New AI and Cell Therapy Advances Show Promise for Overcoming Treatment Resistance

New AI and Cell Therapy Advances Show Promise for Overcoming Treatment Resistance
Why this is good news

    New research is tackling cancer's ability to resist treatment, using artificial intelligence and engineered immune cells.

  • AI Targets "Undruggable" Protein.The protein GRB2 was previously considered impossible to block with drugs. AI helped design a way to target it, which could stop cancers from hiding and evading therapy.
  • Predicts Relapse in Leukemia.Doctors previously had limited ways to know which leukemia patients would relapse after a stem cell transplant. A new test can identify high-risk patients much earlier, allowing for proactive treatment.
  • Engineers Safer T Cells.Engineered immune cells (CAR-T) can cause dangerous side effects. A new method creates "switchable" CAR-T cells that doctors can control, potentially making this powerful therapy safer to use.
  • Overcomes Common Resistance Pathway.Many cancers become resistant by using a specific survival pathway called RAS. A new combination therapy blocks this pathway, offering a strategy to treat cancers that have stopped responding.

Researchers have unveiled a suite of innovative strategies aimed at overcoming some of the most persistent challenges in cancer care, from predicting risk to engineering immune cells and breaking through treatment resistance. The findings, presented at a major medical meeting, highlight progress on multiple fronts.

In the realm of artificial intelligence, one study demonstrated how AI can identify new drug targets. Scientists used AI-driven structural approaches to tackle GRB2, a protein long considered "undruggable" that helps cancer cells hide damaged DNA and evade immune attack. They developed a first-in-class small molecule that locks GRB2 in its inactive state, which in lab studies helped expose cancer cells and enhanced sensitivity to PARP inhibitors. In another application, a digital twin model called OncoTwin was designed to predict individual patient responses to therapies for ALK-positive lung cancer, potentially guiding more personalized treatment plans.

Significant advances were also reported in cell therapy and targeted treatments. Engineers developed a novel platform called NK-TCR, which combines natural killer cells with T cell receptors to create immune cells capable of targeting specific antigens inside tumor cells. Early models targeting NY-ESO-1 and PRAME antigens showed strong antitumor activity in multiple myeloma. Separately, a first-in-class strategy uses a peptide-linked NRP1 antibody to redirect the body's existing antiviral T cells to attack solid tumors expressing the NRP1 protein.

Further research provided new tools for prediction and detection. A study of 2,121 patients found that tracking CA19-9 trajectories in individuals with new-onset diabetes could help identify those at highest risk for underlying pancreatic cancer. For thyroid cancer, a new gene expression signature called PRECISE was developed to identify tumors likely to have poorer outcomes, using data from a cohort followed for a median of 14 years. Additional studies clarified mechanisms of resistance in cancers like gliomas and cervical cancer, pointing to new potential therapeutic targets.

These diverse studies represent a move toward more precise, effective, and adaptable cancer interventions. While further clinical validation is needed, the work provides a hopeful outlook for developing next-generation treatments that are smarter, more targeted, and capable of outmaneuvering a cancer's defenses.

This article is for informational purposes only and does not constitute medical advice. The information presented is based on published research and official announcements. Always consult a qualified healthcare professional before making any medical decisions.

← Back to all stories
Medical Disclaimer: Content on Curative News is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.