BioNTech (BNTX) investor relations material

Strategic vision and AI integration

• AI is fully integrated across R&D, business operations, and manufacturing, focusing on oncology and infectious diseases, leveraging in-house and collaborative innovation with InstaDeep and global AI centers.

• The company advances personalized cancer vaccines and combination therapies, using AI to predict neoantigens, optimize vaccine design, and enable rapid, individualized mRNA vaccine production.

• Pursues a multi-platform oncology approach with 16 clinical programs and over 20 ongoing Phase 2 or 3 trials, including partnerships with major pharma companies such as Bristol Myers Squibb for novel bispecific antibodies.

• The holistic approach aims to create a future pharmaceutical model integrating immunity, cancer understanding, and therapeutic development through modular, universally aware AI systems.

• Combines immunomodulators, mRNA cancer immunotherapies, and targeted therapies for synergistic, potentially curative cancer treatments.

AI infrastructure and model innovation

• InstaDeep has built a robust AI infrastructure, including the Kyber supercomputer (~500 PetaFLOPS) and Alchor/AIChor platform, enabling large-scale model training, high-throughput experimentation, and 75% GPU utilization.

• The Nucleotide Transformer (NTv3) model, trained on over 150,000 species, sets new standards in genomics with multi-species, multimodal, and generative capabilities, outperforming state-of-the-art models in gene finding and regulatory element detection.

• Bayesian flow networks (AbBFN2) and ProtBFN enable flexible, multimodal antibody and protein design, optimizing multiple properties and reducing experimental timelines from months to minutes.

• Machine learning interatomic potentials (MLIP) provide quantum-level accuracy for molecular simulations at a fraction of traditional costs, accelerating drug and material discovery.

• AI models generate synthetic data, improve through reinforcement learning, and co-design hardware for accelerated progress.

Data strategy and applications

• Data-driven strategies combine large-scale open-source data with high-quality, in-house clinical and experimental data to enhance model performance and discovery.

• InstaNovo technology enables discovery of novel tumor antigens from the dark proteome, with next-generation models doubling peptide identification rates and increasing accuracy by 10-15%.

• Advanced digital pathology platforms and foundation models cluster and annotate histology data, increasing pathologist efficiency fivefold and improving model performance.

• AI-assisted de novo protein and nanoparticle design enables creation of novel vaccine platforms, validated by successful lab assembly and antigen functionalization.

• AI-guided TCR engineering pipeline achieves up to 50,000-fold affinity enhancements, enabling new therapeutic modalities and robust in vivo efficacy.