Insilico Medicine has developed a new class of small molecule inhibitors targeting diacylglycerol kinase alpha (DGKα) designed to restore T cell function and overcome resistance to immune checkpoint blockades in solid cancers. The latest results from this program have just been published in the Journal of Medicinal Chemistry, describing the discovery and comprehensive preclinical evaluation of Compound 10, a novel, potent, selective and orally administered DGKα inhibitor. The compound exhibits a differentiated pharmacokinetic and safety profile and strong combination activity with anti–PD-1 and anti–CTLA-4 therapies in multiple syngeneic tumor models.
The paper, titled " Design, Synthesis and Biological Evaluation of Novel, Potent, Selective and Orally Available DGKα Inhibitors for the Treatment of Tumors, " details how Insilico combined AI-guided target validation and generative chemistry to move from a highly lipophilic reference molecule, which showed suboptimal solubility and oral pharmacokinetics, to a series of more drug-like DGKα inhibitors, culminating in Compound 10. In preclinical studies, Compound 10 showed sub-nanomolar enzymatic potency, robust T cell activation in human and mouse systems, improved solubility and oral exposure versus the starting scaffold, and the ability to synergistically boost the antitumor efficacy of PD-1 and CTLA-4 checkpoint inhibitors, while maintaining good tolerability.
When trying to tackle the longstanding problem of checkpoint inhibitor resistance in highly refractory cancers, Insilico first turned to its AI-powered target discovery platform, PandaOmics , to search for fresh angles on responder versus non-responder biology. The team analyzed multiomics datasets from patients with melanoma and clear cell renal cell carcinoma, focusing on cohorts with documented responses to checkpoint inhibition. Using a combination of disease relevance scores and disease-agnostic filters for druggability, novelty, and predicted safety, PandaOmics highlighted DGKα as one of the top immune-oncology targets associated with checkpoint resistance. Although DGKα was already a known and mechanistically compelling node, previous efforts to drug it have not yielded substantial clinical or commercial success, which only sharpened the team's resolve to apply Insilico's structure-based design and generative chemistry tools to overcome these challenges to create a new generation of DGKα inhibitors.
DGKα is a lipid kinase that converts diacylglycerol (DAG) into phosphatidic acid (PA), shifting DAG-dependent signaling such as Ras/ERK and PKC pathways that drive tumor cell proliferation and survival, and dampening T cell receptor signaling in the tumor microenvironment to promote immune evasion. Overexpression of DGKα has been reported in tumors such as glioblastoma and melanoma and is associated with chemotherapy resistance and an immunosuppressive tumor microenvironment, where T cell activation and effector cytokine production are blunted. In T cells, DGKα acts as a brake on T cell receptor signaling, promoting an anergic, low-function state that is increasingly recognized as a key contributor to primary and acquired resistance to PD-1/PD-L1 therapy. Inhibiting DGKα can shift this balance back toward effective antitumor immunity by both reactivating intratumoral T cells and disrupting pro-survival signaling in tumor cells, offering a dual-mechanism strategy to reverse immune anergy and deepen responses to checkpoint blockades.
Historically, DGKα has been challenging to drug. First-generation inhibitors achieved nanomolar potency but failed to translate due to poor selectivity, off-target kinase activity, short half-lives and low oral bioavailability. More recently developed DGKα inhibitors have advanced into early clinical testing in combination with checkpoint inhibitors, but patent data indicating relatively high lipophilicity and limited disclosure of structural and pharmacokinetic data raise potential developability concerns. Together, these factors underscored the need for a new generation of DGKα inhibitors with improved selectivity, pharmacokinetics, and overall drug-likeness.
Because no experimental DGKα co-crystal structures were available to use as a basis for novel drug design, the team turned to AlphaFold-predicted DGKα models and homology modeling to build structure-based hypotheses. Insilico's generative chemistry platform Chemistry42 , together with molecular modeling, explored modifications around scaffolds that would reduce lipophilicity (cLogP) while preserving structural pieces critical to its interactions with the DGKα binding pocket. Guided by these models, the medicinal chemistry team carried out systematic structure–activity relationship (SAR) studies to reduce lipophilicity, introduce targeted hydrophilic substituents, and fine-tune the drug's hydrophobic tail, all while maintaining enzymatic potency and T cell activation. This iterative process yielded a series of DGKα inhibitors with reduced cLogP and sub-nanomolar IC₅₀ values. Among these, Compound 10, emerged as the optimal balance between potency, T cell activation, and developability.
The paper further reports that Compound 10 displays improved in vivo pharmacokinetics in mice relative to the reference molecule, with less than one-eighth the clearance rate, a half-life more than 2.5-times longer, and markedly higher oral exposure. In vitro ADME profiling showed enhanced solubility in biorelevant media, acceptable cell permeability, low to moderate microsomal clearance, good plasma stability, no meaningful inhibition of major CYP isoforms, and an hERG safety margin consistent with further development. Compound 10 enhanced stimulation of human and mouse immune cells in vitro, including increased secretion of activating cytokines such as IL-2 and IFN-γ, greater T cell proliferation, and stronger T cell activation. These data support its role as a potent immunomodulatory agent with the potential to boost antitumor immune responses.
Given this immune-activating profile, the team evaluated orally administered Compound 10 in MC38 and CT26 syngeneic mouse tumor models in combination with immune checkpoint inhibitors. Adding Compound 10 to anti–PD-1 and anti–CTLA4 therapy produced a clear, dose-dependent boost in antitumor activity, with tumor growth inhibition (TGI) up to 93% at doses up to 3 mg/kg, with a confirmed synergistic effect with both checkpoint inhibitors. All regimens were well tolerated with no significant body weight loss and low brain penetrance, mitigating concerns about DGKβ-related CNS effects, together supporting its potential as a backbone agent for combination immunotherapy.
Harnessing state-of-the-art AI and automation technologies, Insilico has significantly improved the efficiency of preclinical drug development. While traditional early-stage drug discovery typically requires 3 to 6 years, from 2021 to 2024 Insilico nominated 20 preclinical candidates, achieving an average turnaround - from project initiation to preclinical candidate (PCC) nomination - of just 12 to 18 months per program, with only 60 to 200 molecules synthesized and tested in each program.
References
[1] Lu, H. et al. "Design, Synthesis and Biological Evaluation of Novel, Potent, Selective and Orally Available DGKα Inhibitors for the Treatment of Tumors." Journal of Medicinal Chemistry (2025). https://doi.org/10.1021/acs.jmedchem.5c01943
About Insilico Medicine
Insilico Medicine, a leading and global AI-driven biotech company, utilizes its proprietary Pharma.AI platform and cutting-stage automated laboratory to accelerate drug discovery and advance innovations in life sciences research. By integrating AI and automation technologies and deep in-house drug discovery capabilities, Insilico is delivering innovative drug solutions for unmet needs including fibrosis, oncology, immunology, pain, and obesity and metabolic disorders. Additionally, Insilico extends the reach of Pharma.AI across diverse industries, such as advanced materials, agriculture, nutritional products and veterinary medicine.
