Purpose: To investigate associations between body composition phenotypes and survival in women with ovarian cancer.
Sample and setting: Retrospective observational study of 73 women diagnosed with epithelial ovarian cancer between August 2020 and March 2024, treated at Peter MacCallum Cancer Centre.
Procedures: Sociodemographic and treatment details were extracted from medical records. Body composition phenotypes (low skeletal muscle index [SMI], low skeletal muscle density [SMD], high total adipose tissue [TATI], sarcopenic obesity [low SMI, high TATI]) were assessed from pretreatment CT images using sex-and BMI-specific cut-off points. Survival time was defined from diagnosis to death, with records censored at data extraction. Cox proportional hazards models were fitted to assess the association between body composition phenotypes and survival. Models were adjusted for age, stage and comorbidities.
Results: Mean age was 62 (12.7), 86% (n=60) had an advanced stage (III/IV), the median comorbidity index was 2 [1-3], and 85% (n=62) were postmenopausal. Across the sample, 71% (n=52) of patients were treated with neoadjuvant chemotherapy, 81% (n=59) surgery, 53% (n=39) adjuvant chemotherapy, 35% (n=26) targeted therapy, 7% (n=5) hormone therapy, and 3% (n=2) radiotherapy. The body composition analysis identified 61% (n=44) had low SMI, 58% (n=42) had low SMD, 58% (n=40) had high TATI, and 26% (n=18) had sarcopenic obesity. Women with normal SMI (n= 28, HR=.59, 95%CI: .23-1.53, p= 0.28), normal SMD (n= 30, HR= .50 95%CI: .20-1.3, p= 0.14) and those without sarcopenic obesity (n= 51, HR= .46, 95%CI: 0.095-1.14, p= 0.095) had lower hazards for mortality compared to those with altered body composition. However, wide confidence intervals and non-significant p-values indicate substantial uncertainty.
Conclusion and clinical implications: Normal body composition may be protective for survival. However, further analysis will be completed in a larger sample to confirm these associations and provide insight into high-risk phenotypes to guide clinical care.