RATIONALE: Monoclonal antibodies, such as rituximab, can block cancer growth in different ways. Some block the ability of cancer cells to grow and spread. Others find cancer cells and help kill them or carry cancer-killing substances to them. Drugs used in chemotherapy work in different ways to stop the growth of cancer cells, either by killing the cells or by stopping them from dividing. Giving rituximab together with combination chemotherapy may kill more cancer cells.

PURPOSE: This phase II trial is studying how well giving rituximab together with combination chemotherapy works in treating patients with newly diagnosed, HIV-associated Burkitt's lymphoma.

OBJECTIVES:

Primary

• Determine the efficacy of rituximab, cyclophosphamide, vincristine, doxorubicin hydrochloride, and high-dose methotrexate (R-CODOX-M ) alone or alternating with rituximab and ifosfamide, etoposide phosphate, and high-dose cytarabine (IVAC) and intrathecal CNS prophylaxis in patients with newly diagnosed, previously untreated, HIV-associated Burkitt's lymphoma or atypical Burkitt's lymphoma.
• Determine the safety of this regimen in these patients.

Secondary

• Evaluate downstream effectors of apoptosis as mechanisms of chemotherapy resistance and prognosis and perform exploratory analysis of their relationship to treatment effect.
• Evaluate multi-drug resistance gene expression as a mechanism of chemotherapy resistance and prognosis and perform exploratory analysis of their relationship to treatment effect.
• Confirm the use of flow cytometry in the identification of occult leptomeningeal disease and determine whether abnormal flow cytometry is predictive when CNS cytology is negative for malignant cells.
• Determine the biologic and prognostic significance of Epstein-Barr virus (EBV)-positive Burkitt's lymphoma in the highly active antiretroviral therapy era and perform exploratory analysis of their relationship to treatment effect.
• Compare genotyping in patients with HIV-associated Burkitt's lymphoma with that of patients who are HIV-negative and determine whether they are uniform in their genetic profile or whether some cases are more like diffuse large B-cell lymphoma.
• Determine if EBV detection in cerebrospinal fluid at diagnosis is predictive of leptomeningeal disease.

OUTLINE: This is a prospective, multicenter study. Patients are stratified according to risk category (low-risk vs high-risk). Patients with low-risk disease receive 3 courses of R-CODOX-M chemotherapy as described below. Patients with high-risk disease receive 4 alternating courses of R-CODOX-M/IVAC chemotherapy as described below in an A/B/A/B sequence.* Courses repeat every 21-28 days in the absence of disease progression or unacceptable toxicity.

NOTE: *In patients presenting with anasarca, pleural effusion, or ascites, methotrexate can pool causing difficulties with clearance; in this case, treatment may be given in a reverse sequence: B/A/B/A.

• Regimen A (R-CODOX-M chemotherapy): Patients receive rituximab** IV and doxorubicin hydrochloride IV over 15 minutes on day 1, cyclophosphamide IV over 30-60 minutes on days 1 and 2, pegfilgrastim subcutaneously (SC) on day 3, vincristine IV on days 1 and 8, high-dose methotrexate IV over 2-4 hours on day 15, and leucovorin calcium IV beginning 24 hours after the start of methotrexate and continuing every 6 hours until the methotrexate level is less than 50 nmol/L. Patients receive CNS prophylaxis comprising methotrexate intrathecally (IT), cytarabine IT, and hydrocortisone IT on day 1. Patients with high-risk disease receive an additional dose of cytarabine IT on day 3. Patients also receive filgrastim (G-CSF) SC once daily on days 3-9. Once the methotrexate levels drops below 50 nmol/L, patients resume G-CSF SC once daily beginning on approximately day 18 and continuing until blood counts recover.
• Regimen B (rituximab and IVAC chemotherapy): Patients receive rituximab** IV on day 1, ifosfamide IV continuously and etoposide IV continuously over 24 hours on days 1-5, and high-dose cytarabine IV over 1-3 hours twice daily on days 1-2. Patients receive CNS prophylaxis comprising methotrexate IT and hydrocortisone IT on day 5. Patients also receive pegfilgrastim SC once 24-48 hours after completion of chemotherapy OR G-CSF SC beginning on day 6 and continuing until blood counts recover.

