Background: When initial therapy for cHL fails, the standard approach is multiagent salvage chemotherapy followed by autologous stem cell transplant (ASCT). The success of ASCT can be predicted by achievement of complete remission (CR) by fluorodeoxyglucose positron emission tomography (PET) following salvage chemotherapy (Moskowitz, et al. Blood 2012;119:1665-70). Unfortunately, about half of patients (pts) are unable to achieve CR with first salvage. The anti-CD30 antibody-drug conjugate BV is effective as a single agent in rel/ref cHL (Younes, et al. JCO 2012;30:2183-9). Pre-ASCT salvage chemotherapy combinations with BV have been explored (LaCasce, et al. ASH 2015, #293; Garcia-Sanz, et al. ASH 2015, #582), as has a PET-adapted sequential approach with BV then ICE (Moskowitz, et al. Lancet Oncol 2015;16:284-92). We hypothesized that concurrent therapy with BV and ICE would be safe and produce high CR rates. We also sought to explore inflammatory microenvironment prognostic factors. (ClinicalTrials.gov #NCT02227199)
Methods: Pts ≥ 18 years old with first relapse or primary refractory CD30+ cHL were eligible. Key inclusion criteria were ≥ 1 FDG-avid site of disease ≥ 1 cm; adequate marrow, hepatic, and renal function; and HIV-. Key exclusions were any prior BV; chemotherapy ≤ 3 weeks of enrollment; and pre-existing neuropathy > NCI-CTCAE Grade 1. All pts signed IRB-approved informed consent forms. Treatment included BV on Days 1 and 8, ifosfamide and mesna 5 g/m2 each on Day 2, carboplatin AUC 5 (capped at 800 mg) on Day 2, and etoposide 100 mg/m2daily on Days 1-3. A 3+3 dose-escalation schema for BV was used to determine the maximum tolerated dose (MTD) with ICE: starting dose was 1.2 mg/kg on Days 1 and 8, escalated to 1.5 mg/kg on Days 1 and 8 or de-escalated to 1.2 mg/kg on Day 1 only, based on rate of dose-limiting toxicity (DLT). Once MTD of BV was established, subsequent pts received this dose. 2 21-day cycles were given with G-CSF support. PET was performed after Cycle 2, with response assigned per Cheson 2007. Stem cells were collected after Cycle 2 at discretion of treating investigator. Peripheral blood (PB) pre- and post-treatment, stem cell (PBSC) product, and (when available) archived formalin-fixed paraffin-embedded tissue from presentation and relapse were collected for correlative studies. Pre-treatment PB PD-L1 levels were measured by ELISA.
Results: To date, 16 pts have enrolled and completed study treatment. Median age was 32 (range, 23-60). All received ABVD-based initial chemotherapy, and 5 (31%) received radiation. 11 (69%) did not obtain CR with initial therapy. Grade 3-4 neutropenia, lymphopenia, and anemia was seen in 2 (12%) pts each; thrombocytopenia in 4 (25%) pts. 6 (38%) pts experienced Grade 3-4 non-heme toxicity, but no single toxicity occurred in > 1 pt. Neuropathy was seen in 5 (31%) pts: 4 Grade 1, and 1 Grade 3. 1 DLT (sepsis) was observed in the 9 pts treated in the dose-escalation phase, defining 1.5 mg/kg on Days 1 and 8 as the MTD of BV with ICE.
Overall response rate for all enrolled patients was 94% (n=14), with 88% (n=14) and 69% (n=11) achieving CR by investigator and central independent radiographic review, respectively. 15 (94%) pts underwent PBSC collection, all of which were successful. 12 (75%) pts proceeded to ASCT immediately following this treatment. With a median follow-up of 6.5 (range, 2-20) months, there have been 3 (19%) relapses (2 were in CR after BV-ICE) and no deaths.
Counts of CD68+ macrophages (15%, 14%), granzyme+ cytotoxic cells (6.2%, 8.7%), and CD20+ B cells (24%, 14%) by immunohistochemistry were not significantly different between presentation and relapse (respectively). Levels of PD-L1 were below the limit of detection (500 pg/mL) in all pts.
Conclusions: Pts with rel/ref cHL can safely receive BV up to 1.5mg/kg on Days 1 and 8 per cycle of ICE. Apart from generally-mild neuropathy, adding BV does not appreciably change the toxicity profile of ICE. Observed response rates by PET are very encouraging. Inflammatory cell composition appears similar at presentation and relapse. Without more sensitive testing, circulating levels of PD-L1 are unlikely to be of use in this setting. Enrollment is ongoing, and updated clinical and correlative data, including the use of next-generation sequencing-based detection of lymphoma-specific DNA sequence as a novel method of response assessment, will be presented at the meeting.
Gopal:Seattle Genetics: Research Funding.
Asterisk with author names denotes non-ASH members.