Incidence and characteristics of infections in patients with multiple myeloma treated with BCMA bispecific antibodies in British Columbia

Abstract Background: Bispecific T-cell engagers (bispecifics) targeting BCMA have improved outcomes in relapsed/refractory multiple myeloma (RRMM) but carry substantial infectious risk. Real-world data on infection incidence, microbiology, prophylaxis, and risk factors remain limited. Methods: We retrospectively analyzed 60 RRMM patients treated with compassionate access single-agent anti-BCMA bispecifics (teclistamab 95%, elranatamab 5%) in British Columbia (May 2023-June 2025), as per standard indications. Baseline characteristics, prophylaxis, and infection outcomes were collected. Infections were graded per CTCAE v5. Univariable logistic regression assessed infection risk factors. To address lead-time bias from early death in non-responders, subgroup analyses were performed in responders (≥VGPR) (n=35). Results: Median age was 67 years (range 41–84), 47% were female, and median time from diagnosis was 4.8 years (range 0.7-15.1). Patients had a median of 4 prior lines of therapy (range 3–10); all were triple-class refractory and 43% were also penta-refractory. Pneumocystis prophylaxis with trimethoprim/sulfamethoxazole was used in 95% of patients, prophylactic anti-microbial in the first 3 months in 80% (doxycycline 54%, levofloxacin 44%, moxifloxacin 2%); valacyclovir in 100%; G-CSF in 48%. In total, 58% received intravenous immunoglobulin (IVIG), 35% of which initiated prior to first infection, either prior to bispecific (15%) or as primary prophylaxis (20%). Median time to start IVIG primary prophylaxis was 3 months (range: 0.1-10.6). During a median follow-up of 11.5 months (range 2.5–26.7), 45% experienced at least one infection. Median time to first infection was 3.3 months (range 0.1-16.2) and 48% of first infections occurred within 100 days. There was a total of 59 infectious events, including upper respiratory tract infections (22%), pneumonia (13%), respiratory syncytial virus (10%), COVID-19 (7%), cellulitis (7%), cytomegalovirus (3%), sinusitis (3%), influenza (3%), dental (3%), febrile neutropenia without a primary source of infection (3%), peritonitis (1.7%), pneumocystis jirovecii (1.7%, in a patient without prophylaxis), and urinary tract infection (1.7%). There were no cases of herpes virus infections, including shingles. Grade ≥3 infections occurred in 17%. Therapy interruption due to infection occurred in 28% with a median next cycle delay of 2 weeks (IQR 1-12). There were 2 infection-related deaths, both in the responder group: one at 1.7 month from complications of COVID-19, including bacterial pneumonia, pulmonary embolism and kidney failure, and one at 5 months from sepsis secondary to febrile neutropenia of unknown origin. Among patients on antibiotic prophylaxis (n=48), 12.5% developed bacterial infections versus 25% without prophylaxis (n=12) (p=0.3). In responders, median time on treatment was 7.9 months (range 2.3-26.1). At least one infection occurred in 66% of responders, which increased to 70% in patients remaining on treatment beyond 100 days (n=30). First infection occurred within the first 100 days in 35%, while most events (81%) were observed thereafter. Initiation of IVIG (before bispecific or as primary prophylaxis) was associated with reduced infection risk (OR 0.11, 95% CI 0.02–0.51, p=0.004). IgG levels below 4g/L were associated with increased infections (OR 5.25, 95% CI 1.22–26.5, p=0.03). There was a non-statistically significant benefit from prophylactic antibiotics (OR 0.48, 95% CI 0.06–2.57, p=0.41). Other variables such as utilization of tocilizumab and corticosteroids for the treatment of cytokine release syndrome or immune effector cell-associated neurotoxicity syndrome, age, sex, line of therapy and white blood cell count at time of therapy initiation were not associated with infection risk. Conclusion: Infections were frequent in RRMM patients receiving anti-BCMA bispecific therapy, particularly respiratory infections. Early prophylactic antimicrobials may reduce bacterial infections but rates remain high beyond 100 days when these are stopped. Early immunoglobulin replacement may further mitigate infection risk across all potential pathogen classes. Valacyclovir effectively prevented herpes virus infections, including shingles. These findings highlight the need for optimized prophylaxis and vigilant monitoring with a low threshold for treatment in this population at high risk for infection.