The use of artificial intelligence (AI) is rapidly expanding in oncologic clinical practice, with applications including expediting administrative tasks, risk-stratification testing, research, and assistance with interpreting pathology and imaging results.
In order to gain a better understanding of how this technology is currently being applied to practice and where it could be heading in the near future, OncLive® spoke with:
Alicia Morgans, MD, MPH, a genitourinary medical oncologist and the Medical Director of the Survivorship Program at Dana-Farber Cancer Institute, as well as an associate professor of medicine at Harvard Medical School, both in Boston, Massachusetts.
Balazs Halmos, MD, a professor in the Departments of Oncology (Medical Oncology) and Medicine (Oncology and Hematology), and the associate director of Clinical Science at the Montefiore Einstein Comprehensive Cancer Center in New York, New York.
Nitin K. Yerram, MD, the codirector of Urologic Oncology and the director of Urologic Research at Hackensack University Medical Center in New Jersey.
Samina Hirani, MBBS, a hematologist/oncologist at the Mayo Clinic in Eau Claire, Wisconsin.
Alexander Itskovich, MD, the medical director of Oncology Services of the Statesir Cancer Center at Atlantic Health CentraState Medical Center in Freehold, New Jersey.
Ruben Mesa, MD, FACP, the senior vice president of Atrium Health and the president and executive director of Atrium Health Wake Forest Baptist Comprehensive Cancer Center, as well as the vice dean for Cancer Programs and the Charles L. Spurr, MD, Professor of Internal Medicine at Wake Forest University School of Medicine, in Winston-Salem, North Carolina.
“These tools don’t replace the physician’s judgment—they augment it,” Itskovich said. “AI can rapidly sift through massive amounts of data, but the interpretation, clinical decision-making, and patient communication remain firmly in human hands. Together, these technologies are not just making my work more efficient—they’re helping to ensure that fewer clinically significant findings slip through the cracks.”
Significant Time Savings Seen With Administrative Tasks and Patient Communication
Due to its ability to quickly synthesize and organize large datasets, AI is already being used to streamline administrative tasks such as dictation, note taking, and patient monitoring and communication. Examples of these tools include Ambient Voice AI and DAX Copilot.
“Ambient Voice AI allows conversations between patients and clinicians to be securely captured and transformed into medical notes in real time,” Itskovich said. “This technology dramatically reduces the administrative burden of manual notetaking, allowing me to listen more attentively and focus on the patient, rather than writing the note. It’s not just a timesaver; it improves patient engagement and accuracy of documentation.”
“Our institution has been piloting the use of DAX Copilot with good success,” Mesa said. “It listens to my discussion with the patient and generates an editable draft of notes. It learns from the style of my prior notes and gets better and better [with additional use].”
AI Makes a Difference in Risk Stratification and Disease Detection
Multiple investigators in the field of prostate cancer indicated that they are presently using the ArteraAI Prostate Test in their clinical practice. “ArteraAI is a multimodal AI model for [patients with] intermediate-risk localized prostate cancer to help predict whether a patient is likely to benefit from the addition of androgen deprivation therapy [ADT] to their radiation treatment with curative intent,” Morgans explained.
In March 2024, the National Comprehensive Cancer Network included ArteraAI in its updated Clinical Practice Guidelines in Oncology for Prostate Cancer, making it the first AI-enabled predictive and prognostic test to be included in the prostate cancer guidelines.1 The tool was classified with a Category 2A recommendation and was supported by level 1B evidence from multiple phase 3 clinical trials.
“We are currently using the ArteraAI Prostate Test to improve the risk-stratification of patients who have been recently diagnosed with prostate cancer,” Yerram added. “This tool helps to better individualize patient care and provides patients with a clear path forward regarding treatment, whether that is through surveillance, surgery, or radiation.”
During the 2025 ASCO Annual Meeting, investigators presented findings from an analysis of the ArteraAI Prostate Test v1.2 for identifying patients who could benefit the most from the addition of abiraterone acetate (Zytiga) with or without prednisolone and/or enzalutamide (Xtandi) to standard-of-care ADT the analysis revealed that patients who were identified by ArteraAI as being in the upper risk quartile of risk derived the greatest benefit with the addition of abiraterone in terms of metastasis-free survival, prostate cancer-specific mortality, and distant metastasis.
AI tools are also starting to be adopted for the analysis of imaging and pathology reports. “Machine learning is transforming how we detect and manage incidentalomas—unexpected findings on imaging performed for unrelated reasons. These findings can range from benign cysts to potentially malignant tumors. In the past, incidentalomas in radiology reports could be overlooked due to the sheer volume of imaging that is performed. Now, AI algorithms can scan imaging reports, and any abnormal findings are moved into a specialized workflow that serves as a safety net for our patients,” Itskovich said.
Enhancing and Expediting the Research Processes
Mesa explained that he uses DAX Copilot to create power point slides based on his published research, perform journal article review, and are working to incorporate AI into the clinical trial matching process. Platforms such as OpenEvidence can be used to create tables and figures, ask clinical questions, perform guideline searches, create exam questions, and conduct literature searches.
“I use OpenEvidence commonly, that’s my number 1 go-to, but I [also use] Grok AI for research purposes,” Hirani commented.
Excitement Is Mounting for the Future of AI in the Clinic
Looking ahead, investigators are highly optimistic about the future of AI in the clinical and the potential effect it could have on improving treatment for patients with cancer. As AI technology continues to improve and tools are refined, they will be more integrated into clinical practice, saving investigators time and benefiting patients through enhanced disease detection.
“I believe the future of AI in the clinic is incredibly bright and it will [have an effect] on everything,” Halmos said. “We’re looking at new biomarkers for antibody-drug conjugates aided by AI, as well as its use for lung cancer screening. It could also be used to break down medical jargon for patients and break down language barriers. AI will certainly add an additional tool with utility to our clinics.”
References
- ArteraAI announced as the first-and-only predictive test for therapy personalization in the 2024 NCCN Guidelines for prostate cancer. News release. ArteraAI. March 4, 2024. Accessed August 15, 2025. https://www.businesswire.com/news/home/20240304893588/en/ArteraAI-Announced-as-the-First-and-Only-Predictive-Test-for-Therapy-Personalization-in-the-2024-NCCN-Guidelines-for-Prostate-Cancer
- Parker CTA, Liu VYT, Mendes L, et al. Multimodal artificial intelligence (MMAI) model to identify benefit from 2nd-generation androgen receptor pathway inhibitors (ARPI) in high-risk non-metastatic prostate cancer patients from STAMPEDE. J Clin Oncol. 2025;43(suppl 16):5001. doi:10.1200/JCO.2025.43.16_suppl.5001