Long-term follow-up and biomarker analyses of relapsed/refractory Hodgkin lymphoma patients treated with brentuximab vedotin and DHAP: the HOVON/LLPC Transplant BRaVE study
Patients with relapsed/refractory Hodgkin lymphoma (r/r HL) generally have a poor prognosis. We earlier showed that brentuximab vedotin (BV) in combination with DHAP (dexamethasone, high-dose cytarabine, cisplatin) is highly effective with manageable toxicity (Kersten, Haematologica 2020). Here we present the long-term follow-up analyses, including serum thymus and activation regulated chemokine (TARC) and quantitative 18F-FDG-PET-CT scan analyses.
Patients were treated with three cycles of BV-DHAP followed by autologous stem-cell transplant (ASCT). Serum was collected at baseline, after each cycle and during follow-up. Soluble (s)TARC was measured by ELISA. Immunohistochemical (IHC) staining was done for TARC. Baseline 18F-FDG-PET-CT scans were analyzed quantitatively for metabolic tumor volume (MTV) and standardized uptake values (SUV). Prognostic value of biomarkers for progression free survival (PFS) was calculated using receiver operating characteristics, area under the curve (AUC) and log-rank survival analysis. Correlations were calculated using Pearson, data was log-transformed if skewness>0.5.
Sixty-five patients (46% primary refractory) were included. Central pathology review confirmed classical HL diagnosis in 58/65 patients. Patients with a HL-mimicker diagnosis (e.g. PTCL) were excluded from biomarker analysis, but not from clinical evaluation per intention to treat. Pre-ASCT complete metabolic response (CMR) was reached in 53/65 patients (82%) with 5 patients having a partial response (PR; 8%). After a median follow-up of 39 months (range 5-46), 3-year PFS was 77% (95%CI: 70-88) and the overall survival was 95% (95%CI: 90-100).
Serum sTARC decreased significantly after 1 cycle of BV-DHAP (P<0.001) and was prognostic for 3-year PFS (AUC 0.71, optimal cutoff 533 pg/mL). This was independent of pre-ASCT PET-response in a multivariable cox-regression (Hazard ratio: 4.6; P=0.036). Patients with absent or weak TARC staining (n=7) of Hodgkin-Reed Sternberg cells had lower baseline serum sTARC levels compared to patients with positive TARC staining (median 608 versus 3701 pg/mL; P=0.042), and had a lower 3-year PFS rate (57% versus 88%; P=0.005). MTV had low prognostic value for PFS (AUC 0.54) but correlated significantly with baseline serum sTARC (r=0.53; P<0.001). SUVmean, SUVmax and number of lesions on 18F-FDG-PET-CT at baseline were prognostic for 3-year PFS (AUC=0.68, 0.71 and 0.69, respectively) and did not correlate with baseline serum sTARC.
Long-term follow-up of r/r HL patients treated with BV-DHAP shows a high pre-ASCT CMR rate and 3-year PFS and compares favorably to other salvage regimens. Serum sTARC can be used as an early response marker already after 1 cycle of BV-DHAP and is independent of pre-ASCT PET-status. Baseline 18F-FDG-PET parameters are predictive for PFS. Risk assessment using different biomarkers could help guide treatment decisions in an era with new targeted treatment options for r/r HL.