CHICAGO – Resection of liver and lung colorectal metastases produced superior survival, compared with resection of liver metastases alone, in a retrospective analysis of 1,260 consecutive patients.
After a median follow-up of 49 months, 5-year overall survival was 40% in patients with colorectal cancer who underwent resection of liver metastases alone, compared with 50% in those who underwent resection of both liver and lung metastases (P = .01), Dr. Eddie Abdalla reported at the annual meeting of the Western Surgical Association.
"Survival following resection of liver and lung metastases is excellent, and may even be better than [survival] for the general population that undergoes resection for colorectal lung metastases alone," he said.
Overall survival at 5 years is about 40% in patients with resected colorectal lung metastases, and ranges from 11% to 31% in the limited studies involving patients with colorectal liver-plus-lung metastases.
The presumed explanation for the improved survival is that patients with liver and lung metastases had less-extensive recurrence (27%) than did the rest of the cohort (69%), said Dr. Abdalla of the University of Texas M.D. Anderson Cancer Center in Houston. Recurrence also was seen in 33% of patients with lung-only metastases and in 22% with liver-only metastases.
Although recurrence in the study was substantial, it can often be retreated, he observed. Among 78 patients who developed recurrence, 33 (42%) were retreated with resection, radiofrequency ablation, or chemotherapy, resulting in a 5-year disease-free survival of 25%.
Extrahepatic disease, long associated with poor outcome, has been considered a contraindication to resection. More recent studies are challenging this belief, citing advances in systemic chemotherapy and surgical technique as well as improved multidisciplinary management and patient selection.
In multivariate analysis, the only significant predictors of worse survival in patients with resection of both liver and lung metastases were a rectal primary tumor (P = .004; hazard ratio, 2.9) and a carcinoembryonic antigen (CEA) level before resection of liver metastases of more than 5 ng/mL (P = .04; HR, 2.1), Dr. Abdalla said. Interestingly, liver metastasis greater than 5 cm, positive margins at resection of liver metastases, synchronous lung metastases, and even disease-free intervals were not significant.
"Occurrence of lung metastases that are subsequently resected is not a poor prognostic factor or a contraindication to resection of colorectal liver metastases," he said.
In all, 32 of the 112 patients with liver-plus-lung metastases had a rectal primary tumor, compared with 254 of the 1,148 patients with liver-only metastases. Their median preoperative CEA levels were 3.2 and 3.7, respectively. Preoperative chemotherapy was used in 69% of patients before resection of liver metastases and in 50% of patients before resection of lung metastases.
Invited discussant Dr. John Brems of Loyola University Medical Center in Maywood, Ill., asked whether the survival outcomes were influenced by selecting patients with a higher rate of chemotherapy utilization in the liver-plus-lung group, and by considering patients for surgery if they had no increase in tumor size on chemotherapy.
"There’s no question that we are selecting patients, and the objective of selection is to deliver treatment to patients who will benefit," Dr. Abdalla said. "Response to chemotherapy appears to be a good way to select patients for resection of colorectal liver metastases."
Dr. Abdalla referenced a recent study he coauthored that reported a significantly longer overall survival among patients with colorectal liver metastases who had an optimal morphological response on imaging to bevacizumab-containing chemotherapy prior to hepatic resection, compared with those who had an incomplete or no response (median, 31 months vs. 19 months) (JAMA. 2009;302:2338-44).
During the discussion, Dr. Anton Bilchik of the University of California, Los Angeles, asked whether the researchers treat disappearing lung and liver lesions in the same manner.
Dr. Abdalla said that they tend to follow the disappearing lung lesion, but resect the disappearing liver lesion. "At this point, if you have a patient with a tiny liver lesion that disappears, it’s difficult to propose a major hepatectomy to take out a central disappearing lesion," he said. "However, we know from follow-up and resection studies that most of those – certainly over 90% – will recur. We don’t have that kind of data in the lung."
None of the speakers disclosed any conflicts of interest.