When treating cancer, few things are more precious than time. Jeanine Bortel, Vice President and Head of AI Portfolio Development at Pfizer, who has developed clinical trials at Pfizer for more than two decades, remembers when she first witnessed the transformative potential of artificial intelligence (AI) put into practice to help address the needs of patients enrolled in cancer clinical trials.

For years, Bortel’s work had run up against an inherent challenge: Many cancer patients miss out on the opportunity to enroll in clinical trials. When someone is diagnosed with metastatic cancer, there’s often little time to sign up for a trial before a treatment plan must be determined. Missing that narrow window denies patients a “potentially life-changing opportunity to be part of a clinical trial,” Bortel says.

Thanks to rapid advances in AI, however, Bortel witnessed a turning point. Her “aha” moment, she recalls, occurred not long ago when she was managing team leaders for breast cancer trials. In the course of their research, machine learning—a type of AI that learns patterns from data it has access to—drastically improved their ability to predict who might be eligible for clinical trials and, in turn, offer them the chance to enroll before their window closed. Suddenly, these patients had access to treatment options that might have been inaccessible without AI.

“It unlocked a major opportunity,” says Bortel. “We have the ability to be transformative in how we identify patients who are likely to become eligible for a trial — something we simply couldn’t do before these new technologies.”

Oncology’s AI-Powered Future

Earlier identification of prospective patients and, in turn, broadening access to clinical trials, is just one of the breakthroughs in oncology made possible by AI. The stakes, of course, couldn’t be higher: Cancer is the second leading cause of death in the United States, claiming more than 600,000 lives a year.1 AI’s extraordinary ability to review and process data is helping enhance experts’ ability to study cancer, design treatments, and run clinical trials. For so long a tantalizing opportunity for oncology, AI is, at last, beginning to accelerate breakthroughs and impact lives.

Though the power and potential of AI are in a league of their own, oncology has leveraged predictive data science for decades. In the 1950s, researchers began using mathematical modeling to better understand the way cancer progresses. Computational biology, which gleans insights from vast sets of cancer data, emerged in the 1970s.2 It wasn’t until the 1990s, according to Tamara Mansfeld, Vice President and Head of AI Portfolio Research at Pfizer, that computational biology began to play a substantial role. Even then, Mansfeld says, it was often difficult for scientists in the lab to validate hypotheses generated by computational models.

Not until much more recently, Mansfeld explains, did insights from AI become actionable in healthcare consistently, thanks to the “convergence” of rapid progress in AI as well as the exponential growth of health data.

When it comes to health technology, Mansfeld adds, “oncology has always been, historically, on the cutting edge.” That’s why recent developments in AI are so promising: not only for what it means for how we combat cancer, but also what it could foreshadow for medicine on a much more global scale.

Understanding and Attacking Cancer

Honing the most sophisticated cancer treatments requires understanding how the disease develops and progresses on a molecular level.

In 2020, Pfizer partnered with the startup, PostEra, which helps researchers discover new, synthetic chemical structures, known as payloads. That work allows Pfizer to deliver new antibody-drug conjugates (ADCs) — powerful approaches to fighting cancer.3,4

PostEra exemplifies why AI is about more than just expediting the conventional way of doing things; it’s also about enhancing the precision and creativity of medicine.

“You wouldn’t be able to design these payloads [as efficiently] without AI,” Mansfeld says. With that tool in hand, “you’re generating the [potential] next generation of medicines for oncology.”

Streamlining Documentation and Enrollment

There’s a pronounced sense of urgency with all medical research, and certainly with cancer.

Just a couple of years ago, Bortel remembers experimenting with large language models to co-author Pfizer’s research and development drafts. They couldn’t cut it — the output was rife with mistakes. But LLMs have made significant leaps to the point that AI now regularly helps generate usable first drafts. Pfizer’s Document Generation ecosystem is slashing the time it takes to produce a first draft of a clinical study report by 40%, reducing the overall manuscript submission timeline by 15%.5

The result, according to Bortel, is more time for science and research, the work that matters most.

In oncology, participation in clinical trials is limited by more than just the tight windows patients have to enroll. Traditionally, the opportunity to participate in a clinical trial isn’t as visible across patient populations, and patients can’t join what they don’t know about.

AI helps to bridge that gap by securely reviewing complex medical data from a variety of clinical data sets to identify cancer patients eligible for clinical trials. Over time, Pfizer aims to increase enrollment rates by more than 20%.5

That’s a win-win — for patients who get the opportunity to enroll in trials they might not have otherwise, and for the trials themselves.

From Pipeline to Patients

AI accelerates the administrative work involved in recruiting patients for clinical trials and the development of innovative medicines, freeing scientists to apply their expertise in the most efficient ways possible. But it’s more than just speed. AI is elevating the precision and accuracy of science. It’s harnessing data in ways never thought possible. All of this is already making tangible improvements in the pursuit of treatment options for cancer patients.

And, in many ways, this is just the beginning.