Paclitaxel/carboplatin (PC) followed by twice daily chemoradiation (hfRT/PC) versus hfRT/PC followed by PC as induction to surgery in stage III non-small cell lung cancer (NSCLC): A randomized phase III trial of the Lung Cancer Study Group Mitteldeutschland.

7541 Background: In multimodal neoadjuvant therapy of stage III NSCLC, the potential impact of the sequencing of the different induction modalities on survival has not been evaluated.,Patients (pts) with biopsy-proven stage III (PET obligatory) NSCLC were stratified (center/technical operability) and then randomized to (arm 1) 2-3 cycles paclitaxel 225 mg/m d 1 / carboplatin AUC 6 d 1, followed by hfRT at 1.5 Gy 2x/d (total 45 Gy) with concurrent paclitaxel 50 mg/m / carboplatin AUC 2 d 1, 8, 15 vs (arm 2) hfRT/PC followed by PC at identical dosages to arm 1, then restaging incl. PET and surgery. The primary (secondary) endpoint was median overall survival (mOS), the secondary endpoint was median progression-free survival (mPFS). Assuming a median survival advantage of 6 months (mo) for arm 2, 104 pts per arm would detect a difference with type I (type II) error ≤ 5% (20%).,210 pts were randomized between 01/01 and 06/05, 207 were evaluable. Median follow-up was 35 mo. The pts were well balanced between treatment arms 1/2: 106/104 pts; stage IIIA 34%/39%; stage IIIB 64%/60%; primary technical operability 24%/23%; PS 0 19%/18%; PS 1 29%/29%. mOS was 543 vs 550 days (p=073). mPFS was 326 vs 343 days (p=0.86). mOS of Stage IIIA was 489 vs 775 days (p=0.14), of stage IIIB 594 vs 546 days (ns). Grade 3 and 4 nonhematologic toxicities were not significantly different between treatment arms, leukopenias were more frequent in arm 2 (8 vs 28 pts).,The sequence of chemotherapy and chemoradiation as multimodal induction therapy before surgery has no impact on mOS or mPFS in Stage III NSCLC. In Stage IIIA there is, however, a trend favoring arm 2. Grade 3 and 4 leukopenias are more frequent in arm 2. No significant financial relationships to disclose.