Cevostamab Monotherapy Safe, Highly Active in Heavily Pretreated Relapsed/Refractory Myeloma

Adam D. Cohen, MD

The first-of-its-kind FcRH5xCD3 bispecific antibody cevostamab (BFCR4350A) demonstrated a manageable safety profile with an overall response rate (ORR) of 53% in heavily pretreated patients with relapsed/refractory multiple myeloma who received active doses, according to data from an early, phase 1 dose-escalation trial (NCT03275103) presented during the 2020 ASH Annual Meeting & Exposition.

Results indicated high clinical activity with the agent in a difficult-to-treat patient population. In cohorts of 34 patients who received target doses of cevostamab at 20 mg or higher (n = 34), 18 patients responded to treatment (53%); 32% had a very good partial response (VFPR) or better rate, and 18% had a complete response (CR) or better.

At the highest dose levels evaluated, which were 90 mg or 132 mg (n = 18), the agent elicited an even higher ORR of 61%, which included a 28% VGPR or better rate, and 11% had a CR or better. Notably, responses were also observed in 7 of 17 patients who were penta-drug refractory, translating to an ORR of 41%; this was also true for 5 of 8 patients who had received prior BCMA-targeted therapy, yielding an ORR of 63% in this population.

Moreover, minimal residual disease (MRD) negativity was observed in the patient who achieved a VGPR or better with the treatment.

“These data establish FcRH5 as a novel target in multiple myeloma and demonstrate the activity of cevostamab monotherapy in this difficult-to-treat, relapsed/refractory patient population,” study author Adam D. Cohen, MD, director of Myeloma Immunotherapy and associate professor of medicine at the Hospital of the University of Pennsylvania, Penn Medicine, said during a virtual presentation of the data.

FcRH5 is a cell surface receptor that is explicitly expressed in the B-cell lineage; the highest expression is noted on terminally differentiated plasma cells, Cohen explained. Because FcRH5 is expressed on myeloma cells with near 100% prevalence, it is a strong target for immunotherapy, he added. The humanized IgG-based T-cell engaging bispecific antibody cevostamab was developed to target both FcRH5 and CD3 on the myeloma cells and the T cells, respectively.

“This leads to an immunologic synapse within the marrow microenvironment, activating the endogenous T cells and leading to myeloma cell death,” Cohen said.

In the ongoing dose-escalation and -expansion phase 1 trial, investigators are examining the safety and efficacy of single-agent cevostamab in patients with relapsed/refractory multiple myeloma.

To be eligible for inclusion, patients needed to have relapsed/refractory myeloma for which either no established treatment is available, appropriate or tolerable. They also needed to have an ECOG performance status of 0 or 1. Notably, previous treatment with CAR T cells, other bispecific antibodies, and antibody-drug conjugates (ADCs) was permitted.

Cevostamab was administered intravenously every 3 weeks for up to 17 treatment cycles until either progressive disease or intolerable toxicity. In this cohort, patients were given a single step-up dose on cycle 1, day 1; this was followed by the full target dose starting on cycle 1, day 8. Patients were then continued at that target dose, to be administered every 3 weeks thereafter starting at cycle 2.

“The step-up dose was intended to mitigate the risk of high-grade cytokine release syndrome [CRS], which is a known toxicity with bispecific antibodies,” Cohen noted.

Participants are hospitalized for the cycle 1, day 1 and day 8 infusions of the agent for up to 72 hours so that they can be monitored for toxicity. Then, they received subsequent infusions as an outpatient, according to Cohen. Patients received premedication with acetaminophen, diphenhydramide, and dexamethasone to mitigate the risk of infusion-related events.

The primary objectives of the trial focused on safety and tolerability, including maximum-tolerated dose (MTD) and dose-limiting toxicities (DLTs), along with establishing the recommended phase 2 dose. Key secondary objectives comprised evaluating the activity of cevostamab, along with pharmacokinetic (PK) and pharmacodynamic biomarkers. The analysis presented during the meeting had an August 18, 2020 clinical cutoff date.

In the single step-up dosing cohort (n = 53), participants had a median age of 62 years and 59% (n = 31) were male. The majority, or 88% (n = 28), had high-risk cytogenetics. Seventeen percent (n = 9) of patients had extramedullary disease at the time of screening. The median time since their first treatment for their disease was 5.7 years (range, 1.2-22.8), and patients had received a median of 6 prior lines of therapy.

