IVIg Use Associated with Ten-Fold Reduction of Serious Infections in Multiple Myeloma Patients Treated with Anti-BCMA Bispecific Antibodies

Abstract

BCMA-targeted bispecific antibodies (BiAb) are efficacious in relapsed/refractory multiple myeloma; however, serious infections have emerged as important toxicities. In this retrospective study, we characterized all infections and their risk factors, and evaluated the impact of infection prophylaxis in patients treated with BCMA-targeted BiAbs. Among 37 patients, 15 (41%) experienced a grade 3-5 infection, with two infection-related deaths during deep remissions. Most (84%) infections occurred during disease remissions. The cumulative probability of grade 3-5 infection increased over time with no plateau. Among responders (n = 26), profound hypogammaglobulinemia occurred in 100% and continued throughout the entire duration of treatment. During periods when patients were receiving intravenous immunoglobulin (IVIg), the rate of grade 3-5 infections was 90% lower than during observation (incidence rate ratio, 0.10; 95% confidence interval, 0.01-0.80; P = 0.0307). No other risk factors for infection were identified. This study demonstrates that profound hypogammaglobulinemia is universal with BCMA-targeted BiAbs, with intravenous immunoglobulin potentially abrogating most of the infection risk.

Significance: To the best of our knowledge, this is the first study to comprehensively analyze risk factors and mitigation strategies to prevent infections in myeloma patients receiving anti-BCMA bispecific antibodies. Profound and prolonged hypogammaglobulinemia was universal among responders, while immunoglobulin replacement was associated with 90% lower rates of grade 3-5 infections. See related commentary by Garfall and Stadtmauer, p. 427 . This article is featured in Selected Articles from This Issue, p. 419.

Figure 1.

Immunoglobulin levels over time among patients responding to bispecific antibody treatment (A) and nonresponders (B). In the box-and-whiskers plots, the solid line is the median, the dotted line is the mean, the box is the interquartile range (Q1–Q3), the whiskers represent the minimum (Q1–1.5 IQR) and maximum (Q3 + 1.5 IQR), and individual dots are outliers. IgG levels were censored at the time of IVIg use; for IgG myeloma, the M-spike was subtracted from the total IgG. IgA levels excluded IgA myeloma patients. Reference ranges: IgG 700–1,600 mg/dL; IgA 70–400 mg/dL; IgM 40–230 mg/dL.

Figure 2.

Time-to-event cumulative probability of developing any-grade infection (A) and grade 3–5 infection (B) from the start of bispecific antibody therapy, as calculated by the Kaplan–Meier method with shaded 95% confidence intervals.

Figure 3.

Infection rates (per patient-year) at various time periods from the start of bispecific antibody therapy, divided by microbiology and severity.

Figure 4.

A, Swimmer's plot for all grade 3–5 infections during the study period, divided into “on-IVIg” and “off-IVIg” periods. B, Forest plot showing the infection incidence rate ratios of “on-IVIg” versus “off-IVIg” periods for various infection categorizations, with P values derived from z-test on regression parameters in the self-controlled case series model. Sensitivity analyses excluded the first 30 days of bispecific antibody therapy to attempt to limit confounding risk factors for infection including uncontrolled myeloma, increased health care contact due to mandatory hospitalizations during step-up dosing, and cytokine release syndrome management.