Artificial intelligence is beginning to reshape medicine by helping researchers personalize cancer treatment, search for overlooked uses of existing drugs and detect biological signals people might miss. Yet the evidence ranges from clinical success to early laboratory work, making careful scrutiny essential.

One of the most striking advances comes from Moderna and Merck, which on Aug. 19 announced positive topline results from a Phase 3 melanoma trial. The study enrolled 1,137 people whose high risk melanoma had been completely removed by surgeons.

The trial combined intismeran autogene, also called V940 or mRNA-4157, with Keytruda. It met its primary endpoint for survival without recurrence and a key secondary endpoint measuring survival without distant metastasis.

The companies described the findings as the first positive Phase 3 readout for an individualized neoantigen therapy and the first positive Phase 3 result for an mRNA cancer therapy. Researchers analyze mutations from a patient’s tumor and use an algorithm to select targets that could help the immune system recognize the cancer.

Here's What They're Not Telling You About Your Retirement

The resulting individualized treatment can encode as many as 34 neoantigens. Moderna has said integrated AI algorithms are used during the V940 development process, supporting an approach intended to train the immune system to recognize characteristics unique to each patient’s cancer.

Earlier findings provide additional context. In a smaller Phase 2b study with longer follow up, intismeran plus Keytruda reduced the risk of recurrence or death by 49% and the risk of distant metastasis or death by 59% compared with Keytruda alone.

However, only topline Phase 3 results have been announced. Merck and Moderna plan to present complete findings at an international medical meeting and share them with regulators, while the study continues tracking overall survival. Intismeran remains investigational and has not been approved by the FDA for melanoma.

AI is also driving the search for new uses for existing medicines. Dr. David Fajgenbaum cofounded Every Cure, a nonprofit that scans biomedical knowledge for connections between drugs and diseases they might potentially treat.

This Could Be the Most Important Video Gun Owners Watch All Year

Every Cure says its system can produce tens of millions of predictions in less than a day. The federal Advanced Research Projects Agency for Health supports this work through MATRIX, which uses machine learning and AI to predict whether drugs approved by the FDA could treat other diseases.

Researchers must still validate promising candidates through laboratory or clinical work. AI does not establish that a medicine will work for another illness, but it can direct researchers toward possibilities worthy of further investigation.

At Columbia University Fertility Center, the Sperm Tracking and Recovery system, known as STAR, combines rapid imaging, an AI detection model and microfluidics. It was designed for patients with azoospermia or cryptozoospermia, conditions in which sperm appear absent or exist in extremely small numbers.

STAR captures and processes about 1.1 million images every hour while searching for possible sperm cells. In one validation sample, embryologists searched for two days without success, while STAR found 44 sperm in about an hour.

Columbia reported STAR’s first pregnancy in March 2025 after the system recovered sperm missed during conventional examination of the same sample. The pregnancy resulted in a healthy delivery, although Columbia says sperm are found in about 28% of patients previously diagnosed with azoospermia.

Researchers elsewhere are using AI to predict disease and test treatments. The University of Hong Kong’s CardiOmicScore examined 2,920 circulating proteins, 168 metabolites and genomic information to estimate the future risk of six cardiovascular diseases, sometimes flagging elevated risk as much as 15 years before symptoms appeared.

At UCLA, scientists combine 3D bioprinting, advanced imaging and AI to study laboratory grown replicas of patient tumors called organoids. Other researchers are investigating speech changes associated with ALS and Parkinson’s disease, but these systems remain experimental and require more clinical evidence.

Patients should ask what role AI plays, whether qualified professionals review its findings and what research supports the technology. They should also ask about regulatory authorization, data storage, access to sensitive health information and whether their information could be used to train or improve an AI system.