Treatment of colorectal liver metastases in Germany: a ten-year population-based analysis of 5772 cases of primary colorectal adenocarcinoma

Abstract

Background: Purpose of this study was to analyse the surgical management and long-term clinical outcome of patients diagnosed with colorectal liver metastases (CLM) over a period of 10 years using data from a German tumour registry.

Methods: Retrospective analysis of 5772 patients diagnosed with colorectal adenocarcinoma between 2002 and 2007. Follow-up was continued until 2012.

Results: 1426 patients (24.7%) had CLM; 1019 patients (71%) had synchronous, 407 patients (29%) developed metachronous CLM. Hepatic resection was performed in 374 of the 1426 CLM patients (26%). A significant increase in liver resection rate from 16.6% for the 2002 cohort to 32% in later cohorts was observed. In centers specialized in liver surgery, CLM resection rates reached 46.6%. However, up to 52% of patients diagnosed with three or less CLM did not undergo liver surgery, although, if resected, patients with 1 CLM show a similar long-time survival as CRC patients who do not develop any CLM. Univariate and multivariate analyses adjusted for age, sex, year of resection, time of CLM diagnosis and number of CLM revealed a significant survival benefit for CLM resection (HR =0.355; CI 0.305-0.414). Furthermore, significant impact on OS was seen for age at diagnosis, perioperative chemotherapy and number of CLM.

Conclusions: We here present the first long-term, population-based analysis of the surgical management of CLM in Germany. Significant increase in hepatic resection rates, translating to a significant benefit in OS, was seen over years. However, we still see a striking potential for further improvements in interdisciplinary CLM management.

Figure 1

Synopsis of patients analysed in the present study. A total of 5772 patients diagnosed with colorectal cancer (CRC) in the years 2002–2007 were included. Of these, 1426 patients developed hepatic metastases and 374 underwent hepatic resection.

Figure 2

Incidence of colorectal cancer in the study population. A Relative (left axis) and absolute numbers (right axis, black line) of colorectal cancer patients with no (chequered), and metachronous (grey), synchronous (hatched) colorectal liver metastases by year of CRC diagnosis. B Relative cumulative incidence of colorectal liver metastases in months after diagnosis of the primary tumour.

Figure 3

Resection rates of colorectal liver metastases. A Resection rate of colorectal liver metastases (CLM) in % by year of CLM diagnosis. B Resection rate of CLM by year of CLM diagnosis in a subgroup of patients diagnosed with 1–3 CLM. C Resection rate of CLM by treatment centre: 2 academic liver centres compared with 28 regional hospitals.

Figure 4

Ten year overall survival analyses. A Comparison of 10-year OS in colorectal cancer cases with (n = 1426) an without (n = 4346) liver metastases (CLM) 2002–2007. B Comparison of 10-year OS in colorectal cancer liver metastases (CLM) patients by number of CLM independent of CLM resection. (1CLM: n = 236; 2-3CLM: n = 147; >3CLM: n = 729; unknown number of CLM: n = 314). C Comparison of 10-year OS in colorectal cancer liver metastases (CLM) patients with (n = 374) and without (n = 1052) curative liver resection.

Figure 5

Ten year overall survival analyses in patients undergoing hepatic resection. A Comparison of 10-year OS - Liver resection yes (n = 124) versus no (n = 112) in patients with singular metastasis. B Comparison of 10-year OS - Liver resection yes (n = 76) versus no (n = 71) in patients with 2–3 metastases. C Comparison of 10-year OS - Liver resection yes (n = 94) versus no (n = 635) in patients with more than 3 metastases. D Comparison of 10-year OS - Liver resection yes (n = 80) versus no (n = 234) in patients with unknown (x CLM) number of metastases.