DA-EPOCH-R is more effective and less toxic than modied CODOX-M/IVAC-R for treating Chinese HIV-infected patients with Burkitt lymphoma: an uncontrolled retrospective clinical trial

BACKGROUND Burkitt lymphoma (BL), commonly associated with human immuno-deciency virus (HIV) in China, is an aggressive B-cell lymphoma with a poor prognosis. Current treatments are insucient and have severe side effects in adult patients with immunodeciency. METHODS We retrospectively studied HIV-positive patients with untreated BL from December 2011 to June 2019. We compared a low-intensity treatment regimen comprising infused etoposide, doxorubicin, and cyclophosphamide with vincristine, prednisone, and rituximab (dose-adjusted EPOCH-R, study group) and a dose-intensive regimen comprising cyclophosphamide, vincristine, doxorubicin, high-dose methotrexate/ifosfamide, etoposide, high-dose cytarabine, and rituximab (CODOX-M/IVAC-R, control group). Differences in survival outcomes, toxicity, and survival-associated factors were analyzed. RESULTS Twenty-eight consecutive patients were enrolled with integrated clinical data, with 18 patients in the DA-EPOCH-R group and 10 in the modied CODOX-M/IVAC-R group. The median age of patients was 40 years (range: 21–60 years). The median baseline absolute CD4+ cell count was 243.5 cells/µL. All patients had high-risk diseases. Overall, grade 3–4 toxicity was observed at least once in 22 patients. The principal grade 3–4 toxic events, namely leukopenia (100% vs. 50%), febrile neutropenia (90% vs. 22.2%), and thrombocytopenia (80% vs. 16.7%), were signicantly higher in the control group than in the study group. One patient in the control group died of septic shock induced by pneumonia. The median follow-up time was 19 months, and the median overall survival (OS) was signicantly improved in the study group compared with that in the control group [18.0 months (95% condence interval [CI], 14.0– 22.0 months) vs.

Improved immune function in the HAART era has led to a reevaluation of full-dose, high-intensity chemotherapy. Moreover, HIV-positive patients with BL and preserved immune function may bene t from more aggressive chemotherapeutic regimens with strong toxicities, such as the modi ed cyclophosphamide, vincristine, doxorubicin, high-dose methotrexate/ifosfamide, etoposide, high-dose cytarabine, and rituximab (CODOX-M/IVAC-R) regimen [4][5][6][7], which is preferentially recommended under current National Comprehensive Cancer Network (NCCN) guidelines. The AIDS Malignancy Consortium 048 trial further prospectively evaluated modi cation of the CODOX-M/IVAC regimen in 34 patients with HIV-associated BL. This trial demonstrated a grade 3-4 toxicity rate of 79%, with a 68% protocol completion rate [5]. The one-year progression-free survival (PFS) rate was 69%, and the overall survival (OS) rate was 72%; the two-year OS rate was 69% [5]. The severe toxicity of CODOX-M/IVAC ± R has gained attention. Chinese patients with HIV-associated BL in the retrospective case series evaluated by Xiao et al. also experienced a 33.3% rate of grade 3-4 infectious complications, with a 13.3% rate of septic shock, when treated with a regimen of CODOX-M/IVAC ± R [8]. An optimal treatment regimen has not been established for HIV-associated BL in China.
Therefore, because of the perceived risk of increased hematologic and infectious complications, HIVassociated BL cases tend to be treated with low-intensity therapy. During the HAART era, dose-adjusted infusional etoposide, infusional doxorubicin, and cyclophosphamide with infusional vincristine, prednisone, and rituximab (DA-R-EPOCH) is another preferred regimen for treating HIV-associated DLBCL under current NCCN guidelines based on the results of multiple phase II clinical trials and retrospective studies [10,[14][15][16]. Moreover, because of severe side effects and inferior outcomes of CODOX-M/IVAC-R in immunode ciency patients, the DA-R-EPOCH regimen was tested as an alternative to improve the toxicity pro le and effectively treat BL [17]. Dunleavy et al. investigated DA-EPOCH-R and a short course regimen with a double dose of rituximab (SC-EPOCH-RR) in 30 untreated patients with BL [17]. In this trial, 11 patients with HIV-associated BL were enrolled in the SC-EPOCH-RR group. The median follow-up period was 73 months in the SC-EPOCH-RR group, and the PFS and OS rates were 100% and 90%, respectively [17]. Despite these advancements, the median OS of Chinese HIV-associated BL patients was approximately 1 year [8,9], treatment of those remains challenging. Additionally, no studies have evaluated the e cacy and tolerability of the DA-EPOCH-R versus modi ed CODOX-M/IVAC-R regimens.
Thus, in the present retrospective study, we compared the safety and clinical outcomes of DA-R-EPOCH versus modi ed CODOX-M/IVAC-R combined with cART in Chinese patients with BL.

