Five Ways AI Is Transforming Cancer Care—and Companies That Are Making It Happen

Artificial intelligence (AI) is on pace to become the most rapidly adopted technology in health care,¹ whether clinicians are ready for it or not. From ambient scribes transcribing doctor-patient conversations, to AI-powered digital twins replacing placebo arms in clinical trials,² AI is being inserted throughout oncology.

Clinicians remain divided on AI’s impact on their work. Approximately one-third of oncologists say AI has improved how they diagnose or stage cancer, another one-third disagree, and the rest are undecided, according to results of a survey by Medscape.³ Patients share in the uncertainty. A review of 19 studies involving more than 2100 individuals with cancer found that most support AI only when it operates under physician supervision.⁴ Their top concerns were depersonalization, privacy, and algorithmic bias.

Despite the doubts, health care investors are going all-in when it comes to AI. Although the broader biotech sector is in the midst of a 20-year low for early-stage financing,⁵ the number of companies using AI for drug discovery and digital health is surging, with half of all investment dollars in diagnostic and tool companies going to these organizations.⁶

Studies have demonstrated AI’s improving ability to summarize notes, scan digitized pathology slides, and even draft prior-authorization letters and predict treatment responses, but keeping up with the latest developments can be challenging. For many oncologists, the problem is no longer access but excess: There are too many tools and not enough time to learn which ones are worth using.

To break down how AI is being used across the field, Evidence-Based Oncology examined where AI is already influencing cancer care across several categories.

Keeping Up With New Research

During her presidential address at the 2025 meeting of the American Society of Clinical Oncology (ASCO), Robin Zon, MD, FACP, FASCO, noted that new medical articles are being published at an average of every 26 seconds.⁷ Keeping up to date with the latest research, clinical trials, and treatment guidelines can be difficult. A wave of new AI tools emerged in 2025 to help. Examples include the following:

• ASCO’s Guidelines Assistant. ASCO partnered with Google to launch an AI-powered chat tool in May 2025 to give oncologists faster, interactive access to the oncology organization’s clinical guidelines.⁸ The system lets users ask questions in plain language and provides cited, guideline-based answers. The tool is available exclusively to ASCO members through the society’s website and mobile app.
• OpenEvidence. This AI-powered research platform is designed to help physicians find and interpret medical studies. The company has signed content agreements with the New England Journal of Medicine and JAMA Network, giving its system access to full-text content from their medical journals. Already used by clinicians at more than 10,000 hospitals, OpenEvidence reports that approximately 40% of US physicians log in to the platform daily, generating more than 8 million clinical searches each month.⁹

Imaging and Pathology

Radiology and pathology are key areas where AI seems to constantly improve. In mammography, results of a randomized trial showed that AI-supported readers detected 17% more cancers than the standard double reading,¹⁰ whereas a multimodal AI system cut recall rates by 32% and reduced radiologist workload by 44% without losing sensitivity.¹¹In colonoscopy, AI-assisted detection achieved a 20% higher adenoma detection rate and reduced miss rate by 55% vs conventional methods.¹² These tools are increasingly receiving regulatory approvals and integration with existing electronic health record (EHR) systems. Companies to watch include the following:

• iCAD. The company’s breast-imaging AI received FDA clearance for an upgrade shown to detect 22% more cancers in dense breast tissue. In April 2025, imaging chain RadNet announced plans to acquire iCAD in a $103 million deal, aiming to make AI-based mammography a standard feature in its centers nationwide.¹³
• PathAI. InJune 2025, the FDA expanded the approved use of the company’s digital-pathology system, allowing its use for primary diagnosis.¹⁴ The change means specialists can make official judgments from digital images, often available immediately, instead of waiting for physical glass slides to be created and shipped.

