Treatment Strategies for Colorectal Liver Metastases Based on Analysis of Prognostic Factors

Kazuo Hatsuse, Hideki Ueno, Tsukasa Aihara, Yasuhiro Oobuchi, Nobuaki Kawarabayashi, Akihumi Watanabe, Keiichi Ishikawa, Youjirou Hashiguchi and Hidetaka Mochizuki

Ist Department of Surgery, National Defense Medical College

Background: To clarify the beneficial candidates of hepatectomy for colorectal liver metastasis based on the analysis of survival predictors. Patients and Methods: Clinical, pathological, and outcome parameters were determined for 92 patients undergoing liver resection for colorectal liver metastases. Prognostic variables are (1) factors of the primary lesion, including site, dominant histological type, tumor depth, lymph node involvement and tumor budding; (2) factors of the metastatic lesion, including time to diagnosis, number, distribution, and size; and (3) factors of treatment, including resection type and hepatic arterial chemotherapy. Results: Univariate analysis showed that 6 variables were statistically significant risk factors influencing survival, namely tumor budding, tumor depth, nodal state, time to diagnosis, number of metastases and prophylactic hepatic arterial infusion chemotherapy. Of these, marked tumor budding (moderate or severe degree), deeper tumor depth (serosa or invasion to other organs and time to diagnosis (synchronous metastasis) proved to worsen survival independently by stepwise regression analysis using Cox's proportional hazards model. Based on the results of this multivariate analysis, we classified patients into 3 groups. Group A included 39 patients without 3 prognostic factors or with 1 unfavorable factor (tumor depth or time to diagnosis). Group B included 21 patients with budding or 2 unfavorable factors other than budding. Group C included 32 patients with budding and 1 other unfavorable factor or all unfavorable factors. Cumulative 5-year survival was 55.5% in group A, 17.5% in group B, and 0% in group C, Extrahepatic recurrence was higher in groups B and C than in group A. Recurrence of remnant liver occurred more often in group C than in groups A or B. Group C, especially those with unfavorable 3 prognostic factors, had no advantege in survival. Conclusion: Patients without the 3 variables or with 1 unfavorable variable other than budding had a highly favorable outcome, and surgical resection is undoubtedly indicated for such patients. Patients with 3 positive variables had very poor prognosis, and resection without additional effective adjuvant therapy is highly questionable.

Key words
colorectal cancer, liver metastases, liver resection, prognostic factors

Reprint requests
Kazuo Hatsuse First Department of Surgery, National Defense Medical College 3-2 Namiki, Tokorozawa, 359-8513 JAPAN

Accepted
June 26, 2001

To read the PDF file you will need Abobe Reader installed on your computer.