2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Saturday, December 12, 2026
05:30 p.m. - 07:30 p.m.
MCCNO - Hall F
Technology and Procedures, Imaging, Clinical Procedures, Treatment Considerations, Combination Therapy, Drug Development, Biological Therapies, Bispecific Antibody Therapy, (CAR)-T Cell Therapies, Immunotherapy, Monoclonal Antibody Therapy, Non-Biological Therapies, Pharmacology, Adverse Events, Artificial Intelligence (AI), Machine Learning, Biological Processes, Metabolism, Clinical Practice (Health Services and Quality), Patient Safety, Quality of Care, Diseases, Lymphoid Malignancies, Lymphomas, Measurable Residual Disease, Non-Hodgkin Lymphoma, B Cell Lymphoma, T Cell Lymphoma, Aggressive Lymphoma, Research, Fundamental Science, Translational Research, Clinical Research, Clinical Trials, Patient-Reported Outcomes, Real-World Evidence, Study Population, Human, Adult, Elderly, Pediatric, Young Adult
PAPER

1890 Comparative effectiveness of belinostat and romidepsin in relapsed/ refractory nodal PTCL: Advanced statistical and machine learning analysis from the PETAL consortium.

Mark N Sorial, PharmD, Carrie van der Weyden, MD2, Wilson Lam3, Jessy Xinyi Han4, Eric Minwei Liu, PhD5, Nazarii Shokun6, Massimo Federico, MD7, Andre Abdo, MD8, Carlos Sergio Chiattone Sr, MD9, Govind Bhagat, MD10, Valentina Nardi11, Teresa Palomero12, Robert Willim, MD13, Meredith Nichols, MD14, Esther Rheinbay, PhD15, Eric Jacobsen, MD16, Enrica Marchi, MD, PhD17, Stefan K Barta, MD18, Tatyana Feldman, MD19, Christina Poh, MD20, Basem William, MD, MRCP(UK), FACP21, Estelle Verburgh, MD, PhD22, Zainab Mohamed23, Nathan Lee Roberts, MD24, Danielle Blunt, MBBS25, Takashi Sakamoto, MD, PhD26, Yumeng Zhang, MD27, Bradley Haverkos, MD28, Mitsumasa Watanabe, MD29, Nora Bennani, MD30, Yun Kyoung Tiger, MD, PhD31, Jonathan Brammer, MD32, Pamela Allen, MD33, Deepa Jagadeesh, MD, MPH34, Maya deepti Srinivasan, MD, MSc35, John Mark L Sloan Jr, MD36, Lauren Shea, MD37, Harsh Rohit Shah, DO38, Nazila Shafagati, MD39, Sonali Smith, MD, FASCO40, Bhagirathbhai Dholaria, MD, FACP41, Changyu Shen42, Salvia Jain, MD43
1 Dana-Faber Cancer Institute, Boston, United States
2 Peter MacCallum Cancer Centre, Melbourne, Australia, University of Melbourne, Parkville, Australia
3 Harvard School of Public Health, Boston, Massachusetts, United States
4 Broad Institute of Harvard and MIT, Cambridge, Massachusetts, United States, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States
5 Massachusetts Institute of Technology, Cambridge, Massachusetts, United States
6 University of Salento, Lecce, Italy
7 University of Modena and Reggio Emilia, Modena, Italy
8 Hospital Alemão Oswaldo Cruz, São Paulo, Brazil
9 Santa Casa Medicine School, São Paulo, Brazil
10 Saint Louis University School of Medicine, St Louis, Missouri, United States
11 Massachusetts General Hospital Cancer Center, Boston, Massachusetts, United States
12 Columbia University Medical Center, New York, United States
13 Beth Israel Deaconess Medical Center, Boston, Massachusetts, United States
14 Cleveland Clinic, Cleveland, United States
15 Massachusetts General Hospital, Boston, United States
16 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
17 Dana-Farber Cancer Institute, Boston, Massachusetts, United States
18 University of Pennsylvania, Philadelphia, United States
19 Hackensack Meridian Health Network, Hackensack, United States
20 City of Hope, Duarte, California, United States
21 OhioHealth, Columbus, Ohio, United States
22 University of Cape Town, Cape Town, South Africa
23 University of Cape Town, Cape Town, South Africa
24 University of Virginia Cancer Center, Charlottesville, United States
25 Royal Adelaide Hospital, Adelaide, Australia
26 Kyoto University Hospital, Kyoto, Japan
27 Moffitt Cancer Center, Tampa, United States
28 University of Colorado, Denver, United States
29 Amagasaki General Medical Center, Amagasaki, Japan
30 Mayo Clinic, Rochester, United States
31 Mayo Clinic, Rochester, United States
32 Ohio State University Wexner Medical Center, Columbus, United States
33 Emory University, Atlanta, United States
34 Cleveland Clinic, Cleveland, United States
35 Boston Medical Center, Boston, United States
36 Boston Medical Center, Boston, United States
37 University of Alabama at Birmingham Medical Center, Alabama, United States
38 Huntsman Cancer Institute, Utah, United States
39 University of Chicago, Chicago, United States
40 University of Chicago, Chicago, United States
41 Vanderbilt University Medical Center, Nashville, United States
42 GlaxoSmithKline, Collegeville, PA, United States
43 Broad Institute of Harvard and MIT, Cambridge, Massachusetts, United States, Harvard Medical School, Boston, Massachusetts, United States, Massachusetts General Hospital Cancer Center, Boston, Massachusetts, United States

