Paper Details

ABSTRACT

Background:  Loncastuximab tesirine-lpyl (Lonca) is an anti-CD19 antibody–drug conjugate with demonstrated clinical activity in relapsed/refractory (R/R) aggressive B-cell non-Hodgkin lymphoma (NHL), including large B-cell lymphoma (LBCL). Across the LOTIS clinical program (NCT03589469, NCT04384484, NCT04970901), Lonca is being evaluated as monotherapy and in combination regimens to improve depth and durability of response (Caimi et al, Haematologica, 2024; Carlo-Stella et al, EHA, 2025; Alderuccio et al, Hematol Oncol, 2025).

Circulating tumour DNA (ctDNA)–based measurable residual disease (MRD) has emerged as a sensitive tool to assess treatment response and identify patients at risk of relapse in LBCL (Roschewski et al, JAMA Onc, 2025; Narkhede et al, Front Onc, 2024).

We conducted a pooled MRD analysis across LOTIS studies to characterize ctDNA response kinetics during Lonca-based therapy and explore its prognostic value and associations between MRD clearance and clinical outcomes.

Methods: Patients treated in LOTIS studies were assessed for ctDNA MRD using the Foresight CLARITY assay. Foresight CLARITY is a tumour-informed, plasma-based, ctDNA MRD assay that tracks patient-specific phased variants to enable ultrasensitive detection of residual lymphoma DNA over time (limit of detection <10?6). The analysis integrated available MRD data from LOTIS-2 (n=31), LOTIS-5 safety run-in (n=8), and LOTIS-7 (n=12) (total N=51). Best overall response (BOR, including 22 patients with complete response [CR], 14 with partial response, and 15 with progressive disease) was used to stratify patients for analyses comparing MRD status across response categories. MRD status was evaluated at baseline, an early treatment landmark (cycle 2 day 1 [C2D1] or cycle 3 day 1 [C3D1], depending on study), and end of treatment (EoT). The prognostic impact of MRD status at each landmark was assessed in relation to overall survival (OS) using time-to-event analyses (Kaplan-Meier method), with study-adjusted Cox proportional hazards models used to account for differences among studies. Exploratory analyses evaluated associations between MRD status, BOR, treatment regimen (Lonca monotherapy vs combination therapy), and baseline clinical characteristics.

Results: MRD dynamics were observable across landmarks, with patients achieving MRD negativity demonstrating improved OS compared with patients who remained MRD positive. The prognostic value of MRD strengthened during treatment, with clear statistically significant OS separation from the early treatment landmark and EoT (early landmark: hazard ratio [HR]=0.16, 95% confidence interval [CI]=0.04-0.56, adjusted P-value=0.0042; EoT: HR=0.05, 95% CI=0.005-0.43, adjusted P-value=0.007).

The CR subset of patients showed the deepest MRD clearance responses throughout time. Among patients achieving CR, 20 out of 22 (~90%) had undetectable ctDNA at EoT and remained in CR at the time of analysis. Two patients with detectable MRD at EoT subsequently relapsed (within 3 months), supporting an association between EoT MRD status and long-term clinical outcome.

Comparable patterns of MRD clearance were observed in CR patients receiving Lonca monotherapy and those receiving combination regimens (rituximab or glofitamab), supporting consistent molecular responses across various Lonca-containing regimens. When correlating MRD data with baseline clinical characteristics, similar patterns were observed in hard-to-treat populations (International Prognostic Index [IPI] ≥3, primary refractory disease, or prior lines of treatment ≥3) vs their respective counterparts, with deep molecular responses observed across risk groups.

Conclusions: In this pooled MRD evaluation in NHL (predominantly LBCL), ctDNA dynamics correlated strongly with clinical response and demonstrated prognostic value throughout treatment. MRD status at EoT identified patients in CR at risk of subsequent relapse, while early treatment MRD (C2D1/C3D1) provided additional prognostic information. MRD clearance patterns were broadly similar between Lonca monotherapy and combination regimens, supporting a central role for Lonca in driving ctDNA MRD clearance. Furthermore, deep molecular responses were observed in clinically high and lower risk patient subgroups. Larger data sets from LOTIS-5 and LOTIS-7 will be analyzed in the future to support these initial findings.