Recurrence dynamics after trimodality therapy (Neoadjuvant concurrent chemoradiotherapy and surgery) in patients with stage IIIA (N2) lung cancer

Introduction: We investigated the timing and patterns of recurrence after the treatment of stage IIIA (N2) non small cell lung cancer via neoadjuvant concurrent chemoradiotherapy followed by surgery. Materials and methods: An institutional database was reviewed retrospectively between 1997 and 2013 (N = 570). Eligible patients had pathologically proven N2 disease, and they completed the planned trimodality therapy with curative intent. The hazard rate function and competing risk analysis were used to evaluate the recurrence dynamics. Results: Among the included patients, 76% had single station N2 involvement and 95% had complete resection. The 5-year overall and recurrence-free survival rates were 47% and 29%, respectively. Of the 290 patients who experienced recurrence, 25 (8.4%) experienced loco-regional recurrence, whereas 238 (80.4%) had distant metastases. The hazard rate function for overall recurrence revealed a peak at approximately 8 months after surgery and a marked decline after 2 years. The peak recurrence frequency of distant metastasis differed at each site, with isolated brain metastases exhibiting the earliest peak (6 months) and a narrow recurrence interval (15 months). A histological comparison revealed a higher recurrence hazard rate for adenocarcinoma than for squamous cell carcinoma but similar pattern of recurrences. Patients with complete responses had a lower cumulative incidence rate of recurrence but a slightly earlier peak of recurrence. Nodal responses to induction therapy demonstrated that patents with ypN0 had the lowest recurrence risk, whereas patients with ypN1 and ypN2 had similar hazard rates and cumulative incidence rates of recurrence. Conclusions: The dynamics of recurrence after trimodality therapy is organ-specific and varies according to pathologic factors. Our finding provides information on selection patients with risk of recurrence and timing of surveillance study.