Outcomes of Burkitt lymphoma (BL) managed in academic (Acad) or community (Comm) centers: real-world evidence (RWE) from 30 US sites.

Abstract

8043 Background: Prior analyses have suggested better overall survival (OS) of cancer patients (pts) treated in Acad rather than Comm hospitals, but these disparities may reflect different patient characteristics. We examined outcomes of pts with BL in a large RWE cohort from 30 US healthcare systems (Evens, ASH 2019) with a mix of Acad and affiliated Comm sites. Methods: We collected clinical data on adults with BL diagnosed in 2009-2018, individually assigned to Acad or Comm principal setting of care. We compared duration of chemotherapy (CTx, incl. standard CODOX-M/IVAC, hCVAD/MA, DA-EPOCH), rates of complete response (CR), progression-free survival (PFS), and OS adjusting for age, sex, HIV, performance status (PS), stage, LDH > 3x upper limit of normal (ULN), involvement of bone marrow or cerebrospinal fluid (CSF), reporting adjusted risk (RR) or hazard ratio (HR) with 95% CI. Results: Among 641 BL pts, 77 (12%) were managed in Comm setting. Comm pts had lower median age (45 vs 48 in Acad, P= .049), less frequent HIV (13% vs 23%, P= .039), less marrow (21% vs 36%, P= .009) or detected CSF involvement (8% vs 15%, P= .11), and less LDH > 3xULN (21% vs 41%, P= .013), with no significant differences in sex, PS, stage, hemoglobin, or receipt of CTx (97% vs 99%). Acad sites more often applied standard intensive CTx regimens (93% vs 85%, P= .03) and rituximab (92% vs 79%, P= .001), without significant difference in median time to CTx ( P= .69) or treatment-related mortality (TRM, P= .16). Pts managed in Comm (vs Acad) sites were less likely to achieve CR (61% vs 75%, P= .03; RR = 0.79 [0.65-0.95]) and had worse 3-year PFS (46% vs 67%, log-rank P= .003; HR = 2.17 [1.51-3.14]) and OS (53% vs 72%, P= .006; HR = 2.20 [1.48-3.25]). There was no significant interaction with age, sex, HIV, PS, or CSF involvement. Excess mortality concentrated in the 1st year of follow-up. CR, PFS, and OS appeared similar between Acad and Comm settings for pts receiving hCVAD or DA-EPOCH, but outcomes were significantly worse in Comm setting for pts receiving CODOX-M/IVAC. Median number of cycles did not differ between Comm or Acad sites, but median duration of CODOX-M/IVAC delivery was significantly longer in Comm setting (113 vs 101 days, P= .023). Conclusions: In this large RWE analysis, superior outcomes of adults with BL in Acad setting were not explained by baseline patient characteristics or TRM. Differences in the use of standard CTx regimens, rituximab, duration of Ctx, and CR rates suggest need for further research on potential barriers to delivery of intensive CTx for BL in a broader Comm setting.