Introduction: Optimal empiric strategies for histone deacetylase inhibitor (HDACi) use in relapsed/refractory (R/R) nodal mature T-cell lymphomas (nMTCL) remain unknown. Comparative analyses using 3 independent methods leveraged multinational datasets to assess romidepsin (romi) vs belinostat (bel) monotherapies (mono) and combinations (combos) to evaluate if differential efficacy exists to facilitate therapeutic decision-making in R/R nMTCL.

Methods: This international cohort study used real-world data from 3 clinical datasets: retrospective PETAL Consortium and GELL, and the prospective T-Cell Project 1.0. Eligible patients had R/R PTCL-NOS, nTFHL (e.g. AITL), or ALCL and received romi- or bel-based regimens (mono or combo) following frontline failure. Primary analyses used inverse probability of treatment weighting (IPTW) to balance confounders (age, pre or post WHO 2022 classification, histology, PIT score 0/1 or 2+, primary refractory disease, relapse <12 m (TTR12), combo vs mono, and therapy line) and compare PFS (HDACi start to progression, next therapy, or death) and OS (HDACi start to death) between romi and bel cohorts. Cox regression estimated pooled HRs with 95% CIs across imputations for missing data. Multivariable Cox adjusting for the same covariables were used to estimate conditional HRs for overall and prespecified clinical subgroups. Given limited benchmark data regarding bel-based combo strategies, an independent machine learning–driven survival synthetic control (SSC) framework was constructed to predict counterfactual survival and enable comparative inference.

Results: The analysis cohort comprised 231 patients across 4 continents, with 188 (81%) treated at US centers. Overall, 188 and 43 patients received romi- and bel-based therapy at any therapy line. Baseline characteristics were balanced between the romi and bel cohorts: male predominance (62% vs 55%) and median age 62 (range 53–71) vs 61 (range 54–70). Less romi patients were primary refractory (43% vs 70%); TTR12 rates were similar (60% romi, 54% bel).

A majority had PTCL-NOS (53% romi, 67% bel), with AITL/TFHL in 44% and 30% and ALCL 4% and 2% respectively. Most common use was 2nd line (65% both groups, range 2-5). Mono was more common than combo (65% romi, 93% bel), with diverse partner agents (commonly duvelisib 17%, pralatrexate 6%, carfilzomib 5%, azacytidine 5%). Subsequent allo-HSCT utilization was low in both cohorts (10% romi, 9% bel). Weighting balanced characteristics effectively, with all standardized mean differences <0.1 except age (0.18), ALCL (0.11), and treatment pre vs post WHO 2022 (0.19).

With primary IPTW analyses, there was no statistically significant difference between romi vs bel in PFS [HR 0.76 (0.37-1.57, p=0.455)] or OS [HR 1.1 (0.52-2.34, p=0.806)]. Results were confirmed with multivariable Cox [PFS HR 0.74 (0.50-1.12, p=0.156); OS HR 0.95 (0.59-1.56, p=0.844)]. Subgroup analyses showed consistent equivalence across histologic subtypes (e.g. TFHL and non-TFHL), lines of therapy, and WHO classification eras. Romi vs bel mono only revealed no significant differences in PFS [HR 0.74 (0.47–1.18, p= 0.204)] or OS [HR 0.77 (0.44–1.32, p= 0.338)]. 

Compared to mono, romi combos improved PFS [HR 0.67 (0.46-0.96, p=0.031)] but not OS [HR 0.73 (0.47-1.11, p=0.140)] and, using SSC, predicted bel combos demonstrated superior outcomes over bel mono for both PFS [restricted mean survival time (RMST) 7.1 vs 3.8m (14.8m horizon), p=0.01] and OS [RMST 10.8 vs 7.8m (15.5m horizon), p=0.038]. Cross-comparison of combo regimens demonstrated no statistical difference between romi and predicted bel for either PFS (p=0.129) or OS (p=0.119). Across the entire cohort, combo therapy prolonged PFS vs mono with multivariable Cox [HR 0.60 (0.39-0.92, p=0.019)] with a non-statistically significant trend toward improved OS [HR 0.68 (0.42-1.09, p=0.109)].

Conclusions: In R/R nMTCL, comparative effectiveness across 3 distinct analytical methods did not demonstrate significant differences for PFS or OS between romi vs bel with therapeutic parity maintained across histologic subtypes, lines of therapy, treatment eras, and mono or combo regimens. Notably, our findings reveal superior PFS with HDACi–based combo strategies vs mono. Subsequent HSCT utilization would have major implications on OS and needs further exploration. These results support clinicians using HDACi according to regional availability and motivate further exploration of HDACi combos.

Carrie van der Weyden, MD: No relevant conflicts of interest to declare.