P-354 Association between tumor regression with tumor location and inflammatory biomarkers in patients with resectable gastric cancer and gastroesophageal junction cancer who underwent perioperative chemotherapy: Multicentric data from Peru

Abstract

FLOT perioperative chemotherapy is the standard treatment in patients with resectable gastric and gastroesophageal junction (GEJ) cancer. There is a lack of information if tumor location or inflammatory biomarkers are associated with tumor regression after chemotherapy treatment. The aim of this study is to determine associated factors with tumor regression after FLOT therapy. This is a retrospective observational study of a single patient cohort. Clinical records of patients with clinical stage II and III gastric and GEJ cancer were recollected, since January 2018 to December 2019, in ALIADA cancer center and Rebagliati Hospital, in Peru. Clinical stage was determinate using AJCC 8th edition and was evaluated before therapy and after surgery to determine tumor regression. Tumor location was evaluated in endoscopy reports; meanwhile NLR and PLR were calculated from blood counts before the beginning of chemotherapy. Clinical records with incomplete data were excluded. The bivariate statistical analysis was performed with Chi-squared test, Odds ratio and logistical regression was performed with a confident level of 95% (p< 0.05). 52 clinical records were evaluated and finally 41 were included. 63% were male and the mean age was 58 years. All patients had adenocarcinoma type histology, 65% of them had diffuse subtype and the rest intestinal subtype. The 63% of patients had gastric cancer distributed as follows: 50% in corpus, 35% in antrum, 10% in cardia and 5% fundus. 37% of patients had GEJ cancer; the 53% of them were siewert II and 47% siewert III. The 90% of patients had clinical stage III and 10% clinical stage II. All patients received four cycles of FLOT and the 82% were operated. 78% of patients had tumor regression after chemotherapy; 12% of them achieved pathological complete response. The 17% of patients had stable disease and 24% had progression of disease, 71% of them progressed to peritoneum. Diarrhea (27%) and emesis (20%) were the most frequent adverse events presented in our population meanwhile neutropenia (8%) and anemia (5%) were the most frequent grade III-IV adverse events. There was no statistic difference between patients who had tumor regression compared to patients who did not have response to chemotherapy with respect to mean NLR (2.61 ± 0.27 vs 2.84 ± 0.31, p>0.05) and mean PLR (2.61 ± 0.27 vs 2.84 ± 0.31, p>0.05). Tumor regression was more frequent in patients with gastric cancer (69% vs 63%, p=0.02). There was no statistic difference in the multivariate analysis (OR = 148, IC95% = 0.34-6.40). Tumor regression was associated with tumor location in bivariate analysis, being more frequent in patients with gastric cancer; however there was no statistic difference in multivariate analysis. Studies with bigger amount of patients are needed to validate this association.