Patient characteristics
A total of 177 MM patients who received teclistamab were included in this analysis. The median age at diagnosis was 63.8 years (range: 27.8–82.6 years), with male predominance (60.5%). The majority of patients were non-Hispanic (97.2%) and White (78.5%), and with 17.5% Black. ECOG performance status at teclistamab initiation was 0 in 26.0%, 1 in 48.0%, 2 in 22.6%, and 3 in 3.4% of patients. The most common heavy chain subtype was IgG (55.4%), followed by IgA (23.7%), with kappa light chain predominance (67.2%) (Table 1).
Table 1 Baseline Patient Demographics and Disease Characteristics (N = 177).
Prior CMV history and baseline characteristics
Prior to teclistamab initiation, 35 patients (19.8%) had documented CMV reactivation. Among these, six episodes were attributed to prior stem cell transplantation, two were symptomatic, and three patients received antiviral treatment. The median prior CMV viral load at time of previous reactivation was 222 IU/mL (range: 35–29,051 IU/mL).
The median time from MM diagnosis to first teclistamab dose was 6.15 years (range: 0.59–23.1 years), reflecting heavily pretreated patients. The majority of patients (85.9%) had a baseline testing with undetectable CMV DNA at teclistamab initiation.
During teclistamab treatment, the median number of CMV surveillance tests performed was 4 (range: 0–44). The median lowest absolute lymphocyte count (ALC) was 0.15 × 10³/µL (range: 0.00–11.00), median lowest total IgG was 395 mg/dL (range: 22–6,955), and median lowest functional IgG was 290 mg/dL (range: 8–1419), demonstrating significant immunosuppression in this population.
Viral panel PCR testing showed coinfections in a subset of patients: three had EBV reactivation, two had HHV-6 reactivation, seven had parvovirus B19 reactivation, and one had adenovirus reactivation (Table 2).
Table 2 Baseline CMV History and Immunologic Parameters During Teclistamab Treatment.
CMV reactivation during teclistamab treatment
CMV reactivation occurred in 38 of 173 patients with CMV testing (22.0%) during teclistamab therapy. The median number of CMV tests performed before initial detection was 3 (range: 0–11). The median initial positive CMV viral load was 87.2 IU/mL (range: 37.1–4807.0 IU/mL), and the median highest viral load during the reactivation episode was 288 IU/mL (range: 45.3–20,000 IU/mL).
Despite detectable viremia, most reactivations were clinically silent. Symptoms included fever, malaise, cytopenia, or transaminitis. Four patients (10.5%) developed symptomatic CMV infection, three patients (7.9%) required antiviral treatment, and notably, no patients developed clinically significant CMV end-organ disease. Eight patients (21.1%) experienced a second CMV viremia episode after initial clearance while continuing teclistamab therapy. We found no association between earlier use of tocilizumab and/or step-up steroid dosing and the risk of CMV reactivation.
CMV reactivation occurred predominantly in the early treatment period. Median time to first CMV reactivation: 5 weeks (range 0.14–29.86 weeks). Approximately 90% of patients who developed CMV reactivation did so within 3.3 months of teclistamab initiation, and around 75% experienced reactivation within 2.3 months, suggesting the highest risk period is within the first 2–3 months of therapy [Fig. 1].
Fig. 1: Cumulative Incidence of CMV Viremia.
Cumulative incidence curve showing time from teclistamab initiation to first CMV viremia detection in patients who developed CMV reactivation during treatment. The curve demonstrates that most reactivation events occurred within the first 3 months of therapy.
Among patients with CMV reactivation, the median lowest hemoglobin was 10.3 g/dL (range: 6.0–13.1), median lowest white blood cell count was 4.45 × 10³/µL (range: 0.78–10.44), median lowest absolute neutrophil count was 1.2 × 10³/µL (range: 0.30–4.48), median lowest absolute lymphocyte count was 0.78 × 10³/µL (range: 0.00–5.40), and median lowest platelet count was 128 × 10³/µL (range: 6–328). Eight patients (21.1%) had elevated liver function tests during CMV reactivation, and the median highest LDH was 257 U/L (range: 21–1347). The median lowest IgG level during reactivation was 423 mg/dL (range: 99–6,630), and median functional IgG was 326 mg/dL (range: 60–1322).
Importantly, no patient discontinued teclistamab due to CMV reactivation. More than half of all patients (61.6%) received primary prophylactic IVIG during teclistamab treatment (Table 3).
Table 3 Detailed Characteristics of CMV Reactivation During Teclistamab Treatment (N = 173).
Risk factors for CMV reactivation
Univariable logistic regression analysis identified three variables significantly associated with CMV reactivation: prior CMV reactivation (p < 0.001, OR = 4.56), increasing age at diagnosis (p = 0.005, OR = 1.06), and longer time from diagnosis to first teclistamab dose (p = 0.010, OR = 0.89). Primary prophylactic IVIG use showed a trend toward lower CMV reactivation (p = 0.074). Sex, CMV status at teclistamab initiation, and other viral reactivations (EBV, HHV-6, parvovirus B19, adenovirus) were not associated with CMV reactivation.
In multivariable analysis including prior CMV reactivation, age at diagnosis, and time to first teclistamab dose, only prior CMV reactivation remained statistically significant (p = 0.005, OR = 3.34), indicating that patients with prior CMV reactivation had more than three-fold increased odds of reactivation during teclistamab therapy. Age at diagnosis and time to first teclistamab dose lost statistical significance in the multivariable model (Table 4).
Table 4 Logistic Regression Analysis of Factors Associated with CMV Reactivation During Teclistamab Treatment.
Survival outcomes
The median overall survival from first teclistamab dose was 24.2 months. CMV reactivation did not significantly impact overall survival (p = 0.673, Fig. 2), with similar survival curves observed in patients with and without CMV reactivation during treatment.
Fig. 2: Overall survival stratified by CMV reactivation status.
Kaplan-Meier survival curves comparing overall survival in multiple myeloma patients receiving teclistamab with and without CMV reactivation during treatment. Median overall survival was 24.2 months. CMV reactivation did not significantly impact overall survival (log-rank test, p = 0.673). Number of patients at risk shown below the x-axis at specified time points.
