Definitions and use of tumour bulk in phase 3 lymphoma trials: a comprehensive literature review

• In phase 3 lymphoma randomised trials, definitions of bulk and its use in the design varies significantly across and within disease subtypes

• Of 87 trials, 32 analysed bulk as a prognostic marker in their study design with only five demonstrating association with survival outcomes.

Tumour 'bulk' has historically been considered an important prognostic marker and clinical tool to guide treatment in patients with lymphoma. However, its use and definitions in trial designs varies significantly and it is unclear how this has influenced the relevance of bulk in contemporary practice. This comprehensive literature review evaluated the definitions, applications and prognostic impact of bulk in phase 3 randomised trials in four major lymphoma subtypes. Overall, 87 studies were identified across follicular lymphoma (FL), diffuse large B-cell lymphoma (DLBCL), peripheral T-cell lymphoma (PTCL) and Hodgkin lymphoma (HL) with a wide range of bulk thresholds employed (5cm, 6cm, 7cm, 7.5cm, 10cm and >1/3 mediastinal mass ratio (MMR)). The most common threshold was: FL; 7cm (58%), DLBCL; 7.5cm and 10cm (44% each), PTCL; 7.5cm (66%) and HL; 1/3 MMR (91%). Bulk threshold was used by trials to determine eligibility (66%), stratification (24%), as a prognostic risk factor (37%) and decision tool for risk-adapted treatment e.g. radiotherapy (29%), however bulk definitions used for these varied both between, and within, lymphoma subtypes, and even within single trials in 25%. Thirty-two studies incorporated bulk in prognostic analyses with only five showing significance for differential survival outcomes. Our analysis demonstrates high inconsistency in thresholds defining tumour bulk and use of bulk in phase 3 lymphoma trials across eligibility, stratification, therapeutic risk-adaptation plus prognostication. This highlights an urgent need for international consensus on definitions of bulk within trials to improve its prognostic and predictive value and refine its application in clinical practice.