Active surveillance versus standard surgery after neoadjuvant chemoradiotherapy for oesophageal cancer: Long-term results of a cluster-randomised controlled trial (SANO)

SANO

Patient / Population Intervention / Exposition Vergleich Endpunkt

In patients with oesophageal cancer and a complete clinical response after neoadjuvant chemoradiotherapy, active surveillance showed no statistically significant difference in five-year overall survival compared with standard surgery.

N

309Patienten

Design

Multicentre stepped-wedge cluster-randomised non-inferiority trial (2017 to 2021) in patients with locally advanced oesophageal cancer and a complete clinical response after neoadjuvant chemoradiotherapy, comparing active surveillance with standard surgery (NTR-6803).

Endpunkt

Five-year overall survival in this long-term report. The trial's original non-inferiority comparison was two-year overall survival. Five-year disease-free survival and the cumulative incidence of distant metastases were also assessed.

ErgebnisFive-year overall survival 50% (95% CI 34-57) with active surveillance vs 54% (95% CI 46-64) with standard surgery; HR 1.17 (95% CI 0.77-1.77). Disease-free survival HR 1.39 (95% CI 0.94-2.06). Five-year cumulative incidence of distant metastases 46% (95% CI 39-53) vs 29% (95% CI 21-37); P<.01. In total, 85 of 198 patients (43%) under active surveillance were spared non-beneficial oesophagectomy. Median follow-up 76 months (IQR 67-86).

Gangaram-Panday SS, Van der Wilk BJ, Lagarde SM, et al. Presented at ESMO Congress 2026, abstract LBA78, 23 October. Not yet published.

Diskussion & Kritik

With a median follow-up of 76 months, this long-term analysis found five-year overall survival not statistically significantly different between active surveillance and standard surgery, and 43% of the surveillance group avoided non-beneficial oesophagectomy. A non-significant difference is not proof of equivalence: the overall survival hazard ratio of 1.17 has an upper confidence limit of 1.77, and the abstract states no non-inferiority margin for the five-year analysis. The cumulative incidence of distant metastases was significantly higher with active surveillance (46% vs 29%), and the disease-free survival hazard ratio of 1.39 had an interval reaching 2.06, which the abstract does not explain. Allocation was by stepped-wedge cluster randomisation rather than individual randomisation, and the abstract reports no quality-of-life, salvage surgery or safety data for this follow-up.