Abstract
PURPOSE: To establish the relationship between the number and site of p53 genomic mutations in metastatic colorectal cancer, and the response to hepatic arterial floxuridine. METHODS: Liver metastasis biopsies were collected, at the time of laparotomy for hepatic arterial cannulation. in 28 patients with metachronous colorectal liver metastases. p53 Gene mutations were assessed using reverse transcription, nested polymerase chain reaction, single strand conformational polymorphism and gene sequencing. Chemotherapy response was determined from computerised liver tomograms after 4 months of treatment. RESULTS: Liver metastasis p53 mutation was identified in 21 (75%), and p53 "hot spot" mutation in 11 (39%) patients. There was a significantly lower prevalence (Fisher's, P=0.001) of patients with p53 "hot spot"-mutated liver metastases in stable disease and partial response (5/22) than in progressive (6/6) disease groups. Significantly fewer (Mann-Whitney U, P=0.002) p53 "hot spot" mutations/biopsy were observed in liver metastases from stable disease and partial response (median 0, iqr. 0-0) than in progressive (median 1, iqr 1-2) disease patients. p53 "Hot spot"-mutated liver metastases were associated with significantly shorter survival times (logrank P=0.05) after hepatic arterial floxuridine. Significant response or survival-time differences by total or L2/L3 zinc-binding site p53 mutations were not detected. CONCLUSIONS: The results support a role for p53 "hot spot" mutations in colorectal liver metastasis resistance to fluorinated pyrimidines, and suggest that the presence of such mutations may be a contra-indication to treatment of colorectal liver metastases with hepatic arterial floxuridine.