Burden and characteristics of infections in patients treated with bispecific antibodies and the impact of immunoglobulin replacement

Abstract Bispecific antibodies (BsAbs) targeting BCMA or GPRC5D have significantly improved outcomes in relapsed/refractory multiple myeloma (RRMM) but are associated with increased infection risk. Infections occur in up to 60% of patients in clinical trial and real-world settings, driven by profound and prolonged T-cell and B-cell dysfunction, hypogammaglobulinemia, prior therapies, and corticosteroid use. Both common microbial and viral infections as well as opportunistic infections have been reported. Risk-based prophylaxis, including antiviral and PJP prophylaxis, immunoglobulin (Ig) replacement, and close monitoring, are recommended. Vaccination strategies and antimicrobial stewardship also play key roles in infection prevention during BsAb therapy. In this analysis we aimed to describe infection incidence and characteristics as well as the role of Ig replacement in a cohort of BsAb treated patients from a single center in Athens, Greece (Department of Clinical Therapeutics). The analysis included 108 consecutive patients with RRMM treated with bi-specific antibodies (n=90 targeting BCMA and n=18 GPRC5d) starting treatment between 1/2023 and 4/2025. The data on infections were prospectively collected. All patients received prophylaxis for VZV and PJP; CMV monitoring was individualized based on baseline assessment of viral load. Polyclonal IgG in patients with IgG myeloma was calculated by subtracting monoclonal protein from total IgG. Ig replacement strategy changed over time and was adopted routinely later during the study period. The median number of prior lines of therapy was 3 (IQR 2-4). The median time on BsAb was 7.5 months (IQR 3.3-16.5) and the median follow up time was 8 months. At least one infection occurred in 71 (65.7%) patients; the cumulative risk of infection (accounting for death or disease progression as competing events) was 52%, 63% and 71% at 3, 6 and 9 months, respectively. This suggests a higher early risk for infection. The median time to first infection was 81 days (95%CI: 45-117) and 75% of infectious episodes occurred within the first 3 cycles of therapy. The most frequent infection sites were upper (26%) and lower (27%) respiratory tract, GI (7%) and urinary tract (5%); the etiology was identified as microbial in 48% and viral in 18%. Most infections were grade 2 (52% of all patients) but in 13% were grade ≥3 (including grade 5 in 5% of the cohort). In 13 (18%) patients BsAb therapy was discontinued (in 5 due to death) after their first infectious episode. Only one patient was neutropenic at the time of 1st infection but 41/71 (58%) had ALC <1000/mm3 and 38% had increased LDH levels. At first infection, 70% of patients achieved disease remission (25% were in CR/sCR, 24% in VGPR, 21% in PR). BCMA targeting BsAbs were associated with higher probability of infections (70% vs 44%, p=0.037); which occurred earlier than in non-BCMA BsAb (median time to 1st infection 74 vs 169 days). Higher number of prior lines was associated with numerically higher incidence of infections (p=0.070). The median number of infectious episodes per patient was 1 (IQR 0-2), with 15% of patients having 3 or more Grade 2-5 infections. The incidence rate of infections was 12.3 per 100 person-months (12.3 per 100 person-months for BCMA-targeting and 12.1 per 100 person-months for GPRC5d-targeting). 80% of the patients in our cohort received Ig supplementation; 40% started Ig replacement prophylactically (before any infection); 39% of patients with infections were on Ig replacement at the time of 1st infection but 58% had polyclonal IgG < 400 mg/dl at the time of the infection. An infection occurred in 73% of patients not receiving prophylactic Ig supplementation vs 22.5% of patients receiving Ig supplementation (HR:0.14, p<0.001). The use of prophylactic Ig reduced mostly grade 2 (9% vs 43%) (p<0.001) but had less impact on grade 3-5 infections (11% vs 12%). Both upper and lower respiratory tract infections were reduced with prophylactic Ig (12% vs 35% and 16% vs 34%) of both microbial (11% vs 37%) and viral origin (9% vs 23%).The risk of infection is significant among patients with MM treated with BsAbs, especially BCMA targeting. Early prophylactic use of Ig supplementation decreases infection burden in patients treated with BsAbs; however, further optimization of prophylactic strategies is required. Physicians and patients should be alert as serious infections can occur despite prophylactic measures.