By Jim Foote

Despite decades of progress in oncology, ranging from molecular diagnostics to targeted therapies, cancer remains one of medicine’s most complex and costly challenges. While genomic sequencing and AI-assisted analytics have improved disease classification, biomarker identification, and the discovery of drugs that may help, most patients are still treated using standardized protocols driven by population-level data rather than individual tumor behavior. As cancer incidence is projected to rise sharply in the coming years, this trial-and-error approach is increasingly unsustainable for providers and healthcare systems alike and is quickly becoming unacceptable for patients.

A new class of technology, AI-driven Functional Precision Medicine (FPM), is emerging to address this gap. By combining patient-derived tumor biology with advancements in proprietary cell enrichment processes, automation, robotics, and artificial intelligence, FPM platforms enable oncologists to rapidly test how an individual patient’s cancer responds to hundreds of FDA-approved drugs and combinations, delivering ranked treatment options within days.
Why precision oncology has fallen short
Cancer’s biological complexity means that patients with the same diagnosis, and even similar genomic profiles, often respond very differently to the same therapy. While genomic profiling has become a cornerstone of precision oncology, it primarily offers probabilistic guidance, identifying mutations that may be targetable and identifying drugs that may work on that mutation. But genomics alone lacks one final step: confirming the selected drug or treatments will actually work for a given patient.

As a result, oncologists frequently rely on standardized treatment pathways and sequential regimens, potentially exposing patients to ineffective therapies, unnecessary toxicity, and costly delays. This challenge is especially true in relapsed or refractory cancers, where time-sensitive decisions can determine outcomes. In many cases, the cost of trial-and-error is not abstract. Each failed line of therapy can mean additional hospital visits, avoidable adverse events, and loss of clinical momentum. For patients, it often means spending limited time and energy on treatments that were never likely to work.

The problem is not a lack of therapeutic options. Hundreds of oncology drugs are approved today. The challenge is determining, quickly and accurately, which therapy will work for each individual patient.

Artificial Intelligence Homepage