Multicenter Study of Risk-Adapted Therapy With Dose-Adjusted EPOCH-R in Adults With Untreated Burkitt Lymphoma

Abstract

Purpose: Burkitt lymphoma is an aggressive B-cell lymphoma curable with dose-intensive chemotherapy derived from pediatric leukemia regimens. Treatment is acutely toxic with late sequelae. We hypothesized that dose-adjusted etoposide, doxorubicin, cyclophosphamide, vincristine, prednisone, and rituximab (DA-EPOCH-R) may obviate the need for highly dose-intensive chemotherapy in adults with Burkitt lymphoma.

Methods: We conducted a multicenter risk-adapted study of DA-EPOCH-R in untreated adult Burkitt lymphoma. Low-risk patients received three cycles without CNS prophylaxis, and high-risk patients received six cycles with intrathecal CNS prophylaxis or extended intrathecal treatment if leptomeninges were involved. The primary endpoint was event-free survival (EFS), and secondary endpoints were toxicity and predictors of EFS and overall survival (OS).

Results: Between 2010 and 2017, 113 patients were enrolled across 22 centers, and 98 (87%) were high risk. The median age was 49 (range, 18-86) years, and 62% were ≥ 40 years. Bone marrow and/or CSF was involved in 29 (26%) of patients, and 28 (25%) were HIV positive. At a median follow-up of 58.7 months, EFS and OS were 84.5% and 87.0%, respectively, and EFS was 100% and 82.1% in low- and high-risk patients. Therapy was equally effective across age groups, HIV status, and International Prognostic Index risk groups. Involvement of the CSF identified the group at greatest risk for early toxicity-related death or treatment failure. Five treatment-related deaths (4%) occurred during therapy. Febrile neutropenia occurred in 16% of cycles, and tumor lysis syndrome was rare.

Conclusion: Risk-adapted DA-EPOCH-R therapy is effective in adult Burkitt lymphoma regardless of age or HIV status and was well tolerated. Improved therapeutic strategies for adults with CSF involvement are needed (funded by the National Cancer Institute; ClinicalTrials.gov identifier: NCT01092182).

FIG 1.

Treatment was risk stratified based on pretreatment characteristics. Patients were considered low risk if they had all of the following: stage I or II disease, Eastern Cooperative Oncology Group performance status (ECOG PS) of 0 or 1, normal serum lactate dehydrogenase (LDH) levels, and no tumor mass with a diameter ≥ 7 cm. Low-risk patients were treated with two cycles of dose-adjusted etoposide, doxorubicin, cyclophosphamide, vincristine, prednisone, and rituximab on days 1 and 5 (DA-EPOCH-RR), followed by an interim positron emission tomography (PET) scan. If the PET scan was considered negative (neg), these patients received only one additional cycle of DA-EPOCH-RR and no CNS prophylaxis. If the PET scan was considered positive (pos), patients were treated for a full six cycles of therapy and CNS prophylaxis with intrathecal (IT) methotrexate was given. Patients were considered high risk if they had any of the following: stage III or IV disease, ECOG PS of 2-4, elevated serum LDH levels, or any tumor mass ≥ 7 cm. High-risk patients were treated with six cycles of DA-EPOCH-R (rituximab on day 1 only) along with either CNS prophylaxis or active CNS therapy with IT methotrexate, as indicated.

FIG 2.

Patients were enrolled and treated according to baseline risk category. Fifteen low-risk patients were enrolled, and 13 (87%) received only 3 treatment cycles per protocol. One low-risk patient discontinued after one cycle, and another received four cycles. Among 98 high-risk patients, 80 (82%) received the planned six treatment cycles. The reasons for not completing protocol therapy included physician discretion (n = 3), disease progression (n = 4), early toxicity-related death (n = 5), second malignancy (n = 1), and nonmedical reasons (n = 5). Study analysis was based on intention to treat and included all patients. DA-EPOCH-R, dose-adjusted etoposide, doxorubicin, cyclophosphamide, vincristine, prednisone, and rituximab on day 1 only; DA-EPOCH-RR, dose-adjusted etoposide, doxorubicin, cyclophosphamide, vincristine, prednisone, and rituximab on days 1 and 5; MTX, methotrexate.

FIG 3.

Kaplan-Meier estimates of the event-free survival (EFS) and overall survival (OS) of patients enrolled with Burkitt lymphoma. Median follow-up was 58.7 months. (A) The 4-year EFS of all patients (N = 113) was 84.5% (95% CI, 76.3% to 90.1%). (B) The 4-year OS of all patients (N = 113) was 87.0 (95% CI, 79.0% to 92.1%). (C) The 4-year EFS of low-risk patients (n = 15) was 100%. (D) The 4-year EFS of high-risk patients (n = 98) was 82.1% (95% CI, 72.7% to 88.5%).

FIG 4.

Kaplan-Meier estimates of the 4-year event-free survival (EFS) according to prognostic variables. (A) EFS of negative interim positron emission tomography (PET) scans (n = 51) versus positive interim PET scans (n = 34) in high-risk patients was 90.0% (95% CI, 77.7% to 95.7%) and 78.7% (95% CI, 60.5% to 89.2%; P = .12), respectively. (B) EFS of HIV-negative (n = 85) versus HIV-positive (n = 28) patients was 84.5% (95% CI, 74.8% to 90.7%) and 84.9% (95% CI, 64.5% to 94.0%; P = 1.00), respectively. (C) EFS across age group categories of 18-39 (n = 43) versus 40-59 (n = 41) versus ≥ 60 years (n = 29) was 81.1% (95% CI, 65.8% to 90.1%), 87.5% (95% CI, 72.5% to 94.6%), and 85.4% (95% CI, 65.6% to 94.3%; P = .77), respectively. (D) EFS according to high-/high-intermediate–risk (HighInt) score on the International Prognostic Index (IPI) risk score (n = 51) versus low-/low-intermediate–risk (LowInt) IPI risk score (n = 62) was 88.2% (95% CI, 75.5% to 94.5%) and 81.5% (95% CI, 69.1% to 89.3%; P = .29), respectively. (E) EFS for patients with no CSF involvement (n = 102) versus CSF involvement (n = 11) was 89.9% (95% CI, 82.0% to 94.4%) and 45.5% (95% CI, 16.7% to 70.7%; P = .0004), respectively. (F) EFS for patients with no involvement of the CSF or bone marrow (n = 69) versus involvement of either (n = 29) was 92.4% (95% CI, 83 to 97) and 58.6% (95% CI, 39 to 74; P = .0001), respectively.