Japan's New Energy and Industrial Technology Development Organization (NEDO) has selected a five-party consortium led by NEC Corporation and Taiho Pharmaceutical to build a quantum computing platform for cancer immunotherapy research, with the project running from September 2026 to March 2029.

The program — formally titled "Development and Demonstration of a Computational and Evaluation Platform for Next-Generation Cancer Immunotherapy Using Quantum Computing Technology" — falls under NEDO's Research and Development Project of the Enhanced Infrarastructures for Post-5G Information and Communication Systems, in its Large-Scale Demonstration for Use Case Creation track. Beyond NEC and Taiho, the partners are the Japanese Foundation for Cancer Research (JFCR), the National Institute of Advanced Industrial Science and Technology (AIST), and Waseda University, where Professor Nozomu Togawa of the Faculty of Science and Engineering serves as principal investigator.

The consortium will combine quantum computing, AI, drug discovery and immunology expertise to establish a computational and evaluation platform for designing and validating neoantigen candidates — molecules regarded as promising targets for next-generation cancer immunotherapy.

Why neoantigens, and why MHC class II

Cancer immunotherapy has advanced considerably, with immune checkpoint inhibitors now among the most widely used approaches. But not all patients benefit sufficiently from them, which drives the search for new modalities. Neoantigens — which arise specifically in cancer cells and appear on the cell surface bound to major histocompatibility complex (MHC) molecules — are a leading candidate for closing that gap, because they are absent from healthy tissue.

The project zeroes in on CD4-positive T-cell responses induced through MHC class II molecules. These responses are expected to activate other immune cells and sustain immune reactions over time, making them a key driver of durable antitumor immunity. The catch is combinatorial complexity: amino acid sequence characteristics, MHC class II binding and cell surface presentation interact in ways that remain poorly understood, and no systematic method exists today for designing neoantigen sequences that induce CD4-positive T-cell responses through MHC class II while accounting for all of these factors simultaneously.

Evaluating the vast number of possible amino acid sequence combinations, plus the biological processes of antigen formation, presentation and immune recognition, is exactly the class of multifactorial optimization problem the consortium wants to attack with quantum hardware.

Compute plus wet lab, in a closed loop

The workflow is a feedback loop. AI models will predict and score immune responses to neoantigen candidates. Quantum computing technology will be applied to the design of amino acid sequences flanking the neoantigen core region, exploring a broader design space than conventional methods and generating diverse candidates with enhanced immunogenicity. Immunological experiments will then validate the optimized sequences, and the experimental results feed back into the computational models and AI prediction methods. The stated goal is a new drug discovery pipeline that tightly couples computational candidate design with experimental validation.

The compute substrate is ABCI-Q, an infrastructure integrating quantum computing, high-performance computing and AI that is being developed by the Global Research and Development Center for Business by Quantum-AI Technology (G-QuAT) at AIST. ABCI-Q gives the project a hybrid architecture rather than a standalone quantum machine — quantum processors handle the combinatorial sequence optimization while HPC and AI carry the prediction and evaluation workloads.

For NEDO, the project doubles as a drug discovery use case demonstrating how quantum technology can serve the Post-5G infrastructure program's broader mandate of use case creation. The partners say they aim to accelerate drug discovery innovation using quantum technologies originating in Japan — a framing that ties the research both to industrial policy and to the country's push to commercialize domestic quantum capability.

NEC, founded in 1899, brings roughly 110,000 employees and its portfolio of AI, security and communications technologies to the effort; the company positions such projects as part of creating social value through technology.

If the platform proves out by March 2029, it would give Japan's pharmaceutical and quantum computing sectors a reusable template: a validated pipeline in which quantum-designed candidates are ranked by AI and confirmed in the lab, potentially shortening the path from sequence design to immunotherapy candidates.