AUTHORS
Ken Zou, BA, Praveen Ramakrishnan Geethakumari, MD, MS, MBA, FACP2, Elif Yilmaz, MD3, Muhammad Abdullah Humayun, MD4, Farrukh T Awan, M.D.M.S.,M.B.A.5, Heather Wolfe, MD6
AFFILIATIONS
1 University of Texas at Southwestern Medical Center, Dallas, TX, United States
2 University of Texas at Southwestern Medical Center, Dallas, TX, United States
3 University of Texas at Southwestern Medical Center, Dallas, TX, United States
4 University of Texas at Southwestern Medical Center, Dallas, TX, United States
5 University of Texas at Southwestern Medical Center, Dallas, TX, United States
6 University of Texas at Southwestern Medical Center, Dallas, TX, United States
ABSTRACT
Background
Richter transformation (RT) is a rare but severe complication of chronic lymphocytic leukemia (CLL), with a dismal median survival of 6-12 months and no standardized treatment. Much of the foundational RT outcome data (Wang et al. 2019, n=204) comes from the chemoimmunotherapy era, before widespread adoption of novel agents. Despite increasing use of novel and cellular therapies, comparative real-world data on contemporary frontline strategies remain limited. We sought to characterize real-world treatment patterns and outcomes in patients with RT.
Methods
We conducted a single-center retrospective study of patients with CLL and subsequent biopsy-confirmed RT diagnosed between 2006 and 2025 at a tertiary academic center. All histologic subtypes and treatment patterns were collected for descriptive purposes in our RT cohort. Comparative outcome analyses were focused on patients with large B-cell lymphoma (LBCL) histology who received RT-directed treatment; patients who received best supportive care only (BSC) were described secondarily. Frontline treatment patterns (L1) were collected and classified into: chemoimmunotherapy (CIT), CIT + Novel Agent (C+N), or Novel Agent-only (NAO). We compared overall (OS) and progression-free survival (PFS) between these groups, which were estimated by Kaplan-Meier and compared by log-rank test.
Results
Among all patients (n=62) at RT diagnosis, median age was 69 years, median time from CLL diagnosis to RT was 37 months, and 63% received prior CLL treatment. High-risk molecular CLL features were common, including unmutated IGHV (20/31, 65%), TP53 aberration (del(17p) or TP53 mutation) (18/40, 45%), and complex karyotype (22/53, 42%).
Histologic subtype was predominantly LBCL (51/62, 82%) vs Hodgkin lymphoma (HL) (8/62, 13%) vs Other (3/62, 5%). Among LBCL cases, double-expressor (MYC+BCL2 by IHC) phenotype was present in 18/40 (45%), non-GCB cell-of-origin (Hans) in 32/40 (80%), RT-clone TP53 abnormality in 13/23 (57%), and median Ki-67 was 80%.
At median follow-up of 64.9 months (reverse Kaplan-Meier), overall median OS was 17.6 months, and PFS was 8.0 months. Across LBCL-type RT (n=51), median OS was 17.0 months; this group included all BSC patients (n=3; median OS 2.1 months). Among HL patients (n=8), L1 overall response rate (ORR) was 88% (7/8), median PFS and OS were 12.2 months and 75.2 months, with ~55% survival plateau through 5+ years.
Among response-evaluable LBCL patients (n=44), L1 ORR was 73% and differed significantly by frontline arm (p=0.001): C+N (16/17, 94%), CIT (15/21, 68%), NAO (1/6, 17%). For all treated LBCL (n=48), median OS was 19.6 months and did not differ by arm (p=0.226): C+N 26.5, CIT 20.3, NAO 5.9. Median PFS was 7.3 months overall and differed significantly (p=0.047): C+N 10.6, CIT 9.5, NAO 3.7. Among those who received prior CLL therapy (n=29), there were significant differences between the three treatment arms for OS (C+N 26.5, CIT 8.2, NAO 5.9; p=0.006) and PFS (C+N 10.6, NAO 3.7, CIT 2.1; p=0.011). Significant univariable prognostic factors for OS included prior CLL therapy (mOS 14.3 vs 60.1; p=0.002), ECOG ≥2 (7.7 vs 20.7; p=0.012), and CLL-clone TP53 aberrancy (12.0 vs NR; p=0.019).
Novel therapy exposure across any line was common. BTKi (23/48, 48%) and venetoclax (13/48, 27%) were the most frequent and often used frontline (12/23 and 11/13 at L1, respectively). Any-line exposure to CAR-T (9/48, 19%), bispecific antibodies (7/48, 15%), IMiDs (6/48, 13%), ADCs (6/48, 13%), and checkpoint inhibitors (6/48, 13%) occurred predominantly in relapsed/refractory settings.
Conclusions
In this contemporary RT cohort, frontline use of chemoimmunotherapy + novel agent(s) was common. Combination chemoimmunotherapy with novel agent(s) appeared to improve ORR and PFS among LBCL patients. Median OS for LBCL-type RT (17.0 months) exceeded the 12.0-month benchmark from the largest prior CIT-era LBCL-type series, potentially reflecting greater use of novel agents, improved supportive care, and enrichment of our cohort for de novo (previously untreated) RT, which has been associated with improved outcomes. Among those with prior CLL therapy, combination CIT with novel therapies resulted in improved OS and PFS. Prior CLL therapy, poor performance status, and CLL-clone TP53 aberrancy appeared to be adverse prognostic factors. Our findings provide real-world evidence supporting prospective evaluation of CIT plus novel-agent combinations as frontline RT treatment.
DISCLOSURES
Ken Zou, BA: No relevant conflicts of interest to declare.