Background A regional cancer hospital has been identified to be crucial in the management of malignancies of undefined primary origin (MUO) and cancer of unknown primary (CUP). This hospital primarily consists of oncologists with expertise in CUP, pathologists, and interventional radiologists. Early consultation or referral of MUO and CUP to a cancer hospital is deemed important. Methods This study retrospectively collected and analyzed the clinical, pathological, and outcome data of all patients (n = 407) referred to the Aichi Cancer Center Hospital (ACCH) in Japan over an 8-year period. Results In total, 30% of patients were referred for a second opinion. Among 285 patients, 13% had non-neoplastic disease or confirmed primary site and 76% had confirmed CUP (cCUP), with 29% of cCUP being identified as favorable risk. In 155 patients with unfavorable-risk CUP, 73% had primary sites predicted by immunohistochemistry (IHC) and distribution of metastatic sites, whereas 66% of them received site-specific therapies based on the predicted primary sites. The median overall survival (OS) was found to be poor in patients with MUO (1 month) and provisional CUP (6 months). In addition, the median OS of 206 patients with cCUP treated at the ACCH was 16 months (favorable risk, 27 months; unfavorable risk, 12 months). No significant difference was noted in OS between patients with non-predictable and predictable primary-sites (13 vs 12 months, p = 0.411). Conclusion The outcome of patients with unfavorable-risk CUP remains to be poor. Site-specific therapy based on IHC is not recommended for all patients with unfavorable-risk CUP.

Background: Metastatic small bowel adenocarcinoma (mSBA) is a rare condition with a poor prognosis. Enough data are not available on second-line chemotherapy (SLC) for mSBA. The aim of this study was to explore the efficacy and safety of SLC for mSBA patients.

Results: sT+ and pT+ was observed in 225 patients (44.9%) and in 132 patients (26.3%), respectively. Regarding the concordance between sT and pT, sT+/pT+ was observed in 108 patients (21.6%), sT+/pT-in 117 patients (23.4%), sT-/pT+ in 24 patients (4.8%), and sT-/pT-in 252 patients (50.3%), respectively. Both sT and pT were selected as independent risk of peritoneal recurrence, with the hazard ratio of 5.04 (95% CI, 2.15-11.83) in sT and 5. 09 (2.30-11.30) in pT by multivariable analysis. Multivariable analysis showed both sT and pT were independent factors for RFS with hazard ratio of 2.09 (1.52-2.87) in sT and 2.14 (1.54-2.99) in pT. The 5-year RFS according to the subgroup was 50.0% in sT+/pT+, 63.2% in sT+/pT-, 62.5% in sT-/pT +, and 81.7% in sT-/pT-. Survival in sT+/pT-was almost identical to that in sT-/pT+.Conclusions: Macroscopic serosal invasion diagnosed at the time of surgery showed a similar predictive value for peritoneal recurrence as pathologically diagnosed serosal invasion. Establishment of novel staging system incorporating macroscopic serosal invasion is warranted.Clinical trial identification: NCT00112099.

Objectives: Insufficient oral intake in advanced gastric cancer (AGC) limits the use of several drugs. We aimed to determine the oral intake status of patients with AGC during later-line chemotherapy. Materials and Methods:We retrospectively evaluated data of patients with AGC who experienced disease progression during first-line chemotherapy administered from January 2012 to December 2018 in a single institution. We defined "insufficient oral intake" as requiring daily intravenous fluids or hyperalimentation. Multivariate logistic regression was performed to identify oral intake-related factors.Results: Among 589 included patients, at disease progression during firstline, second-line, and third-line chemotherapy, 78.3% (461), 53.3% (314), and 30.4% (179) of patients, respectively, exhibited sufficient oral intake. Fourth-line chemotherapy was initiated for 22.2% (131) of patients, with 20.0% (118) exhibiting sufficient oral intake. During second-line and thirdline chemotherapy, 11/67 (16%) and 2/39 (5%) patients, respectively, exhibited improvements in oral intake; 85/428 (19.9%) and 70/259 (27.0%), respectively, exhibited deteriorations in oral intake. Factors correlated to deterioration in oral intake during second-line chemotherapy were poor Eastern Cooperative Oncology Group Performance Status (odds ratio, 4.32; P < 0.001), moderate or severe ascites (1.96; P = 0.045), peritoneal metastasis (2.12; P = 0.029), prior palliative surgery (3.41; P = 0.003), and high neutrophil-to-lymphocyte ratio (3.09; P < 0.001); those correlated to deterioration in oral intake during third-line chemotherapy were poorly differentiated pathology (2.52; P = 0.025) and high neutrophil-to-lymphocyte ratio (2.65; P = 0.006). Conclusion:As later-line chemotherapy is ineffective in improving oral intake in patients with AGC, careful adaptation of regimens is required for patients at risk for impaired oral intake.

126 Background: Sarcopenia and muscle loss during chemotherapy (Cx) have a poor prognosis in metastatic colorectal cancer (mCRC). It is unclear whether cachexia during Cx has a survival impact. Methods: mCRC patients (pts) receiving first-line Cx at a single institution between Jan 2010 and Jun 2018 were retrospectively evaluated. Pts receiving doublet Cx with bevacizumab or anti-EGFR agents, ECOG Performance Status (PS) 0–2, and abdominal computed tomography (CT) before and after initiating first-line Cx at least once were included and classified pts into those with (cachexia group) or without (non-cachexia group) cachexia. The skeletal muscle index (SMI) was calculated from the CT cross-section area at L3 divided by the length squared. Muscle loss was defined as a < 5% reduction in the SMI. The association between muscle loss and cachexia during Cx, time to treatment failure (TTF) and overall survival (OS) was determined by univariate and multivariate analysis including muscle loss, primary tumor location, ECOG PS, number of metastatic sites, ALP, WBC, LDH, KRAS status, and BRAF status as independent variables. Results: Of 562 included pts, 185 were eligible and 69 (37%) experienced cachexia. Differences in all patient characteristics such as muscle loss, ECOG PS 2, KRAS mutant, and BRAF mutant (28%, 8%, 32%, and 10% with cachexia group and 27%, 3%, 32%, and 8% with non-cachexia group) were not significant. Median follow-up was 26.8 months. Muscle loss was not associated with TTF (13.3 vs. 15.5 months, HR = 1.05; 95% CI: 0.76–1.45, p = 0.76). OS was shorter (24.4 vs. 29.9 months, HR = 1.23; 95% CI: 0.86–1.76, p = 0.25), but the difference was not significant in univariate and multivariate analysis. However, cachexia group presented significantly shorter TTF [median TTF 11.9 vs. 16.9 m; HR 1.52, 95% CI: 1.12-2.07, p < 0.01; adjusted HR (aHR) 1.53, 95% CI: 1.12-2.11, p < 0.01] and shorter OS (median OS 21.4 vs. 34.1 m; HR 1.81, 95% CI: 1.29-2.55, p < 0.01; aHR 1.97, 95% CI: 1.38-2.81, p < 0.01) than non-cachexia group. Conclusions: Cachexia during Cx may have a negative impact on survival in pts with mCRC.