Identifying the optimal treatment strategy in patients with resectable non-cardia gastric cancer.

BACKGROUND

Multimodal treatment strategy including perioperative chemotherapy (PEC), postoperative chemoradiation therapy (POCR), and postoperative chemotherapy (POC) has been accepted as the standard of care in gastric cancer (GC). The ideal sequence and type of therapy remain undetermined.

METHOD: The National Cancer Database was examined from 2006 to 2016 to identify patients with resectable non-cardia gastric cancer. Patient outcomes were compared based on the receipt of PEC, POCR, and POC. This comparison was repeated in a sub-group of patients who received optimal treatment. Optimal treatment was defined as initial chemotherapy within 45 days of diagnosis, resection within 45 days of diagnosis, negative margins, adjuvant chemotherapy within 90 days of resection and standard radiation dose (45 Gy). Kaplan-Meier test, log-rank test, and multivariable analysis (MVA) were performed.

RESULTS

We identified 9589 patients. Median survival was greater in the PEC group followed by POCR and POC (60.6, 42.3, and 31.2 months, respectively). On MVA, factors associated with worse overall survival included age above median (≥ 63 years), Charlson-Deyo score of ≥ 1, non-academic/research program, poorly differentiated/undifferentiated grade, positive margins, and positive lymph nodes. Both PEC and POCR were associated with improved survival when compared to POC (HR 0.78 and 0.79; p < 0.001). When compared with PEC, no significant difference was noted with POCR (HR 1.01; p = 0.987). These results were maintained in optimally treated cohort (n = 3418).

CONCLUSION

In patients with resectable non-cardia gastric cancer, both perioperative chemotherapy and postoperative chemoradiation therapy were associated with improved survival when compared to postoperative chemotherapy. No difference was noted between perioperative chemotherapy and postoperative chemoradiation therapy. These results were maintained in the optimally treated cohort.