The CAR-Hematotox As a Risk Model to Predict Early Complications and Outcome after Bispecific T-Cell Engager Therapy in Relapsed/Refractory Multiple Myeloma

Introduction:

The bispecific T-cell engagers (BTCE) teclistamab and talquetamab are transforming the treatment landscape of relapsed/refractory multiple myeloma (RRMM). Based on promising efficacy and safety results in the pivotal MajesTEC-1 and MonumenTAL-1 trials, both antibodies were recently approved by FDA and EMA. However, the significant clinical benefits can be compromised by severe complications, including CRS, ICANS, cytopenias and life-threatening infections. To date, there are no established models for suitable risk stratification and identification of vulnerable patients prior to treatment initiation. The CAR-HEMATOTOX (HTX) score was originally developed to predict long-lasting neutropenia following anti-CD19 CAR T-cell therapy, but has since then emerged as a risk model for other immune-related toxicities, infections and survival across cellular immunotherapies. We therefore aimed to evaluate the HTX score to predict toxicities and outcomes in RRMM patients undergoing BTCE therapy.

Methods:

This retrospective observational study included 128 RRMM patients who had received at least one step-up dose of either standard-of-care teclistamab (n=70) or talquetamab (n=58) at two major German myeloma centers until June 2024. The median follow-up time was 6.0 months (95% CI 4.6-7.3). Treatment response was determined by investigators according to IMWG criteria. Near complete remission (nCR) was defined as serological CR without bone marrow status. CRS and ICANS were graded according to ASTCT criteria, cytopenias according to CTCAE v5.0. Infections and their severity were defined as previously described (Rejeski et al., JITC 2022) and assessed only in case of clinical, radiological or microbiological evidence. Severe infections required hospitalization or intravenous anti-infective therapy. The HTX score (Rejeski et al., Blood 2021) includes markers of bone marrow reserve (neutrophil count, hemoglobin, platelet count) and inflammation (CRP, ferritin) and was calculated prior to the first step up dose with a leniceny period of up to 5 days.

Results:

The median age was 65 years (range 35-87), and 52 patients were female (41%). Among evaluable patients, 30% had an ECOG ≥ 2 (n=38/126), 33% had ISS stage 3 (n=37/111), 49% had extramedullary disease (n=51/105) and 45% had high risk cytogenetics (n=48/107), including del(17p), t(4;14) and/or t(14;16). The overall response rate (≥ PR) was 59%, with 27% patients achieving a nCR or better (n=34). CRS affected 87 patients (68%), including two grade 3 cases. Six patients developed ICANS, with one grade 3 event. 32% of the examined patients (n=32/100) had a severe infection.

Sixty-one of 113 evaluable patients were assigned to the HTX ≥ 2 group. These patients showed a significantly longer median duration of hospitalization (12 vs. 10 days; p=0.006) and days on intravenous antibiotics (6 vs. 2 days; p=0.006) during the initial step-up period. Moreover, a HTX ≥ 2 was associated with a significantly increased rate of CRS grade ≥ 2 (36% vs. 12%; OR 4.33; 95% CI 1.59-11.53; p=0.004) and tocilizumab administration (43% vs. 17%; OR 3.55; 95% CI 1.43-8.16; p=0.004). Regarding cytopenias until the end of cycle 1, the HTX ≥ 2 group was characterized by a higher rate of CTC grade ≥ 3 neutropenia (31% vs. 9%; p=0.007), thrombocytopenia (46% vs. 0%; p<0.0001) and anemia (55% vs. 8%; p<0.0001). In addition, these patients had a higher need for G-CSF stimulation (23% vs. 10%; p=0.08), platelet (30% vs. 2%; p<0.0001) and red blood cell transfusions (48% vs. 12%; p<0.0001) during the treatment. In a univariate Cox proportional-hazards model including 84 evaluable patients, we observed an increased risk for the time to severe infection among HTX ≥ 2 patients (90-days 47% vs. 15%; HR 2.29; 95% CI 1.02-5.17; log-rank p=0.04). A HTX ≥ 3 (n=30/117) was associated with an inferior progression-free (6-months PFS 34% vs. 66%; HR 2.39; 95% CI 1.31-4.36; log-rank p=0.004) and overall survival (6-months OS 68% vs. 92%; HR 3.76; 95% CI 1.52-9.27; log-rank p=0.002).

Conclusion:

Our study provides first evidence for the HTX score as a suitable model for risk stratification prior to BTCE therapy. Such scores can potentially influence clinical decision-making regarding inpatient step-up dosing, frequency of clinical follow-ups and use of prophylaxes and supportive measures.