Impact of Body Composition on Treatment Toxicity and Outcomes in Patients With Primary Mediastinal Large B-Cell Lymphoma

Abstract

Primary mediastinal large B-cell lymphoma (PMBL) is a rare entity that predominantly affects young female patients and typically presents as a large and compressive anterior mediastinal mass. Accumulating evidence suggests relationships among PMBL patient body composition (BC), cancer outcomes, and treatment-related toxicities. The aim of this study was to evaluate the impact of BC on PMBL patients using PET-CT images acquired pretreatment. Two hundred nineteen patients were included in an ancillary analysis of a multicenter retrospective LYSA cohort of treatment-naïve adult PMBL patients who received first-line treatment with ACVBP, CHOP14 or CHOP21 plus anti-CD20. Anthropometric parameters were assessed from the baseline PET-CT image using two methods: (i) manual segmentation at the L3 level and (ii) automatic software-based multislice measurements with Anthropometer3DNet. The median age was 35.4 years (range 18-88 years), and the median body mass index was 23.8 kg/m² (15.6; 40.8). Overall, 137 patients were treated with R-ACVBP, 44 received R-CHOP14, and 38 were treated with R-CHOP21. Patients with low lean body mass had a higher incidence of febrile neutropenia, both in the overall cohort (25% vs. 12.6%, p = 0.02) and in the R-ACVBP subgroup (35.7% vs. 19.4%, p = 0.03). Univariate analysis showed that in patients treated with R-ACVBP, subcutaneous low adiposity, determined by 3D measurements, was associated with overall survival (OS) (p = 0.04). At 3 years, the OS (95% CI) was 96% (93-100) in above-median adiposity patients and 86% (78-95) in below-median adiposity patients. Low lean body mass (LBM), assessed from the pretreatment PET-CT images using automatic Anthropometer3DNet software, may serve as a predictive marker for acute treatment-related toxicity in PMBL patients, particularly those receiving the dose-intensive R-ACVBP regimen. Additionally, depletion of the subcutaneous fat mass was correlated with an increased risk of mortality, highlighting the importance of a comprehensive BC assessment in this patient population.

Keywords: body composition; dose‐intensive treatment; febrile neutropenia; outcomes; primary mediastinal large B‐cell lymphoma.

FIGURE 1

Flow chart. PMBL: Primary Mediastinal B‐cell Lymphoma; LYSA: Lymphoma Study Association; PET‐CT: Positron Emission Tomography–Computed Tomography; CT: Computed Tomography; L3: Third lumbar vertebra, 2D: two‐dimensional; 3D: three‐dimensional.

FIGURE 2

Plot of the correlations among different anthropometric parameters. SAT (3D): 3D‐derivedsubcutaneous adipose tissue. LSAI (2D): 2D‐derived L3 subcutaneous adipose tissue index. VAT (3D): 3D‐derived visceral adipose tissue. LVAI (2D): 2D‐derived L3 visceral adipose tissue index (2D). MBM (3D): 3D‐derived muscle body mass. FBM (3D): 3D‐derived fat body mass. LBM (3D): 3D‐derived lean body mass. LSMI (2D): 2D‐derived L3 skeletal muscle tissue index.

FIGURE 3

Overall survival (A) and progression‐free survival (B) in the R‐ACVBP‐treated subgroup according to subcutaneous adipose tissue (SAT) status. Kaplan–Meier survival curves are shown for patients with low versus high SAT as defined by sex‐specific median values.