Dose-escalation study of ABBV-838, an antibody–drug conjugate, in patients with relapsed and refractory multiple myeloma

Multiple myeloma (MM) is characterized by an uncontrolled proliferation of plasma cells in the bone marrow and is currently incurable. Despite therapeutic advances in MM, there remains a high unmet medical need, particularly in the relapsed and refractory (R/R) setting. Clinical studies using antibody–drug conjugates (ADCs) to target epitopes, which are highly expressed on MM cells, have demonstrated significant antitumor activity in patients with R/R MM.¹ Notably, belantamab mafodotin, a clinically advanced B-cell maturation antigen (BCMA)-targeted ADC, demonstrated anti-myeloma activity and a tolerable safety profile in patients with R/R MM in the DREAMM-2 study. The results were previously covered by the MM Hub —here.  

ABBV-838 is an ADC that is made up of a humanized recombinant immunoglobulin G1κ directed against the CD2 subset 1 epitope (CS1) and is conjugated to a cytotoxic agent. The cell-surface glycoprotein, CS1, is expressed in ~90% of MM bone marrow cells and less in other cells, therefore it is an attractive therapeutic target.¹

Ravi Vij, Washington University School of Medicine in St. Louis, St. Louis, US, and colleagues conducted a first-in-human study (NCT02462525) investigating the safety profile, pharmacokinetics, and preliminary activity of ABBV-838, in patients with R/R MM.¹

Study design¹

• Primary objectives:
  • Assess the safety, tolerability, and pharmacokinetics of ABBV-838
  • Establish the maximum tolerated dose (MTD) of ABBV-838 alone or in combination with other treatments
• Secondary objective:
  • Assess the preliminary activity of ABBV-838
• Exploratory objectives:
  • Assess the pharmacodynamics of ABBV-838
  • Find any prognostic biomarkers associated with efficacy and safety

Dosing¹

• This phase I/Ib, open-label study enrolled a total of 75 patients and was designed in four stages:
  • Dose escalation: 3+3 design (n = 32)
• Dosing started from 0.6 mg/kg and increased up to 6.0 mg/kg
• ABBV-838 was administered intravenously every three weeks (Q3W)
• Three to six patients were enrolled per dose level
  • Dose expansion (n = 29)
    • Patients received 5.0 mg/kg of ABBV-838 Q3W
  • Alternative dosing regimen (weekly [Q1W] and every two weeks [Q2W]; n = 14)
    • Eight patients received 1.5 mg/kg Q1W
    • Six patients received 3.0 mg/kg Q2W
  • Combination therapy cohort
    • This cohort was not enrolled as the study did not meet the defined efficacy endpoints

Results¹

Patient characteristics

Patients were enrolled to the study (Table 1) if they were ≥ 18 years of age, were not eligible for stem cell/bone marrow transplant or had relapsed after autologous or allogeneic stem cell/bone marrow transplant, and had received at least three prior lines of therapies. Patients must have previously received a proteasome inhibitor (PI) and an immunomodulatory drug (IMiD) or be double refractory to a PI and ImiD with disease progression on, or within, 60 days of last therapy.

Table 1. Patient characteristics¹

Safety¹

All (100%) patients discontinued from the study at the data cut-off on April 3, 2018, with the most common reason for discontinuation being progressive disease per International Myeloma Working Group definition (69.3%) and adverse events not related to progression (16.0%).

• Dose-limiting toxicities were observed in three patients:
  • During dose escalation:
    • Febrile neutropenia was experienced by one patient at dose 6.0 mg/kg Q3W
    • Cholestasis and hepatotoxicity were experienced by one patient at dose 1.5 mg/kg Q1W
  • In the expansion cohort:
    • Grade 3 leukopenia was experienced by once patient at dose 5.0 mg/kg Q3W
• Treatment-emergent adverse events (TEAEs) were reported in 74 patients
• The most common TEAEs were neutropenia and anemia (n = 21 in both cases), followed by fatigue (n = 20) and nausea (n = 19)
• Grade 3/4/5 TEAEs were observed in 55 patients across all treatment groups
  • The most common were neutropenia (n = 15), anemia (n = 14), and leukopenia (n = 10)
  • Grade 3/4/5 TEAEs that were considered to have a relationship to ABBV-838 treatment were reported by 30 patients across all treatment groups
• Serious TEAEs were observed in 27 patients across all treatment groups, with progression being the most common (n = 9)
• Three patients experienced a TEAE leading to death, none of which were considered related to ABBV-838 treatment
• The MTD was not reached
• The recommended dose for the expansion cohort was determined to be 5.0mg/kg Q3W

Pharmacokinetics¹

• Fifteen days subsequent to ABBV-838 monotherapy, there was effective engagement of ABBV-838 with the CS1 target, with ≥ 50% CS1 receptor occupancy as the dosing of ABBV-838 increased from 0.6 mg/kg to 6.0 mg/kg. However, there was insufficient data to conclude the dose proportionality.

Efficacy¹

Evaluation of all 75 patients demonstrated a limited and modest efficacy:

• Eight patients achieved an objective response (95% CI, 4.7–19.9)
• Two patients achieved a very good partial response
• Six patients achieved a partial response
• Best overall response of stable disease was achieved by 52 patients
• Fourteen patients had progressive disease
• Overall duration of response: 1–11 months
• Median duration of response: 4 months

Conclusion¹

In conclusion, the results from this phase I dose-escalation study of the ADC ABBV-838 in patients with R/R MM demonstrate that ABBV-838 has an acceptable safety profile and is well tolerated.  However, ABBV-838 failed to show robust antitumor efficacy. A limitation of this study was the small cohort of patients that may skew any interpretation of data.  As the defined endpoints were not reached and the set efficacy criteria were not achieved, enrolment of patients into this study (and into the combination therapy arm) was discontinued.

Currently, the difference in clinical efficacy between belantamab mafodotin and ABBV-838 is yet to be elucidated — the investigators attribute this to an array of possibilities, such as the difference in therapeutic target, exposure of the drug, or the conjugated toxin. The authors explain there is a discrepancy in membrane permeability potential between the two drugs; ABBV-838 is conjugated to a membrane-soluble cytotoxin (monomethyl auristatin E), whereas belantamab mafodotin is conjugated to a membrane-impermeable toxin (monomethyl auristatin F [MMAF]). This difference could potentially allow for greater accumulation of MMAF in MM cells and, therefore, greater exposure of belantamab mafodotin.