All patients had previously received a proteasome inhibitor and an immunomodulatory drug. Eighty-one percent (n = 43) had received prior treatment with a CD38-targeted antibody, while 21% (n = 11) had received a BCMA-targeted therapy in the form of CAR T cells (n = 6/11) or the ADC belantamab mafodotin-blmf (Blenrep; n = 5/11).

The majority of patients, or 72% (n = 28), were triple-class refractory. Moreover, just under half, or 45% (n = 24) were penta-refractory. Ninety-four percent (n = 50) of patients were refractory to their last line of treatment.

Additional data from the trial showed that the median time to first response and best response was 29.5 days (range, 21-105) and 57.5 days (range, 21-272), respectively. To date, responses with the agent were observed regardless of target expression level of FcRH5 in all participants evaluated.

At the time of data cutoff, MRD was evaluable in a total of 7 patients; 6 of these patients with a VGPR or better showed MRD negativity via next-generation sequencing at a sensitivity of less than 10-5, noted Cohen.

A total of 8 patients who were treated at active dose levels of 20 mg or higher had a duration of response (DOR) of longer than 6 months. Several patients had a DOR of over 1 year, added Cohen. Notably, 4 patients continued to experience a response, even after treatment discontinuation. The median follow-up in responders was 10.3 months (range, 2.7-19.5), but this was shorter for those who received treatment at the highest dose levels, according to Cohen.

“If you look at the 132-mg dose cohort, you’ll see that 5 of 7 patients respond [to treatment] and all 5 of these [patients] have ongoing responses at the time of data cutoff,” Cohen said.

In terms of PK, serum concentrations peaked following infusion and declined in a multi-exponential fashion. At the cycle 1, day 1 step-up dose, as well as the cycle 1, day 8 dose, a generally linear increase in exposure with increasing doses was observed. “Overall, the PK behavior is supportive of the every-3-week dosing regimen,” noted Cohen.

Regarding safety, at a median follow-up of 8.1 months, grade 3/4 hematologic toxicities included decreased platelet counts (25%), anemia (19%), neutropenia (15%), and decreased lymphocyte count (15%). The most commonly reported non-hematologic effect was CRS, which was reported in 76% of patients, but most were grade 1 or 2.

Of the 76% of patients who experienced CRS, 34% (n = 18) were grade 1, 40% (n = 21) were grade 2, and 2% (n = 1) were grade 3. CRS effects included pyrexia (74%; n = 39), hypotension (30%; n = 16), tachycardia (26%; n = 14), chills (15%; n = 8), confusional state (13%; n = 7), and hypoxia (9%; n = 5). The median time to onset of CRS ranged from 6 hours to 12 hours. “The vast majority of patients had resolution of CRS within 1 day,” noted Cohen.

Any neurological effect was reported in 28% (n = 15) of patients and all were grade 1 (19%; n = 10) or grade 2 (9%; n = 5). Here, the median time to onset ranged from 12 hours to 24 hours. Neurological events that were reported in 2 or more patients included confusional state (13%; n = 7), headache (8%; n = 4), aphasia (6%; n = 3), and cognitive disorder (4%; n = 2). Notably, all of these events occurred in the setting of CRS and resolved with CRS resolution. Only 25% of patients required tocilizumab (Actemra) and 17% received steroids.

“Infusion-related reactions were seen in roughly one-quarter of patients and other AEs were reported, but almost all of these were grade 1 or 2,” said Cohen.

Five patients (9%) experienced toxicities that resulted in treatment discontinuation, but only 2 of these were determined to be related to treatment: 1 patient had pneumonitis and 1 had meningitis at a center that was currently experiencing a Listeria outbreak.

No treatment-related grade 5 toxicities were observed. Only 1 patient in the 90-mg cohort experienced a DLT of grade 3 pneumonia. The MTD has not yet been reached.

“Dose-escalation and -expansion efforts are ongoing,” concluded Cohen. “In a single step-up dosing cohort, we are also exploring a double step-up dosing cohort to try to further decrease the incidence of CRS.”

Reference

Cohen AD, Harrison SJ, Krishnan A, et al. Initial clinical activity and safety of BFCR4350A, a FcRH5/CD3 T-cell-engaging bispecific antibody, in relapsed/refractory multiple myeloma. Presented at: 2020 ASH Annual Meeting & Exposition; December 5-8, 2020; Virtual. Abstract 292.https://bit.ly/2JwU4Q4.