Blocking IL1RAP Sensitizes Pancreatic Tumors to Chemoimmunotherapy in New Trial and Mouse Data

Pancreatic ductal adenocarcinoma resists chemotherapy and checkpoint immunotherapy in large part because the tissue around the tumor, not only the tumor cells, stays chronically inflamed and keeps T cells out. A June 2026 research letter in JCI Insight, from a University of Miami team working with the Swedish biotech Cantargia AB, traces that resistance to a shared coreceptor called IL1RAP and shows that blocking it restores drug sensitivity in mice and in post hoc analysis of a completed human trial.

IL1RAP Marks a Chemoresistant Circuit Across Tumor, Stroma, and Myeloid Cells

IL-1 family cytokines, including IL-1α, IL-1β, IL-33, and IL-36, all converge on the IL-1 receptor accessory protein IL1RAP, and high tumor IL1RAP expression already correlated with shorter survival in pancreatic ductal adenocarcinoma (PDAC) before this study. To test whether that correlation reflects an active resistance mechanism, the JCI Insight research letter reanalyzed transcriptomes from the COMPASS trial, comparing 39 chemotherapy-resistant advanced PDAC tumors against 156 chemotherapy-responsive tumors. IL1RAP-related pathways were enriched in the resistant group.

The team then turned to the Human Tumor Atlas Network's PDAC single-cell atlas, a dataset covering roughly 655,000 cells from 79 patients. IL1RAP expression was elevated not just in tumor and acinar cells but across immune and myeloid populations and stromal cancer-associated fibroblasts (CAFs), and that elevation was specifically pronounced in chemotherapy-resistant samples across all three compartments. That distribution matters for treatment design: a biomarker confined to tumor cells would only ever flag tumor-intrinsic resistance, while one expressed across the surrounding stroma and immune infiltrate points toward a network-level target rather than a cancer-cell-only one.

Duration of Response to Nadunolimab Tracks Stromal and Immune IL1RAP, Not Tumor Cells Alone

IL1RAP already had a clinical readout before this letter. The phase 1/2a CANFOUR trial (NCT03267316) tested nadunolimab, an anti-IL1RAP antibody developed by Cantargia, alongside gemcitabine and nab-paclitaxel in 76 patients with advanced or metastatic PDAC. The published trial results reported a median overall survival of 13.2 months and found that patients with high baseline tumor-cell IL1RAP had better survival on the combination.

The new JCI Insight analysis re-scored biopsies from that same trial by IL1RAP expression on stromal/CAF and immune cells rather than tumor cells alone. Elevated stromal IL1RAP (34 biopsied patients) and elevated immune IL1RAP (19 patients) each associated with a longer duration of response to nadunolimab plus chemotherapy: stromal IL1RAP-high carried a hazard ratio of 0.41 for a shorter response (95% CI, 0.18–0.94; n = 24), and immune IL1RAP-high carried a hazard ratio of 0.36 (95% CI, 0.13–1.02; n = 19). The compartment-specific read is mechanistically more precise than the trial's original tumor-cell-only stratification, since it points at the myeloid-stromal network rather than the cancer cell itself, but it also rests on subgroups covering well under half of the trial's 76 enrolled patients, and the immune-compartment interval crosses 1.0, meaning that stratum alone does not reach clean statistical separation.

Blocking IL1RAP Reprograms Myeloid Cells and Rescues Exhausted T Cells in Mouse Tumors

To test causation rather than correlation, the team turned to the Ptf1aCre/+;LSL-KrasG12D/+;Tgfbr2fl/fl (PKT) mouse model, which reproduces the dense, inflamed stroma seen in human PDAC and, by single-cell sequencing, showed the same broad IL1RAP module expression across tumor, stromal, and myeloid compartments found in the human atlas. Treating tumor-bearing PKT mice with a murine version of nadunolimab (10 mg/kg, three times weekly) slowed tumor growth (P = 0.019) and epithelial proliferation (P = 0.002) compared with an isotype control, and reduced both stromal fibrosis on Sirius red staining (P = 0.0003) and desmoplasia on Masson's trichrome staining (P = 0.0088).

Flow cytometry showed the treated tumors had fewer CD11b+ myeloid cells (P = 0.0077) and more CD3+ T cells (P = 0.0033). The same direction of change appeared in two paired human biopsies collected before and after nadunolimab treatment: immunofluorescence showed fewer myeloid cells and more granzyme B– and Ki67-positive CD8 T cells inside the tumor tissue itself, not at its margins. Within the CD8 population, mNadu-treated mice showed more Ly108+CD69+/– memory-progenitor exhausted cells (P = 0.0278) and fewer terminally exhausted Ly108–CD69+ cells (P = 0.0084), a shift linked elsewhere to better responsiveness to checkpoint blockade.

Single-cell RNA sequencing of the treated tumors traced this shift to specific transcriptional programs. Myeloid cells downregulated immunosuppressive genes including Arg1, Cybb, Chil3, Cd177, and Camp along with IL-1 and TNF signaling, and upregulated type I interferon response and antigen presentation pathways. T cells moved toward memory and stem-like programs marked by T cell receptor engagement, interferon-gamma response, and IL-2-driven CD8 activation. Immunofluorescence confirmed the same pattern at the protein level, with fewer CD206+F4/80+ M2-like macrophages and more MHC-II+F4/80+ M1-like macrophages and CD3+ T cells in treated tumors (all P < 0.05).

A Four-Drug Combination, a Funder-Linked Trial, and What Remains Unproven

The mechanistic case built toward a combination test. PKT mice treated with mNadu plus gemcitabine/paclitaxel and an anti-PD1 antibody survived longer than mice given the isotype control or chemoimmunotherapy alone (median 54 days versus 43 days for either comparator, P < 0.001). mNadu alone also outperformed the isotype control (50 days versus 43 days, P = 0.023), suggesting IL1RAP blockade contributes a survival benefit of its own rather than only enabling the other two drugs to work.

The authors report that a neoadjuvant trial combining nadunolimab with chemoimmunotherapy in patients with operable PDAC is near deployment, though the letter does not name a registry number, so that trial's design and timeline cannot yet be checked against a public listing. A related 2024 paper in the Journal for Immunotherapy of Cancer, a companion mechanistic study from an overlapping Cantargia-affiliated group, offers independent support for the same biology: blocking IL1RAP on cancer-associated fibroblasts suppressed IL-1-driven neutrophil and monocyte recruitment in a separate PDAC model.

Two limitations bear directly on how much weight this evidence can carry. The JCI Insight paper is a Research Letter, a shorter format built for early, hypothesis-generating findings rather than a full clinical study, and its human confirmation of the mechanism rests on immunofluorescence from two paired biopsies. It also carries a direct funding relationship to the drug it evaluates: corresponding author Jashodeep Datta discloses funding from Cantargia AB, nadunolimab's developer, and the work itself was supported in part by a Cantargia research contract alongside NIH and Department of Defense grants. None of that invalidates the biology, which is corroborated across an independent trial dataset, an independent mouse model, and a separately published mechanistic paper, but it is the kind of vendor tie that belongs in the open rather than left for a reader to find elsewhere.