Optimal radiation dosing in concurrent neoadjuvant chemoradiation for resectable esophageal cancer: a meta-analysis

This is the first meta-analysis to study optimal radiation dose in the setting of concurrent neoadjuvant chemoradiotherapy (cnCRT) for esophageal cancer (EC). We sought to compare outcomes between high dose radiotherapy (HDRT) [>48.85 Gy biologically effective dose (BED)] group and low dose radiotherapy (LDRT) (≤48.85 Gy BED) for patients with EC receiving cnCRT.

Medline, Embase, and Cochrane databases were searched independently by two members of our team on August 07, 2017. Articles were screened using Covidence. Study quality was assessed via CONSORT. Eligible studies had to be randomized controlled trials (RCT) comparing cnCRT vs. surgery alone in full-text English. Those with induction or sequential chemoradiotherapy were excluded. We captured data points including radiation dose, hazard ratios (HRs) for overall survival (OS), and treatment-related mortality (TRM). We analyzed HRs for OS and risk ratio (RR) for TRM and corresponding 95% confidence interval (CI) as the summary statistic. We used both fixed- and random-effects models in the presence of heterogeneity. The primary outcome was OS; secondary endpoint was treatment related mortality (TRM). We compared outcomes by HDRT vs. LDRT. To minimize chemotherapy heterogeneity, we performed a pre-planned analysis excluding the CROSS trial.

The eleven included studies contained a total of 1,697 patients. Eight hundred forty-eight were randomized into the cnCRT. Of these 848 patients, 287 received HDRT and 561 received LDRT. HR for OS was not statistically different between LDRT (HR 0.67; 95% CI, 0.55-0.8) and HDRT (HR 0.68; 95% CI, 0.45-0.91). Excluding the CROSS trial, there was still no difference in outcomes between LDRT and HDRT. TRM was similar between LDRT and HDRT.

With no difference in OS or TRM between LDRT and HDRT, 48.85 Gy BED cnCRT may be a sufficient radiation dose for cnCRT for patients with EC fit for surgery.