Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • ADVERTISEMENT FEATURE Advertiser retains sole responsibility for the content of this article

Compugen: from code to cure

Immuno-oncology (IO) therapeutics are a promising way of treating cancer but, despite significant advances, most patients do not respond to current treatments. The development of such treatments is complex and relies on understanding a multidimensional spectrum of interactions between immune, tumor and stromal cells infiltrating the tumor microenvironment and draining lymph nodes—all of which can vary significantly among patient populations. Precise characterization of patient subpopulations and their predicted response to immunotherapies is therefore required in the discovery and development of precision medicine in cancer.

Compugen, a data-science-driven biotechnology company focused on target and biomarker discovery in IO, has developed proprietary predictive computational solutions to untangle this complexity, discover untapped drug targets, and support the drug development process to extend the reach of cancer immunotherapies to new patients.

Data science code

Compugen’s cloud-based analytics approach integrates public-domain genomics and clinical data with proprietary omics data, such as proteomics and spatial single-cell transcriptomics. It uses a unique analytics pipeline to standardize the data and augment it with machine-learning approaches and other algorithms developed in-house. The output of this process is used to predict novel IO-specific targets, biomarkers and mechanisms of action.

The computational-driven hypotheses are tested and validated by an internal wet-lab experimental group and might further inform the program’s clinical strategy and guide the stratification of patients in Compugen’s clinical trials. Consented research data from these clinical-trial participants is integrated back into the discovery cycle, providing an additional layer of proprietary data. “This tight in-house integration of computational prediction with experimental validation is one of Compugen’s strengths and has proved to be essential in our IO discoveries, clinical-stage programs, and pipeline progression,” said Anat Cohen-Dayag, president and CEO of Compugen.

Validated discovery cycle

Compugen’s discovery cycle has already yielded several potentially first-in-class drug candidates, currently in pre-clinical and clinical development.

The company identified PVRIG and TIGIT as immune checkpoints in the DNAM axis, an important signaling pathway in T- and NK-cell function, which led to the discovery and development of novel immunotherapies (Fig. 1). “We believe the DNAM axis is a foundational immuno-oncology axis,” said Cohen-Dayag. “TIGIT came from our computational capabilities, and we have identified PVRIG as a new inhibitory immune-checkpoint pathway—ground-breaking work that is evidenced by our numerous publications and granted patents.”

How Compugen goes from code to cure

Fig. 1 | How Compugen goes from code to cure. Novel IO targets and biomarkers are identified via proprietary computational solutions and validated experimentally towards guidance of clinical programs and patients’ stratification strategy. The validation and clinical data are integrated back into the discovery cycle to enhance the predictive computational models. APC, antigen-presenting cell; ML, machine learning.

The company’s clinical pipeline includes two proprietary product candidates, COM701 and COM902. COM701, a potential first-in-class anti-PVRIG antibody for treating solid tumors, is undergoing phase 1 studies as a single agent and in dual and triple combinations targeting PVRIG, TIGIT and PD-1. COM902, a potential best-in-class anti-TIGIT antibody, is undergoing phase 1 studies as a single agent and in combination with COM701 to treat solid and hematological tumors.

“Our preclinical findings suggest that, in tumor types where the PVRIG, TIGIT and PD-1 inhibitory pathways are operative, simultaneously blocking them may improve anti-tumor immune responses,” explained Cohen-Dayag. “If clinical trials are successful, combination therapy involving PVRIG and TIGIT may expand the reach of cancer immunotherapies to patient populations and cancer indications currently unresponsive or refractory to existing treatments.”

Preliminary data for COM701 are promising, suggesting potent immune activation with the triple blockade of PVRIG, TIGIT and PD-1, which complements the early signals of anti-tumor activity reported in Compugen’s studies. With COM902, the company was the first to present early signals of monotherapy anti-tumor activity for an IgG4 anti-TIGIT antibody, with low Fc-effector function. COM902 also avoided depletion of CD8+ T cells—the most effective anti-cancer immune subset—supporting its strategy of developing an antibody with low Fc-effector function.

In addition, Compugen’s therapeutic pipeline of early-stage IO programs address various mechanisms of immune resistance, including myeloid targets.

Partnering

Compugen’s proprietary analytics solutions and insights have resulted in strategic IO collaborations with pharma and academia, including long-standing partnerships with Bristol Myers Squibb, Bayer, AstraZeneca and Johns Hopkins University. Compugen’s partnered programs include: an antibody targeting ILDR2 (bapotulimab) in phase 1 development, an additional novel target discovered by Compugen, licensed to Bayer under an R&D collaboration and license agreement; and a TIGIT/PD-1 bispecific (AZD2939) derived from COM902 in phase 1/2 development through a license agreement with AstraZeneca for the development of bispecific and multispecific antibodies.

“We are open to forming additional collaborations involving our computational capabilities and pipeline programs,” said Cohen-Dayag. “With our data science and computational capabilities—and as the first company with clinical programs targeting both PVRIG and TIGIT—we are uniquely positioned to help bring novel IO treatment options to cancer patients.”

Search

Quick links