Increased teclistamab dosing interval improves T-cell diversity and reduces infection risk, while T-cell exhaustion remains minimal irrespective of schedule

Abstract

Abstract Introduction: Bispecific antibodies are among the most effective treatments for relapsed/refractory multiple myeloma (RRMM), inducing rapid and durable remissions. Yet, most agents, including teclistamab, were developed with short dosing intervals and indefinite treatment. Given the depth of responses, the rationale for continuous, dose-dense therapy is limited. At the same time, infectious complications remain a major challenge, sometimes requiring treatment interruptions and leading to fatal outcomes. Interestingly, many patients who discontinue therapy due to toxicity remain in remission. In practice, dose interval extensions (e.g., weekly (q1w) to biweekly (q2w)/monthly (q4w)) have become common even within ongoing trials. However, the immunologic consequences of extended dosing remain poorly understood. While longer dosing intervals are thought to reduce T-cell exhaustion, this remains a hypothesis that has not been formally tested in clinical samples. To address this, we conducted a multiomics study comparing immune effects of q1w, q2w, and q4w teclistamab dosing. Methods: We biobanked peripheral blood mononuclear cells (PBMCs, n=82) from 25 RRMM patients who transitioned from q1w (n=33) to q2w (n=12) and/or q4w (n=30) teclistamab dosing. As a control, samples from 7 patients were collected at baseline. We performed single-cell multimodal sequencing (scRNA/CITE/TCR; n=22 samples, 10 patients, 10X Genomics), 14-color flow cytometry (n=68), and in vitro cytotoxicity assays using MM.1S target cells and patient-derived T cells (n=20) across varying effector-to-target ratios. Results: We first evaluated infection rates across treatment schedules dosing and observed a significantly lower incidence of infections in patients receiving q4w compared to q1w (any-grade infections: 39.2% vs. 69.2%, p=0.01, median follow up 15.7 months). Importantly, progression-free survival was not inferior in the q4w cohort (median 22.7 months). Next, we performed a quantitative analysis using flow cytometry, showing elevated leukocytes counts along with a higher number of T cells in the q4w samples compared to the q1w samples (median 1292 vs 731, p=0.04). ScRNA/CITE-seq analysis of 118743 high-quality T cells showed no significant differences in exhaustion signature (PDCD1, B3GAT1, KLRG1, LAG3, SLAMF6, HAVCR2) across dosing schedules. Only a small subset of pre-terminally exhausted T cells (TCF7+) was detectable from the clustering (0.85% of all T cells), while terminally exhausted (TCF7-) T cells were virtually absent, challenging the relevance of T-cell exhaustion in this setting, a finding that was confirmed by flow cytometry. From in vitro cytotoxicity assays, no significant differences in killing capacity were observed between T cells from patients on q1w, q2w, or q4w dosing schedules. Consistent with partial T-cell recovery during longer treatment-free intervals, we observed increased frequencies of naïve T cells (+40%), central memory T cells (+20%), and CD45RA+ effector memory T cell (TEMRA) subsets (+40%), in the q4w group. To explore this further, scTCR sequencing revealed greater clonotypic T-cell diversity in q4w compared to q1w samples (5349 vs. 2172 clonotypes, p=0.02). Mapping TCR sequences to known viral epitopes demonstrated a restricted antiviral repertoire in q1w samples (Shannon Index (SI): median=1.04; IQR 0.17-1.86), largely confined to a few dominant viral specificities. In contrast, q4w samples showed broader viral TCR diversity (SI: median=2.19; IQR 1.52-2.86), potentially contributing to the lower infection rate observed in this group. We further applied a tumor reactivity score (Friedrich et al., 2022) to assess the frequency of MM-reactive T cells across schedules. As expected, tumor-reactive T cells markedly increased from baseline to q1w dosing (+27%), followed by a modest, non-significant decline with q2w (+26%) and q4w intervals (+22%). Conclusion: Our data support an initial dose-dense treatment phase to expand tumor-reactive T cells, followed by extended dosing intervals in responders to promote immune recovery, restore T-cell diversity, and reduce infections without compromising anti-myeloma activity. Notably, T-cell exhaustion, a potential contributor to infections, was virtually absent in our PBMC samples across all dosing schedules.