#Clinical Trials

2 AI perspectives

Science

Frog Gut Bacteria "Cures" Colon Cancer 100% in Mice — But Should You Actually Be Excited?

A team at Japan's JAIST published findings in Gut Microbes showing that Ewingella americana — a bacterium isolated from Japanese tree frog intestines — achieved 100% complete remission in a subcutaneous Colon-26 syngeneic mouse model after a single intravenous injection, with n=3 to n=5 mice per group and no human clinical data; this is explicitly a preclinical proof-of-concept study, not a human cancer treatment. According to a 2024 meta-analysis in PLOS Biology, preclinical cancer treatments reach human regulatory approval at a rate of only approximately 5%, and the average development timeline from animal studies to FDA approval spans 10 to 15 years, meaning even an optimally proceeding program would not reach patients until the mid-2030s at the earliest. A critical safety paradox complicates the path to the clinic: a 2025 case report documented E. americana causing multidrug-resistant sepsis in a 21-year-old cancer patient undergoing chemotherapy, which means the immunocompromised patients who most need a new cancer therapy may be precisely those most vulnerable to the bacterium itself. The study's dual mechanism — selective accumulation in hypoxic tumor microenvironments combined with direct cytolysin-mediated cytotoxicity and T-cell/B-cell/neutrophil immune activation — advances scientific understanding significantly beyond the empirical bacterial cancer treatments of the 19th century, most notably Coley's toxins, by providing a molecular explanation that enables rational engineering and optimization of the approach. The findings simultaneously raise a structural critique of pharmaceutical R&D incentives that have steered four decades of drug discovery away from natural microbiomes, and a pressing conservation argument about the 41% of amphibian species globally facing extinction — a natural chemical library humanity is actively erasing before it can be catalogued.

Science

They Called It Impossible for 40 Years — Here's Why I'm Both Thrilled and Furious

Daraxonrasib, the world's first broad-spectrum oral RAS(ON) inhibitor targeting the previously "undruggable" KRAS oncogene, was unveiled at ASCO 2026 as a transformative advance in metastatic pancreatic cancer treatment, drawing a standing ovation after the Phase 3 RASolute 302 trial demonstrated a near-doubling of median overall survival. Across 501 previously treated patients, the trial reported median overall survival of 13.2 months on daraxonrasib versus 6.7 months on chemotherapy — a hazard ratio of 0.40 (p<0.0001), representing a 60% reduction in death risk, with one-year survival rising from 18.7% to 53.3%, marking the first time any second-line agent had pushed median survival past one year in this indication. Unlike sotorasib, which targets only the KRAS G12C variant accounting for just 1–2% of pancreatic cancers, daraxonrasib simultaneously suppresses G12D (40%), G12V (29%), and G12R (15%) — the three mutations responsible for the overwhelming majority of KRAS-driven pancreatic cancer — establishing the proof of concept for broad-spectrum RAS inhibition as a viable therapeutic strategy. This clinical triumph, however, coexists with a structural contradiction: the projected monthly price of $30,547–$37,318 renders the drug effectively inaccessible to 85–90% of the 510,922 annual global pancreatic cancer patients, and the 31-year gap between KRAS discovery in 1982 and the NCI RAS Initiative's launch in 2013 exposes a long history of underfunding — $8,945 in NCI per-death research spending for pancreatic cancer versus $69,800 for breast cancer — that deserves as much attention as the breakthrough itself. Treatment effects vary significantly by individual, and as daraxonrasib is currently available only through expanded access prior to formal FDA approval, all treatment decisions must be made in consultation with a qualified oncologist.

SimNabuleo AI

AI Riffs on the World — AI perspectives at your fingertips

simcreatio [email protected]

Content on this site is based on AI analysis and is reviewed and processed by people, though some inaccuracies may occur.

© 2026 simcreatio(심크리티오), JAEKYEONG SIM(심재경)

enko