Peter Hosein, M.D., co-leader of Sylvester’s Gastrointestinal Cancer Site Disease Group, called the approval a monumental shift in the treatment of pancreas cancer. The significance lies not only in the survival benefit itself but in what it signals about the future of the disease: pancreatic cancer, long treated as a monolith to be attacked with blunt cytotoxic chemotherapy, is increasingly being understood at the molecular level, allowing drugs designed for specific biological vulnerabilities to be matched with the patients whose tumors carry them. Targeted therapies of this kind have transformed outcomes in lung cancer, breast cancer and leukemia, and daraxonrasib’s success suggests that pancreatic oncology is finally entering that same era of precision medicine.

Even as treatment advances, earlier detection remains the field’s most urgent unmet need, and a new grant from the Pancreatic Cancer Action Network aims to address it. The organization has awarded Sylvester researchers 225,000 dollars over two years to help establish a blood test for pancreatic cancer based on circulating tumor DNA, or ctDNA. For patients with lung cancer, colon cancer and certain other solid tumors, liquid biopsies that detect fragments of tumor DNA shed into the bloodstream have become routine tools, helping oncologists identify molecular weaknesses in a tumor without invasive procedures and adjust treatment as the disease evolves. Pancreatic cancer patients have not enjoyed that advantage; invasive tissue biopsies remain the standard, and the tumors’ biology has made reliable ctDNA detection difficult.

The PanCAN-funded project seeks to change that by establishing ctDNA analysis as a dependable way to measure the molecular makeup of metastatic pancreatic cancer, one of the deadliest malignancies known. If successful, the approach could allow clinicians to track tumor evolution in near real time through simple blood draws, matching patients to targeted agents as new mutations emerge and monitoring treatment response without repeated surgical intervention. The work complements the daraxonrasib story in an important way: targeted therapies are only as useful as the molecular diagnostics that identify eligible patients, and liquid biopsies could dramatically widen the funnel of patients who benefit from drugs like the newly approved pill.

Shifting from treatment to prevention, Sylvester investigators have published sobering new data on alcohol’s role in cancer mortality. A study by the center’s researchers, first presented at the annual meeting of the American Society of Clinical Oncology and now published in The Lancet Regional Health – Americas, found that alcohol-attributable cancer deaths in the United States doubled between 1990 and 2023. Perhaps more striking than the overall trend is its demographic reach. Alcohol-linked cancer deaths are not confined to older adults or to liver cancer, the organs and populations most commonly associated with drinking-related disease in the public imagination.

Among adults aged 20 to 54, the study found that colorectal cancer was the leading alcohol-related cancer death in men, while breast cancer ranked first among women. Those findings carry substantial public health implications, because they indicate that alcohol’s carcinogenic influence operates across a wide swath of tumor types and age groups, and that the burden is falling on populations who may not perceive themselves as being at risk. Alcohol is an established carcinogen, classified as such by international cancer research bodies, and the mechanistic pathways are well described: its metabolite acetaldehyde damages DNA, alcohol consumption impairs nutrient absorption and hormone regulation, and it acts as a solvent that enhances the penetration of other carcinogens into tissues. The doubling of deaths over three decades suggests that awareness and policy responses have not kept pace with the accumulating evidence.

Against that backdrop, a second Sylvester study examined whether communication strategies could close the awareness gap. Writing in Cancer, a journal of the American Cancer Society, researchers led by Taghrid Asfar, M.D., MSPH, professor of public health, reported that online warnings about alcohol and cancer prompted stronger responses when they named specific cancers and included AI-generated images depicting health consequences. Many people remain unaware that alcohol is linked to several cancers, Asfar noted, and the study suggests that specificity matters: generic warnings about harm are less effective than messages that identify concrete diseases and pair that information with vivid visual depictions of what those diseases entail.

The work sits at the intersection of behavioral science and emerging technology, testing whether generative artificial intelligence can be harnessed to produce health communication that is both scalable and persuasive. If AI-generated imagery can meaningfully improve public understanding of alcohol’s cancer risks, health agencies could deploy tailored warning campaigns at low cost across digital platforms, reaching younger audiences precisely where the mortality data show the burden is growing. The approach also raises questions the field will need to address, including how to ensure that synthetic images used in public health messaging remain accurate and non-misleading, but the core finding, that specificity plus visual evidence improves risk comprehension, offers a practical template for campaigns aimed at other preventable risk factors as well.

Artificial intelligence is also reshaping basic discovery at Sylvester, where scientists have used computational methods and advanced laboratory experiments to illuminate a previously hidden corner of the human proteome. The protein universe contains hundreds of millions of members, yet the vast majority remain poorly characterized, a biological darkness that may conceal clues to health and disease. In research published in Nature, the Sylvester team identified a protein that helps cells form physical connections and share resources with one another, revealing an entirely new layer of cellular communication. Daniel Isom, Ph.D., the study’s senior author, said the discovery suggests there is another layer of biology that has been hiding in plain sight.

