Drugs used in chemotherapy, such as decitabine, work in different ways to stop the growth of cancer cells, either by killing the cells or by stopping them from dividing. Valproic acid may stop the growth of cancer cells by blocking some of the enzymes needed for cell growth. Giving decitabine together with valproic acid may be an effective treatment for non-Hodgkin's lymphoma. This phase I trial is studying the side effects and best dose of decitabine and valproic acid in treating patients with relapsed or refractory aggressive B-cell non-Hodgkin's lymphoma.

PRIMARY OBJECTIVES:

I. Determine the minimally effective pharmacological dose (MEPD) of single-agent decitabine in patients with relapsed or refractory aggressive B-cell non-Hodgkin's lymphoma.

II. Determine the maximum tolerated dose of valproic acid when administered with the MEPD of decitabine in these patients.

III. Determine the MEPD of valproic acid when administered with decitabine in these patients.

IV. Determine the toxic effects of decitabine alone and in combination with valproic acid in these patients.

SECONDARY OBJECTIVES:

I. Determine the response rate in patients treated with these drugs. II. Determine the pharmacokinetics of these drugs in these patients.

OUTLINE: This is a dose-escalation study. Patients are assigned to 1 of 2 treatment stages.

STAGE 1: Patients receive decitabine IV over 1 hour on days 1-5 or 1-10. Treatment repeats every 28 days for up to 12 courses in the absence of disease progression or unacceptable toxicity.

STAGE 2: Patients receive decitabine as in stage 1 and valproic acid orally (PO) thrice daily (TID) on days 5-21. Treatment repeats every 28 days for up to 12 courses in the absence of disease progression or unacceptable toxicity.

For both stages, patients who achieve an objective response (complete response [CR], unconfirmed CR, or partial response) may discontinue study treatment and undergo stem cell transplantation, if eligible.

PROJECTED ACCRUAL: Approximately 18-42 patients (18 for stage 1 and 24 for stage 2) will be accrued for this study.

Inclusion Criteria:

• Histologically or cytologically confirmed aggressive B-cell non-Hodgkin's lymphoma (NHL), including any of the following subtypes:

  • Mantle cell lymphoma

  • Diffuse large cell lymphoma

  • Burkitt's lymphoma

  • Transformed NHL* arising from a previously diagnosed low-grade lymphoma, including any of the following:

    • Follicular lymphoma
    • Small lymphocytic lymphoma
    • Chronic lymphocytic leukemia

• Relapsed or refractory disease

  • Relapsed or refractory disease must have occurred during the most recent prior therapy

• Has accessible tissue for biopsy OR evidence of ≥ 50% bone marrow involvement AND willing to undergo serial biopsy

• Not eligible for OR refused curative stem cell transplantation

• No active or untreated CNS lymphoma

• Performance status - ECOG 0-2

• Absolute neutrophil count ≥ 1,500/mm^3

• Platelet count ≥ 75,000/mm^3

• AST and ALT ≤ 2.5 times upper limit of normal

• Bilirubin ≤ 1.5 mg/dL

• Creatinine ≤ 2.0 mg/dL

• No symptomatic congestive heart failure

• No unstable angina pectoris

• No cardiac arrhythmia

• Not pregnant or nursing

• Negative pregnancy test

• Fertile patients must use effective contraception

• No known HIV positivity

• No ongoing or active infection

• No other uncontrolled illness

• No psychiatric illness or social situation that would preclude study compliance

• Prior stem cell transplantation allowed

• Recovered from all prior biologic therapy-related toxicity

• Recovered from all prior chemotherapy-related toxicity

• No other concurrent chemotherapy unless it is used in the chronic daily setting for other medical conditions, including pulmonary, rheumatologic, or adrenal disorders

• No concurrent corticosteroids unless they are used in the chronic daily setting for other medical conditions, including pulmonary, rheumatologic, or adrenal disorders

• Recovered from all prior radiotherapy-related toxicity

• No concurrent palliative radiotherapy

• Recovered from all prior therapy-related toxicity

• No concurrent anticonvulsants, including valproic acid (except as used in this study)