eComment. Should persistent N2/N3 non-small cell lung cancer be treated by surgery?

We have read with interest the paper by Steger et all. [1] on the efficacy of surgery in persistent N2/N3 non-small cell lung cancer (NSCLC) after induction therapy, and we are very surprised by some questionable authors' choices.

First of all, they treated stage IIIB-N3 patients with trimodal therapy, despite the recent indications to administer only chemoradiotherapy (CRT) and avoided surgery (National Comprehensive Cancer Network guide-lines version 3, 2012). Then, they operated on patients with persistent N2 and even N3-staged cancer, despite the adverse consensus [2]. Many trials demonstrated that surgery should be limited to IIIA-N2 patients with evident nodal down-staging after induction CRT due to poor prognosis in persistent N2 [2] and surgical complications related to induction therapy. Recently Gomez-Caro [3], reporting his surgical experience in patients after induction therapy, suggested to avoid right pneumonectomy due to high postoperative complications. Moreover, he confirmed that down-staging of mediastinal nodes is the strongest predictor of survival and that incomplete resection offers no benefit. That being said, invasive restaging is considered mandatory prior to surgery [4].

Surprisingly Steger et al. [1] excluded only patients with disease progression after induction CRT from surgery and treated all patients with resectable tumours, both N2 and N3. They never performed surgical restaging and, in fact, operated on 30 patients with persistent N2, and 5 patients with persistent N3. Lastly, they did not administer adjuvant therapy in any patient. Furthermore, we were also surprised that 44 pneumonectomies were performed on 83 procedures despite the well-known complications after induction therapy. Anyway, we congratulate the authors for the hospital mortality, which was 2.4% and the acceptable morbidity rate (58%). Focusing on the results, ypN2/N3 had a significantly worse prognosis than ypN0/N1, confirming the common experience. We also note that initial T-stage and N-stage were not significant survival indicators. Again, these evidences should have led the authors to conclude that surgical restaging is essential in prognosis determination and must be always performed. Nevertheless, the authors defended surgery in persistent N2/N3, since median progression-free interval was 17 months, and 13 patients had no recurrences (follow-up not specified). They attributed this last point to surgery. They also reported that overall outcomes in ypN2/N3 are consistent with those of patients treated with definitive CRT, concluding that surgery does not worsen overall prognosis but in singular cases may improve it. Their conclusions could prompt us to reconsider the role of surgery in advanced NSCLC. However, we must underline that their overall outcomes are not encouraging. It would be more interesting to investigate why a small group of ypN2/N3 presented with a better prognosis, in order to identify further prognostic criteria [5].

Finally, we disagree in considering N2/N3 patients a homogeneous population with regard to prognosis and underline that current guidelines recommend adjuvant therapy after induction therapy, especially in cases of residual tumours.

Conflict of interest: none declared