Short Diagnosis to Treatment Interval Is Associated with Poor Outcomes in Relapse or Refractory Diffuse Large B-Cell Lymphoma

Background

Short diagnosis to treatment interval (DTI) is associated with aggressive clinical features and poor outcomes in newly diagnosed diffuse large B-cell lymphoma (DLBCL), indicating selection bias in clinical trials (Maurer et al, JCO 2018). However, there is no data on the significance of DTI in relapsed/refractory DLBCL (R/R DLBCL). This study aimed to evaluate the impact of DTI in R/R DLBCL and identify factors influencing DTI.

Method

We retrospectively analyzed 184 patients diagnosed with R/R DLBCL who received second-line therapy at our institutions between January 2015 and March 2022. All patients were treated with R-CHOP like regimen at initial diagnosis. The primary end point was the clinical significance of DTI for overall survival (OS) from second-line therapy. DTI was defined as the time from the date of physician-assessed diagnosis for R/R DLBCL to the initiation of treatment. Patient characteristics were compared by Fisher’s exact test. The optimal cutoff of DTI for OS was calculated using receiver operating characteristic (ROC) analysis. OS was analyzed using the Kaplan-Meier method, and compared using the log-rank test. Factors independently associated with OS were extracted by multivariate analysis using a Cox proportional hazards model.

Results

Of the 184 patients with R/R DLBCL (median age 74 years, range 25-92), 100 (54%) were male, 109 (59%) had elevated lactate dehydrogenase (LDH) levels, 55 (30%) had ECOG performance status (PS) ≥2, 78 (42%) had International Prognostic Index of 3-5 (IPI 3-5) at relapse, and 45 (24%) presented with central nervous system involvement confirmed through MRI or cerebrospinal fluid analysis. Biopsies were performed at relapse in 71 (39%) patients. Forty-seven (26%) patients experienced disease progression during therapy or within 6 months of completion therapy (i.e., primary refractory). The second-line regimens were 39 patients of ICE, 38 of MTX-based regimens, 30 of ESHAP, 17 of CHASE, 15 of CHOP like, and 29 of others, respectively. Additionally, 27 patients underwent autologous stem cell transplantation (ASCT), and 4 received chimeric antigen receptor T-cell therapy. The response to salvage therapy was 71 (39%) patients with CR, 33 (18%) with PR, 10 (5%) with SD, and 70 (38%) with PD.

In a median follow-up of 13.8 months (range 0.6-97.6), the 2-year OS was 44.1% and the median OS was 16.7 months among overall patients. The median DTI was 17 days (range 0-160), with 41 patients having DTI ≤ 7 days, 82 patients ≤ 14 days, 112 patients ≤ 21 days, and 139 patients ≤ 28 days. Across all these DTI cutoffs, short DTI was significantly associated with poor prognosis (p=0.029, 0.022, 0.005, and <0.001, respectively). The optimal cutoff of DTI for OS determined by ROC analysis was 28 days (AUC 0.584). The 2-year OS was significantly poor in patients with DTI ≤ 28 days (DTI≤28) compared with DTI > 28 days (35.2% vs. 69.7%, HR: 2.8, p<0.001). Patients with DTI≤28 were significantly associated with elevated LDH, PS ≥ 2, IPI 3-5 at relapse, and primary refractory. Patients with DTI≤28 tended to have IPI 3-5 at initial diagnosis and CNS relapse more frequently. While the CR rate was lower in patients with DTI≤28 across all salvage regimens (31% vs. 62%, p<0.001), achieving CR in this group was associated with prolonged OS (HR: 0.18).

In univariate analysis, risk factors for short OS were DTI≤28, elevated LDH, PS ≥ 2, advanced stage, IPI 3-5 at relapse, no ASCT, and primary refractory. In multivariate analysis using DTI≤28, IPI 3-5 at relapse, no ASCT, and primary refractory, the following were identified as independent risk factors: DTI≤28 (HR: 1.8, p=0.030), IPI 3-5 at relapse (HR: 3.0, p<0.001), and primary refractory (HR: 2.3, p<0.001).

Conclusions

Short DTI is independently associated with poor outcomes in R/R DLBCL, suggesting potential for selection bias in clinical trials of salvage therapy. Our study demonstrated a significant correlation between DTI and IPI at relapse. As short DTI is related to high total metabolic tumor volume in newly diagnosed DLBCL (Alig et al, JCO 2021), short DTI reflects the aggressiveness of DLBCL and IPI at relapse may be useful to predict the urgency of initiating therapy. In this study, achieving CR mitigated the negative impact of DTI≤28, although the CR rate was low in this group. In conclusion, the significance of DTI on prognosis was confirmed in R/R DLBCL, and new therapeutics may be needed to improve response rates for patients with DTI≤28.