Muscle Fat, Not Muscle Mass, May Signal Early Setbacks in Radioactive Tumor Therapy

For patients with neuroendocrine tumors of the gut and pancreas, one of the most elegant weapons in modern oncology is a radioactive molecule that homes in on cancer cells like a guided missile. The treatment, known as peptide receptor radionuclide therapy, or PRRT, delivers lutetium-177 to tumor cells that display somatostatin receptors on their surface, sparing much of the surrounding healthy tissue. Most patients benefit, but a stubborn subset sees their disease march forward almost immediately despite the therapy. A new study from LMU University Hospital in Munich set out to discover whether the answer to who fails early might be hiding not in the tumor itself, but in the patient’s own muscle.

The research, published in the European Journal of Nuclear Medicine and Molecular Imaging, examined whether body composition measured on routine CT scans could predict which patients would experience early progression after PRRT. The idea rests on a growing body of evidence in oncology: it is not merely how much muscle a patient carries, but the quality of that muscle, that shapes outcomes across a striking range of cancers. Previous studies have linked fatty infiltration of muscle, a condition called myosteatosis, to poorer survival after pancreatic surgery, worse responses to chemoembolization in liver cancer, and altered activity of immunotherapies in metastatic melanoma.

Myosteatosis is best understood as muscle that has been quietly invaded by fat at the microscopic level. On a CT scan, healthy skeletal muscle appears relatively dark because it contains abundant water-dense contractile tissue. When fat infiltrates the muscle fibers, the tissue’s average density drops, which radiologists measure in Hounsfield units on the standard gray-scale of CT imaging. Crucially, researchers can extract these measurements from virtually any abdominal CT, including the CT component of a PET/CT scan that every PRRT patient already receives, simply by analyzing a slice at the level of the third and fourth lumbar vertebrae, where the psoas muscles anchor the spine.

The Munich team, led by first authors Elma Dema and Christian Dascalescu under the senior guidance of Rudolf Werner and Sophie Siegmund, screened 321 consecutive patients who underwent PRRT for somatostatin receptor-positive gastroenteropancreatic neuroendocrine tumors. Complete imaging and follow-up data were available for 301 of them. Using BMI-adjusted cut-offs for psoas muscle attenuation, the investigators defined which patients had myosteatosis at baseline, before a single radioactive dose had been administered. Their primary endpoint was deliberately stringent: early progression as seen on somatostatin receptor-targeted PET imaging after up to four consecutive treatment cycles, dichotomized as progressive disease versus disease control.

The headline result was, in one sense, a negative one. Traditional quantity-based measures of body composition, including skeletal muscle area and related indices that simply quantify how much muscle tissue surrounds the spine, showed no association whatsoever with early progression. Muscle bulk alone, it seems, tells clinicians nothing useful about whether a neuroendocrine tumor will shrug off PRRT within the first treatment window. This finding matters because muscle quantity is still the most commonly reported body composition parameter in oncology literature, often treated as a stand-in for overall nutritional reserve.

The quality-based measure told a more intriguing story. Only eight of the 301 patients, a mere 2.7 percent, met the criteria for baseline myosteatosis, yet those eight appeared to fare strikingly worse. In univariable analysis, myosteatosis was associated with roughly an eight-fold increase in the odds of early PET-based progression, with an odds ratio of 7.8 and a 95 percent confidence interval running from 1.74 to 34.78, yielding a nominal p-value of 0.0073. In a separately specified model adjusted for baseline clinical characteristics, the association held steady at an odds ratio of 8.1. But here the statistical plot thickens considerably.

When the researchers applied the Benjamini-Hochberg procedure, a standard correction that controls the false discovery rate when multiple hypotheses are tested simultaneously, the p-value drifted to 0.1073, crossing above the conventional significance threshold. The association was also attenuated when the number of completed PRRT cycles was added as a post-baseline treatment exposure variable, landing at an odds ratio of 5.7 with a confidence interval that narrowly grazed 1.0 and a p-value of 0.0538, just shy of the 0.05 benchmark. With only eight affected patients, the confidence intervals around these estimates were wide, which is the statistical signature of a hypothesis that cannot yet be called a finding.

The authors are admirably blunt about this limitation. They conclude that the results should be regarded as exploratory and hypothesis-generating, and that they do not establish any clinical utility for myosteatosis in individual risk stratification. In plain terms, no one should yet order a CT-based muscle analysis to decide who gets PRRT or how closely they should be monitored. The study also reveals a practical puzzle: myosteatosis was far rarer in this cohort than the 30 to 50 percent prevalence reported in some other cancer populations treated with PRRT, suggesting that the BMI-adjusted thresholds used here may flag only the most severe cases, or that neuroendocrine tumor patients differ fundamentally in their metabolic profiles.

Why might fat-infiltrated muscle and treatment resistance be connected at all? The mechanisms remain speculative, but several plausible pathways have been described in the wider literature. Myosteatosis is considered a marker of systemic metabolic derangement, reflecting chronic inflammation, insulin resistance, and altered mitochondrial function within muscle tissue. These same processes could plausibly influence tumor biology, modulate the delivery of radiopharmaceuticals, or shape how the body tolerates and responds to radiation. Some researchers have even proposed myosteatosis as an early and targetable biomarker of cancer prognosis, a measurable vulnerability that nutritional or exercise interventions might one day reverse before treatment begins.

For the field of nuclear medicine, the study is a careful lesson in statistical humility and a genuine invitation to dig deeper. Early progression after PRRT occurs in roughly 8 percent of patients, and identifying this minority before therapy would allow clinicians to consider intensified surveillance, alternative sequencing, or combination approaches for those at highest risk. The Munich data suggest that the search should move beyond how much muscle a patient has and toward the finer architecture of that muscle, ideally with larger, prospective cohorts and standardized measurement protocols. Until then, the radioactive missile remains a remarkably effective weapon, even if the secrets of the rare patients it fails are still written, faintly, in their own flesh.

Subject of Research: Association between CT-based body composition and early progression after peptide receptor radionuclide therapy in gastroenteropancreatic neuroendocrine tumors

Article Title: Baseline body composition and early PRRT outcome in GEP-NETs: an exploratory analysis

Article References: Dema, E., Dascalescu, C., Holzgreve, A., Cyran, C. C., Widjaja, L., Gildehaus, F. J., Herr, F. L., Spitzweg, C., Auernhammer, C. J., Ingrisch, M., Brendel, M., Ricke, J., Werner, R. A., & Siegmund, S. C. (2026). Baseline body composition and early PRRT outcome in GEP-NETs: an exploratory analysis. European Journal of Nuclear Medicine and Molecular Imaging. https://doi.org/10.1007/s00259-026-08182-w

Image Credits: AI Generated

DOI: 10.1007/s00259-026-08182-w

Keywords: PRRT, neuroendocrine tumors, body composition, myosteatosis, lutetium-177, PET/CT, sarcopenia, radiomics, nuclear medicine, cachexia, psoas muscle, tumor progression