eunji Choi, Julie T. Wu, Allison W. Kurian, Michelle C. Odden, Manisha Desai, Leah M. Backhus, Heather A. Wakelee, Ann N. Leung, Joel W. Neal, Summer S. Han. Nationwide patterns of incidence and surveillance of second primary lung cancer among long-term lung cancer survivors using SEER Medicare data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4792.