Prognostic impact of neoadjuvant chemoradiation in cT3 oesophageal cancer - A propensity score matched analysis.

Background: The prognostic effect of neoadjuvant treatment in advanced oesophageal cancer is still debated because most studies included undefined T-stages, different radio/chemotherapies or different types of surgery.

Objectives: To analyse the prognostic impact of neoadjuvant chemoradiation in patients with clinical T3 oesophageal cancer and oesophagectomy.

Methods: In a retrospective study 768 patients from two centres with cT3/Nx/M0 oesophageal cancer and transthoracic en-bloc oesophagectomy were selected. Clinical staging was based on endoscopy, endosonography and spiral-CT scan. Propensity score matching using histology, location of tumour, age, gender and ASA-classification identified 648 patients (n=302 adenocarcinoma (AC), n=346 squamous cell carcinoma (SCC)) for the intention-to-treat analysis comparing group-I (n=324) patients with planned oesophagectomy and group-II (n=324) patients with planned neoadjuvant chemoradiation (40Gy, 5-FU, cisplatin) followed by oesophagectomy. The prognosis was analysed by univariate and multivariate analyses.

Results: In the intention-to-treat analysis group-I had a 17% and group-II a 28% 5-year survival rate (5-YSR) (p<0.001). After excluding patients without oesophagectomy the 5-YSR of group-II increased to 30%. The results were more favourable for patients with AC (5y-SR of 38%) compared to SCC (22%) (p=0.060). In group-II patients with major response (n=128) had a 41% 5-YSR compared to 20% for those with minor response (n=155, p<0,001). In multivariate analysis neoadjuvant chemoradiation was a favourable independent prognostic factor.

Conclusion: Neoadjuvant chemoradiation followed by oesophagectomy results in 11% higher 5-YSR than surgery alone for patients with cT3/Nx/M0 oesophageal cancer. This effect is due to the substantial prognostic benefit of the major responders.