Injecting immunotherapy directly into melanoma tumors reduced severe toxicity in early trial

Direct injection of immunotherapy into melanoma tumors substantially reduced severe side effects compared with standard intravenous therapy in patients with untreated metastatic melanoma, according to results from a phase 1b trial published April 2026 in Nature. The study was led by Caroline Robert, M.D., Ph.D., head of dermatology at Gustave Roussy and co-director of melanoma research at INSERM 981 Paris-Sud University, in Paris.

Over the past decade, immune checkpoint inhibitors have transformed treatment for metastatic melanoma, helping drive long-term survival rates well above historical levels. The combination of the anti-CTLA4 therapy Yervoy (ipilimumab) and the anti-PD1 therapy Opdivo (nivolumab) is considered one of the most effective first-line treatment options for advanced disease.

However, the regimen is also associated with high rates of severe immune-related toxicities, with prior studies showing grade 3 or 4 treatment-related adverse events in roughly 60% of patients receiving the combination.

To explore whether toxicity could be reduced without sacrificing efficacy, the researchers evaluated a lower-dose intratumoral formulation of ipilimumab combined with intravenous nivolumab in 61 patients with previously untreated advanced melanoma. The investigational approach aimed to preserve antitumor activity while limiting systemic exposure by injecting a dose of ipilimumab that was 10 times lower directly into tumor lesions while maintaining high local concentrations.

Patients were randomly assigned to receive either intratumoral ipilimumab plus intravenous nivolumab or standard intravenous administration of both agents. As part of the study, researchers performed 162 intratumoral injections across 46 lesions, including lymph node, cutaneous, liver and lung metastases.

The study met its primary endpoint, showing substantially lower rates of grade 3 or 4 treatment-related adverse events at six months in the intratumoral group compared with the intravenous group, 22.6% versus 57.1%, respectively. No grade 3 or higher procedure-related adverse events or severe injection-site toxicities were reported in the intratumoral arm.

Although the study was not powered for efficacy comparisons, investigators reported encouraging response rates. Objective response rates reached 65.7% in injected lesions and 50% in uninjected lesions, suggesting the treatment may trigger both local and systemic immune responses.

“By using interventional radiology to directly target the core of the tumour, we have shown that it is possible to trigger a powerful immune response while halving toxicity for the patient,” first author Lambros Tselikas, M.D., Ph.D., said in a press release. Tselikas is a professor of interventional radiology at the University of Paris-Saclay and deputy head of anesthesia, surgery and interventional medicine at Gustave Roussy.

Researchers also identified immune features that appeared linked to better outcomes. Patients who responded to treatment had higher baseline levels of activated regulatory T cells and certain immune-related macrophages within tumors. The findings may help researchers identify which patients are most likely to benefit from checkpoint inhibitor combinations.

The authors concluded that intratumoral anti-CTLA4 approaches may have potential in oligometastatic and early-stage cancers, particularly in situations where local disease control is important. They also suggested that rapid analysis of fresh tumor biopsies could help identify patients most likely to benefit from checkpoint blockade therapies.