Abstract 4207: Elucidating drivers of the Hispanic paradox in non-small cell lung cancer

Abstract

Abstract The Hispanic paradox refers to findings in the United States (US) showing similar or better health outcomes in Hispanics than in non-Hispanic whites (NHWs) despite lower average socioeconomic status (SES). This paradox has been reported for non-small cell lung cancer (NSCLC) using data from national cancer registries, with Hispanics experiencing lower mortality rates than Non-Hispanics (NHs). However, these registries often lack critical covariate data such as smoking. To this end, we developed the Lung Cancer Clinical Cohort at Montefiore Medical Center (LC3MMC) in the Bronx, NY, to describe the growing but understudied US Hispanic population and to further elucidate factors in lung cancer survival. Subjects in LC3MMC were ≥18 years old with no prior cancer history who were diagnosed with incident primary lung carcinoma of any histology and of stage 1-4, between 2004-2017. Demographic and clinical data were obtained from MMC’s clinical systems, and tumor-related information was obtained from MMC/Einstein’s Cancer Registry. Of the 5102 subjects in LC3MMC, the mean age at diagnosis was 68 (±12) years, 50% were male, 80% were ever-smokers, 72% had known ethnicity [NHW: n=1278, Non-Hispanic Black (NHB): n=1222, Hispanic: n=855], and 80% had known histology (NSCLC: n=3554, SCLC: n=505). Based on log-rank tests, overall survival was greater in Hispanics compared to NHs (all subjects: p=0.01, NSCLC only: p=0.01) and did not differ between NHWs and NHBs (all: p=0.26, NSCLC: p=0.21). Cancer subtype (χ2: NSCLC vs. SCLC: p=0.18) and treatment rates (yes vs. no) [χ2: surgery (all: p=.08, NSCLC: p=0.12), radiation (all: p=.85, NSCLC: p=0.55), chemotherapy (all: p=0.26, NSCLC: p=0.23)] did not vary by ethnicity. Multivariate Cox proportional hazards modeling compared survival between Hispanics and NHs while adjusting for age; gender; stage; adenocarcinoma histology; smoking status; receipt of surgery, chemotherapy, radiation, and palliative care; marital status; and SES. In NSCLC patients, Hispanic ethnicity was associated with decreased risk of death (HR=0.86, 95% CI=0.74-0.99). When limiting the model to Hispanics, stage (regional: HR=2.81, 95% CI=1.72-4.57; distant: HR=6.78, 95% CI=4.24-10.85; unknown: HR=4.17, 95% CI=2.42-7.18), receipt of surgery (HR=0.53, 95% CI=0.39-0.72) and receipt of chemotherapy (HR=0.74, 95% CI=0.57-0.97) were associated with survival. Hispanic ethnicity was not associated with better survival in SCLC subjects (adjusted HR=0.97, 95% CI: 0.68-1.39). We have observed the Hispanic paradox in NSCLC in one of the largest US academic medical centers. Clinical and social factors do not fully explain the improved survival in this group. Our results provide the basis for detailed exploration of the demographic, cultural, and molecular drivers of lung cancer survival disparities among different racial/ethnic groups in the Bronx. Citation Format: Madelyn Klugman, Xiaonan Xu, Mindy Ginsberg, Thomas Rohan, H. Dean Hosgood. Elucidating drivers of the Hispanic paradox in non-small cell lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4207.