Loncastuximab in High-Risk and Heavily-Pretreated Relapsed/Refractory Diffuse Large B-Cell Lymphoma: A Real World Analysis from 21 US Centers

Background: In the LOTIS-2 study, patients (pts) with relapsed/refractory (R/R) DLBCL treated with loncastuximab-tesirine (lonca), a CD19 directed antibody-drug conjugate, demonstrated an overall response rate (ORR) and complete response rate (CRR) of 48.3 % and 24.1%. However, there is a paucity of data evaluating outcomes with lonca in the real-world setting (RWS). Hence, we performed a multicenter retrospective study to describe pt characteristics and clinical outcomes in R/R DLBCL pts receiving lonca in this setting.

Methods: This retrospective study included pts with R/R DLBCL treated with commercial lonca at 21 academic centers in USA. Clinicopathologic data, treatment outcomes and adverse event (AE) data were collected. Progression-free survival (PFS) and overall survival (OS) were estimated using Kaplan-Meier method and characteristics associated with survival and CR calculated using Cox proportional hazards model and logistic regression. Response was assessed per institutional standards.

Results: 187 pts were analyzed with a median follow-up of 12.5 months (mo). Median age was 68 years (range 22-95), 64% male, and 85% white (Table 1). Most common histology was DLBCL (55% de novo, 22% transformed from low-grade) and 19% were double hit (DH). Thirty-two percent (n=59) had primary refractory disease, 17% (n=31) prior autologous transplant, and 60% (n=112) prior CAR T-cell therapy (CART). Median number of treatment lines before lonca was 4 (1-11) with 81% (n=151) receiving lonca in 4^th line (4L) or later and 8 pts treated off-label in 2^nd line. More pts receiving lonca in ≥4L had prior CART (72% vs 8%, p <0.001). CD19 status was confirmed prior to lonca in 128 pts (68%) with 109 (58%) confirmed CD19+.

The ORR/CRR were 33% and 14%, respectively. Median PFS was 2.1 mo (95%CI=1.8–2.6) and median OS 4.6 mo (95%CI=3.7-5.8). 12-month PFS and OS were 12% and 20%. The CRR in 2L/3L, 4L, and >4L were 15%, 13%, and 15% with ORR of 44%, 26%, and 33%, respectively. Figure 1 shows PFS stratified by response with median PFS not reached in those achieving CR. Patients with a CR to last therapy prior to lonca had superior median PFS (8.8 vs 2.0 mo, p<0.01) and OS (10.8 vs 4.5 mo, p =0.01). Factors associated with achieving CR to lonca included CR to last prior therapy (OR 8.3, p<0.01) and nonGCB subtype (OR 3.9, p=0.02). No pts with bulky disease (>10 cm) had objective response. In multivariable analysis (MVA) including factors in Table 1, elevated LDH (HR 1.8, p=0.02; 95%CI=1.1-2.9) and bulky disease (HR 1.7, p=0.03; 95%CI=1.1-2.6) were associated with inferior PFS while CR to last therapy was associated with superior PFS (HR 0.2, p=0.02; 95%CI=0.05-0.76). Bulky disease (HR 2.3, p<0.01; 95%CI=1.4-3.5), HGBL histology (HR 6.1, p<0.01; 95%CI=2.6-14.5), and elevated LDH (HR 2.0, p<0.01; 95%CI=1.2-3.3) were associated with inferior OS, while CR to last therapy (HR 0.4, p=0.05; 95%CI=0.13-0.98) was associated with superior OS in MVA.

Neither CD19 status nor prior CART were associated with PFS, OS, or CR. In pts receiving lonca after CART, ORR/CRR were 31% and 15%, respectively, and median PFS/OS were 2.0mo (95%CI=1.6-2.7) and 4.6mo (95%CI=3.2-6.1). In pts who received tafasitamab (n=15) and CART (n=7) after lonca, CR was 13% and 29%, respectively.

The most commonly documented AE was cytopenias (45%), followed by peripheral edema, and rash (27% and 27% respectively). Infection was seen in 5% and AEs led to lonca discontinuation in 14%, including peripheral edema (24%) and rash (28%).

Conclusions: In this heavily pre-treated population enriched with prior CART exposure, HGBL, and DHL, ORR and CRR to lonca were lower than previously reported. Nonetheless, we found that pts who achieve a CR to lonca have favorable outcomes and factors associated with achieving a CR include lack of bulky disease, nonGCB subtype, and achieving a CR to most recent therapy prior to lonca. Receipt of prior CART did not negatively impact outcomes to lonca. Furthermore, responses were seen with CD19-directed therapies following lonca failure, suggesting sequencing of CD19 therapy pre and post lonca was successful.