Decision Support and Precision Oncology

AI systems are being integrated into clinical reasoning, acting as copilots for treatment selection, dosing, adverse-event prediction and enrollment matching. Real-world examples include researchers at Weill Cornell Medicine, who used an AI model to personalize chemotherapy for patients with bladder cancer by analyzing tumor images and gene-expression data, outperforming traditional single-data approaches.¹⁵A Nature Cancer study introduced an AI agent that integrated multimodal data, including imaging, pathology, and genomic results, to match oncologists’ treatment choices in simulated cases.¹⁶ Companies to watch include the following:

• Lunit. In March 2025, the National Cancer Institute (NCI) announced it would use Lunit’s AI biomarker-analysis tools across NCI clinical trials to identify tumor-immune features and support biomarker discovery in immunotherapy research.¹⁷
• Tempus AI. Tempus uses AI to link patient test results and clinical data with treatment decisions. Results of a study published in JCO Precision Oncology showed that the company’s data model could help oncologists choose more effective first-line chemotherapy for patients with advanced pancreatic cancer.¹⁸ Tempus was selected by the federal government’s Advanced Research Projects Agency for Health to support a new national program exploring how AI and biomarker testing could guide treatment choices for metastatic lung, breast, and colorectal cancers.¹⁹

Matching Patients to Clinical Trials

AI is starting to make clinical trial recruitment faster and more precise. New tools can scan pathology reports, laboratory results, and clinician notes to flag patients who meet study criteria, work that once required days of manual review. Some systems can even generate referral letters or outreach messages automatically. These tools are not only helping patients, they also are saving clinician time, said Zach Koontz, MD, president, managing partner, and director of clinical research at Pacific Cancer Care in Monterey, California. His organization is working with N-Power Medicine to query their patient population for potential enrollment into clinical trials.²⁰

”My daily workflow has improved and saved me hundreds of hours of time on documentation/charting,” Koontz wrote in an emailed statement. “We are currently enrolling into a clinical trial faster and with more ease than ever before, which allows us to get our patients potential access to studies and therapeutics for which they otherwise would have to travel great distances.”

Companies to watch include the following:

• OncoLens. Used by more than 225 cancer centers in the US and internationally, OncoLens uses AI to analyze structured and unstructured patient data, helping identify people who may be eligible for new diagnostics, therapies, or clinical trials. In one case study at a community hospital in Kentucky,²¹ integrating the platform with the EHR led to a 350% increase in genetic testing orders for patients with breast cancer simply by flagging those who met National Comprehensive Cancer Network (NCCN) testing criteria but had not been referred.
• Deep 6 AI – Deep 6’s platform searches structured and unstructured patient data to find candidates for oncology trials based on diagnosis, stage, or biomarkers. Acquired by Tempus earlier in 2025, the tool includes more than 750 provider sites and 30 million patient records, helping trial sites identify 25% more eligible patients and recruit up to 3 times faster than traditional methods.²²

Patient Tools

Physicians are used to patients arriving at their offices with information from “Dr Google,” but now they’re increasingly asking “Dr ChatGPT” to weigh in. Nearly 1 in 3 individuals in the US now turn to AI tools for health information, according to results of a nationwide survey of 2000 adults by Censuswide.²³ Usage rises to nearly half of those aged 16 to 34 years; they cite speed, convenience, and limited access to care as key motivations. Still, clinicians should pay careful attention to the information their patients present: Results of a recent study found that AI chatbots produced unsafe or clinically inappropriate answers in nearly one-third of patient-posed questions.²⁴ Some tools to watch include the following:

• Outcomes4Me is an AI-enabled app that helps patients with cancer understand their treatment options, manage symptoms, and access clinical trials using data drawn from real-time guidelines and genomic insights. In collaboration with Labcorp, Outcomes4Me showed that digital education could significantly improve patient understanding of HER2-low testing in metastatic breast cancer, prompting more patients to discuss biomarker testing with their doctors.²⁵ The platform now supports more than 280,000 users and integrates NCCN guidelines for patient-friendly summaries.²⁶
• CureWise is a patient-facing platform that helps individuals organize records and query multiagent virtual tumor board models to generate focused questions and discussion points for their clinicians (eg, tests to consider, trial options). The company was started after the founder’s rare blood cancer was missed by his physicians, so he decided to build this tool to help others.²⁷

These examples demonstrate how AI is rapidly becoming an integral part of nearly every step in cancer care. With AI drawing $4 billion, or two-thirds of all digital health investment, in the first half of 2025,²⁸ and regulators formally recognizing the role of AI in drug development, medical devices, and clinical trials,²⁹ it may soon be unusual not to see AI incorporated in some aspect of oncology practice.

