Post-Therapeutic Tumor Losses of B-Cell Receptor Components CD20/CD19, a New Challenge in the Management of Patients with Relapsed/Refractory Diffuse Large B-Cell Lymphoma

Introduction

More than 50% of patients (pts) with relapsed or refractory diffuse large B-cell lymphoma (RR-DLBCL) will experience disease escape and to determine optimal therapy for these pts remains a major challenge. Several immunotherapeutic approaches were developed for RR-DLBCL targeting the B-cell lineage surface components CD20 or CD19 with monoclonal antibodies, antibody drug conjugated, T-cell engagers and CD19-chimeric antigen receptor T cells (CAR-T cells). Antigen tumor CD20 and CD19 losses has been reported as a therapeutic resistance mechanism in RR-DLBCL, however the frequency of CD20 and CD10 losses and impact on outcome are not determined. This study aims to investigate incidence, outcome, clinical and molecular characteristics according to CD20 and CD19 tumor cell surface expression in a large cohort of pts with RR-DLBCL.

Methods

The LNH-EP1 study is a prospective, observational cohort study, of adult pts, with histologically confirmed RR-DLBCL, after at least one prior line of systemic treatment, including a combination of anti-CD20 immunotherapy plus anthracycline-based chemotherapy in France. The clinical characteristics of pts were recorded at relapse including previous therapies, histopathological characteristics with cell of origin (COO) and the CD20 and CD19 expression determined by immunochemistry.

The primary outcome was one year overall survival (1y-OS) according to CD20 and CD19 expression and secondary endpoint was 1y-OS according to COO at the time of RR-DLBCL biopsy. A Kaplan–Meier method was used to estimate OS and a log-rank test to compare survival curves. To investigate the clinical significance of loss of CD20 on membrane expression, an individual propensity score was calculated including the potential confounders factors associated with CD20-loss or CD20-weak expression (number of previous lines therapies and main demographics such as age and gender) and a matched cohort (1:3) was built by matching on propensity scores.

Matched (tumor and germline samples) targeted next-generation sequencing (NGS) sequencing was performed with a custom in-house panel of 140 genes. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and in accordance with good clinical practice described in the E6 guidelines of the International Council for Harmonization. All the pts included in the study consented and signed a written informed consent.

Results

Overall, 172 pts (63 female; 109 male) with a median age (range) of 66 (26-87) years were included between 2013 and 2024. At the time of tumor biopsy, median number of previous lines of therapy was 2 (1-9), 33 (19%) pts previously underwent autologous stem cell transplantation and 30 (17%) received CAR-T cells. 88 (51%) pts had GCB and 76 (44%) non-GCB RR-DLBCL. 78% (134) of pts had advanced Ann Arbor stage, 67% (115) had elevated LDH. The median of follow-up period in surviving pts was 43.0 months (range 1.8–100.3). The median overall population OS was 11.0 [95%CI: 8.1–12.6] months.

All 172 pts were examined to ascertain the presence of CD20-positive tumor at the time of DLBCL diagnosis. CD20-loss or CD20-weakness on RR-DLBCL tumor were observed in 30/172 (17%) and 7/172 (4%) pts, respectively. In the matched cohort (n= 160), 1y-OS of pts with CD20-loss was significantly lower compared to pts with CD20-positive biopsy (36% [95%CI= 23-57] versus 49% [95%CI= 40-59]; p <0.0001). Pts with available CD19-staining (n= 37), CD19-negative (22%; n= 8) and CD19-positive (78%; n= 29) did not show different 1y-OS (33% [95%CI= 12-96] versus 39 % [95%CI= 23-66]; p= 0.3).The 1y-OS was similar in pts with GCB and non-GCB RR-DLBCL (45% [95%CI= 36-57] versus 44% [95%CI= 33-57], respectively; p= 0.64). The five most recurrently mutated gene were TP53 (32%), KMT2D (31%), PIM1 (31%), CREBBP (29%) and BCL2 (23%). The comprehensive molecular analysis of the 140 investigated genes according to the surface antigenic positivity of CD20 and CD19 will be detailed at the meeting.

Conclusion

Post-therapeutic CD20 and CD19 tumor losses reached up to 17% and 22% of RR-DLBCL pts, respectively. CD20 loss was associated with a poor prognosis and a larger cohort and longer follow-up period are needed to determine the impact of CD19 loss on pts outcomes. Testing for CD20 and CD19 positivity could prove beneficial in adapting subsequent therapies for pts with RR-DLBCL.