Patients with CNS involvement (leptomeningeal and/or intraparenchymal) at diagnosis do not receive CNS prophylaxis as above. Instead, these patients receive a combination of sequential liposomal cytarabine and methotrexate IT or via an Ommaya reservoir on day 1 and then every 14 days as tolerated until completion of systemic chemotherapy.

NOTE: **Rituximab may be given up to 3 days before a chemotherapy course and anytime during the course for 3 (low-risk disease) or 4 (high-risk disease) total doses.

Patients undergo blood and cerebrospinal fluid collection and tumor biopsies periodically during study treatment for correlative studies of prognostic biomarkers predictive of survival (e.g., c-flip protein expression; p53 mutations [by immunohistochemistry (IHC)]; multidrug resistance gene expression [by IHC]; and Epstein-Barr virus in tumor DNA or cerebrospinal fluid [by polymerase chain reaction]); genotyping of Burkitt's lymphoma; and flow cytometry as a tool (by staining) for detecting occult positivity of leptomeningeal disease in Burkitt's lymphoma.

After completion of study treatment, patients are followed every 4 months for at least 2 years.

PROJECTED ACCRUAL: A total of 34 patients will be accrued for this study.

DISEASE CHARACTERISTICS:

• Histologically confirmed Burkitt's lymphoma (BL) or new WHO 2009 criteria B-cell lymphoma unclassified (with features intermediated between difuse large B-cell lymphoma and BL)

  • Any stage disease
  • Newly diagnosed disease

• Meets 1 of the following criteria for disease risk:

  • Low-risk disease, defined by 1 of the following:

    • Stage I with a single focus of disease < 10 cm AND normal lactate dehydrogenase (LDH) level
    • Totally resected intra-abdominal disease only AND normal LDH post surgery

  • High-risk disease, defined as not meeting criteria for low-risk disease

• Measurable or nonmeasurable disease

• HIV-positive confirmed by enzyme-linked immunosorbent assay and Western blot OR by measurable HIV viral load

• No visceral Kaposi's sarcoma

PATIENT CHARACTERISTICS:

• Karnofsky performance status 40-100%

• Not pregnant or nursing

• Negative pregnancy test

• Fertile patients must use effective contraception

• LVEF ≥ 50% by MUGA or echocardiogram

• Creatinine ≤ 1.5 mg/dL OR creatinine clearance ≥ 60 mL/min

• Absolute neutrophil count ≥ 1,000/mm³

• Platelet count ≥ 50,000/mm³ (unless related to lymphoma)*

• Direct bilirubin ≤ 2.0 mg/dL OR total bilirubin ≤ 3.5 mg/dL AND direct bilirubin normal (if elevated bilirubin secondary to antiretroviral therapy)

• AST and ALT ≤ 3 times upper limit of normal

• No other malignancy within the past 5 years except curatively treated cutaneous basal cell or squamous cell carcinoma, carcinoma in situ of the cervix, or cutaneous Kaposi's sarcoma

• No other medical illness unrelated to non-Hodgkin's lymphoma, including any of the following:

  • Uncontrolled infection (including opportunistic infection)
  • Chronic renal insufficiency
  • Myocardial infarction within the past 6 months
  • Unstable angina
  • Cardiac arrhythmias other than chronic atrial fibrillation

• Patients with active hepatitis B infection are eligible provided they receive concurrent dual antiviral therapy NOTE: *Patients with bone marrow involvement are eligible irrespective of blood count

PRIOR CONCURRENT THERAPY:

• See Disease Characteristics

• No prior therapy for this disease except for 1 of the following :

  • Seven consecutive days of steroids alone or in combination with a non-CHOP regimen necessary for patient stabilization (e.g., cyclophosphamide and steroids steroids for normalization of disease-related hyperbilirubinemia)
  • One course of CHOP or fractionated CHOP (e.g. CODOX) with or without rituximab

• No epoetin alfa or filgrastim (G-CSF) within 24 hours of study chemotherapy

• No concurrent zidovudine