References

1. Flinn R. Doctors embrace AI scribes as solution for burnout. April 17, 2025. November 7, 2025. https://wewillcure.com/insights/ai-and-machine-learning/care-delivery/doctors-embrace-ai-scribes-as-solution-for-burnout
2. Akbarialiabad H, Pasdar A, Murrell DF, et al. Enhancing randomized clinical trials with digital twins. NPJ Syst Biol Appl. 2025;11(1)10): doi:10.1038/s41540-025-00592-0
3. Schneider ME. Oncologists divided on AI’s value in diagnosing cancers. Medscape. September 19, 2025. Accessed October 13, 2025. https://www.medscape.com/viewarticle/oncologists-divided-ais-value-diagnosing-cancers-2025a1000oxq
4. Hilbers D, Nekain N, Bates A, Nunez JJ. Patient attitudes toward artificial intelligence in cancer care: scoping review. JMIR Cancer. 2025;11:e74010. doi:10.2196/74010
5. Glasner J. Biotech share of US funding hits lowest point in Crunchbase history. September 29, 2025. Accessed November 7, 2025. https://news.crunchbase.com/venture/biotech-us-funding-share-lowest-2025/
6. AI deal activity remains strong in healthcare amid decline in fundraising; Silicon Valley Bank releases 16th edition of health care investments and exits report. News release. Silicon Valley Bank. July 29, 2025. Accessed November 7, 2025. https://bit.ly/4otT7G8
7. Zon R. President’s address: driving knowledge to action, building a better future. American Society of Clinical Oncology. June 1, 2025. Accessed November 7, 2025. https://connection.asco.org/do/2025-president-s-address-driving-knowledge-action-building-better-future
8. Introducing ASCO Guidelines Assistant. American Society of Clinical Oncology. Accessed November 7, 2025. https://www.asco.org/practice-patients/guidelines/assistant
9. OpenEvidence, the fastest growing application for physicians in history, announces $210 million round at $3.5 billion valuation. News release. OpenEvidence. July 15, 2025. Accessed November 7, 2025. https://www.openevidence.com/announcements/openevidence-the-fastest-growing-application-for-physicians-in-history-announces-dollar210-million-round-at-dollar35-billion-valuation
10. Eisemann N, Bunk S, Mukama T, et al. Nationwide real-world implementation of AI for cancer detection in population-based mammography screening. Nat Med. 2025;31(3):917-924. doi:10.1038/s41591-024-03408-6
11. Park J, Witowski J, Xu Y, et al. A multi-modal AI system for screening mammography: integrating 2D and 3D imaging to improve breast cancer detection in a prospective clinical study. arXiv. Updated April 11, 2025. November 7, 2025. https://arxiv.org/abs/2504.05636
12. Anderer S. Meta-analysis: AI-assisted colonoscopy increases detection of polyps, adenomas. JAMA. 2024;332;(23):1968. doi:10.1001/jama.2024.23621
13. Axente A. RadNet, Inc. to acquire iCAD Inc to accelerate AI powered early detection and diagnosis of breast cancer. April 15, 2025. DeepHealth. Accessed November 7, 2025. https://deephealth.com/press-releases/radnet-inc-to-acquire-icad-inc-to-accelerate-ai-powered-early-detection-and-diagnosis-of-breast-cancer/
14. PathAI receives FDA clearance for AISight Dx platform for primary diagnosis. News release. PathAI. News release. June 30, 2025. Accessed November 7, 2025. https://www.pathai.com/news/pathai-receives-fda-clearance-for-aisight-dx-platform-for-primary-diagnosis
15. Karniadakis N, Lee J, Liu J. AI model integrates pathology and gene-expression data to personalize chemotherapy for bladder cancer. Weill Cornell Medicine Newsroom. April 2025. Accessed November 7, 2025. https://phs.weill.cornell.edu/news/ai-meets-oncology-new-model-personalizes-bladder-cancer-treatment
