Comprehensive Mediastinal Nodal Characterization Optimizes Outcome Prediction: Results From a Single-institution Trimodality Regimen in Mediastinoscopy-Defined Stage III Non-Small Cell Lung Cancer (NSCLC)

Abstract

Purpose/Objective(s): To report our experience with a trimodality approach to mediastinoscopy-defined stage III NSCLC. Materials/Methods: Following pre-treatment evaluation, our institutional protocols have consisted of neoadjuvant accelerated hyperfractionated chemoradiotherapy (AHFX-ChRT), followed by curative resection, followed by adjuvant AHFX-ChRT. AHFX-ChRT consisted of 30Gy/20fx at 1.5Gy/fx BID (> 6h interfraction interval). Chemotherapy consisted of cisplatin/paclitaxel or carboplatin/paclitaxel. Pretreatment evaluation included history and physical exam, heme and chemistry profiles, pulmonary function tests, computed tomography of chest and upper abdomen, positron emission tomography (PET), bronchoscopy, and mediastinoscopy (meds). Results: Between 2/99 and 1/05, a total of 121 patients (pts) with pathologically confirmed NSCLC were treated in a similar fashion. The study sample consists of 78 (64.5%) males and 43 (35.5%) females with a median age of 61 years (range 31-78). The median Karnofsky performance status (KPS) was 90 (range 70100). 47.9% and 21.5% of pts had adenocarcinoma and squamous cell carcinoma, respectively (resp). 57.9% of pts had right-sided tumor. Meds-defined stage was IIIA in 57.9%, IIIB in 31.4%, other in 5.8%; 6 pts did not undergo meds. 100 pts (82.6%) completed neoadjuvant AHFX-ChRT and curative resection (pneumonectomy was performed in 29 pts). There were 4 postoperative deaths (4%). Pathologic complete response (pCR) was noted in the primary in 21% and 41% in the nodes; pCR in both was 16%. Overall, 72 pts (72%) completed the protocol as above. The median follow-up for these pts is 27.7 mos. Median survival time (MST) and 2-year OS were 29.9 mos and 63.5%, resp. 2-year LC rate and 2-year metastaticfree survival were 85.2% and 70.3%, resp. For pts with meds-defined stage IIIA and stage IIIB disease, MST was 34.5 mos and 26 mos; 2-year OS was 69% and 50.3%, resp (p=NS). For pts with pCR in the nodes, more than half were still alive at the time of analysis while those with residual nodal disease had a MST of 28.8 mos; 2-yr OS was 70.8% versus 60.2%, resp (p=0.0079). On univariate analysis, KPS (analyzed as a continuous variable), absence of PET positive nodal stations, pCR in the nodes, and percentage of nodes positive at resection (analyzed as a continuous variable) were prognostic for OS. On multivariate analysis (MVA), KPS and absence of PET positive nodal stations remained significant. Mediastinal stage, pCR in the primary, margin status, age, sex, histology, side of tumor, number of mediastinal positive stations (none versus single versus multiple), and type of curative resection (pneumonectomy versus other) did not correlate with OS. Percentage of nodes positive at resection and sex (female versus male) were significantly correlated with local control on MVA. Percentage of nodes positive at resection was the only factor significantly correlated with metastatic-free survival on MVA. Conclusions: This protocol employing neoadjuvant AHFX-ChRT followed by curative resection and adjuvant AHFX-ChRT is feasible and effective, comparing favorably with previously reported data for mediastinoscopy-defined stage III NSCLC. Comprehensive mediastinal nodal characterization optimizes outcome prediction.