Patients with stage III non–small cell lung cancer (NSCLC) are a heterogeneous group at high risk for local and distant relapse. It is generally agreed that optimal treatment selection occurs in a multidisciplinary team setting after optimal staging. Adjuvant chemotherapy is now considered the standard of care for patients with resectable disease; the NSCLC Collaborative Group meta-analysis concluded that adjuvant chemotherapy improved survival in locally advanced NSCLC irrespective of whether it was performed after surgery alone or after surgery plus radiation therapy (RT) [NSCLC Meta-analyses Collaborative Group. Lancet. 2010]. The role of neoadjuvant or adjuvant RT, however, remains controversial. A landmark meta-analysis of 9 randomized trials with > 2000 patients concluded that postoperative RT (PORT) was detrimental in patients with early-stage, completely resected NSCLC but that its role in patients with N2 disease was unclear [PORT Meta-analysis Trialists Group. Cochrane Database Syst Rev. 2003].

Cécile Le Péchoux, MD, Institut Gustave-Roussy, Villejuif, France, reviewed the evidence concerning the role of neoadjuvant or adjuvant RT in improving outcomes for patients with stage III disease. She noted that patient selection and treatment have changed considerably since the publication of the PORT meta-analysis. In addition, neoadjuvant or adjuvant chemotherapy has become the standard of care; surgical and RT techniques have improved; and positron emission tomography scanning is being used to refine patient selection.

Prof Le Péchoux summarized the results of 4 large randomized trials of neoadjuvant RT and 2 of PORT in patients with stage III NSCLC, none of which showed a survival benefit (Table 1). In the most recently completed trial—SAKK 16/00 [Pless M et al. Ann Oncol. 2014; NCT00030771]—RT increased rates of response, complete resection, and pathologic complete response but failed to improve local control, event-free survival, or overall survival. In a 2005 update of the PORT meta-analysis, there was an 18% relative increase in the risk of death for PORT when compared with surgery alone [Burdett S, Stewart L. Lung Cancer. 2005]. A 2013 update using new statistical methodology found that the effect of PORT varied by stage and nodal status [Burdett et al. Lung Cancer. 2013], but Prof Le Péchoux stated that this analysis was underpowered.

Although findings from meta-analyses do not support a role for RT after complete resection, excess toxicity, poor local control, and the use of older techniques may have contributed to excess mortality, she said. Population-based studies have found superior 5-year survival in completely resected N2 patients who received PORT when compared with patients who received surgery alone or surgery plus chemotherapy [Mikell JL et al. J Thorac Oncol. 2015; Robinson CG et al. J Clin Oncol. 2015], but these studies cannot be considered robust evidence in favor of PORT, she said.

Both preoperative RT and PORT, however, have been shown to reduce local recurrence rates [Pless et al. Ann Oncol. 2014; Le Péchoux C. Oncologist. 2011]. Technical advances may enhance the ability of RT to improve local relapse-free survival, disease-free survival, and possibly overall survival. New evidence to reassess the role of PORT using modern radiation techniques may emerge from the international randomized LUNG ART trial [NCT00410683], Prof Le Péchoux said. In this ongoing phase 3 study, an expected enrollment of 700 patients with completely resected N2 NSCLC will be randomized to receive conformal mediastinal PORT or no PORT. Patients may also receive neoadjuvant or adjuvant chemotherapy. The trial’s primary end point is disease-free survival.

Paul De Leyn, MD, PhD, University Hospitals Leuven, Leuven, Belgium, offered a surgeon’s perspective on improving long-term outcomes for patients with stage III disease. Definitive upfront stratification of tumors—as resectable, potentially resectable with an increased risk of incomplete resection, or unresectable—is crucial, he said. According to Prof De Leyn, good survival rates and acceptable morbidity and mortality have been achieved with induction chemotherapy or chemoradiotherapy in selected patients with potentially resectable tumors with an increased risk of incomplete resection.

Prof De Leyn reviewed the evidence from several studies involving patients with potentially resectable N2 disease and one study of patients with unresectable disease, all of whom underwent surgery following induction chemotherapy or chemoradiation therapy (CRT; Table 2). In the German Lung Cancer Cooperative Group study [Thomas M et al. Lancet Oncol. 2008], preoperative CRT increased postsurgical mortality compared with preoperative chemotherapy, primarily due to increased rates of empyema and bronchial insufficiency. The evidence does not support a role for induction CRT for N2 disease, Prof De Leyn said. For Pancoast tumors, however, induction CRT is the standard of care. For stage III tumors deemed unresectable at baseline assessment, the EORTC 8947 trial [van Meerbeeck JP et al. J Natl Cancer Inst. 2007] demonstrated that induction chemotherapy will not render an unresectable tumor resectable, Prof De Leyn said. He recommended that patients whose tumors are deemed unresectable should receive immediate CRT.

Both types of surgery and hospital surgical volume have been shown to influence outcomes. In a systematic review and meta-analysis of 27 studies published between 1990 and 2010, right pneumonectomy following neoadjuvant therapy was associated with significantly higher 30-day (P = .02) and 90-day (P = .03) mortality compared with left pneumonectomy; 90-day mortality for all pneumonectomies was also higher than 30-day mortality [Kim et al. J Thorac Cardiovasc Surg. 2012]. Prof De Leyn noted that these findings highlight the need for careful patient selection for pneumonectomy and reporting of 90-day mortality.

In regard to surgical volume, Bach and colleagues reported in 2001 that 44% of patients who had surgery at the highest volume centers survived 5 years postsurgery, compared with 33% of patients treated at the lowest volume centers (P < .001). A 2013 analysis of data on > 134 000 patients with NSCLC diagnosed in England between 2004 and 2008 found that high procedure volume was strongly associated with improved survival, a higher resection rate, and a higher percentage of resections in patients with higher levels of comorbidity [Lüchtenborg M et al. J Clin Oncol. 2013].

In conclusion, evidence reviewed in this session showed that unanswered questions still surround the role of PORT in the treatment of patients with stage III NSCLC, that the role of induction CRT is dependent on tumor resectability, and that type of surgery and hospital surgical volume are important factors in surgical outcomes.