Palliative external beam radiotherapy for the treatment of tumor bleeding in inoperable advanced gastric cancer

BACKGROUND: To assess the outcomes and prognostic factors associated with palliative external beam radiotherapy (EBRT), administered to patients with advanced gastric cancer. METHODS: Forty-two patients with bleeding gastric tumors that received EBRT for palliation were analyzed. The response to EBRT was assessed by the palliation of tumor bleeding. Patients were classified as either responders, or non-responders to EBRT. The prognostic utility of clinical and dosimetric variables was examined in a multivariate logistic regression model. The optimal dose cutoff to classify the two groups was determined with receiver operating characteristic analysis. RESULTS: The palliation of gastric tumor bleeding after EBRT was achieved in 29 patients (69.0%). The time to resolve tumor bleeding ranged from 1 to 84 days (median, 15 days). The median duration of palliation was 14.9 weeks. The median EBRT dose was 40 Gy in responders vs. 21 Gy in non-responders, with the difference being significant (p < 0.001). The biologically effective dose (using α/β = 10, BED(10)) for responders was significantly higher than the BED(10) for non-responders (median 48 Gy vs. 26.4 Gy, p < 0.001), and the optimal cut off value to separate the two groups was 36 Gy (p < 0.001). The absence of distant metastasis and the use of concurrent chemotherapy generally showed a better EBRT response (p = 0.079 and p = 0.079, respectively). In the multivariate analysis, BED(10) ≥ 36 Gy was the most significant factor associated with EBRT response (p = 0.001). Overall survival (OS) and re-bleeding-free survival was median 12.6 weeks and 14.9 weeks. The responders to EBRT showed superior OS (16.6 vs. 5.1 months, p < 0.001). Neither acute nor chronic toxicities of grade 3 or higher were observed. CONCLUSIONS: EBRT is an effective method for treating tumor bleeding in advanced gastric cancer, and does not induce severe toxicity.