Transarterial chemoembolization (TACE) is a standard treatment for patients with hepatocellular carcinoma (HCC) who are eligible for embolization. However, outcomes with TACE alone remain limited, leading to increasing interest in combining locoregional treatment with systemic therapy.
The STRIDE regimen, consisting of a single dose of tremelimumab combined with durvalumab followed by durvalumab maintenance, has demonstrated efficacy in advanced HCC. The phase 3 EMERALD-3 trial investigated whether STRIDE, with or without lenvatinib, could improve outcomes when combined with TACE in patients with embolization-eligible HCC.
In September 2026, the study titled “Durvalumab and tremelimumab, with or without lenvatinib, combined with transarterial chemoembolisation in participants with embolisation-eligible hepatocellular carcinoma (EMERALD-3): a global, randomised, open-label, sponsor-blinded, phase 3 study” was published in The Lancet Oncology.
Authors: Masatoshi Kudo, Ghassan K. Abou-Alfa, Zhenggang Ren, Jeong Heo, Riccardo Lencioni, Yasuaki Arai, Mohamed Bouattour, Maria A. Gonzalez-Carmona, Yabing Guo, Ai-Bing Xu, Gustavo Vasconcelos Alves, Arunee Dechaphunkul, Gwo Fuang Ho, Yueh Ni Lim, HongTao Hu, Ruibao Liu, Shanzhi Gu, Vikas Ostwal, Witsarut Manasirisuk, Chang-Fang Chiu, Sri Harsha Tekumalla, Manish Shroff, Vimal Dave, Osama Rahma, Joseph P. Erinjeri, and Jia Fan, for the EMERALD-3 Investigators.
The trial was registered as NCT05301842.
How Was EMERALD-3 Designed?
EMERALD-3 was a global, randomized, open-label, sponsor-blinded phase 3 trial conducted at 177 medical sites across 21 countries. The study included adults with confirmed HCC that was not amenable to curative surgery, ablation, or transplantation but was suitable for TACE. Eligible patients had Child–Pugh class A liver function, an ECOG performance status of 0–1, and at least one measurable intrahepatic lesion according to modified RECIST.
Patients who had previously received systemic therapy for HCC were excluded, as were those with major portal vein thrombosis involving Vp3 or Vp4 disease. Between March 2022 and November 2024, 760 patients were randomly assigned to receive:
- STRIDE plus lenvatinib plus TACE: 293 patients
- STRIDE plus TACE: 175 patients
- TACE alone: 292 patients
The STRIDE regimen consisted of a single 300 mg intravenous dose of tremelimumab and durvalumab 1500 mg on day 1, followed by durvalumab 1500 mg every 4 weeks. Patients assigned to the lenvatinib-containing regimen also received oral lenvatinib once daily at 8 mg for a bodyweight below 60 kg or 12 mg for a bodyweight of 60 kg or greater.
In the investigational groups, the first TACE procedure was administered at least 7 days after the first durvalumab dose. Lenvatinib was withheld for 2 days before and 2 days after TACE. TACE technique, timing, and the number of procedures were determined by investigators according to local practice.
The primary endpoint was progression-free survival (PFS) by blinded independent central review according to RECIST version 1.1 for STRIDE plus lenvatinib plus TACE versus TACE alone. Key secondary endpoints included overall survival (OS) for STRIDE plus lenvatinib plus TACE versus TACE and PFS and OS for STRIDE plus TACE versus TACE. The study was funded by AstraZeneca.
Earlier Updates From EMERALD-3
The first positive results from EMERALD-3 were announced in April 2026, when AstraZeneca reported that STRIDE plus lenvatinib and TACE significantly improved PFS compared with TACE alone, while interim OS showed a trend toward improvement. At that time, the STRIDE plus TACE arm had also shown trends toward improved PFS and OS, although these results had not been formally tested.
Detailed results were subsequently presented by Ghassan K. Abou-Alfa, MD, at the 2026 ASCO Annual Meeting as Abstract LBA4000. Median PFS was 13.0 months with STRIDE plus lenvatinib plus TACE versus 9.8 months with TACE alone (HR 0.70), while interim median OS was 39.5 versus 34.7 months (HR 0.84).
Updated tumor response analyses were later presented by Joseph P. Erinjeri at the ESMO Gastrointestinal Cancers Congress 2026 as LBA2. ORR by RECIST v1.1 central review was 36.6% with STRIDE plus lenvatinib plus TACE versus 30.0% with TACE alone, with additional response analyses reported using mRECIST.
STRIDE Plus Lenvatinib and TACE Met the Primary Endpoint
At the first data cutoff on September 2, 2025, 485 PFS events had occurred among the 760 patients. Median PFS was 13.0 months with STRIDE plus lenvatinib plus TACE compared with 9.8 months with TACE alone (HR 0.70; 95% CI, 0.57–0.86; p=0.0007). The trial therefore met its primary endpoint.
Among patients with measurable disease at baseline, the confirmed objective response rate by blinded independent central review according to RECIST version 1.1 was 37% with STRIDE plus lenvatinib plus TACE and 30% with TACE alone. Median duration of response was 15.7 months versus 13.3 months, respectively, while disease control at 20 weeks was achieved in 79% versus 66% of patients.
What Did EMERALD-3 Show for Overall Survival?
