The research drew on the Geriatric Assessment for Patients (GAP70+) study, a multicentre, cluster-randomized clinical trial conducted across 39 community oncology clinics affiliated with the University of Rochester Cancer Center NCI Community Oncology Research Program Research Base. Participants were aged 70 or older, had incurable stage III or IV solid tumours or lymphoma, and were scheduled to begin a new systemic treatment regimen known to carry a high risk of toxicity. From the trial’s usual care arm, the investigators collected abdominal and pelvic CT or PET/CT scans performed as part of routine care, applying strict time windows around enrollment and rigorous quality control. That process yielded 161 patients whose imaging and clinical data could be fully analysed, with a mean age of 76.6 years, nearly 61 percent male, and most carrying gastrointestinal or lung cancers.

What sets this analysis apart from the bulk of prior sarcopenia research in oncology is its adherence to a full clinical definition of the condition rather than a single imaging measurement. The European Working Group on Sarcopenia in Older People, known as EWGSOP2, defines sarcopenia using three criteria: reduced muscle strength, reduced muscle quantity and quality, and reduced physical performance. In this study, strength was assessed with a five-time chair-stand test, where a time exceeding 16.7 seconds signalled weakness. Muscle quantity was quantified from the mid-L3 lumbar vertebra slice of each CT scan using validated automatic segmentation software, producing a skeletal muscle index with sex- and BMI-specific thresholds. Muscle quality was captured through Hounsfield Unit measures of tissue density on the same scans, and physical performance was evaluated with the timed up-and-go test. Meeting the strength and quantity or quality criteria classified a patient as sarcopenic; meeting all three criteria indicated severe sarcopenia.

The prevalence figures were sobering. More than two-thirds of participants, 68.3 percent, had low muscle quantity by imaging criteria, and 56.5 percent met the full clinical definition of sarcopenia. Among those, 59 percent were classified as having severe sarcopenia. Patients with sarcopenia were on average about two years older than those without it and were significantly more likely to have diabetes. More tellingly, sarcopenia tracked closely with broader geriatric vulnerability: those with the severe form were significantly more likely to show impaired nutrition, dependence in activities of daily living and instrumental activities, recent falls and impaired cognition compared with patients whose muscles were preserved. In other words, the clinical sarcopenia label captured a constellation of ageing-related frailty that a simple muscle measurement did not.

When the researchers turned to treatment outcomes, documented through clinician-reported adverse events graded with the Common Terminology Criteria for Adverse Events, version 4.0, a nuanced picture emerged. Overall severe toxicity, defined as Grades 3 through 5, during the first three months of therapy did not differ significantly between patients with and without sarcopenia. But severe nonhematologic toxicities—a category that includes complications such as dehydration, electrolyte disturbances, infections, fatigue and pain, which frequently lead to hospitalization—were substantially more common in the sarcopenic group, affecting 64 percent versus 44 percent. In multivariable logistic regression adjusting for age, tumour site, sex, comorbidities and treatment type, sarcopenia more than doubled the odds of severe nonhematologic toxicity, with an odds ratio of 2.24. Notably, patients who suffered these severe nonhematologic events also had significantly slower chair-stand times, and each additional second on that strength test raised the odds of toxicity by roughly 15 percent.

The mortality findings were even more striking. At one year, only 46 percent of participants with sarcopenia were alive, compared with 60 percent of those without it. After statistical adjustment this difference fell just short of significance in the full cohort, but when the analysis focused on patients starting first-line chemotherapy, sarcopenia was clearly associated with poorer survival, with a hazard ratio of 2.12. For severe sarcopenia the signal was unmistakable: one-year survival was just 40 percent versus 60 percent among patients without sarcopenia, a difference that remained significant after adjustment with a hazard ratio of 2.08. The mortality gap opened early—within three months, nearly 21 percent of patients with severe sarcopenia had died, versus 7 percent of those without the condition. Crucially, when the investigators evaluated skeletal muscle index alone, stripped of the strength and performance components, no significant association with survival appeared at all.

