Early Assessment of Circulating Tumor (ct)DNA and Analysis of TP53 mutations in Patients with Relapsed/Refractory (R/R) Large B-Cell Lymphoma Treated with Glofitamab Monotherapy

Introduction: Glofitamab is a CD20:CD3 bispecific antibody with a 2:1 configuration that has recently been approved for treating R/R LBCL following two or more prior lines of therapy. The pivotal phase II trial (NCT03075696) using fixed-duration Glofitamab therapy demonstrated high complete response (CR) rates and a manageable safety profile in patients with R/R LBCL. Despite these promising outcomes, a subset of patients fails to achieve CR and subsequently experience disease progression. This highlights the importance of identifying response biomarkers and resistance mechanisms to Glofitamab. To address this issue, we evaluated ctDNA and explored the role of TP53 mutations in patients with R/R LBCL treated with Glofitamab.

Methods: Included in this study were 41 R/R LBCL patients treated at our center in the pivotal trial or a compassionate use protocol. We analyzed ctDNA by CAPP-Seq at baseline (n=41) and cycle 3 (n=39) with paired wild-type germline DNA to filter out polymorphisms and sequencing errors. We used a targeted enrichment strategy focused on a panel of 155 genes (370 Kb) recurrently mutated in B-cell lymphoma. Ultra-deep-next generation sequencing was performed on an Illumina platform. A bioinformatic pipeline was applied to call non-synonymous somatic mutations using the somatic function of VarScan2. The ctDNA load was calculated as haploid genome equivalent per ml (hGE/ml) of plasma; the log 10-fold change (log-FC) of ctDNA was determined as the logarithm of the ratio between the loads at cycle 3 and baseline.

Results: The median age of the patients was 67 years (range 22-86), with 61% being male and 76% having advanced-stage disease. The median number of prior treatments was 3 (range 1-7), with 61% of the patients being primary refractory and 20% receiving CAR-T therapy. The median baseline ctDNA load was 331.6 hGE/ml (range 27-14,804). A higher ctDNA value was significantly associated with bulky disease (p=0.001) and high IPI (p=0.04). A greater log-FC was strongly correlated with metabolic CR, both at cycle 3 (p<.0001) or EOT (p<.0001).

At cycle 3, stratification of patients (n=39) based on ctDNA detection revealed a significantly longer median progression-free survival (PFS) in patients with undetectable (n=23) compared to detectable (n=16) ctDNA (26.4 vs. 2.5 months, respectively; p<.0001). Notably, none of the patients with detectable ctDNA at cycle 3 achieved CR at EOT. Furthermore, all 3 patients with a partial metabolic response and detectable ctDNA progressed before EOT. In contrast, 3 out of 4 patients with undetectable ctDNA achieved CR.

In this cohort, genes mutated in more than 21% of R/R LBCL cases included TP53 (44%), KMT2D, PIM1, and IGLL5 (37% each), CARD11 (27%), HIST1H1E and CREBBP (24% each), and BCL2 (22%). Patients with TP53 mutations (44%; 18/41) and those with wild-type TP53 (56%; 23/41) demonstrated comparable PFS. Interestingly, analysis of TP53 mutation status at cycle 3 was possible in 17 of 18 cases and was highly predictive of outcomes. All 8 patients with persistent TP53 mutations experienced progression, whereas 8 of 9 patients with undetectable TP53 achieved CR (p=0.0004).

Conclusions: Analysis of ctDNA in patients with R/R LBCL receiving Glofitamab allows early identification of non-responsive patients. The presence of detectable ctDNA at cycle 3 is a strong biomarker of progression at EOT, specifically in patients with a partial metabolic response. The persistence of TP53 mutations at cycle 3 predicts disease progression, whereas their absence correlates with CR at EOT. These findings highlight the potential of ctDNA in providing clinically meaningful information with a key role in the therapeutic optimization of Glofitamab-containing treatment strategies.