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

Paper Details

Abstracts, authors, affiliations, and disclosures.

Monday, December 14, 2026
10:30 a.m. - 12:00 p.m.
MCCNO - New Orleans Theatre C
Treatment Considerations, Supportive Care, Biological Therapies, Transplantation, Bispecific Antibody Therapy, Monoclonal Antibody Therapy, Non-Biological Therapies, Chemotherapy, Adverse Events, Biological Processes, Pathogenesis, Clinical Practice (Health Services and Quality), Patient Safety, Diseases, Immunology, Immunodeficiency, Infectious Diseases, Viral, Lymphoid Malignancies, Lymphomas, Non-Hodgkin Lymphoma, B Cell Lymphoma, T Cell Lymphoma, Aggressive Lymphoma, Research, Clinical Research, Epidemiology, Real-World Evidence, Registries, Survivorship, Study Population, Human, Adult
PAPER

794 Immortal-time bias and upfront autologous stem-cell transplantation in peripheral T-cell lymphoma: A national landmark analysis

Jingke Yang, MD, PhD, Jingkun Li2, Yan Zhuang3, Iris Luo4, Yiran Wu5, Cindy Cheng6, Isabella Huang7, Ryan Ferguson8, Haibo Cheng9, Jing Luo10, Samantha M Thomas11, Jennifer K Plitcha12, Hui Zhou13
1 Parexel International, Raleigh, North Carolina, United States
2 Harbin Medical University, School of Public Health, Department of Biostatistics, Harbin, China
3 Duke University School of Medicine, Durham, North Carolina, United States
4 University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States
5 University of California San Diego, La Jolla, California, United States
6 Green Hope High School, Cary, North Carolina, United States
7 University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States
8 University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States
9 Virginia Polytechnic Institute and State University, Blacksburg, Virginia, United States
10 Arizona State University, School of Mathematical and Statistical Sciences, Tempe, Arizona, United States
11 Duke Cancer Institute, Duke University School of Medicine, Durham, North Carolina, United States
12 Duke Cancer Institute, Duke University School of Medicine, Durham, North Carolina, United States
13 Central South University, Hunan Cancer Hospital / the Affiliated Cancer Hospital of Xiangya School of Medicine, Changsha, Hunan, China

Background: Upfront autologous stem-cell transplantation (ASCT) is used as a consolidative strategy for selected patients with nodal peripheral T-cell lymphoma (PTCL), but its survival benefit remains uncertain. Observational studies of ASCT are susceptible to immortal-time bias because patients must survive induction therapy and transplant preparation before receiving ASCT. In a national cohort of chemotherapy-treated patients with nodal PTCL, we evaluated temporal trends in upfront ASCT use and examined how the estimated association between upfront ASCT and overall survival changed after applying sequential landmark analyses to reduce immortal-time bias.

Patients and Methods: We conducted a retrospective cohort study using the National Cancer Database. Adults diagnosed from 2004 to 2021 with nodal PTCL, including PTCL not otherwise specified, angioimmunoblastic T-cell lymphoma, and ALK-negative anaplastic large-cell lymphoma, were included if they received chemotherapy as part of first-course treatment and had available survival information. Patients receiving allogeneic or unspecified transplant were excluded. The exposure was registry-coded first-course ASCT. Overall survival was measured from diagnosis in the conventional analysis and from the landmark time in each landmark analysis. Annual ASCT use was summarized by diagnosis year. Multivariable Cox models adjusted for demographic, clinical, histologic, facility-level, diagnosis-year, and area-level socioeconomic factors. Landmark analyses were performed at 4, 5, 6, 7, and 8 months after diagnosis by restricting the risk set to patients alive and under follow-up at each landmark; the 6-month landmark was prespecified as the primary analysis. Restricted mean survival time (RMST) differences and propensity score-matched sensitivity analyses were performed.

Results: Among 8,539 chemotherapy-treated patients with PTCL, 453 (5.3%) received upfront ASCT. ASCT recipients were younger than nonrecipients and were more often treated at academic or research programs. Annual upfront ASCT use was low but increased from 1.9% in 2004 to 7.3% in 2021, with the highest annual use of 8.0% in 2018. In the no-landmark multivariable Cox analysis, upfront ASCT was associated with lower mortality (adjusted HR, 0.48; 95% CI, 0.42-0.55; P<.001). This association progressively attenuated with later landmark times but remained statistically significant. Adjusted HRs ranged from 0.57 at the 4-month landmark to 0.70 at the 8-month landmark. At the prespecified 6-month landmark, the adjusted HR was 0.63 (95% CI, 0.55-0.73; P<.001). At this landmark, RMST differences associated with upfront ASCT were 4.7, 8.0, 11.5, and 15.6 months at 3, 5, 7, and 10 years, respectively. Propensity score-matched analyses showed a similar attenuation pattern. In the 1:1 matched cohort, the HR moved toward the null from 0.56 without landmarking to 0.69 at the 6-month landmark and 0.78 at the 8-month landmark.

Conclusion: In this national cohort of chemotherapy-treated patients with nodal PTCL, upfront ASCT was infrequently used but increased modestly over time. Upfront ASCT was associated with longer overall survival, but the association progressively attenuated with later landmark times, while remaining statistically significant. These findings suggest that conventional diagnosis-anchored analyses may overestimate the survival association of upfront ASCT because of immortal-time bias and patient selection. Residual confounding by induction response and transplant fitness remains an important limitation.

Jingke Yang, MD, PhD: No relevant conflicts of interest to declare.