Metabolic Tumor Volume Influences Response to Brentuximab-Vedotin in Relapsed Refractory Hodgkin Lymphoma

Purpose: The total metabolic tumor volume at baseline (T-MTV0) computed on PET has been proposed as a prognostic factor at staging in Hodgkin lymphoma (HL) and diffused large B cell lymphoma (DLBCL). It has also been described to play a role in non-HL treatment by monoclonal antibodies since it could influence antibody exposure and efficacy in a murine model (Dayde D, et al. Blood 2009) that has been confirmed in DLBCL patients study (Casasnovas O, et al. Abstract Lugano 2015). We hypothesized the metabolic tumor burden could influence the efficacy of a monoclonal antibody-drug conjugated (ADC) such as monotherapy based anti-CD30 ADC brentuximab-vedotin (BV). In the present study, we assessed the pre-treatment baseline total metabolic tumor volume based on PET evaluation in the response to BV treatment as a single agent in patients with RR-HL, as well as its prognostic value.

Methods: A retrospective multi-center study was built from January 2011 to June 2015. Forty-one consecutive heavily pre-treated patients with a diagnosis of relapsed refractory HL (RR-HL) were included. PET was performed at baseline (PET0) and after 4-6 cycles of BV defining two groups of responders, good responder group achieving a complete metabolic response (CMR group, 23 patients), and non-responder group (no-CMR group, 18 patients), using the revised Lugano classification with the 5-point scale visual analysis PET/CT. T-MTV0 was measured with a semiautomatic method using a 41%-SUVmax-threshold. SUVmax at PET0 (SUVmaxPET0) was measured in each patient and represented the hottest lesion independently from the site. To assess the influence of the T-MTV0 on BV efficacy, we compared the baseline metabolic tumor volume between the two groups. The ROC curve was established to determine the optimal cut-off of T-MTV0 to predict treatment failure.

Results: T-MTV0 ranged from 14 cm³ to 213 cm³ in the 41 patients (median 62 cm³; 25th - 75th percentiles 25 - 94 cm³) and SUVmaxPET0 ranged from 4,1 to 19,3 (median 10,4; 25th-75th percentiles 7,8 - 13,7). T-MTV0 was significantly higher in the no-CMR group as compared with the CMR group (median 96 and 30 cm³; 25th-75th percentiles 90 - 139 cm³ and 18 - 38 cm³, respectively; p< 0.01). Considering metabolic sites, the nodal metabolic tumor volume (N-MTV0) represented the main component of T-MTV0 and was significantly higher than the extra nodal metabolic tumor volume (EN-MTV0) in the CMR group (p = 0.025). However, the EN-MTV0 dominated in the no-CMR group and was significantly higher than the N-MTV0 (p = 0.01). SUVmaxPET0 as well as clinical characteristics were not significantly different in the two groups. Furthermore the TMTV0 cut-off value was 62 ml and predictive of treatment failure.

Conclusion: In the present study, we demonstrated that tumor metabolic burden and nodal or extra-nodal localizations influence response to BV as a single agent in RR-HL patients, offering some future research directions in the development of new schedules of antibodies administration in anticancer therapy such as the concept of the individual adjustment of treatment dose to tumor burden. This may guide clinicians in their choice of therapeutic strategy.