Cost-effectiveness analysis of stereotactic body radiotherapy and surgery for medically operable early stage non small cell lung cancer.

BACKGROUND: Projections estimate an increase of 50% of the incidence of lung cancer by 2030. Early-stage non-small cell lung cancer represented 19% of NSCLC cases diagnosed in the US between 2005 and 2011. There is rising evidence in favour of lung cancer screening, which will reduce the occurrence of later-stage lung cancers while raising the incidence of early-stage NSCLC. Current guidelines state that for early-stage NSCLC, surgical resection should be performed, and stereotactic body radiotherapy (SBRT) is an option for patients who are non-medically operable. In this study, we compared the cost-effectiveness of SBRT with lobectomy in medically operable patients.

METHODS: We developed a Markov model based on the survival results of two randomized studies comparing SBRT and video assisted thoracoscopic surgery (VATS) lobectomy in early-stage NSCLC, to describe survival and treatment related complications of patients treated for early-stage NSCLC. This analysis was conducted from the French payer perspective on a lifetime perspective. Utility values, recurrence risks, and costs were adapted from the literature. Deterministic (DSA) and probabilistic (PSA) sensitivity analyses were performed to assess the influence of the assumptions made.

RESULTS: The Markov model developed was consistent with survival data reported in the pool analysis of the randomized studies. SBRT and lobectomy total costs were 9,234.15€ and 10,726.98€, respectively, and the quality-adjusted life expectancies were 16.35 and 15.80 QALYs, respectively. The DSA, run on every assumption made, revealed that the incremental cost-effectiveness ratio was mainly sensitive to the decrement of utility caused by treatment related complications and initial cost of both surgery and SBRT. The PSA showed that SBRT had the highest probability of cost-effectiveness compared to lobectomy.

CONCLUSIONS: This is the first medico-economic study evaluating SBRT and lobectomy in stage I NSCLC based on randomized studies, and our analyses suggest that SBRT is dominant over lobectomy in operable early-stage NSCLC treatment. Deterministic and probabilistic sensitivity analyses confirmed that this result was robust and that it was not modified by the assumptions made in the Markov model building.