That last result may be the study’s most important technical message. Much of the oncology literature has equated sarcopenia with low muscle quantity on imaging, treating the skeletal muscle index as a standalone prognostic marker. This analysis suggests that in older adults that approach is insufficient, because muscle mass declines physiologically with age and is not a linear proxy for what muscles can actually do. The consensus definitions that incorporate grip-equivalent strength measures and performance tests appear to identify a genuinely distinct population of patients whose functional reserve is depleted and who are therefore less able to withstand the catabolic assault of cytotoxic therapy. The authors note this contrasts with patterns seen in younger patients, in whom imaging-based muscle measures have shown stronger associations with outcomes.

The comparison with the limited prior work reinforces the point. Only one earlier study, in men with metastatic castrate-resistant prostate cancer, had evaluated treatment toxicity in older adults using a definition of sarcopenia encompassing strength, quantity and performance; it reported a prevalence of 27 percent, lower than the 57 percent seen here, likely reflecting stricter criteria that resemble the severe sarcopenia category. That study likewise linked sarcopenia to poorer survival and heightened toxicity risk. Two recent studies in gastrointestinal cancer patients that relied on muscle index alone produced inconsistent survival associations, underscoring the value of the multi-domain approach. The new analysis also benefits from real-world breadth: patients were treated in community clinics nationwide rather than tertiary referral centres, and toxicity data met the annotation standards of a randomized trial.

The limitations are worth noting. The cohort was predominantly non-Hispanic white, limiting generalizability to other populations. Scans were not mandated by the protocol, and the strict time windows that ensured data quality reduced the analysable sample. As an exploratory secondary analysis, p-values were not corrected for multiple testing, so the findings are hypothesis-generating and require confirmation in prospective cohorts. Treatment regimens and dosing were left to each patient’s oncologist rather than standardized. Even so, the clinical implications are hard to dismiss. Patients with severe sarcopenia received some of the most intensive treatments in the cohort yet had the worst survival, raising the possibility of overtreatment and suggesting that comprehensive geriatric evaluation should precede decisions about aggressive therapy.

The authors argue that sarcopenia screening should become routine in geriatric oncology, and they point to a practical path. Although CT scans are already available for nearly every cancer patient and software tools exist to quantify muscle from them, the expertise to interpret such measurements is not yet part of routine practice. The geriatric assessment, by contrast, is already recommended by ASCO guidelines and captures exactly the measures—chair-stand strength, timed up-and-go performance, falls, daily functioning—that this study linked to toxicity and death. Integrating those simple tests into treatment planning, and designing intervention trials that incorporate sarcopenia status into treatment selection and dosing, could spare the most vulnerable older adults the cruelest harms of therapy while preserving its benefits for those robust enough to tolerate it.

Subject of Research: Sarcopenia and treatment toxicity and mortality in older adults with advanced cancer

Article Title: Sarcopenia Is Associated With Severe Treatment Toxicity and Mortality in Older Adults With Advanced Cancer: A Secondary Analysis

Article References: Mattick, L. J., Mohammed, M., Li, C.-S., Tylock, R. G., Loh, K. P., Flannery, M. A., Peppone, L. J., Lin, P.-J., Goncalves, M. D., Caan, B. J., Cespedes Feliciano, E. M., Popuri, K., Beg, M. F., Bearden, J. D., Berenberg, J., Katato, K., Mustian, K. M., Mohile, S. G., & Dunne, R. F. (2026). Sarcopenia Is Associated With Severe Treatment Toxicity and Mortality in Older Adults With Advanced Cancer: A Secondary Analysis. Journal of Cachexia, Sarcopenia and Muscle, 17(5), Article e70388. https://doi.org/10.1002/jcsm.70388

Image Credits: AI Generated

DOI: 10.1002/jcsm.70388

Keywords: sarcopenia, geriatric oncology, advanced cancer, chemotherapy toxicity, muscle strength, mortality, CT body composition, EWGSOP2, geriatric assessment, cancer cachexia, older adults, clinical trial