Impact of preoperative bimodality induction including twice-daily radiation on tumor regression and survival in stage III non-small-cell lung cancer

Purpose: The objective of this prospective study was to assess the feasibil ity, toxicity, and efficacy of an intensive trimodality approach in stage I II non-small-cell lung cancer (NSCLC). patients and Methods: Fifty-four patients with NSCLC and biopsy-proven N2 n odes (IIIA; n = 25) or N3 nodes or T4 lesions (IIIB; n = 29) were administe red two initial cycles of ifosfamide, carboplatin, and etoposide; subsequen t radiotherapy (45 Gy, twice-daily 1.5 Gy) with concurrent carboplatin and vindesine; and surgery if the patient's disease was resectable or conventio nal radiotherapy (16 Gy, 2 Gy/d) if the patient's disease was not resectabl e or incompletely resectable. Results: Thirty-seven patients (69%) responded to preoperative induction. F orty of 54 patients (74%) had disease that was resectable, with 34 (63%) co mplete resections (RO). A substantial pathologic response (tumor regression [TR] > 90%) was achieved in 27 of 54 patients (50%) and is revealed as an independent predictor for long-term survival after surgery. Five treatment- related deaths (9%) occurred. With a median follow-vp period of 44 months, calculated survival rates at 3 years were 35% for patients with stage IIIA disease, 26% for patients with stage IIIB disease, and 56% for patients wit h RO disease and TR > 90%. Conclusion: This trimodality approach is feasible and results in encouragin g 3-year survival rates in prognostically unfavorable patients with stage I II NSCLC. Patients experiencing a 90% degree of pathologic TR were most lik ely to achieve long-term survival. (C) 1999 by American Society of Clinical Oncology.