BackgroundAtrophic gastritis can be classified according to characteristic mucosal patterns observed by Blue LASER Imaging (BLI) in a medium-range to distant view.AimsTo facilitate the endoscopic diagnosis of Helicobacter pylori (HP)-related gastritis, we investigated whether atrophic mucosal patterns correlated with HP infection based on the image interpretations of three endoscopists blinded to clinical features.MethodsThis study included 441 patients diagnosed as having atrophic gastritis by upper gastrointestinal endoscopy at Nishikawa Gastrointestinal Clinic between April 1, 2015 and March 31, 2016. The presence/absence of HP infection was not taken into consideration. Endoscopy was performed using a Fujifilm EG-L580NW scope. Atrophic mucosal patterns observed by BLI were classified into Spotty, Cracked and Mottled. Image interpretation results were that 89, 122 and 228 patients had the Spotty, Cracked and Mottled patterns, respectively, and 2 patients an undetermined pattern. Further analyses were performed on 439 patients, excluding the 2 with undetermined patterns.ResultsThe numbers of patients testing negative/positive for HP infection in the Spotty, Cracked and Mottled pattern groups were 12/77, 105/17, and 138/90, respectively. The specificity, positive predictive value and positive likelihood ratio for endoscopic diagnosis with positive HP infection based on the Spotty pattern were 95.3%, 86.5% and 8.9, respectively. In all patients with the Spotty pattern before HP eradication, the Cracked pattern was observed on subsequent post-eradication endoscopy.ConclusionsThe Spotty pattern may represent the presence of HP infection, the Cracked pattern, a post-inflammatory change as seen after HP eradication, and the Mottled pattern, intestinal metaplasia.

No abstract

PURPOSE ERBB2 copy number (CN), measured using next-generation sequencing, is a predictive biomarker for trastuzumab efficacy in human epidermal growth factor receptor 2 (HER2)–positive advanced esophagogastric and gastric cancer (AGC). We aimed to investigate the association of ERBB2 amplification and gene coalterations with response and resistance to trastuzumab-combined chemotherapy. METHODS The SCRUM-Japan GI-SCREEN was a comprehensive genomic profiling project of GI cancer tissues using Oncomine Cancer Research Panel and Oncomine Comprehensive Assay. From 885 patients with AGC who successfully underwent gene profiling, 74 with ERBB2 amplification (CN ≥ 4.0) and who received first-line trastuzumab-combined chemotherapy were selected, and ERBB2 CN and gene coalterations were assessed. RESULTS ERBB2 CN did not differ in tumor response to trastuzumab-combined chemotherapy (one-way analysis of variance test, P = .37). Multivariate analysis using the Cox proportional hazard model revealed that ERBB2 CN (continuous log2-converted CN, hazard ratio, 0.76; 95% CI, 0.62 to 0.93; P < .01) and receptor/oncogene amplifications in the HER2 signaling pathway (hazard ratio, 2.5; 95% CI, 1.2 to 5.3; P = .01) were significant predictors for progression-free survival (PFS). ERBB2 variants coexisted in five patients (7%) and were missense mutations. Two patients with low variant allele frequencies (VAFs; 8%, 12%) showed high ERBB2 CN (55, 80) and durable response (≥ 20 months), whereas three patients with high VAFs (66%–90%) showed low ERBB2 CN (8-11) and no response with short PFS (1-10 months). CONCLUSION ERBB2 CN and gene coamplification in the HER2 signaling pathway were positive and negative predictors of PFS in trastuzumab-treated HER2-positive AGC patients, respectively. HER2-positive AGC patients with a high VAF of ERBB2 showed poor outcomes and may need HER2 tyrosine kinase inhibitors and trastuzumab deruxtecan.

No abstract

480 Background: FOLFIRINOX has been shown superior to GEM in terms of OS, PFS and/or RR for treatment of metastatic PC in Accord 11, and this regimen has become the standard of care for the first-line chemotherapy. However, the major disadvantage of FOLFIRINOX is significantly higher toxicity, and, in particular, quite higher incidence of neutropenia(78%) and febrile neutropenia(22%) was observed in the PII clinical trial of FOLFIRINOX performed in Japan (LOHP-PII-05, JSCO 2013). Thus, in clinical practice for the Japanese patients, it should be needed to make modification on this regimen in terms of dose reduction without compromising in its effectiveness. We evaluated the safety and efficacy of modified(m) FOLFIRINOX in Japanese patients with unresectable locally advanced PC(LAPC) and/or metastatic PC(MPC). Methods: The modifications to FOLFIRINOX include discontinuation of the bolus 5-FU and dose-reduction(25%) of irinotecan. 20 patients with unresectable LAPC and/or MPC (PS 0-1) were treated with m FOLFIRINOX given every 2 weeks, and toxicity and efficacy were evaluated in these patients who received more than 3 courses. All patients did not receive prophylactic filgrastim. Results: Patient characteristics were as follows: LAPC/MPC 4/9; median age 65 yrs(range 39~75); male/female 15/5. Median number of cycles was 5.5 (range 1~17). Grade 3/4 toxicities included neutropenia(25%), nausea(30%), anorexia(35%) and diarrhea(20%). Grade 1/2 toxicities were sensory neuropathy(45%) and vomiting(40%). Febrile neutropenia was noted in 2 patients(10%). Toxic death was not observed, and 12 patients(60%) required dose reduction due to toxicity. Hematologic toxicities were decreased, however, those concerned with gastrointestinal toxicities were observed in a bit higher incidence compared to those in LOHP-PII-05. Response by CA19-9 was evaluable in 11 patients. 9 of 11 patients (82%) showed a decline in CA19-9 levels from baseline, and a greater than 50% decline was noted in 5 of 11 patients(45%). Conclusions: m FOLFIRINOX seems to have an improved safety profile even without prophylactic filgrastim and to maintain its efficacy.