TOPLINE:

Patients with large B-cell lymphoma (LBCL) who had stable or progressive disease at the end of treatment or relapsed within 3 months had markedly worse response rates and survival outcomes than those who relapsed later, suggesting that these patients represent a distinct high-risk group with primary refractory disease.

METHODOLOGY:

• Shorter time to relapse after first-line therapy predicts worse outcomes and guides second-line treatment choices in patients with relapsed or refractory LBCL. However, definitions of primary refractory disease vary across trials, and trials that have evaluated risk across the full time‑to‑relapse spectrum are limited.
• To propose a standardized definition of primary refractory disease, researchers conducted a secondary analysis of a phase 3 study involving 555 patients (median age, 55 years) with relapsed or refractory LBCL who were eligible for autologous stem-cell transplant. The study compared two salvage chemotherapy regimens — rituximab, dexamethasone, high-dose cytarabine, and cisplatin vs rituximab, gemcitabine, dexamethasone, and cisplatin.
• Patients were stratified into six groups based on the time from the end of frontline chemoimmunotherapy to relapse: those with stable or progressive disease at the end of treatment, those with relapse within 3 months, those with relapse at 3-6 months, at 6-12 months, at 12-24 months, and those with relapse more than 24 months after the end of treatment.
• The study endpoints included the overall response to salvage chemotherapy, transplantation, event-free survival, and overall survival, with a median follow-up duration of 8 years. In this analysis, researchers evaluated associations between time-to-relapse categories and survival outcomes.

TAKEAWAY:

• The overall response rate to salvage chemotherapy was 23%-26% for patients with stable or progressive disease or those who relapsed within 3 months compared with 72% for patients who relapsed after 24 months. Transplantation rates followed a similar pattern (34%-37% in the former group compared with 67%-75% in those who relapsed at or after 6 months).
• Five-year event-free survival was lowest among patients with stable or progressive disease (17%) and those with relapse within 3 months (17%), followed by those with relapse at 12-24 months (39%) and those with relapse beyond 24 months (47%). A similar trend was observed for 5-year overall survival.
• Patients with stable or progressive disease and those with relapse within 3 months had a higher risk for inferior event-free survival than those with relapse after 24 months (adjusted hazard ratio [aHR], 3.5, and aHR, 3.3, respectively; P < .0001 for both). The risk decreased progressively for patients with relapse at 3-6 months or 6-12 months.
• In a subgroup analysis using stable or progressive disease as the comparator, patients who relapsed within 3 months had similar event-free survival outcomes (HR, 0.94; P = .67), whereas relapses at 3-6 months (HR, 0.7; P = .03) and 6-12 months (HR, 0.5; P < .0001) were associated with improved event-free survival. In a cause‑specific analysis, patients meeting the primary refractory disease criteria had a much higher rate of early lymphoma‑related mortality within the first year than those with relapse after 24 months (69% vs 18%).

IN PRACTICE:

The results extend the observations of a previous trial by providing a more “comprehensive evaluation across the full spectrum of [time to relapse] in a prospective trial cohort,” the authors wrote. “[Primary refractory disease], defined here as [stable/progressive disease] at [end of treatment] or relapse within 3 months, should be recognized as a distinct high-risk subgroup with markedly different outcomes from those who relapse later,” they added.

SOURCE:

The study, led by Inna Y. Gong, Princess Margaret Cancer Centre in Toronto, Ontario, Canada, was published online in a research letter in Blood Advances.

LIMITATIONS:

The survival estimates from this analysis may not apply to contemporary LBCL care, which increasingly incorporates immunotherapies. The study did not use PET-based response assessment, which likely underestimated the proportion of patients with residual metabolic disease and may have misclassified some patients. Additionally, only patients with confirmed relapsed or refractory disease were enrolled, which limited the capture of the full spectrum of partial responders and may have introduced selection bias. 

DISCLOSURES:

The LY.12 trial received support from grants provided by the Canadian Cancer Society to the Canadian Cancer Trials Group. Gong received support from the Clinician Investigator Program (Ministry of Health, Ontario), the Eliot Philipson Clinician Scientist Training Program (University of Toronto), Hold’em for Life Fellowship (University of Toronto), and the Canadian Institutes of Health Research. Michael Crump disclosed receiving institutional research funding to Princess Margaret Cancer Centre from Epizyme and Syneos Health and consulting fees from Canada’s Drug Agency and Kite-Gilead. Some authors declared receiving research funding or honoraria from and having other ties with various sources, including AbbVie, Amgen, AstraZeneca, and others. Full disclosures are noted in the original article.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.