Introduction: Two Practice-Defining Trials

ASCO 2026 brought two late-breaking presentations that are reshaping the treatment of hepatocellular carcinoma and pancreatic cancer. In HCC, the EMERALD-3 trial establishes that adding a STRIDE-based immunotherapy regimen to TACE significantly improves progression-free survival in intermediate-stage disease; important questions about which regimen and which patients benefit most remain open. In pancreatic cancer, the news is more definitive: daraxonrasib, a first-in-class pan-RAS inhibitor, has produced the most dramatic improvement in second-line pancreatic cancer outcomes in decades and is now the new standard of care for previously treated metastatic disease.

EMERALD-3: Advancing Beyond TACE in Intermediate-Stage HCC

Transarterial chemoembolization has been the global standard of care for patients with embolization-eligible unresectable hepatocellular carcinoma for many years. The outcomes with TACE alone are meaningful but limited: overall response rates of approximately fifty percent, median PFS of seven to ten months, and median OS of twenty to forty months depending on the patient population. The rationale for combining TACE with systemic immunotherapy and anti-VEGF or TKI therapy is biologically sound: local regional therapy enhances immune activation through antigen release and changes to the tumor microenvironment, and checkpoint inhibition plus anti-VEGF therapy increases T-cell penetration into the tumor.

Two prior studies set the stage. EMERALD-1 (durvalumab plus bevacizumab plus TACE versus TACE alone) showed significant PFS improvement (15.0 versus 8.2 months) and a higher ORR (44% versus 30%). LEAP-012 (pembrolizumab plus lenvatinib plus TACE versus TACE alone) similarly showed PFS improvement (14.6 versus 10.0 months) with higher ORR. But the question both left open was: is a triplet (checkpoint inhibitor plus TKI plus TACE) better than a doublet (checkpoint inhibitor plus TACE)?

EMERALD-3 directly addresses this. The trial enrolled patients with confirmed HCC, not amenable to curative treatment but amenable to embolization, Child-Pugh A, ECOG 0-1, no extrahepatic disease, no prior systemic therapy. Patients were randomized to three arms: STRIDE (durvalumab plus tremelimumab) plus lenvatinib plus TACE (the triplet); STRIDE plus TACE (the doublet); or TACE alone. The primary endpoint was PFS for the triplet versus TACE alone.

Results: The triplet significantly improved median PFS: 13.0 months versus 9.8 months (HR 0.70, p=0.0007). The OS data are immature, with a trend toward improvement (39.5 versus 34.7 months) that is not yet mature enough to conclude benefit. Looking at the doublet versus TACE alone, the doublet arm showed a median PFS of 12.9 months versus 8.1 months, strikingly similar to the triplet.

The critical comparison: When the triplet and doublet arms are placed side by side, the PFS curves look nearly identical. The triplet produced meaningful additional toxicity: more grade 3 and 4 adverse events and over thirty percent drug discontinuation in the triplet arm compared to the doublet. Quality of life data were not presented at ASCO, which is an important gap. Subsequent therapy data were also not presented, making it difficult to know how much of any eventual OS benefit might be attributable to TACE plus systemic therapy, maintenance systemic therapy, or subsequent lines of treatment.

Open questions: Does every embolization-eligible patient need TACE plus IO? Which patients benefit most from the combination? Subgroup analyses by tumor burden and organ function are needed. Does this data extend to TARE (Y90 radioembolization)? A single-arm phase 2 study (EMERALD-Y90) has completed enrollment and results are awaited.

Clinical bottom line: A STRIDE-based regimen combined with TACE improves clinical outcomes for patients with intermediate-stage HCC and has the potential to become a new treatment option; the triplet does not clearly outperform the doublet and comes with substantially more toxicity. This regimen is not yet ready for universal adoption; patient selection and shared decision-making about the risk-benefit tradeoff are essential until QoL and OS data mature.

Daraxonrasib: A New Standard of Care in Previously Treated Pancreatic Cancer

Pancreatic ductal adenocarcinoma is the hardest disease in oncology. Approximately ninety percent of PDAC tumors harbor an oncogenic RAS mutation. Second-line chemotherapy (FOLFIRI, gemcitabine plus nab-paclitaxel, or FOLFOX) produces a median OS of approximately six to six and a half months. Nothing has changed that calculus for decades.

Daraxonrasib changes it.

Mechanism

Daraxonrasib is an oral, potent, pan-RAS multi-selective inhibitor and the first in its class. Unlike the mutation-selective allele-specific inhibitors such as adagrasib and sotorasib (which target only G12C), daraxonrasib inhibits all GTP-bound mutant RAS variants including G12, G13, and Q61 mutations as well as wild-type RAS. Its mechanism is distinct from the covalent binding of prior agents: daraxonrasib acts as a molecular glue, forming a tri-complex with cyclophilin A intracellularly. This binary complex then binds to the effector-state (RAS-on) configuration of mutant and wild-type RAS, causing steric occlusion that prevents downstream cellular signaling through the PI3K and MAP kinase pathways.

Phase 1/2 Context

Phase 1/2 data published in the New England Journal of Medicine shortly before ASCO established proof of concept. In the second-line setting at the 300 mg recommended phase 2 dose, the response rate was approximately 35%, with median PFS of 8.5 months and median OS of 13 to 15 months, numbers that already exceeded anything previously seen in second-line pancreatic cancer.

