A Single-Institution Retrospective Cohort Study of Patients Treated with R-EPOCH for Richter's Transformation of Chronic Lymphocytic Leukemia

Background: Richter's transformation (RT), the occurrence of an aggressive lymphoma in patients with prior chronic lymphocytic leukemia (CLL), occurs in up to 10% of CLL patients. Anthracycline-based chemoimmunotherapy remains standard, but outcomes are poor (median survival approximately 12 months). R-EPOCH (rituximab, etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin) has demonstrated superior efficacy to R-CHOP in HIV-associated diffuse large B-cell lymphoma, Burkitt's lymphoma, and primary mediastinal B-cell lymphoma. We conducted a single-institution retrospective cohort study to characterize efficacy and toxicity of R-EPOCH as first-line treatment for RT.

Methods: The study included patients treated with R-EPOCH as first-line therapy for histologically confirmed RT at the Ohio State University between January 1^st 2006 and May 31^st 2014. Characteristics of the CLL and RT, ECOG performance status, and laboratory assessment of bone marrow and organ function were abstracted from medical records. Toxicity of R-EPOCH was assessed by reviewing the frequency of adverse events (AE). R-EPOCH treatment outcomes were assessed as documented by the treating physician and not secondarily verified. Progression free survival (PFS) and overall survival (OS) durations were calculated from date of Richter's biopsy until date of event (PFS: relapse/death; OS: death), censoring patients without event at last follow-up. PFS and OS estimates were obtained using the Kaplan-Meier method. Univariable proportional hazards models were used to model PFS and OS as a function of each clinical variable.

Results: The study included 46 patients. Median age at RT diagnosis was 67 (range 38-83). Median number of months from CLL to RT diagnosis was 53 (range 0.4-198). A median of 3 (range 0-13) prior CLL treatments had been given with 10 (22%) patients receiving ibrutinib as the most recent. The majority of patients where CLL risk was evaluable demonstrated poor-risk disease: 24/43 (56%) complex karyotype, 20/41 (49%) del(17)(p13.1), 27/32 (84%) unmutated IGHV. The majority of evaluable patients had an ECOG performance status of 0 or 1 (18/28, 64%). Table 1 shows RT characteristics for these patients. Treatment with R-EPOCH was started a median of 5 days after RT diagnosis (range 0-110). The majority of patients did not complete 6 cycles: 16 (35%) completed 1 cycle, 10 (22%) 2 cycles, 5 (11%) 3 cycles, 4 (9%) 4 cycles, 2 (4%) 5 cycles, and 9 (19%) 6 cycles. Of 131 cycles given, 114 (87%) were fully evaluable for AE with results in table 2. Outcome of R-EPOCH treatment was known for 44 patients: 9 (20%) achieved complete response, 8 (18%) clinical response documented by the treating physician, 14 (32%) progressive disease, and 13 (30%) died without (known) lymphoma progression. With a median follow up of 39 (range 13-54) months, 9 (20%) patients were alive without lymphoma progression. All patients with lymphoma progression died. Median PFS was 3.5 months (95% CI: 2.0-7.6) and median OS was 5.9 months (95% CI: 3.2-10.3) (Figure 1). In univariable analysis, risk of death was higher for patients with complex CLL karyotype (HR 4.38, p=0.0002), del(17)(p13.1) (HR 3.04, p=0.003), higher number of CLL treatments (HR 1.16, p=0.004), higher bilirubin (HR 1.68, p=0.04), and higher serum creatinine (HR 1.65, p=0.05).

Conclusions: Patients with RT treated with first-line R-EPOCH had poor PFS (median 3.5 months) and OS (median 5.9 months) with only 20% of patients alive at last follow up. Characteristics of underlying CLL influenced outcomes of R-EPOCH with worse PFS and OS in deletion 17p and complex karyotype patients. Better therapies for RT are urgently needed, especially in patients with poor risk CLL.

View largeDownload slide

View largeDownload slide

Close modal