Why This Study Matters
Immune checkpoint blockade (ICB) has transformed outcomes for many patients with metastatic melanoma, but a substantial share of patients never respond, or develop resistance over time, and systemic ICB can also cause severe, sometimes treatment-limiting immune-related toxicity. One strategy under investigation is to activate antigen-presenting cells directly inside the tumor — turning immunologically "cold" tumors "hot" — using an intratumoral CD40 agonist, with the goal of priming a systemic antitumor immune response that reaches tumors that were never injected.
This first-in-human trial pairs the intratumoral CD40 agonist sotigalimab with systemic pembrolizumab in ICB-naive patients with metastatic melanoma, reporting both clinical outcomes and an extensive multiomic characterization of how the combination remodels the tumor immune microenvironment.
Study Design
In the phase I dose-escalation portion (14 patients), intratumoral sotigalimab was given at 0.1 mg, 0.5 mg, 1 mg, 3 mg, or 10 mg, each combined with intravenous pembrolizumab 2 mg/kg every 3 weeks. In the phase II expansion (18 patients), sotigalimab was given at the RP2D of 10 mg with the same pembrolizumab regimen. Sotigalimab was injected into a single designated lesion every 3 weeks for up to 4 doses, while pembrolizumab continued until disease progression or unacceptable toxicity.
Secondary objectives evaluated the immunologic impact of the combination in blood and in injected and non-injected tumor biopsies, using T-cell receptor (TCR) sequencing, multiplex immunofluorescence, imaging mass cytometry, single-cell multiome profiling, and gene-expression panels.
Patient Population
Median patient age was 64.5 years (range, 32–81); 84% (27/32) were male, and all had ECOG performance status of 0 (56%) or 1 (44%). Eleven of 30 evaluable patients (34%) had BRAF-mutant disease. Seventeen patients (53%) had an elevated serum LDH level, 17 (53%) had stage IV M1a/M1b disease, 7 (22%) had stage IV M1c disease, and 8 (25%) had stage III disease. PD-L1 positivity (≥1% tumor cell membranous staining) was seen in 8 patients (25%), negative in 11 (34%), and unknown in 13 (41%). Liver metastases were present in 12 of 32 assessed patients (38%).
Primary Endpoint Results
Median overall survival across all patients was 38 months, with survival rates of 94%, 92%, 75%, and 55% at 6, 12, 24, and 36 months, respectively. Median progression-free survival was 16 months, with rates of 62.5%, 55%, 31%, and 28% at the same time points. The 6-month disease control rate was 62.5% (20/32), comprising 5 complete responses, 9 partial responses, and 6 cases of stable disease. The maximum tolerated dose was not reached at any dose level tested.
Subgroup Analyses
Clinical responses occurred regardless of baseline PD-L1 status, LDH level, or BRAF mutation status — a pattern the authors note differs from what is typically seen with anti-PD-1 monotherapy, where these features often predict poorer response.
| Subgroup | Response Rate | n |
|---|---|---|
| PD-L1 positive | 62.5% | 5/8 |
| PD-L1 negative | 54.6% | 6/11 |
| LDH: ULN–2×ULN | 57% | 4/7 |
| LDH: >2×ULN | 60% | 6/10 |
| BRAF-mutant | 60% | 6/10 |
The authors are explicit that these subgroup comparisons are cross-trial rather than from a randomized, head-to-head design, and describe the findings as hypothesis-generating rather than confirmatory, given the small, non-randomized cohort. A biomarker analysis using multiplex immunofluorescence found that responders had significantly higher baseline densities of CD11c+ myeloid cells (P=0.01) and CD8+ T cells (P=0.04) than non-responders. Tumors with high baseline levels of both cell populations together showed improved overall and progression-free survival, though neither population alone correlated with outcome — a possible combined biomarker for future patient selection that the authors say still requires validation.
Safety Profile
| Grade 3/4 Treatment-Related Adverse Event (n=32) | Rate | n |
|---|---|---|
| Any | 25% | 8/32 |
| Colitis | 6% | 2/32 |
| Injection site reaction | 6% | 2/32 |
| Elevated ALT | 6% | 2/32 |
| Elevated AST | 3% | 1/32 |
| Pruritus | 3% | 1/32 |
| Maculopapular rash | 3% | 1/32 |
| Decreased platelet count | 3% | 1/32 |
| Grade 3/4 Immune-Mediated Adverse Event (n=32) | Rate | n |
|---|---|---|
| Any | 16% | 5/32 |
| Colitis | 6% | 2/32 |
| Pruritus | 3% | 1/32 |
| Maculopapular rash | 3% | 1/32 |
| Elevated ALT | 3% | 1/32 |
No patients discontinued treatment or died from treatment-related adverse events, and no dose-limiting toxicities occurred at any dose level. The authors specifically note the absence of cytokine release syndrome — a toxicity often associated with CD40-targeted agents — as a reassuring safety signal for this intratumoral delivery approach.
Interpretation and Broader Context
The trial’s extensive correlative work supports the intended biology: within 24 hours of sotigalimab injection, tumor biopsies showed activation of NF-κB and MAPK signaling, upregulation of antigen-presenting-cell genes, and increased infiltration of activated dendritic cells and macrophages. Single-cell sequencing of peripheral blood showed early (day 3–8) activation of circulating antigen-presenting cells followed by T-cell activation by day 21, and TCR sequencing showed expansion of new, shared T-cell clones across both injected and non-injected tumors — consistent with the "in situ vaccination" concept the therapy is designed around, rather than simply an amplification of a pre-existing immune response.
The combination was well tolerated, with no dose-limiting toxicities, treatment-related deaths, or discontinuations and a safety profile similar to pembrolizumab monotherapy.
— Discussion, Cancer Discovery (2026)
The authors frame the clinical results as promising but preliminary: comparisons to historical trials such as KEYNOTE-006 and ECHO-301/KEYNOTE-252 are cross-trial rather than head-to-head, and they caution that the numerically favorable outcomes "should be interpreted cautiously due to differences in study design, patient populations, and sample size." The dosing schedule was also limited to four doses into a single lesion, and the authors note a more prolonged regimen might further improve outcomes.
Limitations
This was a small (n=32), open-label, single-arm, non-randomized phase I/II study with no concurrent anti-PD-1-monotherapy comparator arm, so it cannot establish that adding sotigalimab improves outcomes over pembrolizumab alone. All comparisons to historical trials are cross-trial rather than head-to-head, and the biomarker subgroup findings — including the suggestion of activity in PD-L1-negative and high-LDH patients who typically respond poorly to anti-PD-1 monotherapy — are exploratory and hypothesis-generating, drawn from subgroups as small as 7 to 11 patients. The dosing regimen was capped at four intratumoral doses into a single lesion, and the study does not address whether a more prolonged schedule would change outcomes.
In this small, single-arm, non-randomized phase I/II trial, adding the intratumoral CD40 agonist sotigalimab to pembrolizumab produced a 50% objective response rate at the recommended phase II dose in ICB-naive metastatic melanoma, including responses in tumors that were never injected, with no dose-limiting toxicities or cytokine release syndrome. The results are encouraging but preliminary: without a randomized comparator arm, the trial cannot show that the combination outperforms pembrolizumab alone, and the biomarker and cross-trial comparisons the authors highlight remain hypothesis-generating pending larger, randomized confirmation.