RATIONALE: Drugs used in chemotherapy, such as etoposide and cyclophosphamide, work in different ways to stop the growth of cancer cells, either by killing the cells or by stopping them from dividing. Thalidomide, celecoxib, and fenofibrate may stop the growth of cancer cells by blocking blood flow to the cancer. Celecoxib also may stop the growth of cancer cells by blocking some of the enzymes needed for cell growth. Giving combination chemotherapy together with thalidomide, celecoxib, and fenofibrate may kill more cancer cells.

PURPOSE: This phase II trial is studying how well giving etoposide and cyclophosphamide together with thalidomide, celecoxib, and fenofibrate works in treating young patients with relapsed or progressive cancer.

OBJECTIVES:

Primary

• Evaluate the activity of etoposide, cyclophosphamide, thalidomide, celecoxib, and fenofibrate, in terms of prolonging the time to disease progression, in young patients with relapsed or progressive cancer.

Secondary

• Determine, preliminarily, the biologic activity of this regimen, in terms of tumor response and overall survival, in these patients.
• Determine the toxicity of this regimen in these patients.
• Evaluate different radiographic techniques as markers of tumor response in these patients.
• Evaluate the predictive ability of in vitro correlative studies as markers of tumor response.

STATISTICAL DESIGN: Patients were classified into one of 8 strata according to diagnosis: leukemia/lymphoma, bone tumors, neuroblastoma, high grade glial tumors, low grade glial tumors, ependymoma, medulloblastoma/PNET, and miscellaneous. A two-stage design for each disease stratum was planned. The accrual goal at the end of the two-stage design was 20 subjects for each stratum. A stopping rule was applied after the accrual of the first 10 eligible subjects enrolled in each disease stratum. If 1 or more patients in the first 10 evaluable patients were alive and progression-free at 27 weeks and have tolerated therapy then accrual to stage two would proceed. Among 20 patients within a stratum, if 3 or more patients met primary endpoint then regimen would be considered successful. The probability of concluding the treatment is feasible is 0.95 if true success rate is 30% and 0.07 if true succes rate is 5%. Overall accrual target was 80-160 patients. Please see published manuscript (Robison et al Pediatr Blood Cancer 2014) for results within disease strata.

DISEASE CHARACTERISTICS:

• Histologically confirmed cancer (at diagnosis or relapse), including any of the following:

  • Leukemia and/or lymphoma (closed to accrual)

  • Bone tumor (e.g., Ewing's sarcoma or osteosarcoma) (closed to accrual)

  • Neuroblastoma (closed to accrual)

  • High-grade glial tumor

  • Low-grade glial tumor

  • Ependymoma

  • Medulloblastoma and/or primitive neuroectodermal tumor (PNET)

  • Miscellaneous tumor (closed to accrual)

  • Brain stem glioma, defined as intrinsic tumors of the pons causing diffuse enlargement

    • Brain stem glioma that progressed after radiotherapy does not require histological confirmation

    • Duration of symptoms at the time of diagnosis must be < 3 months

      • Symptoms should consist of cranial nerve deficits, ataxia, and/or long tract signs

• Relapsed or progressive poor prognosis disease for which no available curative therapy exists

PATIENT CHARACTERISTICS:

• Karnofsky performance status 50-100% OR Lansky play scale 50-100% (for infants)
• Life expectancy > 2 months
• Platelet count > 75,000/mm^3 (transfusion independent)
• Absolute neutrophil count > 1,000/mm^3 (in patients without bone marrow disease)
• Hemoglobin ≥ 9.0 g/dL
• Creatinine < 1.5 mg/dL OR creatinine clearance or glomerular filtration rate ≥ 70 mL/min
• Bilirubin ≤ 1.5 mg/dL
• SGPT ≤ 3 times normal
• SGOT ≤ 3 times normal (4 times normal for patients on ranitidine hydrochloride)
• Alkaline phosphatase ≤ 3 times normal
• Not pregnant or nursing
• Negative pregnancy test
• Fertile patients must use effective double-method contraception during and for 2 months after completion of study treatment
• Must be willing to participate in the Celgene STEPS® program
• Recent thromboembolic disease (e.g., deep vein thrombosis or pulmonary embolism) allowed if patient is clinically stable and the thromboembolic event occurred > 3 weeks prior to study entry
• No active infection
• No active uncontrolled cardiac, hepatic, renal, or psychiatric disease ≥ grade 3
• No known allergies to sulfonamides
• No concurrent illness that would obscure toxicity or dangerously alter drug metabolism
• No other serious medical illness

PRIOR CONCURRENT THERAPY:

• See Disease Characteristics
• Recovered from prior therapy
• Prior chemotherapy and/or radiotherapy allowed
• Prior celecoxib allowed
• Prior standard-dose IV etoposide and cyclophosphamide administered in 3-week courses allowed
• No prior oral therapy with etoposide, thalidomide, cyclophosphamide, or fenofibrate for > 2 months in duration
• No other concurrent investigational agents
• No other concurrent nonsteroidal anti-inflammatory drugs
• Concurrent steroids and/or antiseizure medications allowed