ASCO, Google Cloud Launch AI-Powered Tool to Help Doctors Access Guidelines

Today, the American Society of Clinical Oncology (ASCO) and Google Cloud announced the launch of an artificial intelligence (AI) tool that aims to help doctors access the group’s evidence-based guidelines in a fraction of the time it takes now.

Known as ASCO Guidelines Assistant, the tool uses Google Cloud’s Vertex AI platform and Gemini models to help clinicians quickly pinpoint needed information from ASCO’s full library of evidence-based clinical guidelines.

A joint statement provided to The American Journal of Managed Care^® states,” Clinicians can expect a more dynamic user experience, allowing them to ask follow-up questions for intuitive, easy integration of expert information into their patient care plans.”

Clifford Hudis, MD, FACP, FASCO | Image credit: ASCO

The statement cites the time pressures on oncologists as the field faces both a shortage of physicians and growing number of older Americans—factors that combine to give doctors less time to hunt through guidelines at the point of care. Community oncology practices, in particular, are handling larger patient volumes—while handling a range of cancer types on any given day.

“The ASCO Guidelines Assistant addresses this urgent need in oncology by providing clinicians with a solution that removes the accuracy concerns often associated with publicly available AI tools,” the statement said. “By drawing solely from ASCO’s evidence-based, published clinical practice guidelines, the tool offers clinicians ready access to timely, trustworthy information.”

AI is being used in cancer care in multiple ways, from helping doctors diagnose specific types of skin cancer, to interpreting the status of a tumor microenvironment.¹ However, as authors from Perelman School of Medicine wrote in 2024, current tools have suffered from inconsistent performance once they reach real-world settings.²

"AI tools that have been integrated into clinical workflows in oncology clinics can analyze medical records and help doctors make more informed decisions, saving time and optimizing care," the authors wrote in Cancer. "However, the use of AI tools is limited by their inconsistent performance after deployment. Biases associated with algorithm development and implementation can lead to inaccurate predictions that burden healthcare systems, care teams, and individual patients."

Treating patients according to evidence-based guidelines is increasingly important, as payers are more likely than ever to not approve regimens that don’t fit within guidelines or clinical pathways.

“This partnership unlocks the value of ASCO's trusted guidelines knowledge," ASCO CEO Clifford Hudis, MD, FACP, FASCO, said in the statement. "Google Cloud’s AI expertise is the engine that will drive faster access to critical information, empowering oncology professionals to make rapid, evidence-based decisions for their patients."

In a presentation today, Hudis clarified that the ASCO Guidelines Assistant is not a decision-support tool. In a demonstration, he put in a query to trigger a response, which appears onscreen alongside a box that pulls up the relevant section of the ASCO guidelines. Rather than offering an instruction for care, Hudis noted that the tool quickly directs the physicians to the relevant guideline, allowing them to make decisions themselves.

Thomas Kurian | Image credit: LinkedIn

Officials from ASCO and Google Cloud said the tool is built under principles of reliability and transparency. To ensure accuracy, any information delivered will be backed by citations and information sources. The dynamic design allows users to ask follow-up questions and offer feedback.

The tool will be tested during the upcoming ASCO Annual Meeting among all attendees from May 30 to June 3, and then limited to ASCO members.

"Generative AI is poised to be a powerful ally for the global cancer care community,” Thomas Kurian, CEO, Google Cloud, added in the statement. “Think of it as providing clinicians with an AI-powered assistant that never tires, always learns, and can sift through mountains of data in seconds. Our collaboration with ASCO is about harnessing that potential to improve outcomes for cancer patients everywhere."

References

1. Luchini C, Pea A, Scarpa A. Artificial intelligence in oncology: current applications and future perspectives. Br J Cancer. 2022;126:4–9. https://doi.org/10.1038/s41416-021-01633-1

2. Kolla L, Parikh RB. Uses and limitations of artificial intelligence for oncology. Cancer. 2024;130(12):2101-2107. doi: 10.1002/cncr.35307.