A Pilot Study of Weekly Brentuximab Vedotin in Patients with CD30+ Malignancies Resistant to Brentuximab Vedotin every 3 weeks

Increased use of brentuximab vedotin (BV) for the treatment of CD30-bearing lymphomas will lead to more patients with BV-refractory disease. BV is efficacious as a single agent and approved for treatment of relapsed or refractory classical Hodgkin lymphoma (CHL) and systemic anaplastic large cell lymphoma (ALCL) (Chen et al, 2016; Pro et al, 2017), as well as untreated advanced CHL (Connors et al, 2018). Despite promising overall response rates (ORR), most patients suffer progressive disease (PD) following single agent BV (Chen et al, 2016; Pro et al, 2017). The mechanisms of resistance are unknown. CD30 expression appears to persist following relapse after BV (Nathwani et al, 2012). Retreatment in patients with BV-sensitive disease can result in objective responses (Bartlett et al, 2014). In prior clinical data, weekly BV dosing was tolerable with an ORR of 59% and a higher complete remission rate than in a phase I trial of 21-day dosing (Fanale et al, 2012). We hypothesized that modifying BV dose and schedule may result in ongoing efficacy. We report the results of a prospective, single-arm phase II clinical trial of dose-dense BV in patients with BV-resistant CD30-positive lymphoma.

This trial included patients with less than a partial response (PR) to ≥ 2 cycles of standard BV and patients with PD on BV despite prior response. Eligibility criteria included age ≥18 years, absolute neutrophil count ≥1.0 × 10⁹/l, platelet count ≥100 × 10⁹/l, at least near normal renal/hepatic function, measurable disease (>1.5 cm), resolution of all non-haematological BV-related adverse events to < grade 2, Eastern Cooperative Oncology Group performance score ≤2, no transplant within 100 days, no active central nervous system disease, and no BV-related neuropathy > grade 1. A priori, an absolute response rate of ≥ 20% was deemed sufficiently efficacious to warrant further study. The Fred Hutchinson Cancer Research Center/University of Washington Cancer Consortium Institutional Review Board approved this trial and written informed consent was obtained. Multiparameter flow cytometry quantified CD30 expression on the malignant cells (Fromm et al, 2017) and defined Natural Killer (NK) and T cell subsets within the inflammatory infiltrate.

Patients received BV 1.2 mg/kg (up to 120 mg) intravenously on days 1, 8 and 15 every 28 days for up to 4 cycles. Study treatment was discontinued for progression, grade 4 non-haematological toxicity, dose-limiting toxicity (including impaired renal or liver function), delay of > 4 weeks, or if less than 2 of the 3 planned doses were given per cycle. Response was based on the Revised Response Criteria for Malignant Lymphoma (Cheson et al, 2007).

Eight patients were enrolled between March 2013 and May 2016. The median age was 42 (27–68) years. Patients received a median of 6 prior lines of treatment and 63% had undergone autologous transplant, with one undergoing prior allogeneic transplant (Table I). Histologies included CHL (7) and ALK-negative systemic ALCL (1). Patients had received between 2 to 16 (median 5) prior cycles of standard dosed BV. There was a median of 11 (3 to 36) months from the last dose of standard BV to the start of this study. All patients had PD on standard BV, though 3 had a transient PR.

Table I:

Patient Characteristics and Response

Patient	Age (years)	Gender	Histology	Prior Lines of Therapy	History of ASCT	Best response to prior BV	Final response to prior BV	Cycles of weekly BV	Best response to weekly BV
1	27	M	CHL	3	N	PR	PD	2	PD
2	41	M	CHL	7	Y	PR	PD	2	PD
3	34	F	CHL	5	N	PD	PD	2	SD
4	30	M	CHL	4	Y	PD	PD	4	SD
5	56	M	CHL	8	Y (+ prior allo)	SD	PD	3	SD
6	49	F	CHL	7	N	PD	PD	4	SD
7	68	F	CHL	6	Y	PR	SD	3	PR
8	42	M	ALCL	4	Y	PD	PD	2	PD

ALCL: anaplastic large cell lymphoma; allo: allogeneic; ASCT: autologous stem cell transplantation; BV: brentuximab vedotin; CHL: classical Hodgkin lymphoma; F: female; M: male; N: no; PD: progressive disease; PR: partial response; SD: stable disease; Y: yes.

The median number of cycles received was 2.5 (range 2 to 4), with 2 patients completing all 4 planned cycles. Reasons for early discontinuation were peripheral neuropathy (n=1) and PD (n=5). One patient missed multiple doses due to neuropathy despite dose reduction and treatment delay. No other patients required a dose reduction.

There were 2 serious adverse events, which were unrelated to BV (gastric perforation due to PD and hypotension/confusion due to opiates) and one ≥ grade 3 non-haematological toxicity (hypokalaemia). New or worsening neuropathy was seen in 5 (63%) patients: 3 developed Grade 1 neuropathy and 2 developed Grade 2 neuropathy.

Seven of 8 patients were evaluable for response. One patient withdrew before the first restaging examination; this patient subsequently died from complications related to progressive CHL and was scored as a non-responder. Five had reductions in the target lesions. One CHL patient achieved PR after 3 cycles of weekly BV, but discontinued due to grade 2 neuropathy. This yielded an ORR of 13%, triggering the early stopping rule. One patient responded longer on study compared to standard BV (4 versus 2 months). At last follow-up (median 19 months), 2 patients were alive. Both have progressed after additional lines of treatment including anti-programmed cell death 1 therapy.

CD30 expression at the time of enrolment ranged from dim to intermediate by flow cytometry (Fromm et al, 2017) (Table SI). Flow cytometry-quantified CD30 surface density was compared in 3 paired samples pre- and post-standard dosed BV. CD30 density, as well as NK and T cell infiltrates within the involved lymph nodes, did not show a significant change or correlation with response to standard dosed BV or intensified BV in this study.

To our knowledge, this is the first trial of BV in standard BV-refractory disease. Despite the size of this trial, it addresses a number of critical points regarding the current use of BV. First, it is notable that accrual was slow, primarily due the number of patients that were not considered for screening due to persistent peripheral neuropathy from prior BV. This complication may limit the use of BV in subsequent lines of treatment.

Second, weekly dose-dense BV was tolerated with modest adverse side effects. The transient reduction in disease burden raises the idea of using weekly BV as a short-term bridge to other therapies. Two patients transitioned from weekly BV to a previously unavailable therapy. However, this single-agent strategy is unlikely to be effective as most patients progressed rapidly. Given the minimal toxicities, dose-dense BV could be combined with other agents. Ongoing clinical trials are investigating the combination of BV and a PDL-1 (CD274) inhibitor, nivolumab, in the relapsed setting (Herrera et al, 2018) (NCT02927769; NCT02572167). Our data provide a benchmark for the maximum expected single agent contribution of BV in patients with truly BV-refractory disease.

Finally, this study confirms that CD30 persists even in patients with the most BV-resistant disease. Loss or over modulation of the BV target is unlikely to be the mechanism of resistance. The persistence of CD30 in this setting lends hope that it can be harnessed as a target for alternative therapeutic options as we move into the era with increasing numbers of CHL and ALCL patients with relapsed disease demonstrating resistance to BV.

Figure 1:

Kaplan-Meier Curve for overall survival