Patients And Methods
This prospective study was approved by the ethics committee of Beijing Ditan Hospital, Capital Medical University, Beijing, China. All patients provided written informed consent.

Patients
Twenty-eight patients were consecutively enrolled between December 2011 and June 2019 from a top-tier comprehensive hospital. All patients had documented HIV infection and were newly diagnosed with untreated BL according to the World Health Organization (WHO) criteria [18]. Fluorescence in situ hybridization for MYC using a break-apart probe was performed for patients in whom the immunophenotype and morphology were not obvious for a BL diagnosis.
All patients underwent diagnostic procedures, molecular evaluations, and treatment with DA-R-EPOCH or modi ed CODOX-M/IVAC-R. Pre-treatment evaluations included patients' demographic characteristics, physical examination, Eastern Cooperative Oncology Group (ECOG) performance status (PS), complete blood count, blood chemistry, and computed tomography or magnetic resonance imaging scans of the abdomen and pelvis. All patients were staged according to the Ann Arbor staging system. The disease risk was evaluated according to NCCN guidelines. Patients were regarded as high-risk based on the following features: stage I and abdominal mass or extra-abdominal mass > 10 cm or Ann Arbor stage II-IV. Patients not meeting these criteria were regarded as low-risk [19].
The eligibility criteria were as follows: adequate cardiac function (ejection fraction ≥ 50%), ECOG PS of 0-3; and renal (serum creatinine of ≥ 1.5 mg/dL or creatinine clearance ≥ 60 mL/min) and hepatic function (aspartate aminotransferase and alanine aminotransferase ≤ 3 × the upper limit of normal and direct bilirubin level of ≥ 2.0 mg/dL) unless they were secondary to antiretroviral therapy. In this case, the total bilirubin cutoff was ≥ 3.5 mg/dL, provided that the direct bilirubin was normal. Additionally, absolute neutrophil count ≥ 1.5 × 10 9 /L and platelet count ≥ 50 × 10 9 /L were necessary unless they were disease-related. Cases of primary CNS BL or atypical BL were excluded. Prior malignancies were excluded unless the patient had been disease-free more than 5 years other than having curatively treated cutaneous basal cell or squamous cell carcinoma, in situ cervical carcinoma, or cutaneous Kaposi sarcoma. Non-measurable diseases (e.g., isolated bone marrow involvement) were allowed. Patients lost to follow-up or who refused systemic treatment were excluded.
Rituximab (375 mg/m 2 ) was administered on day 1 (before continuous infusion began). For subsequent cycles, cyclophosphamide was dose-adjusted according to speci ed guidelines based on nadir counts.
Eighteen patients were scheduled to be administered DA-EPOCH-R for a total of eight cycles. These patients were administered prophylactic granulocyte colony-stimulating factor with lgrastim or peg lgrastim. Cycles were repeated every 21 days. Treatment and supportive care options are summarized in Table S1.
Ten patients were considered as high-risk and planned to be administered modi ed CODOX-M/IVAC-R in an R-CODOX-M/IVAC/R-CODOX-M/IVAC sequence for a total of 4 cycles [5]. R-CODOX-M cycles were to span between 21 and 28 days. Treatment and supportive care options are summarized in Table S2.