The finding demonstrates how machine learning can guide experimental biology toward targets that traditional hypothesis-driven research might never have flagged, scanning enormous sequence spaces for proteins with predicted structural features and then validating their functions in the laboratory. Because the newly characterized protein governs how cells connect and exchange resources, it could prove relevant to cancer, where communication between tumor cells and their neighbors shapes growth, survival and drug resistance. Rounding out the center’s recent output, a multicenter study published in Blood Cancer Discovery, conducted with Memorial Sloan Kettering Cancer Center and Moffitt Cancer Center, identified loss of the RHOA gene as a mechanism by which some large B-cell lymphomas evade CAR T-cell therapy, disrupting key immune responses and offering a route toward predicting and overcoming treatment resistance. Meanwhile, Sylvester researcher Melissa Lopez, Ph.D., MS, RDN, nutrition lead in lifestyle medicine at the center’s Survivorship and Supportive Care Institute, received an NCI ENICTO pilot grant to test whether routine clinical and nutritional markers from standard lab results can predict which patients will suffer the heaviest symptom burden during and after chemotherapy, a question that could make the chemotherapy journey more predictable for the many patients for whom it remains anything but uniform. Together, the September and October findings sketch a portrait of a cancer center working simultaneously at the bedside, in the clinic, and deep inside the dark proteome, with each advance feeding the others.

Subject of Research: Recent advances in pancreatic cancer therapy, alcohol-related cancer mortality, ctDNA diagnostics, AI-based protein discovery and CAR T-cell resistance research at Sylvester Comprehensive Cancer Center

Article Title: Sylvester Cancer Tip Sheet for September and October, 2026

Article References: Sylvester Cancer Tip Sheet for September and October, 2026. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: pancreatic cancer, daraxonrasib, FDA approval, targeted therapy, ctDNA, liquid biopsy, alcohol and cancer, artificial intelligence, dark proteome, CAR T-cell therapy, lymphoma, chemotherapy side effects

The story of this breakthrough begins with the RAS family of proteins, the quintessential molecular villains of pancreatic cancer. RAS mutations are present in more than 90 percent of pancreatic ductal adenocarcinoma cases, with the KRAS G12D, G12V, and G12R subtypes being the most frequent. For roughly four decades, these proteins were branded “undruggable.” The problem was structural: RAS proteins present a smooth, near-spherical surface lacking the classical deep pockets that small-molecule drugs typically exploit for binding. The arrival of KRAS G12C inhibitors cracked that impasse, proving that a chemical handle could be found on this notoriously slippery target. Yet the victory was narrow. G12C mutations account for only 1 to 2 percent of pancreatic cancers, limiting the clinical reach of those agents to a vanishingly small fraction of the patient population that desperately needs them.

Daraxonrasib, developed from the investigational compound RMC-6236, takes a fundamentally different pharmacological route. Rather than locking onto a single mutant allele in its inactive, GDP-bound state, the drug employs what its developers call RAS(ON) multi-selective inhibition. It is a non-covalent tri-complex inhibitor: the molecule first binds intracellular cyclophilin A to form a binary complex, and that complex then selectively attaches to the active, GTP-bound conformation of RAS proteins. Because it targets the switched-on state that drives cancer signaling, it can simultaneously cover a broad spectrum of RAS isoforms — KRAS, NRAS, and HRAS — and mutant alleles including G12D, G12V, G12R, G13, and Q61. This breadth transforms the calculus of RAS therapy. A strategy that once addressed a sliver of patients now reaches more than 90 percent of those with pancreatic cancer, marking a fundamental transition from allele-specific inhibition to broad-spectrum suppression of the RAS signaling engine.

The clinical evidence underpinning this shift has accumulated with unusual speed and rigor. In the phase I/II trial reported by Wolpin and colleagues, 168 patients with previously treated advanced RAS-mutant pancreatic cancer were enrolled. Among those with RAS G12 mutations receiving the 300-milligram dose as second-line therapy, the objective response rate reached 35 percent, the disease control rate was a striking 92 percent, and median duration of response, progression-free survival, and overall survival were 8.2, 8.5, and 13.1 months, respectively. Across all RAS-mutant patients carrying G12, G13, or Q61 alterations, the response rate was 29 percent with a median overall survival of 15.6 months. Grade 3 or higher treatment-related adverse events occurred in 30 percent of patients, predominantly rash and gastrointestinal toxicities, both of which proved manageable with standard clinical interventions.

