GNAS is frequently mutated in both low-grade and high-grade disseminated appendiceal mucinous neoplasms but does not affect survival.

We analyzed a series of 55 disseminated appendiceal mucinous neoplasms treated at our institution for GNAS and KRAS mutations in an attempt to correlate mutation status with clinicopathological findings and patient survival. GNAS mutations (p.R201H, c.602G>A and p.R201C, and c.602C>T) were identified in 17 (31%) of 55 of disseminated mucinous neoplasms and were found in 8 (35%) of 23 low-grade mucinous neoplasms, 7 (37%) of 19 high-grade mucinous adenocarcinomas lacking a signet ring cell component, and 2 (15%) of 13 high-grade mucinous adenocarcinomas with a signet ring cell component. All 7 mucinous adenocarcinomas composed of pure (>95%) signet ring cells harbored wild-type GNAS. There was no significant association between GNAS mutations and sex and age (both with P > .05) or between GNAS mutations and individual adverse histologic features including cytologic grade, destructive invasion, tumor cellularity, angiolymphatic invasion, perineural invasion, and signet ring cells (all with P > .05). KRAS mutations were identified in 22 (40%) of 55 disseminated mucinous neoplasms. GNAS-mutated disseminated appendiceal mucinous neoplasms more frequently harbored concurrent KRAS mutations compared with GNAS wild-type tumors (65% versus 29%, P = .018). GNAS mutations were not significantly associated with overall survival (both with P > .05). Only overall tumor grade was an independent predictor of overall survival in the multivariate analysis (P = .01). Our results indicate that GNAS mutations are frequently identified in both low-grade and high-grade disseminated appendiceal mucinous neoplasms indicating that GNAS mutation status cannot be used to distinguish between low-grade from high-grade appendiceal mucinous neoplasms.

Outcome analysis of a decade-long experience of isolated hepatic perfusion for unresectable liver metastases at a single institution.

OBJECTIVE: To evaluate outcomes of isolated hepatic perfusion (IHP) on isolated liver metastases (LMs). BACKGROUND: Isolated unresectable LMs are often the main determinant of overall survival (OS) for colorectal cancer (CRC) and other solid malignancies. We hypothesized that IHP can be performed safely and yield impressive responses for a variety of solid tumor pathology, using different perfusion agents. METHODS: Retrospective review of a prospectively collected database of patients undergoing IHP for unresectable solid tumor LM. RESULTS: Between 2003 and 2012, IHP was completed in 91 patients. Primary tumor pathology was CRC = 54, non-CRC = 37 (ocular/cutaneous melanoma = 32, cholangiocarcinoma = 3, appendiceal = 1, and breast = 1). IHP employed Melphalan (n = 69) (CRC = 32, non-CRC = 37), Oxaliplatin (n = 10) (CRC), or Oxaliplatin + 5FU (n = 12) (CRC). Hepatic arterial infusion (HAI) pumps were placed in all CRC patients. There were 3(3.3%) perioperative deaths. Response rates for CRC, melanoma, and cholangiocarcinoma were 68.2%, 57.1%, and 100% respectively. Response rates for CRC patients using 5FU + Oxaliplatin, Oxaliplatin, or Melphalan were 83.3%, 66.7%, and 60.9%, respectively. Median OS for the CRC patients (from IHP date) was 23 months (95% confidence interval: 15-28 months). On univariate analysis, receipt of HAI-FUDR (floxuridine) within 1 year of IHP was the only factor associated with improved OS (P = 0.043) in CRC patients. CONCLUSIONS: IHP results in excellent response rates for patients with unresectable liver metastasis from solid tumors. Improved local control for CRC patients undergoing IHP-HAI may improve survival.