Validation of prognostic scoring systems for patients undergoing resection of colorectal cancer liver metastases.

BACKGROUND

Several prognostic scoring systems have been established for patients undergoing resection of colorectal cancer (CRC) liver metastases; however, comparative analyses of their prognostic relevance is still lacking in the literature. The aim of the present study was to assess the predictive value of five published scoring systems in an independent patient cohort for the purpose of external validation.

METHODS

A total of 281 patients underwent liver resection for CRC liver metastases at our institution between January 2002 and January 2008. The predictive value of the Nordlinger score, Memorial Sloan-Kettering Cancer Center (MSKCC) score, Iwatsuki score, Basingstoke index, and Mayo scoring system was assessed in this patient set. Furthermore, clinical and pathologic parameters included in the assessed scoring systems were analyzed by means of univariate and multivariate analyses.

RESULTS

The disease-specific survival at 1, 3, and 5 years was 94.6%, 61.8%, and 33.7%, respectively. Of the assessed scoring systems, only the MSKCC score (P = .006) and the Iwatsuki score (P = .01) provided a statistically significant stratification of patients with regard to survival. The predictive value was particularly evident for patients grouped within the high-risk categories. None of these patients was alive at 3 years after surgery. The 3-year survival rates for high-risk patients in the remaining three scoring systems was > 50%.

CONCLUSIONS

In our patient cohort, survival was only predicted by MSKCC and Iwatsuki scores. These findings highlight the importance of validating scoring systems in independent patient groups.