Liver cancer took nearly four times longer to progress after yttrium-90 (Y90) radioembolization than after conventional transarterial chemoembolization (cTACE), according to a single-center, randomized, phase II trial of 45 patients reported in the December issue of Gastroenterology (2016 Aug 26. doi: 10.1053/j.gastro.2016.08.029).
Median time to progression remained unreached more than 26 months after patients underwent Y90 treatment, but was only 6.8 months in the cTACE group (P = .001), Riad Salem, MD, and his associates at Northwestern University,Chicago, reported. Slow accrual limited the study size, but a post-hoc analysis showed that Y90 would have a 97% chance of significantly outperforming chemoembolization if the study had reached its enrollment target, even if the difference in time to progression was less pronounced. Furthermore, Y90 significantly outperformed chemoembolization in a competing risk analysis that accounted for liver transplantation and death, the researchers said.
Conventional transarterial chemoembolization is used in intermediate-stage liver cancer when ablation is contraindicated. However, retrospective studies have favored Y90 radioembolization, a minimally invasive procedure in which a clinician implants radioactive micron-sized particles loaded with Y90 inside blood vessels supplying a tumor. To further study this approach, the investigators randomly assigned patients with unresectable, unablatable hepatocellular carcinoma without vascular invasion, who had Child-Pugh scores of A or B, serum bilirubin levels up to 2 mg/dL, and liver enzymes up to five times the normal upper limit, to undergo selective Y90 at a dose of 120 Gy, or lipiodol-based chemoembolization at a dose of 75 mg/m2.
Source: American Gastroenterological Association
Of 179 eligible patients, 134 (75%) declined to participate in research, opted for other trials, or chose one protocol over the other. Consequently, only 21 patients were assigned to cTACE, while 24 underwent Y90. The groups resembled each other clinically and demographically at baseline, although Y90 patients tended to have more portal hypertension and higher serum bilirubin levels. No patients died within 30 days after treatment. Each group had one case of common femoral artery pseudoaneurysm. The Y90 patients tended to have more fatigue (P = .08), and had higher rates of diarrhea (P = .03) and hypoalbuminemia (P less than .001).
Despite the small group sizes, patients were about 88% less likely to progress at a given time point after Y90, compared with cTACE (hazard ratio, 0.12; 95% confidence interval, 0.03-0.56; P = .007). To explore what might have happened had the study reached target enrollment, the researchers added another 79 hypothetical patients at the 5.1-fold higher hazard ratio (0.625) that they had used in the power calculation. The results showed that Y90 had a 97% chance of statistically outperforming cTACE under these conditions.
Inverse probability of censoring weighting, which is performed to control for dependent censoring between groups, also showed that time to progression was significantly longer with Y90 than with cTACE, the investigators said. “While the relatively low sample size is acknowledged, the seminal studies establishing cTACE as the standard of care were also limited in sample size, [were] single center, and enrolled mostly Child-Pugh A patients,” they emphasized. “Our time to progression results favoring Y90 are in line with other uncontrolled retrospective reports in patients with compromised liver function, [but] our study validates such findings with prospective randomized level I evidence.”
The National Institutes of Health and the SIR Foundation provided funding. Dr. Salem and two coinvestigators reported serving as advisors to BTG. The other coinvestigators reported having no conflicts of interest.