The science behind IMM60 begins with a molecule that has fascinated immunologists for decades. Alpha-galactosylceramide, a lipid originally isolated from a marine sponge, is the archetypal activator of iNKT cells. These cells occupy a peculiar niche in the immune system: they carry an invariant T cell receptor, a fixed molecular sensor that recognizes lipid fragments displayed by a specialized presenting molecule called CD1d. When the right lipid locks into CD1d and engages the iNKT cell receptor, the result is a rapid, coordinated burst of immune activity. Unlike conventional T cells, which need weeks to expand and specialize against a new threat, iNKT cells respond within hours, releasing signaling proteins that recruit and energize other immune players, including cells that kill tumors directly.

Alpha-galactosylceramide itself, however, has proved awkward as a drug. It is chemically unstable, difficult to formulate, and its potency fades during storage and in the bloodstream. IMM60 was engineered to be a more robust cousin: a synthetic lipid agonist with enhanced potency and stability, encapsulated within liposomes to protect it from degradation and deliver it efficiently to the CD1d presentation machinery. The idea is that a stable, liposomal formulation could finally translate the dramatic immune activation seen in laboratory studies into a therapy that works reliably in patients, potentially priming tumors for attack by the immune system and working hand in hand with existing immunotherapies.

The IMP-MEL trial enrolled fifteen patients with advanced melanoma or non-small cell lung cancer, and fourteen of them received treatment, six with melanoma and eight with lung cancer. Together they received sixty-six infusions of liposomal IMM60, with a median of five infusions per patient and a range of two to six. Some patients received the drug on its own, while others received it in combination with pembrolizumab, an antibody that blocks the PD-1 checkpoint and is already a mainstay of immunotherapy for both tumor types. The combination strategy rests on a simple logic: IMM60 sounds the immune alarm, while pembrolizumab removes the molecular brakes that tumors use to silence the responding cells.

Safety was the trial’s central question, and the answer was reassuring. No dose-limiting toxicities were observed at any of the doses tested, and the study did not need to define a maximum tolerated dose, the ceiling that phase I trials exist to find. Treatment-related adverse events did occur, affecting ten of the fourteen treated patients, or seventy-one percent, but every one of them was graded as mild or moderate, meaning grade one or two on the standard toxicity scale, and all were self-limiting. For a drug designed to deliberately provoke the immune system, the absence of severe inflammatory side effects is a notable achievement, since excessive immune activation has derailed many immunotherapy candidates before them.

The pharmacokinetic data, which track how the drug moves through the body, showed clean, predictable behavior. Exposure increased in proportion to dose, with the maximum plasma concentration showing a strong linear fit across dose levels, and the drug lingered in circulation with a mean residence time of about seven hours. That predictability matters enormously for a first-in-human agent, because it means clinicians can anticipate how the body will handle each dose rather than contending with erratic spikes or unexpected accumulation. The liposomal delivery appears to have tamed a molecule that, in its earlier form, was notoriously difficult to manage pharmacologically.

More important than how the drug circulated was whether it engaged its biological target, and here the trial provided direct molecular evidence. Within twenty-four hours of infusion, the researchers observed downregulation of the iNKT cell T cell receptor on the surface of circulating cells, a well-recognized sign that the cells have encountered their target ligand and internalized their sensor, confirming that IMM60 reached and stimulated its intended audience in living patients. Alongside this, the drug triggered a measurable wave of chemokines, the signaling molecules that guide immune cells to sites of inflammation and tumors. Monocyte chemoattractant protein-1, known as MCP-1 or CCL2, peaked at twenty-four hours with a geometric mean fold-change of 4.2 times baseline, a statistically significant rise. Interferon-gamma-inducible protein 10, or IP-10, rose faster, peaking at four hours with a 2.4-fold increase, also statistically significant.

Notably, the broad family of pro-inflammatory cytokines, the molecules most associated with the dangerous systemic immune storms that can accompany overzealous immunotherapy, showed no consistent elevation over baseline. This pattern, a targeted rise in chemokines that recruit immune cells without a indiscriminate inflammatory surge, is precisely the profile a rational iNKT agonist should produce. It suggests the drug can orchestrate immune traffic toward tumors while sparing patients the collateral damage of systemic inflammation, a balance that has been the holy grail of this therapeutic class since alpha-galactosylceramide first entered human testing years ago.

