Surgery versus SABR for resectable non-small-cell lung cancer - Authors' reply - Part 9 of 9

This is the authors' reply to the letters submitted to The Lancet Oncology. They begin by stating that the strength of their analysis was that patients were randomised, thereby avoiding selection bias. With regards to low accrual, one major factor was failure of thoracic surgeons to participate. In addition, stage I NSCLC is relatively uncommon. They state that the median follow-up was long enough to discover recurrences and that there were enough patients to accurately analyze the data from a statistical standpoint. For patients in the ROSEL trial who did not undergo biopsy, clinical staging via PET scan has been shown to have a low false positive rate in the Netherlands. When considering VATS versus thoracotomy, cancer outcomes are not improved by VATS, readmission for complications of VATS is similar to thoracotomy, and nodal upstaging for clinical stage I NSCLC is more common after thoracotomy, although extent of lymph node dissection does not affect outcomes. Therefore the possible superiority of VATS may not lead to a clinically significant improvement in the surgical treatment of these patients. With regards to comorbidities, all patients underwent cardiopulmonary testing to determine their surgical candidacy. Zhang and colleagues had disagreed with the use of SABR and EBUS for concern that lymph node involvement would be understaged and those patients would miss out on adjuvant therapy. The original article showed no difference in regional recurrence between the two groups. Opitz and colleagues graded the original article, but Chang and colleagues replied that the randomised design of the two trials translates to a higher level of evidence than prior studies. Chang clarifies that the 30-day surgical mortality was 0%, but the 90-day mortality was 3.7% which is less than recent published literature. The authors respond to Dearman and colleagues' suggestion regarding NHS England and CtE by stating that there is already a considerable amount of data comparing SABR versus surgery. The authors finish by stating that they continue to support their conclusion that SABR is a non-invasive standard treatment alternative to surgery for patients unable to undergo surgery and should also be considered a viable treatment option for patients who are surgical candidates.