Radiographic and High-Throughput Sequencing (HTS)-Based Response Assessment after Brentuximab Vedotin (BV) Plus Ifosfamide, Carboplatin, and Etoposide (ICE) for Relapsed/Refractory (Rel/Ref) Classical Hodgkin Lymphoma (cHL): Updated Results of a Phase I/II Trial

BACKGROUND: The likelihood of long-term remission following autologous stem cell transplant (ASCT) for rel/ref cHL is predicted by response to pre-ASCT salvage therapy (Blood 2012;119:1665-70). Salvage regimens incorporating the anti-CD30 antibody-drug conjugate BV, administered sequentially and concurrently with chemotherapy, show promise in improving positron emission tomography (PET)-negative complete remission (CR) rates, transplant eligibility, and outcomes (Lancet Oncol 2015;16:284-92; ASH 2015, #293; ASH 2016, #1109; ASH 2016, #1834). While pre-ASCT PET is a strong predictor of outcome, novel HTS-based assays in development imply greater sensitivity than imaging and a potential role in management and surveillance of lymphomas. Here, we report updated clinical results from our ongoing phase I/II trial (#NCT02227199) in the context of HTS-based detection of HL pre- and post-therapy, as well as circulating cytokine and tumor microenvironment markers.

METHODS: Patients (pts) ≥ 18 years old with first relapse or primary refractory CD30+ cHL were eligible for this IRB-approved prospective clinical trial. Treatment included BV on Days 1 and 8 at either 1.2 or 1.5 mg/kg (based on 3+3 dose-escalation schema; capped at 150 mg), ifosfamide and mesna 5 g/m² each on Day 2, carboplatin AUC 5 (capped at 800 mg) on Day 2, and etoposide 100 mg/m² daily on Days 1-3. 2 21-day cycles were given with G-CSF support. BV 1.5 mg/kg was selected as the phase II dose based on reported dose escalation data (ASH 2016, #1834). PET was performed after Cycle 2, with response assigned per Cheson 2007. Peripheral blood (PB) pre- and post-treatment, stem cell (PBSC) product, and (when available) archived formalin-fixed paraffin-embedded tissue (FFPET) from presentation and relapse were collected for correlative studies. DNA for HTS was extracted and analyzed by Adaptive Biotechnologies similar to that described previously (Clin Cancer Res 2014;20:4540-8). Pre-treatment PB cytokine levels were measured by Luminex. Immunohistochemistry (IHC) on FFPET identified components of inflammatory microenvironment.

RESULTS: To date, 24 pts have enrolled and 23 are evaluable (1 withdrew consent after 1 cycle). 15 (65%) pts are female. Median age at enrollment was 30 (range, 20-60), with 7 (30%) age ≥ 45. 9 (39%) had advanced-stage disease at initial diagnosis. 14 (61%) did not obtain CR with first-line therapy, all but 1 of which was ABVD-based (the exception received Stanford V). 6 (26%) received prior radiation.

Of 23 evaluable pts, 20 (87%) achieved a PET CR by investigator review, while 16 (70%) were in PET CR by central independent review. Of the 4 with discrepant results between investigator and central review, 2 were adjudicated as CR by a 3^rd independent reviewer, and 2 were sent for biopsy of residual lesion: 1 biopsy was negative, and 1 noted resolution by imaging guidance so biopsy was aborted. 19 of 22 pts (86%) with adequate follow-up have undergone ASCT. 1 pt in CR after BV-ICE has since relapsed (5%), and no deaths have occurred. The toxicity profile remains similar to that reported previously: 11 (48%) pts experienced Grade 3-4 non-heme toxicity, with sepsis/neutropenic fever (4), hyperglycemia (2), and elevated ALT (2) being the most common; 7 pts (30%) developed neuropathy: 4 Grade 1, 2 Grade 2, and 1 Grade 3; only 1 pt (4%) discontinued study therapy due to toxicity (Grade 3 neuropathy).

21 patients had samples tested for disease-specific clonal sequence by HTS. A clone was detectable in PB pre-treatment in only 1 (5%). Post-treatment, none of the PB or PBSC had a detectable clone. Proportions of CD68+ macrophages, granzyme+ cytotoxic cells, and CD20+ B cells by IHC of FFPET from either initial diagnosis or at relapse/enrollment were similar between those who achieved CR and those who did not, as were circulating levels of MIG, HGF, IL2Ra, IL6, and PDL1.

CONCLUSIONS: BV at a dose of 1.5 mg/kg on Days 1 and 8 per cycle of ICE results in high CR rates, is generally well tolerated, and warrants further investigation in pts with rel/ref cHL. Contrary to previous data (Br J Haematol, 2015;169:689-93), HTS was not able to reliably identify disease-specific clone in PB and thus may not be a useful tool in cHL at present. Inflammatory cell composition and selected serum cytokine levels were not predictive of response to BV-ICE. Enrollment is ongoing, and updated clinical and correlative data will be presented at the meeting.