OBJECTIVES: To evaluate the incremental healthcare resource utilization (HRU) and cost associated with recurrence among resected early stage non-small cell lung cancer (eNSCLC) patients.

METHODS: This retrospective observational study identified patients diagnosed with stage IA-IIIB NSCLC between 2010 and 2015 from SEER registry data linked with Medicare claims. Patients enrolled in Medicare Part A, B, and D for ≥7 months prior and ≥12 months after diagnosis were included. Surgery was identified between 1 month prior and 12 months after diagnosis. Recurrence was defined as evidence of secondary malignant neoplasm (excluding lung), surgery, radiation, chemotherapy, immunotherapy, or targeted therapy during the period starting 6 months after surgery or end of adjuvant treatment, until the end of follow-up (i.e. end of enrollment or study period, or death). Patients with recurrence were matched 1:1 with patients without recurrence using exact matching on cancer stage and treatment, and propensity score matching on other patient characteristics. Pseudo recurrence dates were assigned to non-recurrence patients based on surgery or adjuvant treatment dates. HRU (ER, outpatient and inpatient visits) and all-cause costs were measured starting from 2 months prior to the recurrence date until the end of follow-up.

RESULTS: Among the 7,142 patients included in the study, 2,172 (30%) patients were identified with recurrence. A total of 1,612 patients with recurrence were matched to 1,612 patients without recurrence. HRU was significantly higher in the recurrence cohort. Average follow-up cost in the recurrence cohort was $5,889 per patient per month (PPPM) compared to $1,312PPPM in the no recurrence cohort, resulting in a difference of $4,577PPPM (p<0.001) with inpatient costs as the largest contributor.

CONCLUSIONS: Based on a real-world population, recurrence among eNSCLC resected patients is associated with significantly increased HRU and cost. New therapeutic options that decrease recurrence could potentially reduce the economic burden of eNSCLC.