AUTHORS
Chijioke C Nze, MD, MPH, Mike Hernandez, M.S.2, Amy A. Ayers3, Estela Rojas-Neira4, Reid W Merryman, MD5, Viviana Cortiana6, Andrew Ip, MD, MS7, Brian Hess, MD8, Jordan Ritchie, MD9, Paolo Strati, MD10
AFFILIATIONS
1 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
2 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
3 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
4 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
5 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
6 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
7 John Theurer Cancer Center, Hackensack, New Jersey, United States
8 Hollings Cancer Center of the Medical University of South Carolina, Charleston, South Carolina, United States
9 Hollings Cancer Center of the Medical University of South Carolina, Charleston, South Carolina, United States
10 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
ABSTRACT
Background
Tazemetostat (Taz) is a first-in-class, oral, selective inhibitor of mutant (mut) and wild-type (wt) EZH2. It received accelerated approval by the US Food and Drug Administration (FDA) in June 2020 for adult patients with relapsed/refractory (R/R) follicular lymphoma (FL) who received at least two prior systemic therapies. In the pivotal phase 2 registration trial, including 99 patients (pts), overall response rates (ORRs) were 69% (95% CI 53–82) in the EZH2-mut cohort (n=45) and 35% (95% CI 23–49) in the wt cohort (n=54) and complete response (CR) rates were 13% and 4%, respectively. In the pivotal study, 9% pts required dose decrease and 8% discontinued treatment due to toxicity. In March 2026, Taz was withdrawn from all indications following an FDA safety alert regarding an increased risk for secondary myeloid malignancies. Here, we report the first multi-institutional real-world experience with Taz monotherapy in pts with R/R FL.
Methods
This was a multi-institutional retrospective study of adult pts (18) with pathologically confirmed R/R FL (grades 1–3A) who received tazemetostat monotherapy in routine clinical practice (outside of a trial) at 4 academic cancer centers. Pts were initiated on Taz between 09/2020 - 08/2025, with a data cut-off for follow-up: 06/2026. Data were abstracted from electronic medical records at each site. Response was retrospectively assessed according to Lugano 2014 criteria. Descriptive statistics were used to summarize data.
Results
This study included 30 pts. The median age was 62 years (range, 39-93), 22 (73%) pts were male, 28 (93%) white, 29 (97%) non-Hispanic. Twenty-one (67%) pts had stage III/IV disease, and 23 (77%) had grade 1-2 FL. EZH2 mutation testing was performed in 10 pts (33%), of whom 3 (10% of overall) had EZH2-mut disease. High-risk features were common: 15 (50%) pts met POD24 criteria, 16 (53%) were refractory to last line of therapy, 12 (40%) were refractory to a rituximab-containing regimen, and 8 (27%) were double refractory (no response or relapse within 6 mos of rituximab-containing therapy and alkylator-based chemotherapy).
The median number of prior lines of therapy was 3 (range, 1-12); all pts had received prior rituximab (100%). The most frequently used prior agents included bendamustine (67%, n=20), lenalidomide (57%, n=17), anthracycline-based therapy (33%, n=10), PI3K inhibitor (30%, n=9), and radiotherapy (33%, n=10). One pt had undergone prior stem cell transplantation (allogenic). The median time from the end of the prior therapy to Taz initiation was 8.1 mos (range: 0.25–58.1 mos). All pts initiated Taz due to disease progression; none due to toxicity or patient/provider preference.
Taz dose interruptions and reductions due to toxicity noted in 7 (23%) and 8 (27%) of pts, respectively. Reasons for discontinuation included: disease progression (n=14, 47%), toxicity (n=5, 17%), pt preference (n=5, 17%) and unknown (n=4, 13%). Toxicities leading to discontinuation included: liver toxicity (n=1), fatigue + GI symptoms (n=1), COVID (n=1), bone pain + cytopenia (n=1). Three pts (10%) discontinued Taz due to market withdrawal. Three pts discontinued Taz due to radiographic resolution of disease or resolution of symptoms.
The ORR was 30% (9/30), including CR in 2 pts (7%). After median follow-up of 32.2 mos (range, 7.1 – 67.6 mos), 19 (63.3%) pts had relapsed or progressed and 9 (30%) had died, yielding a median PFS was 11.4 mos (95% CI, 7.7 – 19.1 mos) and the median OS was 52.1 mos (95% CI, 33.5 - NA mos). Causes of death included: Disease progression (n=4, 13%), infection (n=2, 7%) and unknown (n=3, 10%). Two (7%) pts were diagnosed with myeloid malignancies (1 MDS [7 mos], 1 AML [18 mos]) after initiating Taz. Both pts are deceased. The pt with MDS died within 1 month of MDS diagnosis due to progressive cytopenias. The other pt died 15.8 mos after MDS/AML diagnosis from disease progression/lymphoma transformation.
Conclusion
In this multi-institutional real-world cohort, tazemetostat showed similar toxicity and efficacy as what was reported in the pivotal registrational trial. Notably, myeloid malignancies were observed in 7% of pts in this heavily pretreated group, which is higher than what was seen in the pivotal FL trial but similar to what was noted in SYMPHONY-1 trial with Taz+ Lenalidomide + Rituximab. Longer follow-up in larger cohorts is needed to understand the risk of secondary myeloid cancers for Taz and other EZH2 blockers in FL.
DISCLOSURES
Chijioke C Nze, MD, MPH: GRAIL, Inc. - Current equity holder in publicly-traded company; Cardinal Health - Honoraria; Bristol Myers Squibb Foundation - Research Funding; Daiichi Sankyo - Research Funding; Beigene - Research Funding