Effect of pre-biopsy steroids on diagnostic yield in diffuse large B-cell lymphoma (DLBCL).

Abstract

7072 Background: DLBCL is an aggressive lymphoma, and patients (pts) often require urgent steroid administration for symptom relief or organ compromise prevention. Steroids before biopsy are avoided due to concerns about diagnostic accuracy. This supposition remains underexplored, with limited evidence supporting this practice. Methods: A retrospective chart review of pts with a diagnosis of DLBCL at Brown University Health between 2015 and 2024 was conducted for baseline demographics, steroid administration within 30 days of first biopsy, type and dose of steroids, markers of disease severity, type of biopsy, and biopsy results. Exclusion criteria included relapsed/refractory DLBCL. Statistical analysis was conducted using Chi-square test, T-test, and logistic regression. Results: 365 pts met inclusion criteria, of whom 65 received steroids prior to their first biopsy. Both steroid-treated and steroid-naive pts had similar baseline demographics (age) and markers of disease severity (LDH, “Double HIT” status, IPI score, and Stage), (p > 0.05). Both groups had similar rates of diagnostic first biopsies (p > 0.05). After initial negative biopsies, similar rates of diagnostic repeat biopsies were observed. Neither group had a significant difference in treatment delay from initial negative biopsy to start of chemotherapy (p > 0.05). Logistic regression analysis showed no statistical significance in the relationship between total dose of steroids and the likelihood of a diagnostic biopsy result (p = 0.07). Type of biopsy influenced diagnostic yield: fine needle aspiration (n = 32) was inferior with 28% diagnostic biopsies compared with core needle, excisional or incisional biopsies (n = 284) at 88% (p < 0.001). Conclusions: No significant differences were observed in biopsy success rates or treatment delays between steroid-treated and steroid-naive patients. These results support the safe use of corticosteroids when clinically indicated. Notably, biopsy type, rather than steroid exposure, was the primary determinant of diagnostic success, with fine needle aspiration yielding significantly lower diagnostic rates. These findings reinforce the importance of biopsy selection and support corticosteroid use without compromising diagnostic accuracy. Characteristics Steroids-treated (n=65) Steroid-naive (n=283) P value Age (years), mean (95% CI) 67 (64 - 70) 67 (65 - 69) 0.805 Advanced Stage, proportion (95% CI) 64.15% 71.48% 0.286 LDH (U/L), mean (95% CI) 392 (297 - 487) 364 (315 - 413) 0.637 Percent “Double Hit” 13.85% 9.54% 0.304 Percent first biopsy diagnostic 86.67% 81.79% 0.219 Percent repeat biopsy diagnostic 87.50% 97.92% 0.116 Days from first negative biopsy to treatment, mean (95% CI) 39.5 (12.6 - 66.4) 50.7 (42.9 - 58.5) 0.327 Total steroid dose in prednisone equivalents (mg), mean (95% CI) 261 (193 - 329) 0 N/A Total days on steroids, mean (95% CI) 6.8 (4.5 - 9.1) 0 N/A