Muscle loss peaks three months after pelvic radiotherapy in gynecological cancer

CT-defined sarcopenia increased from 23% before pelvic RT to 42% at three months, while bedside bioimpedance missed many cases.

KEY POINTS

  • This prospective cohort study from Amsterdam UMC, the Netherlands, followed 31 women with gynecological cancer receiving curative-intent pelvic radiotherapy. Diagnoses included 21 cervical, six endometrial, and four vaginal cancers, with a median age of 57 years. Muscle mass was evaluated before treatment, during the final week of radiotherapy, and at three and six months using L3 CT-derived lumbar skeletal muscle index (LSMI) and bioelectrical impedance analysis (BIA).
  • Radiotherapy consisted of 45 Gy in 25 fractions using VMAT, with nodal boosts up to 2.2–2.3 Gy per fraction when indicated. Selected patients also received brachytherapy, concurrent chemotherapy, or deep hyperthermia. CT-defined sarcopenia was operationally defined using LSMI <38.5 cm²/m², while BIA applied an appendicular skeletal muscle index threshold of <5.5 kg/m². Muscle strength and physical performance were not assessed.
  • CT-defined sarcopenia increased during treatment and peaked three months afterward. Prevalence was 23% (7/31) at baseline, 32% (10/31) during the final week of radiotherapy, and 42% (13 patients) at three months, before declining to 29% (9/31) at six months. However, an improvement in the proportion meeting the sarcopenia threshold did not necessarily indicate recovery of muscle mass to pretreatment levels.
  • Substantial muscle depletion persisted even when patients no longer met the binary sarcopenia threshold. An exploratory reduction in LSMI exceeding 4.5% relative to baseline was observed in 11 patients during the final treatment week, 15 patients at three months, and 13 patients at six months. Median LSMI change was −1.6%, −4.6%, and −3.0%, respectively. The 4.5% threshold was extrapolated from a different body-composition assessment method and has not been validated as a clinical intervention threshold for CT-derived LSMI.
  • Bedside bioimpedance substantially underestimated CT-defined muscle depletion. Depending on the assessment time point, BIA sensitivity ranged from 0% to 50%, despite specificity of approximately 85–94%. At three months, BIA identified sarcopenia in only one patient, compared with 13 patients by CT. Treatment-associated changes in hydration and body-fluid distribution may partly explain the disagreement, although the study did not directly establish the mechanism.
  • Baseline and disease characteristics were associated with sarcopenia during treatment. Women classified as sarcopenic during the final EBRT week had lower median BMI (23 vs 30 kg/m²; p=0.013), and FIGO stage III was more frequent (70% vs 24% among patients without sarcopenia, with p=0.027 for the overall stage distribution). At three months, estimated five-year overall survival was 77% in patients with sarcopenia versus 100% without, but only three deaths occurred in the entire cohort, preventing reliable prognostic inference.
  • The findings are exploratory and do not isolate radiation-induced muscle loss. Changes may reflect underlying disease, chemotherapy, previous surgery, nutritional intake, reduced activity, and treatment-related symptoms. The cohort was small and heterogeneous, with no untreated control group or functional muscle testing. Consequently, the observed body-composition changes cannot be attributed to radiotherapy alone, and the clinical benefit of nutritional or exercise-based interventions was not tested.

CLINICAL TAKEAWAY

Substantial skeletal muscle depletion can develop during pelvic radiotherapy for gynecological cancer and persist for months after treatment, even when body weight remains relatively stable. CT-based muscle assessment identified substantially more cases than the BIA method tested, suggesting that conventional weight measurements and this bioimpedance approach may be insufficient for detecting muscle loss. Larger prospective studies incorporating nutritional, functional, and treatment-related variables are needed before specific screening or intervention protocols can be recommended.

SOURCE

International Journal of Radiation Oncology, Biology, Physics

Browse more research Suggest a correction