IMPACT OF NEOADJUVANT THERAPY TIMING ON SHORT- AND LONG-TERM SURVIVAL FOR GASTRIC ADENOCARCINOMA PATIENTS
Kurt S. Schultz*, Susanna W. de Geus, Teviah Sachs, Michael Cassidy, Sing Chau Ng, David McAneny, Jennifer F. Tseng
Boston Medical Center; Boston University School of Medicine, Boston, MA
Introduction: For several solid tumors, such as esophageal, pancreas, and rectal cancer, longer time between neoadjuvant chemoradiation (nCRT) and surgery increases the rate of pathologic complete response (pCR). pCR is associated with increased survival, but this survival benefit has not been shown with longer nCRT-surgery time intervals for these cancers. The purpose of this study was to determine the impact of nCRT-surgery time intervals on short and long-term overall survival for gastric cancer patients. Secondary objectives were to determine the effect of nCRT-surgery time intervals on pCR and negative resection margins.
Methods: A retrospective study was conducted of gastric adenocarcinoma patients who received nCRT prior to gastrectomy from 2006-2014 using the National Cancer Database. Patients were grouped into the following nCRT-surgery terciles: 15-42 days, 43-55 days, and 56-90 days. Multivariate analyses were performed to determine clinicopathologic factors that predicted 90-day mortality. The Kaplan-Meier method and the Cox proportional hazards model were used to determine the effect of nCRT-surgery time interval and other independent factors on overall survival.
Results: Of 3,053 gastric cancer patients who received nCRT (2,579 men and 474 women, median age: 62.0 years), the mean nCRT-surgery time interval was 50.6 +/- 15.2 days (median, 49 days). On multivariate analyses, longer nCRT-surgery time intervals did not negatively affect 90-day mortality (tercile 2: p=0.677; tercile 3: p=0.060, ref: tercile 1) or the rate of negative resection margins (tercile 2: p=0.308; tercile 3: p=0.258, ref: tercile 1). Longer nCRT-surgery time intervals were associated with increased rates of pCR (tercile 3 vs. tercile 1: odds ratio, 1.33; 95% confidence interval [95% CI], 1.00-1.76; p=0.047). Unadjusted, the median overall survival for terciles 1-3 was 36.6, 37.5, and 31.9 months, respectively (log-rank, p=0.030) (Figure 1). Cox proportional hazards modeling demonstrated a higher hazard of death for tercile 3 vs. tercile 1 (hazard ratio, 1.16; 95% CI, 1.02-1.32; p=0.020).
Conclusions: Longer time intervals between neoadjuvant therapy and surgery resulted in increased rates of pCR without affecting short-term overall survival. However, long-term overall survival was worse for nCRT-surgery time intervals > 55 days. Future studies may be warranted to determine the optimal timing of surgery following nCRT for gastric cancer patients.
Figure 1. Overall survival based on the time interval between neoadjuvant chemoradiation and gastrectomy
Back to 2020 Abstracts