Radiation Oncology Advances Take Center Stage as MD Anderson Presents Nearly 70 Abstracts at ASTRO 2026

BOSTON — At the 2026 Annual Meeting of the American Society for Radiation Oncology (ASTRO), researchers from The University of Texas MD Anderson Cancer Center are presenting nearly 70 abstracts spanning the full breadth of modern radiation oncology, from the physics of dose delivery to the genomics of tumor behavior. The meeting, which convenes the world’s leading radiation oncologists, medical physicists, and biologists in Boston beginning September 25, 2026, serves as the field’s principal venue for unveiling evidence that can reshape clinical practice. This year, MD Anderson faculty are highlighting two strands of investigation in particular: an unusually comprehensive quality assurance analysis of stereotactic radiosurgery planning, and a long-awaited genomic reanalysis of one of the most influential prostate cancer radiotherapy trials ever conducted. Alongside the scientific program, three faculty members are being honored with major society awards.

The first of the headline findings confronts a blind spot that has persisted in radiation therapy quality assurance for decades. Stereotactic radiosurgery, or SRS, is a technique that delivers very high doses of radiation to small intracranial targets in one or just a few fractions, and its therapeutic window depends entirely on accuracy: the dose must conform tightly to the target while sparing surrounding healthy brain tissue. Historically, quality assurance audits have concentrated on the machines themselves, verifying that linear accelerators and other platforms deliver the dose they are programmed to deliver with precision and accuracy. What those audits did not measure was the quality of the radiation plan that clinicians feed into the machine, even though planning decisions can vary enormously between practitioners and institutions.

To close that gap, a team from MD Anderson’s Image and Radiation Oncology Core, or IROC, analyzed more than 800 radiation treatment plans contributed by over 650 institutions between 2013 and 2025. Crucially, every plan was designed for the same target inside a standardized head phantom, a physical model engineered to measure radiation delivery under controlled and comparable conditions. This design allowed the researchers to isolate the contribution of human and institutional planning choices from the intrinsic capabilities of different treatment hardware. The results were striking: across treatment platforms, the volume of healthy brain receiving a high dose differed by nearly a factor of five. While the physical limits of each machine accounted for part of that spread, the study found that on standard linear accelerators, the planner’s decisions explained more of the variability than the equipment itself. The single most consequential choice was how tightly the radiation field was shaped around the target.

Lian Duan, a Ph.D. candidate in Radiation Physics who presented the findings, emphasized that the implications extend well beyond the audit itself. “The same plan on different machines does not mean the same dose to healthy brain,” Duan said. “Clinics should benchmark against their own equipment and treat that benchmark as a starting point. Careful planning on the same hardware can still improve a plan substantially.” In other words, two clinics operating identical accelerators could be delivering dramatically different doses to their patients’ healthy brain tissue simply because of how their planners conform the field, a variable that no machine-focused quality assurance program would ever capture. Duan conducts the work with Stephen Kry, Ph.D., within IROC, one of only four centers in the United States funded by the National Cancer Institute, part of the National Institutes of Health, to verify and maintain quality standards for radiation therapy across cancer clinical trials. That mandate gives the group a national vantage point on how consistently, or inconsistently, radiation plans are actually built.

The practical takeaway for the field is a call to institutional self-examination. Rather than assuming that possession of modern hardware guarantees optimal plan quality, the MD Anderson team argues that clinics should measure their own plan performance against benchmarks established on their specific equipment, and then treat continuous planning improvement as an ongoing quality objective. Because SRS is used to treat brain metastases, benign brain tumors, and other intracranial lesions where normal tissue dose directly influences the risk of radiation necrosis and cognitive effects, a nearly fivefold difference in high-dose exposure to healthy brain represents a clinically meaningful gap, not a statistical curiosity. The study reframes plan quality as a modifiable variable under the control of every clinic, independent of its capital equipment.

