Résumé
Background: non-small cell lung cancer is the deadliest cancer worldwide with an increasing incidence and a better prognosis for resectable early stages, despite frequent recurrences. As CT-scan-based individual screening is about to be implemented, early-stage NSCLC will become more prevalent. Yet oncologic adjuvant cares are decided with no accurate information about minimal residual disease (MRD). It has been recently shown that nuclear and mitochondrial circulating cell-free DNA (cirDNA) is significantly modified in cancer and might stand as a promising representation of MRD. The analysis of nuclear and mitochondrial cirDNA in blood with quantitative PCR (qPCR), through the use of a multi-analytes test called MiTest, could offer a very sensitive study model of MRD in patients with resected NSCLC. Our main objective is to determine whether or not the lack of normalization of MiTest of cirDNA one month after resection of an early-stage NSCLC is predictive of the relapse rate at 1 year.Method: the LUNGDOC Project is an interventional, prognostic, prospective, single-arm, non-comparative, exploratory study. The primary endpoint is the disease free survival (DFS) probability at 1 year after surgery. Patients with a lack of normalization of MiTest at one month after the resection of a NSCLC have an estimated 1-year probability of relapse of 30% versus 10% in patients for whom MiTest is normalized (HR=1.9). The sample size estimation is 120 patients with a power of 85% and an alpha risk error of 5%. The capability of MiTest to discriminate between patients with or without relapse at 1 year will be assessed by means of ROC curves. The nonparametric Kaplan-Meier estimates will be used to calculate probabilities and plot survival curves.Discussion: we hope to obtain preliminary data proving that the normalization of MiTest at one month postoperatively is a good prognostic factor for NSCLC-resected patients, which could deeply modify management of resected NSCLC by better stratifying prognosis and possibly anticipating relapses. cirDNA analysis may considerably improve our understanding of MRD, enabling more personalised adjuvant therapeutic decisions by identifying subjects with a higher risk of recurrence.