The purpose of the study is to determine if participants who receive the GVHD prophylaxis medication pentostatin will have less severe hepatic toxicities than those receiving MTX. The study is estimated to have sufficient statistical power to ascertain at least a 20% improvement in day 42 NCI CTC grade 2 or above hepatic toxicity-free survival in pentostatin recipients.

Participants will be randomized to receive either methotrexate (MTX) or pentostatin for graft-versus-host disease (GVHD) prophylaxis after receiving an allogeneic bone marrow transplant from an HLA-matched related or unrelated donor. All participants will receive a standard backbone GVHD prophylaxis regimen (tacrolimus and sirolimus) and conditioning (cyclophosphamide/TBI). A risk-adapted approach will be used during conditioning to further minimize the risk of leukemia relapse based on two factors:

1. Lymphoid versus myeloid primary disease.
2. KIR compatibility between donor and host.

Inclusion Criteria:

*Age less than or equal to 21 years old

High risk malignancy as follows:

• High-risk ALL in CR1. Examples include, but not limited to: Induction failure or > 1% leukemic lymphoblasts in the bone marrow on remission date;> 0.1% leukemic lymphoblasts in the bone marrow in week 7 of continuation treatment (i.e. before reinduction I); re-emergence of leukemic lymphoblasts by MRD (at any level) in patients previously MRD negative; persistently detectable MRD at lower levels;early T-cell precursor (ETP) ALL.
• High-risk ALL beyond CR1, or with refractory disease. "Beyond CR1" denotes any CR following CR1, or any relapsed state. "Refractory disease" includes induction failure.
• High-risk de novo AML in CR1.Examples include but are not limited to:evidence of a high-risk genetic abnormality or high-risk MRD features.
• AML beyond CR1, or with refractory disease. "Beyond CR1" denotes any CR following CR1, or any relapsed state. "Refractory disease" includes induction failure.
• Therapy-related AML.
• MDS, primary or secondary, at any stage.
• NK cell lymphoblastic leukemia in any CR
• Biphenotypic bilineage, or undifferentiated leukemia.
• CML in any phase
• Hodgkin lymphoma beyond CR1 or with refractory disease. "Beyond CR1" denotes any CR following CR1, or any relapsed state.
• Non-Hodgkin lymphoma beyond CR1 or with refractory disease. "Beyond CR1" denotes any CR following CR1, or any relapsed state.
• Juvenile Myelomonocytic Leukemia (JMML).
• All patients with prior evidence of CNS leukemia must be treated and be in CNS CR to be eligible for study.
• Has a suitable HLA matched sibling or unrelated volunteer donor available for stem cell donation.A "matched" donor is defined as allele matching at 7/8 to 8/8 HLA loci at A, B, C and DRB1.For the purpose of this study, the term "matched sibling" also refers to an HLA matched family member.
• Does not have any other active malignancy other than the one for which this transplant is indicated.
• Left ventricular ejection fraction > 40%,or shortening fraction > 26%.
• Forced vital capacity (FVC) greater than or equal to 50% of predicted value (corrected for hemoglobin), or if patient is unable to perform pulmonary function testing, pulse oximetry greater than or equal to 92% on room air.
• Creatinine clearance greater than or equal to 70 ml/min/1.73m2
• Karnofsky or Lansky (age-dependent) performance score of greater than or equal to 70.
• Bilirubin less than or equal to 2.5 mg/dL.
• Alanine aminotransferase (ALT), aspartate aminotransferase (AST), and alkaline phosphatase less than or equal to 5 times upper limit of normal
• Not pregnant as confirmed by negative serum or urine pregnancy test within 14 days prior to enrollment.
• Not lactating.
• Has not had a prior allogeneic HSCT.

Exclusion Criteria:

• Pregnant and lactating females are excluded from participation as the short and long-term effects of the protocol interventions and infusion on a fetus or a nursing child through breast milk are not entirely known at this time.