Shifts to Proactive Cancer Care Represent One of Many Benefits of AI Integration

Proactive cancer care, accelerated clinical trials, and the translation of complex concepts to patients all represent shifts in oncology workflows that artificial intelligence (AI) is positioned to underpin, according to Douglas Flora, MD, LSSBB.

“In 2026, [AI is commonly utilized] in back-of-the-store functions. It’s getting good at pattern recognition around revenue cycle management, business intelligence, and prior authorization. There are also great [AI] tools out there for tumor registry, lung nodule recognition, pathology, mammography, and film readings,” Flora, the executive medical director of Oncology Services for the Yung Family Cancer Center at St Elizabeth Healthcare in Edgewood, Kentucky, said in an interview with OncLive^®. “I’m excited for the next year, where [AI tools] will be used for more clinical decision support, electronically reported patient outcomes, and helping me do my job better.”

How has the integration of AI started in oncology? How can the use of AI tools evolve past initial integration?

During the Physicians' Education Resource^® 30th Annual International Congress on Hematologic Malignancies^® in Miami Beach, Florida, Flora’s keynote lecture on AI’s ability to reboot cancer care mapped out roadblocks and issues in the current oncology field, in addition to how AI can help overcome these obstacles and elevate cancer care to new heights of efficiency.¹

However, before AI helps accelerate clinical trials or predict diagnoses, clinicians must familiarize themselves with the tools. “Use it to plan a vacation or buy an anniversary gift. If you are ready for [performing tasks with AI tools like] studying assistance or learning complex topics, that’s even better. When you increase your familiarity with the tools, you start to realize the promise, the power, and the [possibility],” Flora said.

OpenEvidence, an AI platform partnered with the New England Journal of Medicine, JAMA, and NCCN, among others, and the large language model (LLM) Claude were specific tools that Flora highlighted as helpful for completing tasks ranging from drafting complex multidisciplinary tumor board letters to building agents for email inbox management. While first steps help, Flora, a self-proclaimed “AI optimist,” and other proponents of AI-oncology integration have their sights set on larger-scale improvements.

As the oncology field hurtles forward, disparities in key metrics are beginning to widen. Findings published in Cancer Epidemiology, Biomarkers & Prevention outline projections for rates of national cancer incidence, survivors, and care spending from 2020 to 2030.² Specifically, the study found that within the next decade, cancer incidence rates in people aged 65 years or older are projected to rise from 58% to 64%, cancer survivorship is projected to rise by 3.8 million patients, and cancer care spending is projected to rise by $45 billion. Moreover, Flora underscored that in the year 2040, there will be 26 million cancer survivors in the US; as of 2025, there are only 2200 medical oncologists.¹ The 2040 projected cancer survivor population would leave approximately 21,000 patients per oncologist, an increase from the 1200 patients today.

Rapidly scaling trajectories in diagnoses, survivorship, and spending in cancer care positively correlate with the field’s need to make treatments more accessible and efficient. According to Flora and other AI proponents, accessibility and efficiency are both needs suited for AI support. In a letter published in JAMA on regulatory perspectives on AI integration, Haider Warraich, MD, then affiliated with the FDA, and colleagues wrote that, “AI has the potential to reduce the cost of drug development by up to 70% and cut the time required [to develop drugs] by half.”³

Furthermore, Flora peeled back the layers of clinical trials that AI integration could specifically accelerate and expand from start to finish.¹ Regarding enrollment, in 2025, only 3% to 5% of patients enrolled in clinical trials; with AI’s ability to process large data sets and recognize correlations, more patients could be matched to clinical trials in a shorter amount of time, solving issues of access and elevating the reach of enrollment, Flora asserted. While in development, AI can help predict the toxicities of drugs days before crises, Flora added in the presentation. Finally, Flora pointed out AI’s capability for supporting decision-making and assisting with initial assessment, risk stratification, and treatment paths in the clinic.

Ultimately, the additional aid that AI offers across the various facets of cancer care and drug development contributes to an overarching shift to proactivity in the field, according to Flora. Predicting diagnoses and toxicities with the help of AI tools helps lighten the increasing burden of medical oncologists and lower rising costs in areas such as drug development by over $200 million.

How can clinicians maintain a human connection with patients with widespread AI integration?

Proposed AI-integration efforts will increase human touch and connection in cancer care, simultaneously with accessibility and efficiency, according to Flora. In the interview, Flora remarked how AI integration is conducive to a more connected oncology field, both for patients and physicians.

“There is a moment [after being diagnosed with cancer] when patients are never the same. [AI tools] help add order to chaos. As patients become more familiar [with AI tools,] they are more highly educated [about their cancer] than they used to be. It’s no longer [the case] that physicians are the gatekeepers of information. Everyone has access to [AI tools]. I’ve enjoyed the higher levels of discourse with patients who have leaned into [using AI tools for cancer education],” Flora said in the interview.

“Let’s use these [AI tools] intelligently to replace the mundane with the humane so oncologists can once again face their patients and look them in the eye instead of looking at a computer screen and typing. [These tools should] be used in an intelligent fashion to extend physicians’ [reach and enable] them to have more time in the room [with patients] doing what they are uniquely talented to do that a LLM could never replace, like listening, giving a hug, and providing real compassion and empathy for our patients.”

Flora’s presentation echoed the thoughts he shared in the interview, emphasizing how empathy is the most valuable currency physicians can offer to patients, in addition to how expert knowledge is still paramount for oncology, as AI LLMs lack understanding of nuance.

“I would encourage other [oncologists] to recognize that this new technological movement in medicine is going to happen to us, through us, or with us. I would encourage oncologists to dig in and start to study so they can help craft the [AI] tools that best help us and our patients,” Flora added.

References

1. Flora D. Rebooting cancer care. Presented at: 30th Annual International Congress on Hematologic Malignancies; February 25-27, 2026; Miami Beach, FL.
2. Mariotto AB, Enewold L, Zhao J, Zeruto CA, Yabroff KR. Medical care costs associated with cancer survivorship in the United States. Cancer Epidemiol Biomarkers Prev. 2020;29(7):1304-1312. doi:10.1158/1055-9965.EPI-19-1534
3. Warraich HJ, Tazbaz T, Califf RM. FDA perspective on the regulation of artificial intelligence in health care and biomedicine. JAMA. 2025;333(3):241-247. doi:10.1001/jama.2024.21451