Resection Of Breast Cancer Liver Metastases Research Articles

Histological growth pattern as a potential prognostic factor in patients operated for breast cancer liver metastases.

e12576 Background: Indications for surgery in patients with breast cancer liver metastases (BCLM) remain ill-defined. Recently, the histological growth pattern (HGP), such as desmoplastic (dHGP), associated with angiogenesis and inflammation, or replacement (rHGP), associated with vessel co-option and in the absence of immune infiltrate, has been demonstrated as a significant prognostic factor in patients operated for colorectal liver metastases. In BCLM, the distribution and the prognostic value of HGP are poorly documented. We aimed to characterize HGP in a series of patients undergoing liver resection for BCLM and correlate these patterns with postoperative outcomes. Methods: A consecutive series of 58 patients undergoing liver resection for BCLM was reviewed. Clinicopathologic parameters of these patients were analyzed. HGPs were assessed in archival H&E stained tissue sections, according to international consensus guidelines. The proportions of rHGP and dHGP were determined in each metastasis. In case of multiple metastases, mean HGP was calculated for each patient. Patients were categorized as pure (100% rHGP or dHGP) or dominant phenotype ( > 50% rHGP or dHGP, on the entire tumor/normal liver interface). All these factors were correlated with overall and disease-free survivals (OS and DFS). Results: After a mean follow-up of 81 months, 5-years OS and DFS in global population were 49.7% and 22.7%, respectively. No clinicopathologic preoperative factor was found to be predictive for OS or DFS. HGPs were analyzed in a first set of 21 patients. All patients showed some rHGP component. Of these, 11 were pure rHGP (52.4%), 4 dominant-rHGP (19%) and 6 dominant-dHGP (28.6%). Five-years postoperative OS and DFS were of 19.4 and 18.2% in pure rHGP, as compared with 64.8 and 33.3% in patients with some dHGP component (p = 0.089). Conclusions: rHGP appears as the main phenotype of BCLM. First analyzes suggest that surgery for BCLM with pure rHGP is associated with a worse oncological outcome as compared to BCLM with a dHGP component.

Breast cancer liver metastases (BCLM): What may be the role of surgery combined with chemotherapy?

1039 Background: Despite recent treatment improvements, the prognosis of BCLM is still poor. Hepatic resection (HR) has been associated with better outcome in selected patients, but its place in multimodality treatment of BCLM remains controversial. This study aimed to examine the outcome of a large cohort of patients selected for HR of BCLM and to define prognostic factors of survival, in a way to better define the place of surgery. Methods: A standardized questionnaire reviewing the main diagnostic and treatment modalities of primary tumor, liver metastases, response to medical therapies, type of surgical procedures, postoperative outcome, and survival following surgery, was sent to all contributing centers. Results: 460 patients treated with liver resection for BCLM from 1980 to 2000, were collected from 31 hepatobiliary surgery centers. Mean age was 51.8 years. Primary tumor, mainly adenocarcinoma, was treated by resection combined with chemotherapy and/or radiotherapy in most cases. Diagnosis of BCLM was made after an average of 54 months from the treatment of the primary tumor. BCLM were unique in 56% and associated to limited extrahepatic disease in 18.5% of patients. After initial treatment by systemic therapy (70% of patients), HR achieved a R0 resection in 82% of patients and was combined to extrahepatic resection for distant metastases in 9% of patients. Postoperative mortality (= 2 months) was 0.2%. Median survival was 45.4 months after HR, with an overall survival of 41% and 22% at 5 and 10 years, respectively. Disease-free survival rates were 14% and 10%, respectively. Four predictive factors were independently associated to an unfavourable outcome: tumor progression on chemotherapy before surgery (p = 0.0006, RR = 2.9), disease-free interval < 12 months after treatment of the primary tumor (p = 0.0003, RR = 2.1), extrahepatic metastases (p = 0.0002, RR = 1.9) and R2 liver resection (p < 0.0001, RR = 3.0). Conclusions: Inclusion of HR in the multimodality treatment of BCLM is safe and associated with a hope of long term survival (22% at 10 years). Surgery should be discussed on a multidisciplinary basis, particularly when potentially radical, in patients well controlled by chemotherapy with a long disease-free interval, and in the absence of extrahepatic disease. No significant financial relationships to disclose.