AWS and MSK Team Up to Advance Precision Medicine with HPC and AI 

(Have a nice day Photo/Shutterstock)

Earlier this week, Amazon Web Services (AWS) announced a collaboration with global cancer center Memorial Sloan Kettering Cancer Center (MSK). Driven by artificial intelligence, high performance computing, and cloud computing, the project aims to address the projected 15 million annual cancer deaths by 2040 by accelerating breakthroughs in cancer research.

MSK’s mission is ending cancer for life, and this project will allow the organization to unlock insight from its years of clinical research and cancer treatments using AWS technology. This includes using deidentified genomic, imaging, and clinical data to create a powerful source for MSK’s AI driven cancer research and personalized treatment development, the organization said in a statement.

The goal of the collaboration is to build a high quality and updated longitudinal data resource for cancer research at MSK that will also serve as a source for alidating and improving cancer research with its partners.

“MSK’s collaboration with AWS is rooted in our shared vision and commitment to accelerate the pace with which we ideate, create, and bring lifesaving innovations to cancer patients. A critical enabler of this will be harnessing the scale and expertise of the larger Amazon organization to advance our processes and infrastructure related to AI, inclusive of large language models.” said MSK Chief Strategy Officer Dr. Anaeze Offodile II.

The partnership will leverage LLMs through Amazon’s Bedrock and SageMaker AI platforms, which will help MSK manage data quality constraints for building AI models, as well as increase the data density and speed through secure fine-tuning on cohorts of de-identified patient data. Bedrock offers pretrained foundation models for inference when building generative AI applications, while SageMaker AI is designed to help users build, train, and deploy machine learning models, including their own foundation models.

According to a release from MSK, these Amazon platforms will allow its researchers and providers to better track how a patient’s cancer changes over time using computable disease trajectories. Combined with advanced tumor and clinical response predictions, this data-driven approach can help uncover new insights, personalize treatments, and improve patient care, the organization says.

“Pairing MSK’s cancer research expertise and vast data resources from over a century of oncology care with AWS’s advanced cloud and AI capabilities creates a powerful engine for innovation,” said Dave Levy, AWS VP of worldwide public sector and healthcare and life sciences. “MSK and AWS have a shared vision for applying technology to improve patient care and experience, and we’re excited about how this collaboration will accelerate patient-centered discovery and precision medicine for cancer.”

(SOMKID THONGDEE/Shutterstock)

MSK will also leverage the collaboration to expand its other projects, including the Innovation Hub, its startup ecosystem.

The MSK Innovation Hub (iHub), launched in 2020, provides a model for collaboration on digital solutions for difficult oncology challenges. MSK and AWS will scale iHub for increased impact with a dedicated team focused on AI innovation, the organizations say, as well as support technology pilots to improve and validate applications in cancer through dedicated funding and access to AWS AI and tech startup expertise. Startups will work across a wide range of challenges in the oncology space, from developing novel treatments to improving clinical trials to improving the patient experience in the hospital and at home.

The project will also support MSK’s AI-enabled drug discovery efforts that are focused on finding therapeutics for high-unmet need cancers and orphan indications. Orphan diseases are considered rare diseases affecting fewer than 200,000 people in the U.S.

AWS offers a Drug Discovery Workbench where MSK researchers can streamline the early stages of drug development by rapidly screening millions of compounds to identify potential drug candidates. AWS’s HPC resources like its AWS ParallelCluster, designed to dramatically accelerate complex simulations and Cryo-EM data processing, tasks that are crucial for understanding protein structures and drug interactions at the molecular level. The organization aims to create a strong pipeline of clinical-stage therapeutics that will help bring life-saving treatments to cancer patients faster.

With cancer rates projected to rise and the demand for more effective treatments growing, the collaboration between MSK and AWS represents a pivotal step toward harnessing HPC and AI for a major impact in healthcare, offering new hope for patients and researchers alike.