CheckMate 459: Greater OS with nivolumab over sorafenib in advanced HCC

In patients with advanced hepatocellular carcinoma (HCC), first-line treatment with nivolumab may confer greater overall survival (OS) than sorafenib, according to updated results of the phase III CheckMate 459* study.

The researchers enrolled 743 systemic therapy-naïve patients (median age 65 years, 85 percent male, 40 percent from Asia) with advanced HCC not eligible for surgical resection or locoregional therapy or those with progressive disease following these therapies, ECOG PS 0–1, and Child-Pugh class A. They were randomized 1:1 to receive intravenous nivolumab (240 mg Q2W) or oral sorafenib (400 mg BID) until disease progression or unacceptable toxicity.

Patients primarily were BCLC** stage C (82 and 78 percent of nivolumab and sorafenib recipients, respectively), and 75 and 70 percent, respectively, had vascular invasion or extrahepatic spread. Forty-five percent of each group did not have hepatitis, and 33 and 38 percent, respectively, had alpha-fetoprotein levels ≥400 μg/L. More than 50 percent had undergone surgery and/or received local therapy.

At this updated analysis, OS was greater with nivolumab compared with sorafenib (median 16.4 vs 14.8 months; hazard ratio [HR], 0.85, 95 percent confidence interval [CI], 0.72–1.00; p=0.0522; 33-month OS: 29 percent vs 21 percent). [ESMO GI 2020, abstract LBA3]

This benefit continued from the previously reported 12-month (60 percent [nivolumab] vs 55 percent [sorafenib]) and 24-month OS results (37 percent vs 33 percent).

The OS benefit with nivolumab over sorafenib was consistent regardless of baseline PD-L1 status (≥1 percent: median 16.1 vs 8.6 months; HR, 0.80, 95 percent CI, 0.54–1.17; <1 percent: median 16.7 vs 15.2 months; HR, 0.84, 95 percent CI, 0.70–1.01), though it appeared greater among patients with PD-L1 expression ≥1 percent.

The OS benefit with nivolumab vs sorafenib also appeared to be better among those with active or resolved viral hepatitis (median 17.5 vs 12.7 months; HR, 0.72, 95 percent CI, 0.51–1.02 [hepatitis C virus] and median 16.1 vs 10.4 months; HR, 0.79, 95 percent CI, 0.59–1.07 [hepatitis B virus]) compared with uninfected individuals (median 16.0 vs 17.4 months; HR, 0.91, 95 percent CI, 0.72–1.16).

More sorafenib than nivolumab recipients received subsequent systemic therapy (47 percent vs 39 percent), specifically investigational (11 percent vs 3 percent) or immuno-oncologic agents (21 percent vs 2 percent).

Grade 3–4 treatment-related adverse events (TRAEs) occurred more often in sorafenib than nivolumab recipients (50 percent vs 22 percent), and more frequently led to discontinuation in sorafenib recipients (11 percent vs 8 percent). The most common AEs (all grades) in nivolumab recipients were fatigue and pruritus (15 and 13 percent, respectively), and in sorafenib recipients, palmar-plantar erythrodysesthesia syndrome and diarrhoea (49 and 47 percent, respectively).

“No new or unexpected safety signals were observed [and] liver function was preserved over time in the nivolumab compared with the sorafenib arm,” presented Dr Bruno Sangro from the Clinica Universidad de Navarra and CIBEREHD, in Pamplona, Spain, at ESMO GI 2020.

“Patients with advanced HCC not amenable to surgical resection or locoregional therapy have few effective first-line treatment options,” said Sangro. The use of sorafenib in the first-line setting only offers a modest survival benefit, he added.

“[In this study,] nivolumab continued to show a clinically meaningful survival benefit vs sorafenib at a minimum follow-up of 33.6 months,” he concluded.