Burkitt lymphoma (BL) is a highly aggressive B-cell non-Hodgkin’s lymphoma, which often involves extranodal sites or is represented as leukemia (1). It is rare and characterized by rapid tumor cell proliferation, which one deemed related to EB virus infection and chromosomal translocation involving the c-myc oncogene (2, 3).

Current national guidelines recommend short-term, intensive, and multi-drug regimens as the first-line treatment option for adult BL patients (4), including CODOX-M/IVAC, DA-EPOCH, and Hyper-CVAD. GMALL-B-ALL/NHL2002 is also commonly used for the treatment of BL patients and has achieved a promising outcome in already published reports (5–7). Clinical trials and systematic reviews have demonstrated that the anti-CD20 monoclonal antibody rituximab in addition to chemotherapies further provides efficacy benefits in patients with BL in the advent of the immunotherapy era (8, 9). Rituximab has become a component of standard chemotherapy in the treatment of BL for years, and no comparative studies have been conducted to evaluate the efficacy of these first-line regimens so far. Here, we conducted this meta-analysis to identify the better regimen by accessing overall survival (OS), progression-free survival (PFS), and overall response rate (ORR) data in the literature up to December 2021.

BL is an aggressive and chemotherapy-sensitive B-cell non-Hodgkin’s lymphoma categorized into endemic, sporadic, and immunodeficiency-associated subtypes (29). The majority of BL patients obtained long-term survival after some intensive chemotherapy, and the prognosis has further improved owing to the advent of rituximab. Research showed that rituximab, the anti-CD20 monoclonal antibody, can not only prolong the time of disease progression but also extend OS for a variety of B-cell lymphomas, including BL (30). It is still uncertain which is the optimal regimen containing rituximab for BL patients, and we aim to answer this question through a meta-analysis. Previous studies found heterogeneity between pediatric and adult BL patients; thus, this meta-analysis focused on patients no less than 14 years of age (28). A total of 17 studies and 1,258 patients were finally identified through data search and literature screening and divided into four treatment groups as follows: DA-EPOCH, CODOX-M/IVAC, Hyper-CVAD, and GMALL-B-ALL/NHL2002. All included patients were sporadic or immunodeficiency-associated types, probably because endemic BL is prevalent in children of sub-Saharan Africa (29).

CODOX-M/IVAC, developed by Magrath, is a highly effective alternate and the most commonly used regimen as our result showed that patients in almost half of the included studies were treated with this chemotherapy program (31). B-NHL2002, a short intensive chemotherapy program based on a pediatric Berlin-Frankfurt-Münster protocol, was updated and improved several times by the German Multicenter Study Group for Adult ALL (GMALL). Three current reported large prospective trials showed substantial cure rates in adult BL/leukemia whether HIV-negative or HIV-positive (5–7). Hyper-CVAD (hyper-fractionated cyclophosphamide, doxorubicin, vincristine, and dexamethasone alternating with methotrexate plus cytarabine) was also a comparable treatment option among established dose-intensive regimens for BL (18, 23, 26). A retrospective study by Samra et al. reported that the Hyper-CVAD protocol showed highly promising efficacy and safety in high-risk patients with CNS or BM involvement (18). The DA-EPOCH (Risk-adapted etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin) protocol has been widely used in lymphoma and is well tolerated by BL patients of all ages and those with HIV infection (20). Based on the results of our meta-analysis, the DA-EPOCH regimen group might be considered a better treatment option for adult BL patients by comparing the 2-year OS rate among the four regimen groups. No significant difference was observed in the 2-year PFS rate and ORR among these four groups. Interestingly, one RCT performed by Chamuleau reported a similar estimated 2-year OS among R-CODOX-M/R-IVAC and DA-EPOCH-R for high-risk BL; this trial was suspended because of the slow accrual rate. The limited sample size and inclusion of only high-risk populations in this trial may have contributed to the discrepant results between this RCT and our analysis (32).

LDH levels, stage III–IV, CNS, and BM involvement have been widely used for risk stratification and considered as independent influential factors in the poor prognosis of BL. Our result displayed various sources of heterogeneity, including LDH levels, stage III–IV, CNS, and BM involvement, consistent with the previously reported results in strong association with OS rate, further confirming the reliability of our meta-analysis (33–35). Previous studies showed that BL is highly aggressive, especially when with extensive involvement. Patients with BM involvement using the Ann Arbor staging system were diagnosed at stage IV, and those with CNS involvement had no detailed description. The St Jude/Murphy system, which was much more commonly used for staging, suggested that BL patients with BM and/or CNS involvement were divided into stage IV and the high-risk group. Anyhow, the involvement of CNS and/or BM directly related to the advanced stage was a significant factor associated with inferior prognosis. As shown in our results, because of the different proportions of CNS and/or BM involvement in the included studies and the identification of the source of heterogeneity between patients with and without CNS and BM involvement by meta-regression, the conclusion that “the DA-EPOCH regimen had a higher OS rate compared with three other treatment groups” seems to be inapplicable to patients with BM or CNS involvement. Furthermore, studies have shown no difference in clinical and biochemical characteristics and treatment outcomes between HIV-positive and HIV-negative cohorts after rituximab-based chemotherapy (35–38). We agreed that HIV status does not affect the prognosis of adult BL patients. The prevalence of BL is higher in men than in women; however, we found no gender differences regarding prognosis.

In this meta-analysis, we compared the efficacy of different chemotherapy regimens by subgroup analysis and identified the factors that may influence the treatment outcomes. This is the first answer to the question of which is the best treatment option for BL in the rituximab era. The DA-EPOCH-R regimen was a less toxic regimen than other dose-intensive regimens for BL (e.g., CODOX-M-IVAC) (39). DA-EPOCH-R was a preferred option not only because it extends survival time but also because of its better safety and lower medical costs.

The following limitations of this study should be acknowledged. Firstly, included articles were performed by different study types, and most of the included studies were single-arm without RCT studies. Secondly, the small number of included studies and patients may not be representative enough of all adult BL patients; there were only less than 20 patients in some included studies. Thirdly, doses of chemotherapy agents modified by medicine centers and various manufacturers of rituximab cannot be overlooked. Also, the results of our heterogeneity analysis could be different from the real world owing to unavailable data from some cohorts. Furthermore, the applicability of our results in patients with BM and/or CNS involvement remains to be further validated due to the existence of heterogeneity.

In summary, a better treatment strategy for prolonging the survival time of adult BL without CNS and/or BM involvement was first identified through our meta-analysis. We found that DA-EPOCH had a greater OS rate in contrast with CODOX-M/IVAC, GMALL-B-ALL/NHL2002, and Hyper-CVAD groups. More studies, especially RCTs, need to be performed to identify the optimal chemotherapy regimen for patients with extensive organ involvement.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

XL: Data acquisition, writing the original draft, and editing. YL: Data acquisition and analysis. RL: Resources, and data acquisition. JL: Manuscript revision and project administration. XY: Providing guidance, manuscript revision, and funding acquisition. LQ: Methodology, manuscript revision, funding acquisition and final approval of the last version. All authors contributed to the article and approved the submitted version.