The second headline study turns from physics to biology, and to a clinical trial whose results changed standard practice. NRG/RTOG 9601 was a landmark phase 3 prostate cancer study demonstrating that adding anti-androgen hormone therapy to salvage radiation therapy improved outcomes for patients whose prostate-specific antigen rose after radical prostatectomy, signaling that the cancer had returned. The trial’s results reshaped clinical guidelines and made combined hormone therapy plus salvage radiation a common recommendation. But anti-androgen therapy carries significant side effects, and at the time the initial results were published, clinicians had no reliable method for identifying which individual patients would actually benefit from the added treatment. Subsequent research showed that measuring prostate-specific antigen, or PSA, a protein produced by prostate cells, could help guide decisions, yet even with PSA testing, estimating how aggressive a particular patient’s recurrence truly is remains a persistent challenge.

Krishnan Patel, M.D., associate professor of Radiation Oncology at MD Anderson, led an effort to answer that question with modern tools. His team applied a contemporary genomic classifier, a test that evaluates genetic material within the tumor to predict its likely behavior, to archived tumor samples from the landmark trial. The genomic test successfully separated patients into risk groups with markedly different long-term outcomes. Fifteen years after treatment, 75 percent of patients in the low-risk group were still alive, compared with only 25 percent of those in the very high-risk group. That degree of separation, derived from samples collected roughly two decades ago, illustrates how molecular diagnostics can extract clinically actionable prognostic information from historical trial cohorts and validate a modern assay against the gold standard of a completed, practice-changing phase 3 study with long-term survival follow-up.

Beyond prognosis, the analysis carries direct implications for treatment personalization. The study suggests that the genomic classifier can complement, rather than replace, standard blood tests such as PSA. Under current patterns of care, many physicians recommend hormone therapy for men with high pre-radiation PSA levels and withhold it from those with low values. The new findings indicate that a distinct subgroup of patients with low pre-radiation PSA levels but aggressive biological features on genomic testing may nonetheless benefit from hormone therapy, while others with apparently elevated risk by conventional measures may be spared its toxicities. “The ability to apply a modern genomic test to samples from a clinical trial conducted nearly two decades ago is incredibly valuable,” Patel said. “These findings are helping us continue to personalize treatment recommendations, allowing some patients who are unlikely to benefit from hormone therapy avoid its side effects, while giving clinicians greater confidence in identifying those who are most likely to benefit.”

Taken together, the two presentations illustrate the dual trajectory of contemporary radiation oncology: one line of research perfecting the physical delivery of treatment, the other refining the biological selection of which treatments each patient should receive. The SRS planning study shows that substantial gains in safety may be achievable without new machines, purely through better planning practice and systematic benchmarking. The prostate cancer genomic analysis shows that molecular profiling, validated against long-term outcomes from a definitive trial, can move hormone therapy decisions from population-level defaults toward individualized prescriptions. Both studies underscore a theme that runs throughout MD Anderson’s nearly 70 abstracts at this year’s meeting: precision in radiation oncology is as much about decisions and data as it is about dose.

The meeting also serves as an occasion to recognize individual achievement within the department. Bruce Minsky, M.D., professor emeritus of Radiation Oncology at MD Anderson, will receive ASTRO’s Gold Medal, the society’s highest honor. Chelsea Pinnix, M.D., Ph.D., professor of Radiation Oncology, has been named an ASTRO Fellow, and Clifton Fuller, M.D., Ph.D., professor of Radiation Oncology, will receive the society’s 2026 Mentorship Award for his role in developing the next generation of physician-scientists in the field. Additional information on all MD Anderson presentations at the ASTRO Annual Meeting is available through the institution’s dedicated meeting hub, and the full abstract catalog is published by ASTRO as part of the official scientific program.

Subject of Research: Quality assurance in stereotactic radiosurgery planning and genomic prediction of prostate cancer outcomes in salvage radiotherapy

Article Title: ASTRO: UT MD Anderson highlights advances in radiation oncology

Article References: ASTRO: UT MD Anderson highlights advances in radiation oncology. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: radiation oncology, stereotactic radiosurgery, quality assurance, medical physics, prostate cancer, genomic classifier, NRG/RTOG 9601, salvage radiotherapy, androgen deprivation therapy, ASTRO 2026, MD Anderson Cancer Center, PSA