16. Ferber D, El Nahhas OSM, Wölflein G, et al. Development and validation of an autonomous artificial intelligence agent for clinical decision-making in oncology. 2025;6:1337–1349. doi: 10.1038/s43018-025-00991-6.
17. NCI clinical trials get boost from public private partnership. News release. National Cancer Institute. April 22, 2025. Accessed November 7, 2025. https://www.cancer.gov/about-nci/organization/cbiit/news-events/news/2025/nci-clinical-trials-get-boost-public-private-partnership
18. Wenric S, Sangli C, Guittar J, et al. Real-world validation of the Purity Independent Subtyping of Tumors classifier for informing therapy selection in pancreatic ductal adenocarcinoma. JCO Prec Oncol. 2025;9:e2500197. doi:10.1200/PO-25-00197
19. Tempus selected by ARPA-H to provide CRO and testing services for the ADAPT program, advancing precision cancer therapy. News release. Tempus. October 9, 2025. Accessed November 7, 2025. https://www.tempus.com/news/tempus-selected-by-arpa-h-to-provide-cro-and-testing-services-for-the-adapt-program-advancing-precision-cancer-therapy/
20. Uniquely accelerating & de-risking clinical drug development. NPowerMedicine. Accessed November 7, 2025. https://www.npowermedicine.com/
21. OncoLens Analytics enables a 350% Increase in genetic testing at community cancer center. OncoLens. March 1, 2025. Accessed November 7, 2025. https://www.oncolens.com/casestudy/oncolens-analytics-enables-a-350-increase-in-genetic-testing-at-community-cancer-center/
22. Tempus announces acquisition of Deep 6 AI. News release. March 11, 2025. Accessed November 7, 2025. https://www.tempus.com/news/pr/tempus-announces-acquisition-of-deep-6-ai
23. The digital diagnosis: Americans increasingly turn to AI for medical guidance. Drip Hydration. 2025. Accessed November 7, 2025. https://driphydration.com/ai-health-survey/
24. Draelos RL, Afreen S, Blasko B, et al. Large language models provide unsafe answers to patient-posed medical questions. NPJ Digit Med. August 4, 2025. Accessed November 8, 2025. https://arxiv.org/abs/2507.18905
25. Outcomes4Me and Labcorp study reveals digital education can increase HER2-low awareness for metastatic breast cancer patients. News release. Outcomes4Me. November 26, 2024. Accessed November 7, 2025. https://www.prnewswire.com/news-releases/outcomes4me-and-labcorp-study-reveals-digital-education-can-increase-her2-low-awareness-for-metastatic-breast-cancer-patients-302316505.html
26. Outcomes4Me secures $21M in funding to accelerate AI-driven global expansion to transform cancer care. News release. Outcomes4Me. May 29, 2025. Accessed November 9, 2025. https://bit.ly/3Jnl3fR
27. Flinn R. How missed cancer diagnosis led founder to build AI platform for patients. Cure. August 18, 2025. Accessed November 7, 2025. https://wewillcure.com/insights/ai-and-machine-learning/founder-stories/how-a-missed-cancer-diagnosis-led-founder-to-build-ai-platform-for
28. Flinn R. AI-powered companies dominate 2025 digital health funding. Cure. July 16, 2025. Accessed November 7, 2025. https://wewillcure.com/insights/ai-and-machine-learning/investment/ai-powered-companies-dominate-2025-digital-health-funding.
29. Artificial intelligence for drug development. FDA. Updated February 20, 2025. Accessed November 8, 2025. https://www.fda.gov/about-fda/center-drug-evaluation-and-research-cder/artificial-intelligence-drug-development