On the clinical side, the trial was small and early-stage, so dramatic tumor shrinkage was never the realistic expectation, particularly in a population this heavily pretreated. Even so, the signals were encouraging. Stable disease was the best overall response in the majority of assessed patients, and when the investigators examined individual tumor lesions, they found reductions in thirty percent of the lesions evaluated. Among these was a patient with melanoma whose cancer had resisted both BRAF and MEK inhibitor targeted therapy and PD-1 checkpoint blockade, one of the hardest treatment scenarios in modern oncology. Median progression-free survival across the treated group was 2.9 months, a modest figure that nonetheless reflects the refractory nature of the disease in these patients and the fact that the trial’s dose-escalation design was built to find a safe dose, not to maximize efficacy.

The authors conclude that liposomal IMM60 is well tolerated, shows clear evidence of iNKT cell target engagement and early chemokine induction, and produced signs of disease control in heavily pretreated and PD-1-refractory patients, findings they say support further evaluation of the drug in combination with checkpoint inhibition. The trial, sponsored and funded by iOx Therapeutics and conducted at sites including the Churchill Hospital in Oxford and Hammersmith Hospital in London, was prospectively registered in January 2021. As with any phase I study, the road ahead is long: larger trials will need to determine whether the immune activation seen in blood tests translates into meaningful extension of survival, and which doses and schedules best combine IMM60 with checkpoint blockade. But for a therapeutic concept that has struggled for years to make the leap from laboratory promise to clinical reality, this first-in-human dataset offers something precious: proof that a next-generation iNKT agonist can be given safely to cancer patients, can be seen doing its molecular job, and may be stirring the immune system in tumors that have run out of options.

Subject of Research: A first-in-human phase I trial of the liposomal iNKT cell agonist IMM60 with or without pembrolizumab in advanced melanoma and non-small cell lung cancer

Article Title: IMP-MEL: a phase I first-in-human study of IMM60 with or without pembrolizumab in advanced melanoma and non-small cell lung cancer (NSCLC)

Article References: IMP-MEL: a phase I first-in-human study of IMM60 with or without pembrolizumab in advanced melanoma and non-small cell lung cancer (NSCLC). (n.d.). https://doi.org/10.1186/s12967-026-08980-8

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08980-8

Keywords: IMM60, iNKT cells, cancer immunotherapy, melanoma, non-small cell lung cancer, pembrolizumab, phase I trial, liposomal formulation, alpha-galactosylceramide, CD1d, checkpoint inhibitor, chemokines

The biological logic behind the combination rests on a family of proteins called inhibitors of apoptosis, or IAPs. Three members of this family, cIAP1, cIAP2 and XIAP, act inside cells as gatekeepers of cell death and inflammatory signaling. Rather than simply preventing suicide, these proteins actively suppress a form of cell death driven by RIPK1, a signaling kinase that can trigger what scientists call immunogenic cell death. When tumor cells die immunogenically, they release signals that recruit and activate both innate and adaptive immune responses, effectively turning a dying tumor cell into a vaccine. By keeping this pathway in check, IAPs help tumors evade immune surveillance and blunt the effectiveness of drugs that release the brakes on T cells.

Tolinapant was designed to remove that shield. It is a non-peptidomimetic antagonist, meaning it binds the IAP proteins without mimicking the natural peptides that normally regulate them, and it does so with balanced activity against all clinically relevant IAP family members. The intended effect is to lower the threshold at which tumor necrosis factor, or TNF, becomes lethal to the tumor cell. In practical terms, the drug makes cancer cells far more vulnerable to being killed by T cells, the very cells that pembrolizumab unleashes by blocking the PD-1 checkpoint. The trial therefore tested whether priming tumors with tolinapant could convert immunologically cold or resistant cancers into ones that respond to PD-1 blockade.

The ASTEROID trial enrolled patients with advanced solid tumors and followed a classic phase I design. The primary endpoints were safety and the identification of a recommended phase 2 dose, while secondary endpoints captured anti-tumor activity and the pharmacokinetics of the combination. Dosing of tolinapant followed an intermittent schedule of seven days on treatment followed by fourteen days off, an approach intended to balance target engagement against toxicity. Pembrolizumab was administered on its standard intravenous schedule. The combination proved well tolerated overall, an important finding because IAP antagonists can, in principle, sensitize normal cells to TNF-driven inflammation as well as tumor cells.

