Second primary tumors (SPTs) in patients (PTS) with non-small cell lung cancer (NSCLC): A retrospective cohort study

Abstract

7214 Background: Outcomes of PTS with NSCLC are expected to improve with advances in early detection, adjuvant therapy, multidisciplinary management, and molecular-targeted therapy. Improvements in survival may increase the incidence and clinical relevance of second primary tumors (SPTs) in these patients. Because the few studies examining SPTs in NSCLC PTS have been of limited size, we conducted a retrospective, single-institution study to better define clinical characteristics and outcomes in a large cohort. Methods: Subjects were identified through a search of the M. D. Anderson Cancer Center tumor registry. Among 9230 stage I-III NSCLC PTS registered between 1/90–7/02, PTS who had a subsequent cancer diagnosis (excluding non-melanoma skin cancers) were identified. A total of 359 PTS with available medical records formed our study population. A uniform set of demographic and clinical data were abstracted by chart review. Results: Subject characteristics: 63% male, median age 63 yrs (range 37–93), 50%/44% current/former smokers, NSCLC stage: 65% I, 13% II, 22% III. Median follow-up 7.5 yrs. SPT stage: 35% I, 28% II, 25% III, 12% IV. 20%/80% synchronous/metachronous SPTs. Most frequent SPTs: head & neck 16%, prostate 11%, breast 8%, colorectal 7%, gynecologic 7%. Most frequent female SPTs: breast 19%, gynecologic 16%, head & neck 14%, lymphoma 6%, colorectal 5%. Most frequent male SPTs: prostate 17%, head & neck 16%, colorectal 8%, leukemia 8%, esophagus 7%. Median overall, SPT-free, and survival after SPT were 7.1, 2.9, and 2.3 yrs, respectively. Adjusted multivariate analysis demonstrated that older age (P<0.0001), synchronous (vs. metachronous) SPT (P<0.0001), and advanced stage (P=0.044) were significantly associated with poorer cancer-specific survival, while older age (P=0.0019) was significantly associated with poorer SPT-free survival. Conclusions: This study represents the largest series of SPTs in NSCLC PTS. The distribution of SPTs reflects the prevalence of smoking in this population. The median time to SPT development, 2.9 yrs, may aid in the development of screening strategies for NSCLC PTS. This research was supported by an NCI Oncology Research Faculty Development Award. No significant financial relationships to disclose.