AUTHORS
Sanam Ghazi, MD, Yifeng Yang, MD2, Truman Koh, BS3, Lisa Punnen4, Arvind Ramakrishnan5, Rohan Jupelly6, Akul Vyas7, Abdul Moeed8, Faraaz Azam, MD9, Patrick Ruz10, Mussamatt Ferdousi11, Ifeanyichukwu Ozobu, MD12, Benjamin Lee13, Farrukh T Awan, M.D.M.S.,M.B.A.14, Daniel Addison, MD15
AFFILIATIONS
1 University of Texas Southwestern Medical Center, Dallas, Texas, United States
2 University of Texas Southwestern Medical Center, Dallas, Texas, United States
3 University of Texas Southwestern Medical Center, Dallas, Texas, United States
4 University of Texas Southwestern Medical Center, Dallas, Texas, United States
5 University of Texas Southwestern Medical Center, Dallas, Texas, United States
6 University of Texas Southwestern Medical Center, Dallas, Texas, United States
7 University of Texas Southwestern Medical Center, Dallas, Texas, United States
8 University of Texas Southwestern Medical Center, Dallas, Texas, United States
9 University of Texas Southwestern Medical Center, Dallas, Texas, United States
10 University of Texas Southwestern Medical Center, Dallas, Texas, United States
11 University of Texas Southwestern Medical Center, Dallas, Texas, United States
12 University of Texas Southwestern Medical Center, Dallas, Texas, United States
13 University of Texas Southwestern Medical Center, Dallas, Texas, United States
14 University of Texas Southwestern Medical Center, Dallas, Texas, United States
15 University of Texas Southwestern Medical Center, Dallas, Texas, United States
ABSTRACT
Background: T-cell-engaging bispecific antibodies have transformed treatment across hematologic malignancies. Yet, the prognostic value of early response depth in routine practice remains unclear.
Methods: In a retrospective cohort from a U.S. comprehensive cancer center, we identified all patients with hematologic malignancies who received CD3-engaging bispecific antibody therapy between January 2015 and October 2025, with follow-up through May 2026. Only the first eligible treatment episode per patient was included. To reduce immortal-time bias, we performed a day-90 landmark analysis restricted to patients who were alive and progression-free at day 90 and had an evaluable disease-appropriate response assessment at approximately 3 months. Response categories were complete response or better (CR+); partial or very good partial response (PR/VGPR), with VGPR applicable to multiple myeloma; and stable or progressive disease (SD/PD). Progression-free survival (PFS) was measured from day 90 to progression or death. PFS was estimated using Kaplan-Meier methods and evaluated using Cox proportional hazards models adjusted for disease group. Disease-stratified, disease-specific, and timing sensitivity analyses were also performed.
Results: Of 197 patients treated with 7 bispecific agents, 137 were alive and progression-free at day 90; of these, 95 (69.3%) had an evaluable approximately 3-month response and formed the landmark cohort. The landmark cohort had a median age of 61 years (IQR, 52–70), included 42 women (44.2%), and had received a median of 3 prior treatment lines (IQR, 1–4). It comprised 35 patients with acute lymphoblastic leukemia (ALL), 34 with multiple myeloma (MM), and 26 with lymphoma.
At approximately 3 months, 61 patients (64.2%) achieved CR+, 26 (27.4%) achieved PR/VGPR, and 8 (8.4%) had SD/PD. Response distribution differed by disease: all 35 patients with ALL achieved CR+, whereas MM and lymphoma included all three response categories.
Over a median follow-up of 21.1 months (95% CI, 16.5–27.1) after the day-90 landmark, 41 PFS events occurred. Event rates were 17/61 (27.9%) for CR+, 16/26 (61.5%) for PR/VGPR, and 8/8 (100%) for SD/PD. The Kaplan-Meier estimated median PFS from day 90 was not reached for CR+, compared with 277 days for PR/VGPR and 38.5 days for SD/PD (global log-rank P<0.001). After adjustment for disease group, PR/VGPR (adjusted hazard ratio [aHR], 3.04; 95% CI, 1.24–7.43; P=0.015) and SD/PD (aHR, 19.46; 95% CI, 6.57–57.61; P<0.001) were associated with greater risk of progression or death compared with CR+.
The association remained evident in responder-only and disease-stratified analyses, while timing sensitivity analyses preserved its direction but were less precise. Among objective responders (n=87; 33 events), PR/VGPR at 3 months remained associated with inferior PFS compared with CR+ (aHR, 3.06; 95% CI, 1.25–7.47; P=0.014). The PR/VGPR versus CR+ association was similar in the disease-stratified model (HR, 3.16; 95% CI, 1.28–7.82; P=0.013), and after excluding ALL (HR, 3.13; 95% CI, 1.28–7.68; P=0.012). In disease-specific analyses, PR/VGPR at 3 months was associated with inferior PFS in MM (HR, 6.88; 95% CI, 1.41–33.44; P=0.017) and showed a directionally similar but nonsignificant association in lymphoma (HR, 1.84; 95% CI, 0.58–5.82). Excluding potentially late assessments preserved the direction of association but reduced precision (aHR, 2.62; 95% CI, 0.98–7.02; P=0.055), with similar findings using a day-120 landmark (aHR, 2.26; 95% CI, 0.82–6.25; P=0.117).
Conclusions: Achieving CR or better at approximately 3 months was associated with substantially longer subsequent PFS after T-cell-engaging bispecific antibody therapy. Early response depth may provide a pragmatic marker for risk stratification across hematologic malignancies.
DISCLOSURES
Sanam Ghazi, MD: No relevant conflicts of interest to declare.