Patient selection for adjuvant therapy trials in colorectal carcinoma

The aim of this research was to determine how to improve the efficiency of study design in adjuvant therapy trials of colorectal carcinoma. Survival analysis related to pathological variables was performed on 1020 patients who had resections for colorectal carcinoma during an 11 year complete follow-up. It was felt that the median survival of B stage patients was too long to justify their inclusion in survival trials. Practical multi-centre trials could be undertaken with patients who were found to have clinicopathological stage C. Fifty per cent of the total number of stage C patients had primary tumours with either high grade or venous invasion of mural and extra-mural veins. They had a highly significantly (P = 0.0004) poorer prognosis (median survival 27 months) than those with low or average grade and no venous invasion (median survival 49 months). The most efficient trial would select patients whose primary tumours showed lymph node involvement and high grade and venous invasion.