RASolute 302: The Phase 3 Trial

The RASolute 302 trial enrolled adult patients with metastatic PDAC who had received one prior line of fluoropyrimidine-based or gemcitabine-based therapy in the metastatic setting, ECOG 0-1, with documented tumor RAS mutational status. Patients were randomized one to one to daraxonrasib 300 mg orally once daily versus investigator's choice chemotherapy (gemcitabine plus nab-paclitaxel, FOLFOX, or FOLFIRI/5-FU plus nalirinotecan). The dual primary endpoints were OS and PFS in the RAS G12-mutant population, with key secondary endpoints in the overall population.

Primary endpoint: overall survival. Hazard ratio 0.40. Median OS 13.2 months with daraxonrasib versus 6.6 months with chemotherapy. In the overall population (including non-G12 mutations and RAS wild-type), the result was virtually identical: median OS 13.2 versus 6.7 months.

This is a hazard ratio of 0.40, a sixty percent reduction in the risk of death, in a disease where nothing has achieved this level of improvement in decades. The consistency across RAS mutational subtypes and in the RAS wild-type population is notable and supports the use of daraxonrasib regardless of specific RAS variant.

Progression-free survival: Median PFS 7.3 months versus 3.5 months in the G12 population; 7.2 versus 3.6 months in the overall population. Objective response rate approximately thirty-three percent versus eleven percent with chemotherapy.

Patient-reported outcomes: Daraxonrasib significantly delayed time to deterioration in pain and global health status quality of life compared to chemotherapy, meaningful in a disease where most patients are already symptomatic.

Safety: The most common treatment-related adverse events were rash and stomatitis, followed by GI toxicity (nausea, diarrhea). The majority of patients remained on trial; only about 1% of patients in the daraxonrasib group discontinued treatment but dose reductions were needed in about a third of patients (36.1%). The safety profile is consistent with the phase 1/2 experience.

Molecular testing is still essential. An important question was raised at the plenary session: do patients still need molecular testing before receiving daraxonrasib? The answer is yes, unequivocally, for all patients with locally advanced or metastatic disease. Molecular testing remains critical because: (1) actionable alterations other than RAS mutations exist and are preferentially found in RAS wild-type tumors; (2) patients may be eligible for matched therapies or clinical trials beyond RAS inhibition; (3) knowing a patient's specific mutation profile informs other treatment decisions across the disease course. Liquid biopsy is an acceptable alternative when tissue is not available, though tissue-based NGS remains preferred per NCCN guidelines. Resistance to daraxonrasib is being characterized, with early data suggesting KRAS amplification and downstream MAP kinase alterations as common mechanisms.

For Patients

If you have been diagnosed with hepatocellular carcinoma and your disease is at an intermediate stage suitable for TACE, ask your oncologist about clinical trials or approved combinations adding immunotherapy to TACE. The evidence is building that these combinations extend the time before disease progresses, though decisions about which combination to use and whether you are an appropriate candidate require careful individualized assessment.

If you have pancreatic cancer that has been treated with first-line chemotherapy and is continuing to progress, a new oral drug called daraxonrasib has now demonstrated the most dramatic survival improvement ever seen in second-line pancreatic cancer, nearly doubling median survival compared to chemotherapy. Ask your oncologist whether daraxonrasib is appropriate for your situation and make sure your tumor has been tested for molecular alterations, as this information guides the full range of treatment options available to you.

Key Takeaways

  • EMERALD-3: STRIDE plus lenvatinib plus TACE significantly improved PFS versus TACE alone (13.0 vs 9.8 months, HR 0.70) in intermediate-stage HCC; doublet (STRIDE plus TACE) produced nearly identical PFS with less toxicity; OS data immature; not yet ready for universal adoption.

  • Prior TACE plus IO studies (EMERALD-1, LEAP-012) established feasibility; EMERALD-3 asks whether triplet adds to doublet; current data suggest limited additional efficacy but meaningfully more toxicity.

  • Daraxonrasib (RASolute 302): Pan-RAS multi-selective oral inhibitor vs. second-line chemotherapy in metastatic PDAC; OS HR 0.40; median OS 13.2 vs 6.6 months; PFS 7.3 vs 3.5 months; ORR 33% vs 11%; benefit consistent across RAS G12 and overall populations.

  • Daraxonrasib is now the new standard of care for previously treated metastatic pancreatic cancer.

  • Molecular testing remains essential for all metastatic PDAC patients — daraxonrasib does not eliminate the need for NGS testing; RAS wild-type patients also benefited and may have additional actionable alterations.

  • Resistance mechanisms to daraxonrasib are being characterized; KRAS amplification and MAP kinase pathway alterations are emerging.

References

  1. Lau G, et al. EMERALD-3: STRIDE plus lenvatinib plus TACE in intermediate-stage HCC. ASCO 2026. LBA4000.

  2. Sangro B, et al. EMERALD-1: durvalumab plus bevacizumab plus TACE in HCC. Lancet, 2025.

  3. Kudo M, et al. LEAP-012: pembrolizumab plus lenvatinib plus TACE in HCC. Lancet, 2024.

  4. Wolpin BM, et al. RASolute 302: daraxonrasib vs chemotherapy in previously treated metastatic PDAC. ASCO 2026. LBA5.

  5. Halbrook CJ, et al. Pancreatic cancer review. Cell, 2023.

  6. Safyan RA. "ASCO 2026 Updates: Hepatobiliary and Pancreatic Malignancies." Presented at the 2026 Best of Oncology, Seattle, WA, June 2026.