Building on that foundation, O’Reilly and colleagues advanced daraxonrasib into the phase III RASolute 302 trial, an open-label randomized controlled study of 500 patients with previously treated metastatic pancreatic cancer, 91.8 percent of whom harbored RAS G12 mutations. The results were decisive. In the RAS G12 population, daraxonrasib achieved a median overall survival of 13.2 months compared with 6.6 months for chemotherapy, corresponding to a hazard ratio of 0.40 with a P value below 0.001. Median progression-free survival doubled from 3.5 to 7.3 months, again with a hazard ratio of 0.45. The intention-to-treat analysis yielded nearly identical figures — 13.2 versus 6.7 months — underscoring the robustness of the effect. Notably, the response rate in the daraxonrasib arm was 31.6 percent, roughly triple the 11.2 percent seen with chemotherapy. Patient-reported quality of life and time to pain deterioration, metrics that carry enormous weight in a disease defined by debilitation, were also significantly improved.

The safety profile added further weight to the case. Grade 3 or higher treatment-related adverse events occurred in 43.6 percent of patients receiving daraxonrasib, lower than the 57.5 percent observed with chemotherapy. Treatment discontinuation due to adverse events was just 1.2 percent with the targeted agent compared with 11.2 percent with chemotherapy. Rash and diarrhea were the most common side effects, but the majority were grade 1 to 2 and could be managed with routine clinical measures. For a drug that intervenes directly on what was long considered the most intractable target in oncology, this tolerability profile represents a remarkable pharmacological achievement.

Placed in historical context, the magnitude of these results becomes even more apparent. A median overall survival of 13.2 months, achieved in the second-line setting, surpasses the historic benchmark of FOLFIRINOX as a first-line regimen, which delivered 11.1 months. The 42-second standing ovation that greeted the data at the ASCO 2026 plenary session reflected not mere numerical progress but a genuine paradigm shift in treatment strategy. The implications extend to surgical oncology as well: approximately 39 percent of patients in the phase I/II study had previously undergone pancreatic resection, a population for whom effective second-line options have long been lacking. Daraxonrasib now offers these patients a meaningful alternative, and given its robust efficacy in advanced disease, investigators argue that moving the drug into the adjuvant or neoadjuvant setting deserves serious consideration.

Still, a measured perspective is warranted. Both published studies were industry-sponsored, and independent real-world validation remains essential before the results are universally adopted into practice. Although rash and gastrointestinal toxicities were predominantly low-grade, standardized management protocols will need to be established as clinical use expands beyond the controlled environment of a trial. Acquired resistance, an inevitability in targeted therapy, is expected to emerge through several mechanisms, including secondary KRAS mutations, bypass pathway activation through EGFR, HER2, or the PI3K–AKT–mTOR axis, adaptive upregulation of downstream effectors such as RAF or MEK, and tri-complex disruption via RAS Y64 or RAS Y71/BRAF alterations. Research to delineate these escape routes is already underway, and combination strategies — pairing daraxonrasib with chemotherapy, immunotherapy, or other targeted agents — will be critical to sustaining durable responses.

The broader RAS-targeted landscape is also evolving rapidly. Clinical trials of KRAS G12D-specific inhibitors, including VS-7375 and setidegrasib, are actively recruiting patients, and the strategic relationship between pan-RAS inhibitors covering multiple isoforms and mutants and allele-specific agents targeting a single variant — whether complementary or competitive — will be one of the defining questions of the coming years. What is no longer in dispute is the central lesson of this moment. RAS was long regarded as the holy grail of undruggable targets, a protein that defied every attempt at pharmacological conquest. Daraxonrasib has demonstrated that this target is not only tractable but capable of delivering tangible survival benefits and quality-of-life improvements to the patients who need them most. For the oncologists and surgeons who have long confronted the most recalcitrant of malignancies, the concurrent arrival of these two landmark studies marks the formal entry of pancreatic cancer therapeutics into the RAS-targeted era — a long-awaited dawn that has, at last, broken.

Subject of Research: Pan-RAS(ON) multi-selective inhibitor daraxonrasib as second-line therapy for RAS-mutated metastatic pancreatic ductal adenocarcinoma

Article Title: The dawn of RAS-targeted therapy: a landmark breakthrough of daraxonrasib in pancreatic cancer

Article References: Liu, C., & Liu, L. (2026). The dawn of RAS-targeted therapy: a landmark breakthrough of daraxonrasib in pancreatic cancer. Clinical Cancer Bulletin, 5(1), Article 20. https://doi.org/10.1007/s44272-026-00072-4

Image Credits: AI Generated

DOI: 10.1007/s44272-026-00072-4

Keywords: pancreatic cancer, daraxonrasib, RAS-targeted therapy, KRAS mutations, RASolute 302, targeted therapy, clinical trial, oncology, drug resistance, second-line treatment, molecular inhibition, precision medicine