PD005 - PREDICTING DOSE-LIMITING TOXICITY IN PATIENTS WITH CANCER: COMPARISON BETWEEN BIOELECTRICAL IMPEDANCE ANALYSIS AND COMPUTED TOMOGRAPHY

Linked sessions

PD005

PREDICTING DOSE-LIMITING TOXICITY IN PATIENTS WITH CANCER: COMPARISON BETWEEN BIOELECTRICAL IMPEDANCE ANALYSIS AND COMPUTED TOMOGRAPHY

E. Mattavelli1,*, I. Barca1, S. Dierna2, S. Della Valle3, F. De Simeis1, A. Uggè1, G. Greco4, A. Amplatz1, R. Vigorito5, C. Borlotto2,4, P. Pedrazzoli6,7, R. Caccialanza1,8, E. Cereda1 on behalf of ONCO-BIVA Collaborative Group

1Clinical Nutrition and Dietetics Unit, Fondazione IRCCS Policlinico San Matteo, 2Diagnostic Imaging and Radiotherapy Unit, Department of Clinical, Surgical, Diagnostic, and Pediatric Sciences, University of Pavia, Pavia, 3Clinical Nutrition Unit, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, 4Radiology Institute, Fondazione IRCCS Policlinico San Matteo, Pavia, 5Department of Radiology, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, 6Department of Internal Medicine and Medical Therapy, University of Pavia, 7Department of Oncology, Comprehensive Cancer Center, Fondazione IRCCS Policlinico San Matteo, Pavia, 8Department of Oncology and Emato-Oncology, University of Milan, Milan, Italy

 

Rationale: Low skeletal muscle mass - as defined by Skeletal Muscle Index (SMI) - obtained using computed tomography (CT) is a well-established predictor of dose-limiting toxicity (DLT) in patients with cancer. More recently, standardized phase angle (SPA) derived by bioelectrical impedance assay (BIA) has been associated with DLT. However, no studies have directly compared the performance of SMI and SPA in predicting DLT.

Methods: We analysed a subsample of the ONCO-BIVA study. Patients with an available baseline CT scan at L3 suitable for muscle contouring were included (n=118). Skeletal muscle area was quantified using SliceOmatic (v.5) to compute SMI. Low SMI was defined by sex- and BMI-specific first-tertile cut-offs (Table 1). Patients were classified according to SPA as depleted by the literature-based cut-off of <1.65°. Two separate multivariable logistic regression models were built to assess the association of SPA and SMI with DLT, adjusting for NRS-2002 and cancer stage. Model discrimination was evaluated using the Area Under the receiver operating characteristic Curve (AUC), and DeLong test was used to compare AUCs.

Results: DLT occurred in 34 patients (28.8%). Low SPA (n=26.22%) and low SMI (n=39.33%) were both significantly associated with an increased risk of DLT: SPA, OR=15.15 [95% CI, 4.82–47.53], p<0.001; SMI, OR=3.80 [95% CI, 1.53–9.44], p=0.004. SPA showed a numerically higher discriminative ability than SMI, although the difference was only nearly significant (AUCs: 0.815 vs 0.708; p=0.051).

Image:



 

Conclusion: SPA showed a predictive performance comparable to SMI for DLT in patients with cancer, with a numerically higher discriminative ability. Given its feasibility and bedside applicability, SPA may represent a valuable alternative for stratifying DLT risk.

References: Cereda E, et al. ESMO Open (2024) 10.1016/j.esmoop.2024.103666.

Disclosure of Interest: None declared