The recommended phase 2 dose was established at 150 milligrams of tolinapant given on the intermittent schedule alongside pembrolizumab. Within that framework, the investigators observed early but meaningful evidence of efficacy. The overall response rate across the treated population reached 33 percent, a striking figure for a phase I study in a heavily pretreated cohort of mixed tumor types. Particularly notable were responses in estrogen receptor-positive breast cancer, a disease subtype that has historically resisted immune checkpoint blockade and is generally considered immunologically quiet. Seeing durable responses in this setting suggests that lowering the TNF cytotoxicity threshold can genuinely reprogram how tumors interact with the immune system.

Beyond the headline response rate, the trial’s translational component may prove its most enduring contribution. The team performed integrated analyses of tumor biopsies and blood samples to understand why some patients responded and others did not. Responders carried tumors characterized by a TNF and interferon-enriched inflammatory microenvironment, a molecular signature indicating that the intended biology, the sensitization of tumor cells to cytokine-driven killing, was actually occurring in patients. Responders also showed increased tumor antigenicity, meaning their cancers displayed more molecular features that the immune system could recognize as foreign, together with enhanced diversity of T cell receptor clones, a sign that a broader repertoire of T cells had been mobilized against the tumor.

Resistance, by contrast, had a coherent fingerprint of its own. Non-responding tumors were dominated by a niche of immune suppressor cells, specifically tumor-associated macrophages and regulatory T cells, orchestrated by the chemokine receptors CCR4, CCR6, CCR7 and CCR8. These receptors guide the trafficking of immune cells, and their enrichment points to a microenvironment in which suppressive myeloid and regulatory populations crowd out effector T cells, absorbing the benefits of checkpoint release. This finding suggests that patients who fail the tolinapant-pembrolizumab combination might in future be candidates for additional agents that deplete or reprogram macrophages and regulatory T cells, potentially creating triple or quadruple combinations tailored to the biology of each tumor.

The trial was funded by Cancer Research UK and Astex Pharmaceuticals, now part of Taiho Oncology, with drug supplied through Astex and Merck Sharp & Dohme’s investigator-initiated studies program. It was conducted across UK centers including The Royal Marsden, Cambridge University Hospitals and the Institute of Cancer Research’s Drug Development Unit, with laboratory work spanning the Meier Lab, the Centre for Evolution and Cancer and collaborating groups in Australia and Sweden. The study also reflects a growing trend in early-phase oncology: rather than treating phase I trials as pure safety exercises, the investigators embedded deep molecular profiling from the outset, so that every patient’s response contributed to a mechanistic map of the drug’s action.

For the field, the results constitute clinical proof of concept that dual blockade of IAP proteins and PD-1 can sensitize tumors to immunotherapy. They also deliver a practical dividend in the form of candidate biomarkers. The TNF and interferon signature, antigenicity measures and T cell receptor clonality could be developed into diagnostic tests to select patients most likely to benefit, while the CCR-driven suppressor niche signature could identify those who need combination strategies beyond PD-1 blockade. Larger trials will be required to confirm efficacy, define which tumor types gain the most, and validate the biomarkers prospectively. But the ASTEROID data provide a compelling early answer to a question that has frustrated oncologists for a decade: how to make immunotherapy work for the many patients whose tumors currently ignore it. By targeting the machinery tumors use to silence inflammatory cell death, tolinapant appears to reopen a channel that cancer had carefully closed, and the 33 percent response rate, including in a notoriously immunotherapy-resistant breast cancer subtype, is the kind of signal that reshapes research agendas.