At the second data cutoff on February 23, 2026, 305 deaths had occurred, corresponding to 40% OS maturity. Median OS was 39.5 months with STRIDE plus lenvatinib plus TACE compared with 34.7 months with TACE alone (HR 0.84; 95% CI, 0.65–1.09; p=0.1814). This comparison did not meet the prespecified threshold for statistical significance. The estimated OS rates with STRIDE plus lenvatinib plus TACE versus TACE were 83% versus 82% at 12 months, 78% versus 70% at 18 months, 67% versus 62% at 24 months, and 54% versus 48% at 36 months. Overall survival follow-up remains ongoing.
What About STRIDE Plus TACE Without Lenvatinib?
EMERALD-3 also evaluated STRIDE plus TACE without lenvatinib. In this group, median PFS was 12.9 months compared with 8.1 months among the first 175 patients assigned to TACE (HR 0.71; 95% CI, 0.56–0.91). Median OS was not reached with STRIDE plus TACE and was 32.9 months with TACE (HR 0.70; 95% CI, 0.51–0.95).
Among patients with measurable disease at baseline, the confirmed objective response rate was 41% with STRIDE plus TACE compared with 27% with TACE. However, these comparisons require an important statistical consideration.
The trial used a prespecified hierarchical testing procedure. Because the OS comparison between STRIDE plus lenvatinib plus TACE and TACE did not meet the required threshold for statistical significance, the efficacy comparisons between STRIDE plus TACE and TACE could not be formally tested. The STRIDE plus TACE findings should therefore be interpreted as descriptive.
Did the STRIDE-Based Regimens Delay Subsequent Therapy?
Median time to first subsequent therapy or death was 19.4 months with STRIDE plus lenvatinib plus TACE and 16.6 months with STRIDE plus TACE. This compared with 9.2 months among all patients assigned to TACE and 8.6 months among the first 175 patients assigned to TACE. Following treatment discontinuation, subsequent systemic anticancer therapy was received by 29% of patients assigned to STRIDE plus lenvatinib plus TACE, 41% assigned to STRIDE plus TACE, and 53% assigned to TACE.
What Was the Safety Profile?
The safety analysis included 752 patients: 287 treated with STRIDE plus lenvatinib plus TACE, 175 with STRIDE plus TACE, and 290 with TACE. Treatment-related adverse events of any grade occurred in 98%, 96%, and 74% of patients, respectively.
Grade 3 or 4 treatment-related adverse events occurred in 63% with STRIDE plus lenvatinib plus TACE, 49% with STRIDE plus TACE, and 19% with TACE. The most common grade 3 or 4 treatment-related adverse event with STRIDE plus lenvatinib plus TACE was hypertension, occurring in 10% of patients. Post-embolization syndrome was the most common grade 3 or 4 treatment-related adverse event with STRIDE plus TACE and TACE, occurring in 6% of patients in each group.
Serious adverse events occurred in 64% of patients receiving STRIDE plus lenvatinib plus TACE, 51% receiving STRIDE plus TACE, and 23% receiving TACE. Serious treatment-related adverse events occurred in 49%, 36%, and 16%, respectively. Treatment-related adverse events with an outcome of death during the treatment-emergent period occurred in seven patients (2%) receiving STRIDE plus lenvatinib plus TACE, none receiving STRIDE plus TACE, and two patients (1%) receiving TACE.
In the STRIDE plus lenvatinib plus TACE group, these included two cases of myocarditis and one case each of hepatic failure, hemophagocytic lymphohistiocytosis, septic shock, cardiac failure, and an unknown cause. The investigators reported that the overall safety profiles were consistent with those previously established for the individual treatments, with no new safety signals. They also found no evidence of meaningful exacerbation of TACE-related toxicity with the STRIDE-based regimens.
Limitations of EMERALD-3
Several limitations should be considered when interpreting the EMERALD-3 findings. Asian participants were overrepresented in the study population. The trial was open-label, although central randomization and blinded independent central review were used to minimize potential bias. TACE practices also varied across regions and centers, and decisions regarding repeat TACE procedures and subsequent treatments were made by investigators. In addition, the study specifically evaluated TACE and did not assess transarterial radioembolization. Importantly, the OS data remain immature, and final survival follow-up is ongoing.
What Do the EMERALD-3 Results Mean?
EMERALD-3 met its primary endpoint, demonstrating that STRIDE plus lenvatinib plus TACE significantly improved PFS compared with TACE alone in patients with embolization-eligible HCC. Median PFS increased from 9.8 months with TACE to 13.0 months with STRIDE plus lenvatinib plus TACE (HR 0.70; 95% CI, 0.57–0.86; p=0.0007). At the interim OS analysis, median OS was 39.5 months with STRIDE plus lenvatinib plus TACE and 34.7 months with TACE, but this difference did not meet the prespecified threshold for statistical significance.
STRIDE plus TACE without lenvatinib also showed improvements in PFS and OS compared with TACE in descriptive analyses. However, because of the trial’s hierarchical testing procedure, these comparisons could not be formally tested. The investigators concluded that the results support a STRIDE-based regimen as a potential new treatment option for patients with embolization-eligible HCC. Final OS analysis from EMERALD-3 is ongoing.
The full article is available in The Lancet Oncology.
