Survival and prognostic factors
As noted before, median overall survival was 16.5 months. Median progression-free survival was 6.1 months, and median hepatic PFS was 6.7 months.
Factors prognostic for better overall survival included hepatitis B or C virus as the cirrhosis cause versus alcohol (hazard ratio for death, 0.51 for each; P = .0060 for HBV and P = .0007 for HCV); unilobar versus bilobar tumors (HR, 0.67; P = .0422 for left-lobe; HR 0.55; P < .0001 for right); prior surgery (HR, 0.67; P = .0258); prior ablation (HR, 0.65; P = .0394); and curative versus palliative intent (HR, 0.53; P < .0001).
Factors associated with worse overall survival were presence of ascites (HR 1.75, P = .001); presence of extrahepatic disease before TARE (HR, 1.81, P = .0037); tumor burden greater than 5 nodules (HR, 1.67; P = .0073); main portal vein occlusion (HR, 2.14; P = .0064); lobar portal vein occlusion (HR, 1.77; P = .0083); total bilirubin greater than 1.5 mg/dL (HR, 1.69; P = .0094); albumin-bilirubin grade A2 (HR, 1.66; P = .0005); ALB1 grade A3 (HR, 3.92; P < .0001); and BSA/mBSA versus partition-model dosimetry (HR, 1.89; P < .0001).
The safety analysis showed that 36.7% of patients had at least one adverse event, but only 7.1% had at least one grade 3 or greater event.
Grade 3 or greater events were abdominal pain (nine patients), fatigue (six), nausea (three), radioembolization-induced liver disease (three), vomiting (two), and GI ulceration (one). Fifteen additional patients had other unspecified events.
The investigators acknowledged broad inclusion criteria, relatively high rates of loss to follow-up, and differences in national guidelines and local standards of practice as potential limitations to their findings.
In the question-and-answer following the presentation, session comoderator María Varela, MD, PhD, a pathologist in the liver unit at the Hospital Universitario Central de Asturia, Oviedo, Spain, questioned why about one-third of patients received TARE for downstaging, but only 13 underwent subsequent surgical resection.
“We don’t have a detailed analysis of this subgroup of patients who received curative intent as yet, ” Dr. Kolligs said.
Pierre Nahon, MD, from the University of Paris and Hôpital Jean Verdier in Bondy, France, commented that, among this heterogenous population, one of the best indications for TARE is probably localized HCC with adjacent portal vein thrombosis.
He asked whether the investigators had examined overall survival among patients with localized unilobar HCC with adjacent small portal vein thrombosis.
“We find that patients with portal vein occlusion have a worse prognosis in the total group,” Dr. Kolligs replied. “To look into the question whether partial thrombosis with a small tumor might benefit is an interesting question, and we should look into that, but I don’t have any data on that yet.”
Another audience member asked: “According to your data, which patients are the best candidates for radioembolization?”
“According to these data, the best candidates are of course patients with good liver function, ascites should ideally not be present, and what is probably is important is that we identify or include patients without extrahepatic disease,” he said.
The study was sponsored by CIRSE. Dr. Kolligs disclosed speaking activities and consulting for several companies. Dr. Varela disclosed speaking for several companies and advisory board activity for Bayer. Dr. Nahon disclosed honoraria and consulting fees from several companies.