Background:

Early initiation of chemotherapy is widely believed to improve survival in aggressive lymphomas; however, the optimal time-to-treatment initiation (TTI) remains undefined. Prior studies have offered conflicting findings, and the impact of modest treatment delays across histologic subtypes and prognostic groups is poorly understood. Clarifying the relationship between TTI and outcomes is critical to guide clinical decision-making and optimize care pathways.

Methodology:We conducted a retrospective cohort study using the National Cancer Database. Our cohort included patients diagnosed between 2004 to 2022 with aggressive lymphomas: diffuse large B-cell lymphoma (DLBCL: 9680, 9684, 9679), Burkitt lymphoma (BL: 9687), mantle cell lymphoma (MCL: 9673), and peripheral T-cell lymphoma (PTCL: 9702, 9705, 9714). Patients were included if they received multi-agent chemotherapy as first-line treatment and had complete treatment data.

The primary outcome was 5-year overall survival (OS), analyzed via multivariable Cox proportional models. TTI was categorized as ≤30 days, 31–60 days, and ≥61 days. Covariates included age, sex, race, geographic region, insurance status, Charlson-Deyo comorbidity index (CDCC), stage, extranodal involvement, and International Prognostic Index (IPI). Due to database limitations, IPI was used for all lymphoma subtypes, although MIPI and BL-IPI are more appropriate for MCL and BL, respectively.

Sensitivity analyses were conducted among patients who initiated treatment within 30 days of diagnosis, comparing outcomes between TTI ≤14 days and 15–30 days, stratified by IPI risk groups (low/intermediate [0–2] and high [3–5]) within each lymphoma subtype.

Results:A total of 291,396 patients met inclusion criteria: 240,586 (82.6%) had DLBCL, 10,686 (3.7%) had BL, 19,190 (6.6%) had MCL, and 20,934 (7.2%) had PTCL. The median TTI was shortest for BL (8 days, IQR: 3–16) and longest for MCL (23 days, IQR: 8–41).

In the primary survival analysis, delayed treatment was paradoxically associated with improved OS across multiple lymphoma subtypes. For DLBCL, treatment initiated at 31–60 days (HR: 0.79; p < 0.001) and ≥61 days (HR: 0.77; p < 0.001) was associated with significantly lower mortality compared to treatment within ≤30 days. For BL, treatment at 31–60 days (HR: 0.90; p = 0.159) and ≥61 days (HR: 0.76; p = 0.057) showed no significant difference in OS. In MCL, delayed treatment at 31–60 days (HR: 0.64; p < 0.001) and ≥61 days (HR: 0.55; p < 0.001) was associated with improved survival. Similarly, in PTCL, treatment at 31–60 days (HR: 0.77; p < 0.001) and ≥61 days (HR: 0.72; p < 0.001) was linked to significantly lower mortality.

In the sensitivity analysis limited to patients who initiated treatment within 30 days, starting therapy between 15–30 days (compared to ≤14 days) was consistently associated with improved OS in DLBCL, both overall (HR: 0.81; p < 0.001) and within IPI subgroups—low IPI (HR: 0.84; p < 0.001) and high IPI (HR: 0.80; p < 0.001). Among patients with BL, those treated at 15–30 days had better survival compared to ≤14 days (HR: 0.77; p < 0.001), with a significant benefit in the low IPI group (HR: 0.58; p = 0.046), while no significant difference was observed in the high IPI group (HR: 0.77; p = 0.160). In MCL, treatment at 15–30 days was associated with improved survival overall (HR 0.78; p < 0.001); this was significant in the high IPI group (HR 0.73; p = 0.004), but not in the low IPI group (HR 0.76; p = 0.065). For PTCL, initiating treatment at 15–30 days was associated with modestly lower mortality overall (HR 0.90; p < 0.001), though no significant associations were observed in low (HR 0.92; p = 0.475) or high IPI subgroups (HR 0.91; p = 0.309).

Conclusion

In this large national cohort, delayed treatment initiation in aggressive lymphomas was not associated with worse survival; in fact, modest delays often correlated with improved outcomes across subtypes. These findings remained significant after adjusting for stage, IPI, and comorbidities, and were supported by sensitivity analyses among patients treated within 30 days. Notably, patients treated within 14 days had worse survival, suggesting that rapid treatment initiation may reflect unmeasured markers of aggressive disease biology or clinical instability not captured by conventional prognostic tools. Future studies should integrate biological and functional markers to refine risk stratification and optimize therapy timing.

This content is only available as a PDF.
2025
Sign in via your Institution