Subject of Research: A phase I trial of the IAP antagonist tolinapant combined with pembrolizumab in advanced solid tumors

Article Title: IAP antagonist tolinapant (ASTX660) in combination with pembrolizumab in advanced solid tumors: a phase I trial

Article References: Tiu, C., Yau, W., Tenev, T., Lionarons, D., Leung, C., Sanchez-Perez, V., Codacci-Pisanelli, G., Lai, X., Zachariou, A., Hussain, A., Prout, T., Parmar, M., Baikady, B., Goicoechea, M., Layzell, S., Guppy, N., Roxanis, I., Lund, T., Dos Reis, D. C., … Meier, P. (2026). IAP antagonist tolinapant (ASTX660) in combination with pembrolizumab in advanced solid tumors: a phase I trial. Nature Communications. https://doi.org/10.1038/s41467-026-77552-7

Image Credits: AI Generated

DOI: 10.1038/s41467-026-77552-7

Keywords: tolinapant, ASTX660, pembrolizumab, IAP antagonists, immunotherapy, phase I trial, advanced solid tumors, ER-positive breast cancer, immunogenic cell death, PD-1 blockade, tumor microenvironment, biomarkers

Oncolytic viruses occupy a distinctive niche in cancer therapy. Unlike cytotoxic drugs or checkpoint inhibitors, they are replicating agents: they infect tumour cells, hijack their biosynthetic machinery, produce progeny virions and lyse the host cell, releasing new infectious particles that spread to neighbouring malignant cells. Vaccinia virus has long been favoured as a backbone for this approach because it replicates efficiently in the cytoplasm, carries a large genome that tolerates the insertion of therapeutic transgenes, and has an extensive clinical history through smallpox vaccination. The deletion of the thymidine kinase gene is a well-established tumour-selectivity strategy. Because rapidly dividing cancer cells often overproduce the nucleotide substrates that compensate for this metabolic deficit, the engineered virus can complete its replication cycle preferentially inside tumours while replicating poorly, if at all, in normal tissue.

What sets GC001 apart is the second layer of engineering layered onto this backbone. The virus carries an shRNA cassette that suppresses expression of STRIP1, a core component of the STRIPAK complex, a large scaffolding assembly that integrates phosphatases and kinases within the Hippo signalling pathway. The Hippo pathway is a central regulator of organ size, cell proliferation and apoptosis, and its dysregulation has been implicated in tumour progression, metastasis and resistance to cell death across many cancer types. STRIPAK complexes have also been shown to influence how cancer cells migrate and metastasise, which made STRIP1 an attractive target for viral payload design. By delivering a gene-silencing construct directly into infected tumour cells, GC001 aims to strike the tumour with two synchronised blows: the destructive force of viral replication and a molecular disruption of a survival pathway that tumours rely upon.

The preclinical evidence for this strategy is broad. In mouse xenograft models spanning multiple cancer types, including colorectal, pancreatic and gastric cancer, GC001 demonstrated potent and broad-spectrum anti-tumour activity. Colorectal cancer cells showed suppressed growth both in tissue culture and in implanted tumours, and parallel experiments in pancreatic and gastric cancer models confirmed that the effect was not confined to a single tumour lineage. This spectrum matters clinically, because an oncolytic virus that requires a tumour-specific genetic vulnerability present in only a fraction of patients will struggle in late-stage trials. The breadth observed here suggests that the combined mechanism of viral lysis and STRIP1 knockdown operates through pathways that are active across common gastrointestinal malignancies and likely beyond.

Perhaps the most revealing part of the study is mechanistic. The researchers found that GC001 drives a sharp accumulation of reactive oxygen species, or ROS, within infected tumour cells. ROS are chemically reactive oxygen-containing molecules that, at low levels, function as signalling intermediates, but at high levels cause oxidative damage to proteins, lipids and DNA, pushing cells toward death. Critically, the team discovered that neither component alone produced this effect: STRIP1 knockdown by itself did not induce ROS accumulation, and the discussion of the data indicates that the burst of oxidative stress arises from the combinatorial activity of STRIP1 suppression and the oncolytic vaccinia backbone. The ROS surge was traced to two converging sources: degradation of NRF2, the master transcription factor that tumour cells use to mount antioxidant defences, and a reprogramming of mitochondrial metabolism that alters how the infected cell generates energy and handles electron flow.

This dual mechanism has consequences that extend beyond simple tumour cell killing. The NRF2 axis is one of the most commonly exploited antioxidant programmes in cancer, allowing malignant cells to survive the elevated basal oxidative stress that comes with rapid proliferation and metabolic rewiring. By degrading NRF2, GC001 effectively removes the tumour cell’s antioxidant shield at precisely the moment viral replication is placing it under maximal metabolic strain. The mitochondrial metabolic reprogramming component compounds this vulnerability, forcing the cell’s energy-generating organelles into a state that favours ROS production. The authors frame this as a ROS-Hippo pathway interaction: viral replication and cytotoxicity are potentiated through ROS-dependent modulation of Hippo signalling, the very pathway that STRIP1 knockdown targets directly.

