Clinicopathologically diagnosed residual tumor after resection for colorectal cancer. A 20-year prospective study.

BACKGROUND

A lack of comprehensive information exists on the nature, incidence, and prognostic significance of known residual tumor in colorectal cancer patients treated by bowel resection. This study aims to provide this information.

METHODS

A prospective series of 1766 consecutive patients from the Concord Hospital Colorectal Cancer Registry (Concord, Australia) was used for the analysis. Residual tumor was defined as distant metastases diagnosed clinically or pathologically or tumor demonstrated histologically in a line of resection. The pathologic study was highly standardized. Patient follow-up ranged from 6 months to 20.5 years. Survival analysis was by the Kaplan-Meier method. Multivariate models were examined using Cox proportional hazards regression.

RESULTS

The prevalence of residual tumor was 20.9% and the median survival was 11.6 months. In 4.5%, tumor transection alone occurred, 14.5% had distant metastases alone, and 1.9% had both. The difference in survival between the first two groups was at marginal statistical significance (P = 0.076). When each of these two groups was compared with the third group, significant differences were noticed (P = 0.001 and P = 0.004, respectively). Five of 14 pathology variables examined had a significant effect on survival using univariate analysis. On multivariate analysis only tumor transection and distant metastases had significant independent effects.

CONCLUSIONS

Known residual tumor was common in this series: one in five resections. Survival studies show that tumor transection, as defined, is a valid criterion for residual tumor. Survival is significantly reduced when tumor transection and distant metastases both are present. These findings should be heeded when staging colorectal cancer and when stratifying patients for postoperative adjuvant therapy.