Phase III study of concurrent chemortherapy and radiotherapy (CT/RT) vs. CT/RT followed by surgical resection for stage IIIA(pN2) non-small cell lung cancer (NSCLC): Outcomes update of North American Intergroup 0139 (RTOG 9309)

Reviewer: Ryan Smith , MD
Last Modified: May 16, 2005

Presenter: Kathy Albain
Presenter's Affiliation: RTOG
Type of Session: Scientific


  • Patients with stage IIIA NSCLC and N2 (mediastinal) disease have a poor outcome, with long term overall survival consistently reported to be approximately 20%
  • Definitive concurrent chemoradiation is the standard of care in good performance status patients
  • Surgical resection has been demonstrated to possibly improve on these results, but with the cost of increased morbidity and mortality, so there is debate as to whether or not surgery adds any benefit in these patients
  • This study investigates definitive chemoradiation vs. induction chemoradiation followed by surgical resection
  • Results were previously reported, showing an increased progression free survival (PFS) but no change in overall survival (OS)
  • This report presents updated results with longer follow up

Materials and Methods

  • Eligible patients included those with T1-3pN2M0 NSCLC, with ECOG performance status of 0-2
  • Resection was required to be technically feasible with preoperative FEV1 > 2 L or predicted postoperative FEV1 > 0.8 L
  • Patients were stratified by T stage and performance status
  • All patients were randomized initially, then underwent identical induction therapy consisting of thoracic radiation to 45 Gy with concurrent chemotherapy:  cisplatin 50 mg/m2 days 1, 8, 29, and 36 and etoposide 50 mg/m2 days 1-5 and days 29-33
  • Those randomized to the surgical arm underwent resection if there was no progressive disease
  • Those randomized to chemoradiation continued to 61 Gy with 2 additional doses of chemotherapy, without treatment break
  • Median follow up was 81 months, with all patients followed at least 2.5 years


  • 88% of patients randomized to surgery were eligible and 81% underwent resection, with complete responses to induction of 71%
  • 92% of patients randomized to CT/RT were eligible for definitive CT/RT and 80% received full dose of radiation
  • Grade 3/4 toxicities from CT/RT were fairly similar, with the exception of esophagitis rates of 44% in the definitive CT/RT arm vs. 20% in the surgical arm
  • Treatment related death rate in the surgical arm was 7.9%, with 14/15 of these deaths coming from patients who underwent pneumonectomy (26% death rate in patients requiring pneumonectomy).  The majority of these were right sided cases, and the major cause was ARDS
  • Treatment related deaths in the CT/RT arm was 2.1%
  • Patients had an increased PFS in the surgical arm (5 year PFS of 22% vs. 11% and Median PFS of 12.8 months vs. 10.5 months)
  • Median OS was 23.6 months (surgical arm) vs. 22.2 months and 5 year OS of 27% vs. 20% (p=0.24)
  • In a matched control analysis, investigating pneumonectomy patients with matched CT/RT patients, pneumonectomy patients had a poorer survival, with median survival times of 19 months compared to 29 months
  • Patterns of failure in the surgical arm was 10% local failure only and 37% distant metastases vs. 22% local failure and 42% distant metastases in the CT/RT arm
  • Multivariate analysis showed weight loss, male gender, and >1 nodal station involved were predictive for worse survival, but treatment arm was not

Author's Conclusions

  • Longer follow up confirms improved PFS in the surgical arm, but with no improvement in OS
  • There was a trend for increased OS in the surgical arm, with 7% improvement at 5 years
  • N0 status at surgery predicts for better survival
  • Surgical resection after induction CT/RT should be considered in patients with N2 disease if pneumonectomy is not required

Clinical/Scientific Implications

The treatment of stage IIIA NSCLS is controversial.  Standard of care is definitive chemoradiation, though many have advocated surgery in attempts to improve the results.  This study shows an increased PFS but without an effect on OS.  It is easy to see that there will never be an OS benefit in this study because of the number of postoperative deaths in the surgical arm.  However, all but one death occurred in patients who had undergone pneumonectomy, with only 1 death out of 98 lobectomy patients.  This indicates that in carefully selected patients, surgery following induction chemoradiation may improve the results.  However, the key is the selection, and it appears that patients requiring pneumonectomy would be poorly served by attempts at surgical resection. Optimal treatment for this heterogeneous disease will likely continue to be debated.