Phase I results of MEDI2228, an anti-BCMA antibody–drug conjugate, in patients with RRMM

The expression of B-cell maturation antigen (BCMA) occurs at all stages of multiple myeloma (MM); promisingly, BCMA is a tumor-specific antigen that can be targeted repeatedly. MEDI2228 is an antibody–drug conjugate (ADC) that targets the extracellular domain of human BCMA and preferentially binds to membrane-bound BCMA versus soluble BCMA. The ADC is linked to a DNA cross-linking pyrrolobenzodiazepine dimer. It is currently under investigation in relapsed/refractory MM (RRMM), and phase I results were presented by Multiple Myeloma Hub Steering Committee member, Shaji K. Kumar, during the 62nd American Society of Hematology (ASH) Annual Meeting and Exposition.

The first-in-class humanized anti-BCMA ADC, belantamab mafodotin, was approved in 2020 by both the U.S. Food and Drug Administration (FDA) and the European Commission for the treatment of adult patients with RRMM who received ≥ 4 therapies and progressed on the last treatment.

Study design

• A first-in-human, open-label, phase I study (NCT03489525)
• Dose-escalation design (Figure 1)
• Eligibility criteria:
  • Aged ≥ 18 years
  • Confirmed and measurable RRMM
  • European Cooperative Oncology Group (ECOG) performance status ≤ 1
  • Disease progression following treatment with proteasome inhibitors (PIs), immunomodulatory drug (IMIDs), and monoclonal antibodies (mAbs)

• Primary endpoints included safety and tolerability
• Secondary endpoints were preliminary efficacy, pharmacokinetics, and immunogenicity

Patients

Total number of patients included in the study was 82, with a median age of 69 years (range, 40–89). More than half of patients were triple-refractory by the time of enrolment. Baseline characteristics are provided in Table 1.

Table 1. Baseline characteristics in MTD expansion cohort (0.14 mg/kg) and study population¹

Characteristic	0.14 mg/kg cohort (n = 41)	Total (N = 82)
IMID, immunomodulatory imide drug; mAb, monoclonal antibody; MTD, maximum tolerated dose; PI, proteasome inhibitor.
Age, years (range)	69 (40–85)	69 (40–89)
Prior therapies, %
1–4 regimens	48.8	36.6
5–7 regimens	29.2	41.5
≥ 8 regimens	22.0	22.0
IMIDs               Lenalidomide               Pomalidomide               Thalidomide	100 87.8 87.8 19.5	98.8 92.7 82.9 19.5
PIs               Bortezomib               Carfilzomib               Ixazomib	100 95.1 85.4 19.5	100 93.9 78.0 18.3
mAbs               Daratumumab               Isatuximab               Elotuzumab	97.6 90.2 12.2 12.2	97.6 86.6 13.4 11.0
Triple-refractory, %	56.1	57.3

Results

At the data cutoff date, all patients in the maximum tolerated dose (MTD) expansion cohort discontinued treatment. This was mainly due to adverse events (AEs; n = 27), followed by disease progression (n = 10), patient decision (n = 2), investigator decision (n = 1), and death (n = 1). Of note, the most common treatment-related AE (TRAE) leading to discontinuation was photophobia (27%).

Safety

TRAEs were in line with the expected toxicity of the pyrrolobenzodiazepine drug class and mostly Grade 1/2 in severity. Grade 3/4 TRAEs of interest included photophobia (not a class-related AE), thrombocytopenia, pleural effusion, and increased gamma-glutamyltransferase (Table 2).

Table 2. Grade 3/4 TRAEs among different dose levels¹

AE, %	0.025 mg/kg (n = 6)	0.05 mg/kg (n = 9)	0.10 mg/kg (n = 18)	0.14 mg/kg (n = 41)	0.20 mg/kg (n = 5)
AE, adverse event; GGT, gamma-glutamyltransferase; TRAE, treatment-related AE.
Photophobia	0	22.2	11.1	17.1	0
Thrombocytopenia	0	0	11.1	24.4	60.0
Pleural effusion	0	0	11.1	2.4	0
GGT increased	16.7	11.1	0	19.5	0

Ocular toxicity

Oculotoxic events, specifically photophobia, were considered an unexpected event, not treatment-related, based on previous experience. Photophobia of any grade was reported in a total of 36 patients, including the 58.5% of patients enrolled in the 0.14 mg/kg dose level cohort.

Patients experienced increased tearing, blurry vision, and swelling. The median time to onset of these events was about 2 months or 2–3 cycles of treatment. Photophobia improved upon discontinuation in most patients and resolved in four patients; however, some patients were lost to follow-up. The pathophysiology of photophobia remains unknown.

No keratopathy was reported in patients who experienced photophobia in the MTD expansion cohort (0.14 mg/kg).

Efficacy

Responses were observed at all dose levels of MEDI2228. Overall response rate (complete response plus partial response) was highest in the MTD expansion cohort (65.9%; 95% CI, 49.4–79.9) versus other dose levels (≤ 40%). Amongst 23 patients who were triple-refractory to PIs, IMIDs, and mAbs, 17 achieved a partial response or better. The best overall response rates in the MTD expansion cohort are detailed in Figure 2.

Median duration of response was 5.9 months in the MTD expansion cohort; however, it was acknowledged that this could be underestimated due to loss to follow-up. Deep, durable responses were observed even weeks after treatment discontinuation.

Pharmacokinetics

There was a proportional increase in exposure at dose levels ≥ 0.05 mg/kg. Dosing at 3-week intervals (Q3W) did not lead to significant accumulation. The impact of baseline-soluble BCMA levels on pharmacokinetics was minimal, highlighting the specificity of binding to membrane-bound BCMA.

Conclusion

These results suggest that MEDI2228 had a manageable safety profile, with clinical efficacy among several dose levels in a heavily pretreated patient population with RRMM. Thrombocytopenia led to dose limitations in two patients, and 0.14 mg/kg Q3W was selected as the MTD. Photophobia was an unexpected event and led to treatment discontinuations; however, improvements were seen with the discontinuation of MEDI2228. Expansion cohorts with alternative doses and schedule to 0.14 mg/kg Q3W are ongoing with a special interest in ocular toxicity.