Artificial intelligence is beginning to reshape how doctors approach some of the toughest medical challenges, from aggressive cancers to infertility. While many of these tools are still in trials, early successes highlight the potential for AI to help families and patients across the nation.
Personalized Cancer Vaccines Show Promise
Moderna and Merck announced encouraging topline results from a Phase 3 trial of an individualized melanoma vaccine, known as intismeran autogene (V940 or mRNA‑4157), given together with Keytruda. The study enrolled 1,137 patients whose high‑risk melanomas had been surgically removed. The combination met its primary goal of improving recurrence‑free survival and also met a key secondary endpoint of extending distant‑metastasis‑free survival.
The vaccine works by sequencing a patient’s tumor, identifying unique mutations, and using an algorithm to select up to 34 neoantigens that can train the immune system to recognize the cancer. Moderna reports that AI algorithms helped streamline the selection of these targets during development.
Although the data are still preliminary – only topline results have been released – the trial marks the first positive Phase 3 readout for an individualized neoantigen therapy and the first for an mRNA‑based cancer treatment. Earlier Phase 2b data showed a 49% reduction in the risk of recurrence or death and a 59% reduction in distant‑metastasis risk when the vaccine was added to Keytruda.
The FDA has not yet approved the therapy, and full results will be presented at an upcoming medical conference. Nonetheless, the trial represents a significant step toward more precise, patient‑specific cancer care.
AI‑Powered Drug Repurposing Aims to Fill Treatment Gaps
Beyond new drugs, researchers are harnessing AI to find new uses for medicines already on the market. Dr. David Fajgenbaum’s nonprofit Every Cure uses AI to scan biomedical literature and generate millions of drug‑disease hypotheses each day. The federal Advanced Research Projects Agency for Health (ARPA‑H) supports this effort through the MATRIX project, which applies machine learning to predict which FDA‑approved drugs might treat other illnesses.
Every Cure’s approach is exemplified by the story of Kaila Mabus, who was diagnosed with multicentric Castleman disease at age 13. In 2020, doctors tried ruxolitinib—an existing medication for certain blood disorders—and she entered remission within months. While AI did not identify that treatment, the case illustrates how repurposing can dramatically improve outcomes when existing drugs are matched to the right condition.
AI Improves Sperm Detection for Infertility Treatment
At Columbia University’s Fertility Center, researchers have deployed the Sperm Tracking and Recovery (STAR) system, which combines high‑speed imaging, AI detection and microfluidics to locate rare sperm cells in men diagnosed with azoospermia or cryptozoospermia. The system processes roughly 1.1 million images per hour, flagging potential sperm for isolation.
In a validation test, STAR identified 44 sperm in about an hour after two days of manual searching yielded none. The technology has already led to a successful pregnancy in March 2025 for a couple who had struggled for nearly two decades to conceive.
Columbia reports that STAR finds sperm in about 28% of patients previously labeled azoospermic, with roughly 20% of recovered sperm leading to fertilized eggs and about 18% of those developing into quality embryos. While these rates are lower than standard IVF or ICSI, they provide a valuable option for couples facing severe fertility challenges.
AI for Cardiovascular Risk Prediction
Researchers at the University of Hong Kong have created CardiOmicScore, an AI‑driven tool that analyzes blood proteins, metabolites and genetic data to estimate future risk for six major cardiovascular diseases, including coronary artery disease and stroke. Using data from the UK Biobank, the model evaluates 2,920 circulating proteins and 168 metabolites, offering clinicians a deeper view of a patient’s long‑term heart health.
These developments underscore a broader trend: AI can sift through massive datasets far faster than any human, highlighting patterns that may lead to earlier diagnosis, more targeted therapies and, ultimately, better outcomes for patients and families.
While optimism is warranted, experts caution that AI‑driven tools must undergo rigorous clinical testing before becoming standard care. The promise of AI lies in its ability to augment, not replace, the expertise of physicians and the trust of patients.
Original reporting: Fox News (HLL/CB) — read the source article.
