Treatment with STRIDE*, with or without lenvatinib, plus TACE** significantly improves progression-free survival (PFS), objective response rate (ORR), and duration of response (DoR) in patients with embolization-eligible hepatocellular carcinoma (eeHCC) compared with TACE alone, according to the phase III EMERALD-3 trial presented at ESMO GI 2026.
EMERALD-3 achieved its primary endpoint of PFS per RECIST v1.1 by BICR*** for the STRIDE + lenvatinib + TACE regimen, with consistent results across modified RECIST (mRECIST) and investigator assessments, according to Dr Joseph Erinjeri from Memorial Sloan Kettering Cancer Center in New York, US.
He also emphasized that mRECIST is typically the recommended assessment for a tumour response following locoregional therapies (eg, TACE), as it utilizes intratumoral arterial hyperenhancement to identify viable tumour as the target lesion, rather than evaluating the entire lesion as in RECIST v1.1.
The study included 760 patients with eeHCC who were randomized to receive STRIDE + lenvatinib + TACE (arm A: n=293, mean age 67 years), STRIDE + TACE (arm B: n=175, mean age 65 years), or TACE alone (arm C: n=292, mean age 65 years). Baseline characteristics were well balanced across all treatment groups.
STRIDE + lenvatinib + TACE vs TACE only
At the first data cut-off (September 2, 2025), patients treated with STRIDE + lenvatinib + TACE had a significantly longer median PFS per RECIST v1.1 by BICR*** than those treated with TACE alone (13 vs 9.8 months; hazard ratio [HR], 0.70; log-rank p=0.0007). [ESMO GI 2026, abstract LBA2]
Similarly, the median PFS per modified RECIST (mRECIST) also favoured the STRIDE regimen over TACE alone for both BICR (12.9 vs 8.3 months; HR, 0.67) and investigator (10.4 vs 6.7 months; HR, 0.64) assessments.
At 24 months, the PFS rates were higher in the STRIDE + lenvatinib + TACE vs the TACE-only group, as shown by BICR per RECIST v1.1 (30.4 percent vs 19.3 percent) and by the investigator per mRECIST (22.5 percent vs 12.9 percent).]
The BICR- and investigator-assessed ORR were also higher with the STRIDE-based regimen compared with TACE alone (36.6 percent vs 30 percent and 58.9 percent vs 42.5 percent, respectively).
In addition, the median DoR was numerically longer with STRIDE + lenvatinib + TACE compared with TACE alone (15.7 vs 13.3 months [as per BICR] and 13.1 vs 10.2 months [as per the investigator]).
At the second data cut-off (February 23, 2026), overall survival (OS) data had reached a maturity rate of 40.3 percent, showing a trend toward improvement with STRIDE + lenvatinib + TACE compared with TACE alone (median OS, 39.5 vs 34.7 months; HR, 0.84; log-rank p=0.1814).
Notably, Erinjeri stated that approximately 70 percent of patients in the TACE-alone cohort received subsequent anticancer treatment, whereas only 43.3 percent did so in the STRIDE regimen cohort.
STRIDE + TACE vs TACE only
At the second data cut-off, patients treated with STRIDE + TACE compared with TACE alone achieved a longer median PFS in both assessments (12.9 vs 8.1 months; HR, 0.71 per RECIST by BICR and 12 vs 6.6 months; HR, 0.59 per mRECIST by the investigator).
Higher 24-month PFS rates were also observed with STRIDE + TACE than with TACE alone, as per RECIST by BICR (30 percent vs 20.3 percent) and as per mRECIST by the investigator (25.3 percent vs 11.7 percent).
The ORR, as assessed by BIRC and by the investigator, was higher in the STRIDE + TACE group compared with the TACE alone group (40.8 percent vs 27 percent and 55.4 percent vs 41.7 percent, respectively).
Overall, “assessment per mRECIST by the investigator and by BICR showed improvements in PFS, ORR, and DoR with STRIDE, with or without lenvatinib, + TACE vs TACE alone, consistent with findings per RECIST v1.1 by BICR,” said Erinjeri.
“EMERALD-3 is the first phase III study to demonstrate that a STRIDE-based regimen in combination with TACE improves clinical outcomes, supporting its role as a potential new treatment option in patients with eeHC,” he added.
Safety
The incidence of adverse events related to the study drug was highest in the STRIDE + lenvatinib + TACE group (62.7 percent), followed by the STRIDE + TACE group (48.6 percent) and the TACE alone group (18.6 percent).
Nevertheless, Erinjeri noted that “the safety profile was acceptable and consistent with the known safety profiles of STRIDE, lenvatinib, and TACE.”