STRIDE and lenvatinib on top of TACE bought three months of progression-free survival, and no survival gain yet

Design

Phase 3, randomised, open-label, sponsor-blinded; 177 sites in 21 countries

Population

760 adults with Child-Pugh A hepatocellular carcinoma not amenable to surgery, ablation or transplant but suitable for TACE; 72% Asian

Primary outcome

Progression-free survival, STRIDE plus lenvatinib plus TACE versus TACE alone

Effect

Median 13.0 vs 9.8 months, HR 0.70 (95% CI 0.57-0.86), p=0.0007; overall survival 39.5 vs 34.7 months, HR 0.84 (0.65-1.09), p=0.18

EMERALD-3 randomised 760 adults with Child-Pugh A hepatocellular carcinoma that could not be cured by surgery, ablation or transplant but could be embolised, across 177 sites in 21 countries. Three arms: durvalumab plus tremelimumab (the STRIDE regimen) with lenvatinib and TACE, STRIDE with TACE, or TACE alone. Nearly three-quarters of participants were Asian, so the population is closer to Indian practice than most hepatocellular carcinoma trials.

The triplet met its primary endpoint. Median progression-free survival was 13.0 months against 9.8 for TACE alone, HR 0.70 (95% CI 0.57-0.86), p=0.0007. Overall survival at the later cut-off was 39.5 versus 34.7 months, HR 0.84 (0.65-1.09), p=0.18 - not statistically significant, and the trial itself calls the analysis unfinished. STRIDE without lenvatinib showed a similar progression signal, HR 0.71 (0.56-0.91).

The cost is real. Serious adverse events occurred in 64% of the triplet arm against 23% with TACE alone, and seven participants (2%) died of treatment-related causes, including two from myocarditis and one from hepatic failure. That is the trade a patient is being asked to make for a delay in progression that has not yet translated into longer life.

So this is a regimen to raise with a patient who has good liver function, will accept immune-related toxicity, and understands the survival question is open. It is not a reason to change the default for every embolisation-eligible patient, and in Indian practice drug cost will settle many of these conversations before the evidence does.

• Confirm Child-Pugh A and ECOG 0-1 before considering the triplet - the trial did not test it outside those limits
• Baseline thyroid function, morning cortisol, liver enzymes and a cardiac history before the first tremelimumab dose
• Warn explicitly about myocarditis: two of the seven treatment-related deaths were cardiac
• Hypertension was the commonest grade 3-4 event with lenvatinib (12%) - have a plan for it before starting
• Document that overall survival is not yet proven, so the consent conversation holds up if the final analysis is flat

The statistics, in plain English

A hazard ratio of 0.70 for progression means the rate of progressing or dying was 30% lower with the triplet at any given moment; the confidence interval, 0.57 to 0.86, sits entirely below 1.0, so chance is an unlikely explanation. The overall survival interval, 0.65 to 1.09, crosses 1.0 - the data are consistent with a useful survival gain and with none at all. Progression-free survival in an open-label trial is also the softer endpoint: investigators knew who was on treatment when the scans were read.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.