Brentuximab Vedotin with R-CHP Chemotherapy As Frontline Treatment for Patients with CD30 Positive Primary Mediastinal Large B-Cell, Diffuse Large B-Cell, and Grey Zone Lymphomas: Results of a Phase I/II Multisite Trial

BACKGROUND: CD30 is a cell membrane protein detectable by immunohistochemistry (IHC) in approximately 80% of primary mediastinal large B-cell (PMBL), 20% of diffuse large B-cell (DLBCL) and 100% of grey zone lymphomas (GZL). Brentuximab vedotin (BV) is a CD30-directed immunoconjugate with established efficacy in relapsed Hodgkin and some T-cell lymphomas. This targeted agent may have a role in management of patients (pts) with CD30+ B-cell non-Hodgkin lymphomas (NHL).

METHODS: We designed an open label, multicenter phase I/II trial using BV administered concurrently with R-CHP (rituximab 375 mg/m², cyclophosphamide 750 mg/m², doxorubicin 50 mg/m² on Day [D] 1 and prednisone 100 mg on D1-5) every 3 weeks for 6 cycles (C) as frontline treatment for CD30+ PMBL, DLBCL and GZL. Pts with any Ann Arbor stage and at least equivocal CD30 expression on IHC were eligible. Consolidative radiation (RT) was allowed after completion of study at discretion of the treating physician. Phase I utilized a 3+3 de-escalation design with starting dose of BV at 1.8 mg/kg in combination with R-CHP. G-CSF use was based on institutional policies. The primary objective of phase I was assessment of safety and dose limiting toxicity (DLT) defined as any grade (G) 3/4 non-hematologic toxicity observed in C1 requiring dose delay >14 days from the planned D1 of C2. The primary phase II objectives were overall response (ORR) and complete response (CR) rates with secondary endpoints of progression free survival (PFS), overall survival (OS), and safety. Explorative analyses include correlation of response with quantitative CD30 expression and results of a customized Nanostring panel. Revised Response Criteria were used for response assessment (Cheson, 2007). Enrollment began in 1/2014 and completed in 6/2017.

RESULTS: We enrolled 33 treatment naïve pts at 3 centers. One pt was re-classified from GZL to HL prior to starting therapy and one pt withdrew consent. The remaining 31 pts included 23 (74%) with PMBL, 6 DLBCL (19%) and 2 GZL (7%). Median age was 37 years (range 18 -76), 15 (48%) were female, 23 (74%) had bulky disease over 7.5 cm in diameter, and 11 (35%) had stage III-IV disease. No DLT was observed during phase I and BV dose of 1.8 mg/kg in combination with R-CHP was selected for phase II. There were no treatment related deaths. One pt discontinued protocol therapy after concurrent G3 sepsis and G3 left ventricular (LV) systolic dysfunction following C4. One pt discontinued BV following C5 for G2 pneumonitis possibly related to BV. Two pts required BV dose reduction to 1.2 mg/kg due to sensory neuropathy. The hematologic G3/4 toxicities included febrile neutropenia in 4 pts and afebrile neutropenia in 8 pts. There were no G4 non-hematologic adverse events (AE). G3 non-hematologic AEs at least possibly related to the protocol treatment included nausea/vomiting in 1 pt (3%) and sepsis with LV systolic dysfunction in 1 pt (3%). The most common G2 non-hematologic AEs included nausea (16%), sensory neuropathy (13%), diarrhea (10%), motor neuropathy (6%) and anorexia (6%). One pt with a family history of acute myeloid leukemia (AML) developed AML (normal cytogenetics) at 2 years from completion of chemotherapy and is currently in remission after allogeneic stem cell transplant (SCT).

ORR among the 28 pts who underwent end of treatment imaging was 100% (CI: 88%-100%) with 25 (89%) in CR (CI: 72%-98%). With median follow-up of 15 months (range: 5-36), the median PFS and OS were not reached with all pts alive. For the entire cohort, the 1 year PFS is 86% (CI: 63%-95%) and 1 year OS is 100%. In 20 pts with PMBL, the median PFS was not reached with 1 year PFS 87% (CI: 57-97%). For PMBL pts with stage I/II disease (N=13), the 1 year PFS was 100% and for those with stage III/IV 73% (CI: 28-93%). There was no difference in PFS between those with PMBL who received RT (N=12) and those who did not (N=8) with p=0.39. All pts who progressed on this study (N=3) had advanced stage disease. One GZL pt is in CR after autologous SCT; 2 PMBL pts have responded to immune/novel therapies after failing salvage chemotherapy. Ancillary analyses of CD30 expression and Nanostring subtyping are pending.

CONCLUSIONS: BV in combination with R-CHP for CD30+ PMBL, DLBCL, and GZL is a highly active and well-tolerated outpatient regimen. Clinical outcomes utilizing BV in frontline for CD30+ B-cell lymphomas are encouraging and warrant further investigation, especially in pts with PMBL.