gms | German Medical Science

Background: Liver resection for metastases of non-colorectal, non-neuroendocrine tumors (NCNNE) is being performed with an increasing frequency. Although favorable outcomes have been reported by various groups, available prognostic factors are inhomogeneous from center to center and obviously strongly dependent not only on the primary tumor. Indications today are mostly based on individual decisions after interdisciplinary discussion. On the other hand it is likely that many patients are being denied potentially beneficial or even curative surgical therapy, due to the lack of knowledge.

Materials and methods: Patients were identified from our prospective database for liver resections at a single university liver center from 2001 to 2015. Included in this analysis were patients with isolated hepatic metastases from NCNNE tumors. Patients with planned R2 resections (debulking), concomitant peritonectomy and hyperthermic intraperitoneal chemotherapy (HIPEC) as well as liver resection due to direct liver invasion by the primary were excluded from the analysis. Patients were grouped according to the origin of the metastases, histoembryological and clinical considerations. Actuarial survival was estimated by the Kaplan-Meier method. Multivariate survival analysis was performed using the Cox proportional hazard model.

Results: 106 individual liver resections in 100 patients were included in this analysis. Overall 5 year survival was 57%. Median survival was 6.6 years (follow up 0.25-15 year). 5 year survival and significance according to patient characteristics is given in table 1. Metachronous appearance of the metastasis later than 24 months after resection of the primary was associated with highly significant improved survival (p=0.002). Negative hepatic margin tended to result in better survival (p=0.07) although one patient with positive hepatic margin is alive more than 13 years after surgery. Woman had poorer survival than men (p= 0.01). Hepatic resection at time of primary tumor resulted in worse prognosis (p=0.04) Extent of hepatic resection did not influence survival (p=0.52). Subgroup analysis showed significance concerning the embryologic origin of the primary tumor (p=0.009). Survival of mesodermal compared to entodermal tumors was significantly improved (p=0.005). Tumors with genitourinary origin did not display a survival benefit (p=0.85).

Table 1 [Tab. 1]

Conclusion: This study shows that with individual patient selection results after resection liver metastases of NCNNE can achieve 5 year survival rates in the range of almost 60%. This compares favorably to the outcomes generally achieved for classical indications like colorectal liver metastases.An early occurrence of NCNNE metastasis and incomplete resection do seem to be associated with poorer survival. Nevertheless it is likely that especially in patients with late progression of their disease, hepatic surgery is still underutilized. Therefore we strongly encourage discussion of all liver metastases from any entity with a specialized liver surgeon before ruling out resection as a treatment option, preferably in interdisciplinary tumor boards.