At nearly 30 months of follow-up, European investigators found no difference in overall and progression-free survival with thermal ablation instead of surgery, as well as better local control, fewer adverse events, shorter hospital stays, and no treatment-related deaths.
The benefit of thermal ablation was so substantial that the trial was stopped early with about 300 of the planned 600 patients randomized.
Numerous retrospective studies have compared the two approaches, and some have reported better survival with surgery. As a result, although a large number of lesions are amenable to either approach, “the majority of colorectal liver mets [are] still being” resected, said lead investigator and presenter Martijn R. Meijerink, MD, PhD, an interventional radiologist at the Amsterdam University Medical Center, Amsterdam, the Netherlands.
Dr. Meijerink said many of the previous reviews were unreliable due to selection bias because patients only had ablation if their lesions couldn’t be removed surgically. In contrast, all patients in the COLLISION trial were eligible for resection.
“Thermal ablation in experienced centers seems to be at least as good as surgical resection for small liver tumors.” Patients would benefit if it replaced surgery as the standard of care with no compromise in survival, Dr. Meijerink added.
The 296 COLLISION patients were treated at 14 centers in the Netherlands, Belgium, and Italy. They had no more than 12 liver lesions 3 cm or smaller with a median of two lesions. Participants were split equally between the ablation and surgical arms of the trial.
Almost half of the surgeries were laparoscopic, and nearly 60% of the ablations were percutaneous. Recent technological advances were used in the ablation cases, including software to confirm the complete eradication of targeted metastases.
At 28.8 months, there was no difference in overall survival between treatment arms (hazard ratio [HR], 1.051; P = .813) and no difference in local (HR, 0.817; P = .53) and distant (HR, 1.03; P = .836) progression-free survival. Local control — meaning treated lesions didn’t grow back — favored thermal ablation (HR, 0.092; P = .024).
The results held across number subgroup analyses, including by stage, molecular profile, and number of lesions.
“Interestingly, the majority of ablation site recurrences were somehow retreated, and most of them successfully, [while] the majority of resection plane recurrences were not retreated,” Dr. Meijerink said.
Patients with ablation vs surgery spent a median of 1 day vs 4 days in the hospital. Almost 20% of patients in the surgery group had grade 3/4 treatment-related adverse events vs 6% of those in the ablation group, which isn’t surprising, Dr. Meijerink said, because “the needle is less invasive than a knife.”
Three patients (2.1%) died of surgical complications, but there were no treatment-related deaths with ablation.
Major Kenneth Lee, MD, PhD, a gastrointestinal surgeon at the University of Pennsylvania, Philadelphia, who was the study discussant, emphasized the importance of gathering prospective data to compare the two approaches fairly.
“Ablation appears equivalent to resection for small, ideally located colorectal liver mets,” he said. Still, longer follow-up is needed to ensure that cure rates with ablation match those with surgery.
The study was funded by Medtronic-Covidien, a maker of thermal ablation equipment. Among other industry ties, Dr. Meijerink reported receiving honoraria and research funding from Medtronic and advising the company. Dr. Lee didn’t have any disclosures.
A version of this article appeared on Medscape.com.