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
Access, Representation, Population Health, Technology and Procedures, Imaging, Omics Technologies, Treatment Considerations, Combination Therapy, Supportive Care, Biological Therapies, Antibody Therapy, Transplantation, (CAR)-T Cell Therapies, Checkpoint Inhibitor, Immunotherapy, Monoclonal Antibody Therapy, Non-Biological Therapies, Chemotherapy, Adverse Events, Biological Processes, Microbiome, Clinical Practice (Health Services and Quality), Clinical Guidelines, Health Economics, Patient Safety, Practice Models, Quality of Care, Diseases, Lymphoid Malignancies, Lymphomas, Myeloid Malignancies, Hodgkin Lymphoma, Non-Hodgkin Lymphoma, B Cell Lymphoma, T Cell Lymphoma, Indolent Lymphoma, Aggressive Lymphoma, Research, Clinical Research, Clinical Trials, Epidemiology, Patient-Reported Outcomes, Real-World Evidence, Study Population, Human, Adult
PAPER

1870 A retrospective analysis of disease detection strategies in DLBCL

Neil L. Berinstein, MD, Maria Magda Bahena-Garcia, MD2
1 University of Toronto, Toronto, Ontario, Canada
2 University of Toronto, Toronto, Ontario, Canada
Background: Despite the curative potential of first-line immunochemotherapy, approximately 20%–25% of patients with diffuse large B-cell lymphoma (DLBCL) develop relapsed or refractory (R/R) disease, with most events occurring within the first 2 years. Patients who achieve remission typically undergo structured clinical follow-up intended, in part, to facilitate early relapse detection. Although current guidelines generally favor clinically driven surveillance and limit routine imaging in asymptomatic patients, follow-up recommendations remain heterogeneous. Additionally, how the distribution of relapse detection methods varies according to the timing of disease recurrence, and whether the method of detection is associated with subsequent survival, remains uncharacterized.

Methods: We conducted a single-center retrospective cohort study of 296 adults with primary or transformed DLBCL treated with curative-intent first-line therapy at Sunnybrook Health Sciences Centre from 2010 to 2020. Among patients who developed R/R disease, the method of relapse detection was classified as: 1) structured clinical follow-up, 2) symptom-driven presentation, or 3) imaging-based detection. Imaging-based detection was defined as disease identified on end-of-treatment response assessment or subsequent scheduled surveillance imaging in the absence of a preceding symptom-driven evaluation. Events were categorized as: 1) primary refractory disease (<3 months after completion of first-line therapy), 2) early relapse (3–12 months), or 3) late relapse (>12 months). Median follow-up was estimated using the reverse Kaplan–Meier method, OS using the Kaplan–Meier method, and associations with post-relapse OS using multivariable Cox proportional hazards regression.

Results: The median follow-up for the 296-patient cohort was 66.8 months. During follow-up, 84 patients (28.4%) developed R/R disease, whereas 212 (71.6%) remained relapse-free. Among the 84 R/R events, 30 were primary refractory, 25 were early relapses, and 29 were late relapses. Symptom-driven presentation was the most common method of detection (53.6%), while structured clinical follow-up (23.8%) and imaging-based detection (22.6%) accounted for similar proportions of events. Importantly, the distribution of detection methods varied significantly according to the timing of relapse(P=.011). Primary refractory disease was identified in relatively similar proportions across structured follow-up (30.0%), symptom-driven presentation (33.3%), and imaging-based detection (36.7%). This pattern changed with later relapse, as symptom-driven presentation became increasingly predominant, accounting for 56.0% of early relapses and 72.4% of late relapses. In contrast, the contribution of structured follow-up and imaging-based detection declined, with imaging identifying only one late relapse (3.4%). Relapse timing was also strongly associated with post-relapse OS (global P=.002). In multivariable Cox regression, compared with primary refractory disease, patients with early relapse had a 56% lower adjusted hazard of death (HR, 0.44; 95% CI, 0.24–0.82; P=.010), while those with late relapse had a 70% lower adjusted hazard (HR, 0.30; 95% CI, 0.15–0.60; P<.001). Method of relapse detection was also associated with OS in the overall model (global P=.033); however, neither symptom-driven detection (HR, 1.87; 95% CI, 0.97–3.62; P=.063) nor imaging-based detection (HR, 0.88; 95% CI, 0.40–1.92; P=.742) differed significantly from structured follow-up.

Conclusions: Most R/R DLBCL events were identified following symptom-driven presentation, with the method of detection varying significantly according to timing of recurrence. Primary refractory disease was detected similarly across the three methods, whereas nearly three-quarters of late relapses presented symptomatically. Relapse timing was strongly associated with post-relapse survival, while no significant survival differences were demonstrated among the individual detection methods. These findings suggest that timing of disease recurrence  may be more closely related to prognosis than how it is detected and support further evaluation of risk-adapted, symptom-responsive surveillance strategies based on patient prognostic profile and likelihood of early or late relapse.
Maria Magda Bahena-Garcia, MD: No relevant conflicts of interest to declare.