The impact of biomarker status on predicting outcome in patients following cytoreductive surgery and hyperthermic intraperitoneal chemotherapy for colorectal cancer peritoneal metastases


Year:

Session type:

Theme:

Dilraj Bhullar1,Sarah O'Dwyer1,Malcolm Wilson1,Chelliah Selvasekar2,Andrew Renehan1,Paul Fulford2,Omer Aziz1
1The Christie NHS Foundation Trust & The University of Manchester,2The Christie NHS Foundation Trust

Abstract

Background

Colorectal cancer peritoneal metastases (CRPM) have poor median overall survival (OS) of 16.3 months compared to 24.6 months for lung, and 19.1 months for liver (Franko et al. 2016). Cytoreductive surgery with hyperthermic intraperitoneal chemotherapy (CRS/HIPEC) is a treatment for CRPM with curative intent. Whilst biomarkers have been shown to predict response to chemotherapy, their role in predicting response to CRS/HIPEC is not known. We aimed to determine the impact of biomarker status on predicting OS from CRS/HIPEC for CRPM at a national peritoneal tumour centre.

Method

A prospective database was used to collect information on patients undergoing CRS/HIPEC for CRPM. All patients underwent multi-disciplinary team management and were operated on with curative intent following review of CT scans, histology specimens, operation notes, comorbidities, and performance status. Outcomes of interest included tumour subtype, grade, stage, biomarker status, peritoneal cancer index, and completeness of cytoreduction. Median OS from date of CRS/HIPEC was determined using cox regression and Kaplan Meier curves.

Results

Between 2004 and 2017, 195 patients with confirmed histology following CRS/HIPEC for CRPM were identified, with a median follow-up of 19.8 months. Biomarker status available included KRAS (n=108), NRAS (n=79) and BRAF (n=42). Mutation rates were 43.5% for KRAS, 3.8% for NRAS and 14.3% for BRAF. Overall, RAS (KRAS and NRAS) wild-type tumours had a higher OS (35.9 months) compared to mutant tumours (29.8 months). BRAF wild type tumours had a higher OS of 31.1 months versus 17.2 months for BRAF mutant tumours. Univariate analysis identified RAS status as a significant predictor of OS following CRS/HIPEC (HR: 1.797, 95% CI 1.0.39-3.106, p=0.036). Multivariate analysis confirmed KRAS as a negative prognostic indicator for OS (HR: 3.556, 95% CI 1.584-8.030, p=0.002).

Conclusion

Biomarker status predicts outcome following CRS/HIPEC for CRPM, and may be used to personalize therapy in these patients.