Longitudinal CT-derived muscle–fat–bone remodeling following androgen deprivation therapy initiation in prostate cancer
Abstract
Androgen deprivation therapy (ADT) is a cornerstone of prostate cancer treatment and is commonly accompanied by longitudinal changes in muscle, adipose tissue, and skeletal health that may be poorly captured by conventional anthropometric measures. This study investigated whether routine oncologic computed tomography (CT) could provide an integrated longitudinal assessment of muscle, adipose tissue, and vertebral trabecular attenuation following ADT initiation. In this retrospective, single-center longitudinal study, 60 consecutive men with prostate cancer initiating ADT between January 2023 and May 2025 were screened at Antonio Perrino Hospital, Brindisi, Italy, with clinical follow-up available through July 2026. Forty patients with an eligible baseline CT obtained within 30 days before ADT initiation and a technically comparable follow-up CT at 12 ± 2 months were included. At L3, skeletal muscle index (SMI), bilateral psoas attenuation, subcutaneous adipose tissue index (SATI), and visceral adipose tissue index (VATI) were assessed. Trabecular vertebral attenuation was measured at L1–L3. Mean SMI decreased from 48.4 ± 6.9 to 45.5 ± 6.8 cm²/m² (−6.0%; p<0.001), while mean psoas attenuation decreased from 41.0 ± 5.9 to 37.8 ± 6.0 HU (−7.8%; p<0.001). SATI increased by 11.6% (p<0.001), VATI by 6.2% (p=0.018), and mean vertebral trabecular attenuation decreased by 4.4% (p=0.006). BMI changed only modestly and showed a weak relationship with SMI change. Twenty-three patients (57.5%) had an SMI decline ≥5%, a prespecified exploratory threshold. In an exploratory parsimonious adjusted analysis, lower baseline psoas attenuation was associated with subsequent SMI decline ≥5%; however, the limited sample size precludes interpretation as an independent predictive biomarker. Routine oncologic CT demonstrated concurrent longitudinal changes in muscle, adipose tissue, and vertebral attenuation following ADT initiation that were only weakly reflected by BMI. These findings support further evaluation of integrated opportunistic CT phenotyping in larger controlled cohorts.