Patients were administered 10 mg of intrathecal methotrexate on days 1 and 5 of cycle 3; this regimen was repeated every three weeks for a total of six doses (i.e., cycles 3-5) regardless of when DA-EPOCH or CODOX-M/IVAC treatment was stopped. Patients with positive ow cytometry or cytology of cerebrospinal uid test results were administered intrathecal or intraventricular methotrexate twice weekly for two weeks beyond negative ow cytometry for a minimum of four weeks, with consolidation weekly for six weeks and maintenance monthly for six months. Drug combinations for HIV All patients were administered cART before initiation of chemotherapy. The preferred regimen was tenofovir and lamivudine, plus efavirenz, as recommended by the Chinese guidelines for diagnosis and treatment of HIV/AIDS (2018) [20]. In this study, patients who did not start cART until diagnosis of lymphoma initiated treatment with tenofovir and lamivudine, plus efavirenz, whereas those administered cART regimens such as zidovudine and lamivudine, plus nevirapine or efavirenz, were switched to the above regimen prior to chemotherapy to reduce the risk of zidovudine-associated myelosuppression and nevirapine-induced liver injury. All patients were administered cART until their death. Patients were also administered prophylactic treatment for pneumocystis pneumonia if CD4 + cell counts were below 200 cells/µL [20]. Trimethoprim-sulfamethoxazole is the recommended prophylactic agent, and one doublestrength tablet daily is the preferred regimen. Patients with a CD4 count < 50 cells/µL were administered azithromycin 1200 mg once weekly against disseminated Mycobacterium avium complex disease [20]. Endpoints and follow-up The primary endpoint was OS. The secondary endpoints included PFS, clinical tumor response, and adverse events (AEs). OS was calculated from the date of study enrollment until death or the last followup visit. PFS was calculated from the date of study enrollment until the date of progression or the last follow-up visit. The tumor response was assessed by standard whole-body computerized tomographic scanning criteria [21] after cycle 3, and post-treatment (4-8 weeks, and months 6, 12, 18, and 24). AEs were assessed using the National Cancer Institute Common Terminology Criteria for Adverse Events version 4 [22]. Patients were followed every three months for two years post-treatment, and then every six months for years 3-5. Statistical analysis Survival curves were constructed using the Kaplan-Meier method, and statistical signi cance was determined using the log-rank test. For PFS, deaths unrelated to acute treatment or lymphoma were censored. Between-group comparisons were performed using the Student t-test for continuous data and Chi-square (χ 2 ) test for categorical data. Furthermore, an exploratory Cox proportional hazard analysis was conducted to assess the in uence of baseline characteristics on OS. For univariate analysis, the following variables were included to calculate the independent predictors of PFS and OS: age, ECOG PS, B symptom, primary tumor site, treatment regimen, disease risk classi cation, Ann Arbor stage, baseline absolute CD4 + cell count, baseline lactate dehydrogenase elevation, and Epstein-Barr virus (EBV) expression. Hazard ratios (HRs) and 95% con dence intervals (CIs) were estimated using a nonparametric log-rank test, (i.e., Cox proportional hazards model). All reported P-values are two-tailed and presented without adjustment for multiple comparisons. P < 0.05 was considered as statistically signi cant. All statistical analyses were performed using SPSS software (version 17.0, SPSS, Inc., Chicago, IL, USA).