The oxidative stress also appears to serve an immunological function. In both immunocompetent mouse models and humanised mouse models, GC001 stimulated measurable anti-tumour immune responses. This matters because the ultimate success of oncolytic virotherapy is generally believed to depend less on the direct killing of infected cells and more on the conversion of an immunologically cold tumour into an inflamed one. Viral infection releases tumour antigens and danger signals that recruit and activate immune cells, and the resulting response can, in principle, attack tumour cells far beyond the reach of the injected virus. The finding that GC001’s ROS-generating mechanism contributes to these immune responses adds a mechanistic bridge between the virus’s intracellular biochemistry and the systemic anti-tumour immunity observed in the animal models.

Safety, the perennial concern for any replicating virus intended for systemic or intratumoural administration, was addressed directly. In a nonhuman primate study, GC001 did not elicit significant toxicity and was well tolerated. The primate experiments were conducted under the regulatory requirements of the Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee of the National Center for Safety Evaluation of Drugs, and co-author Xin Wang of the National Institutes for Drug Control oversaw the design and performance of these safety studies. Given that vaccinia virus retains its capacity to infect human cells, demonstrating a clean tolerance profile in a species phylogenetically close to humans is a meaningful de-risking step before the phase I trial that the NMPA authorised in 2023.

The study also sits within a rapidly maturing field. Oncolytic viruses have moved from experimental curiosity to approved medicine with talimogene laherparepvec, a modified herpes virus for melanoma, and intratumoral oncolytic herpes virus G47∆ has shown phase 2 efficacy against residual or recurrent glioblastoma. Vaccinia-based platforms have been tested intravenously in humans and are being armed with payloads ranging from interleukin-12 to anti-CTLA-4 antibodies and immune-modulating lectins. Recent work from the same group has explored oncolytic vaccinia viruses expressing CLEC2A and avian lectins that boost viral replication and ROS-mediated killing in pancreatic and gastric cancers. GC001 extends this logic into gene-silencing territory, coupling the viral platform to RNA interference aimed at a specific signalling node rather than an immune-stimulatory cytokine.

There are, of course, familiar caveats. The efficacy data derive from xenografts and mouse models, systems that have historically overpredicted clinical responses in oncology, and the funding of the study by Hangzhou Gongchu Biotechnology, together with the corresponding patent position held by the company and a founder-authored manuscript, warrants the standard attention to conflicts of interest that accompanies commercial virotherapy development. The delivery route, dosing schedule and interaction with existing immunotherapies in humans remain to be established. Even so, the combination of broad preclinical activity, a mechanistically coherent dual-action design, immune activation in immunocompetent models and a favourable tolerability profile in nonhuman primates positions GC001 as a serious clinical contender. As the phase I trial proceeds, the central question will be whether the elegant ROS-Hippo synergy observed in the laboratory survives translation into patients, and whether an armed poxvirus that silences a Hippo pathway scaffold can deliver on the promise that two decades of oncolytic virus engineering has been building toward.

Subject of Research: Oncolytic vaccinia virus therapy engineered to knock down STRIP1 for cancer treatment

Article Title: Efficacy and safety of GC001: an oncolytic vaccinia virus expressing STRIP1 shRNA

Article References: Li, G., Jiang, G., Pang, W., Cai, Z., Liu, R., Zhang, H., Zhu, J., Su, J., Zhu, B., Yu, J., Feng, H., Zhu, Y., Zhang, G., Xu, W., Ye, T., Chen, K., Zhou, Y., Pan, D., Sun, L., … Ning, Y. (2026). Efficacy and safety of GC001: an oncolytic vaccinia virus expressing STRIP1 shRNA. British Journal of Cancer. https://doi.org/10.1038/s41416-026-03618-4

Image Credits: AI Generated

DOI: 10.1038/s41416-026-03618-4

Keywords: oncolytic virus, vaccinia virus, GC001, STRIP1, Hippo pathway, reactive oxygen species, NRF2, cancer immunotherapy, virotherapy, phase I trial, colorectal cancer, mitochondrial metabolism