Patient characteristics
Between December 2011 and June 2019, 41 patients were initially included, of whom 28 were treated with chemotherapy plus rituximab (Fig. 1). Ten patients were administered modi ed CODOX-M/IVAC-R (control group), and 18 patients received the DA-R-EPOCH regimen (study group). All patients were male and administered cART before chemotherapy.
The baseline patients and disease characteristics were listed in Table 1. The median patient age was 40 years (range: 21-60 years). The median baseline absolute CD4 + cell count was 243.5 cells/µL (range: 11-782 cells/µL). All patients were classi ed as high-risk diseases. All evaluable patients were CD20 + CD10 + and had a Ki-67 of > 95%. Most patients (82.1%, 23) presented with bulky disease. Baseline median CD4 + cell counts were comparable between the two groups (198 vs. 188 cells/µL; P = 0.28). Two patients (20.0%) in the control group were on cART at baseline, whereas ve patients (27.8%) in the study group were on cART at baseline. The two groups were well-balanced regarding baseline median CD4 + cell count, age, previous cART, HIV viral load, EBV expression, demographic and disease characteristics (Table 1) .  (Fig. 2B). Overall, 11 patients died during the study. One patient died of treatment-related toxicity characterized by septic shock with pneumonia. During the follow-up period, 10 patients died of systemic disease progression, including one patient who died from CNS disease that was not present at baseline.

Treatment-related Adverse Events
The details of toxicities are shown in Table 2. Overall, grade 3-4 toxicity occurred at least once in 22 patients. The occurrence rates of leukopenia (100% vs. 50%), febrile neutropenia (90% vs. 22.2%), and thrombocytopenia (80% vs. 16.7%) were signi cantly higher in the control group than in the study group.
As a result, grade 4 transfusion level anemia occurred in two patients (11.1%), and thrombocytopenia occurred in three patients (16.7%) in the study group. Additionally, grade 4 transfusion level anemia developed in two patients (20%), and thrombocytopenia developed in four patients (40%) in the control group. Because of severe bone marrow suppression, eight patients (80%) acquired grade 3-4 infection in the control group compared with one (5.6%) in the study group (P = 0.001). Furthermore, one patient died of septic shock induced by pneumonia. Five patients had non-hematologic related grade 3-4 toxicities, including three grade 3-4 elevation of serum transaminase, one grade 3-4 elevation of transaminase combined with grade 3 renal toxicity, and one grade 3 gastrointestinal bleeding.   Table 3 outlines the treatment completion status. Only one patient in the control group completed the treatment scheme. Early termination of therapy in the control group occurred due to disease progression and AEs in 40% and 50% of patients, respectively. AEs leading to early termination were hematologic (n = 2) or infectious (n = 3). Further, 35.3% (6 out of 17) high-risk patients in the study group completed the treatment scheme. Early termination of therapy by high-risk patients in the study group occurred because of disease progression, patient withdrawal, and AEs in 29.4% (n = 5), 23.5% (n = 4), and 11.8% (n = 2) of patients, respectively. AEs leading to early termination were hematologic (n = 1) or infectious (n = 1).  (6) Disease progression (%,n) 40% (4) 29.4%% (5) Early termination (%,n)
Our study had some limitations. As this was a prospective single-center study, investigator bias and selection bias could not be avoided. The use of concurrent cART was not uniformly prescribed. Prior WHO guidelines recommended that cART should not be initiated if the CD4 + cell count was > 350 cells/µL. However, based on Chinese guidelines for HIV [20], all patients were administered cART before chemotherapy regardless of the baseline CD4 + cell count. Finally, our analysis maybe underpowered and could not detect subtle differences in outcomes given the small number of patients in each group.
In summary, the data from our study suggest there is no longer a need for high-intensity, high-toxicity chemotherapy treatment for high-risk BL. We aggressively treated HIV-infected patients with BL and improved OS rates by using DA-R-EPOCH. This regimen minimized drug-related toxicity compared with modi ed CODOX-M/IVAC-R. Thus, given the severe toxicity associated with previous chemotherapy regimens for BL, the results of this uncontrolled retrospective clinical trial are particularly promising. Declarations Figure 1 Trial owchart of participant selection and diagram of treatment.

Figure 2
Kaplan-Meier curves showing progression-free survival (A) and overall survival (B) for the two different treatment regimens: DA-R-EPOCH (study group) and CODOX-M/IVAC-R group (control group).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. TableS1.docx TableS2.docx