The purpose of this research study is:(1) to determine if high doses of chemotherapy without total body irradiation can allow selected stem cells to take and grow,(2) to determine if selected stem cells from the blood or marrow can take and not cause a complication called graft-versus-host disease (GvHD) and (3) to evaluate the side effects of the combination of chemotherapy drugs used for these transplants. In the last 10 years we have developed chemotherapy combinations to be used for this T-cell depleted transplant protocol. By using three chemotherapy drugs (IV busulfan, melphalan and fludarabine), we hope to have a good chemotherapy combination to kill cancer cells, and to make the graft take, without the side effects of total body irradiation. The chemotherapy drugs to be tested in this protocol are busulfan, melphalan and fludarabine, all of which have been used successfully for stem cell transplantation, but not given together as in this specific regimen. This is what is being tested in this study.

Our initial trials in the 1980's with T-cell depleted transplants showed less GvHD, but the overall results of the transplants were not better. The reason for this was that the stem cells did not take and engraft in 15% of our adult patients. This failure of the stem cells to take can leave patients without bone marrow or blood cells necessary for life. Most stem cell transplants were done using bone marrow (BMA) obtained from the donors. However, if we give a medication called G-CSF by shots to the donor, we can collect peripheral blood stem cells (PBSC) and use them for transplant. The advantage of this approach is that we can collect 2-20 times more stem cells than that obtained from the marrow. It has been proven that a larger number of stem cells in the graft make it more difficult for the patient to reject the stem cells. Some donors may be too small to provide peripheral blood stem cells or they may not want to take G-CSF shots. In these cases the donors will have their marrow collected in the operating room under general anesthesia.

Stem cell transplants can lead to a condition known as acute graft-versus-host disease or GvHD. This disease is caused by an assault by certain cells in the marrow or blood (T-cells) of the donor (graft) against your body (the host). These T-cells see your body as foreign and attack it. The disease causes a skin rash, liver disease, and diarrhea. Methods were developed at this institution to prevent GvHD. These methods take out most of the T-cells (responsible for GvHD) from the marrow or blood stem cells before transplant. This is called "T-cell depletion" or "stem cell selection". In this hospital, we use two types of methods of T-cell depletion: one method is used with peripheral blood stem cells and one for bone marrow. Both these techniques have been successful in preventing both acute and chronic GvHD. You will receive a T-cell depleted stem cell transplant.

The trial proposed is a single arm, phase II treatment protocol designed to examine the engraftment, toxicity, graft-versus-host disease and ultimate disease-free survival following transplants derived from (1) HLA-matched siblings or related donors, (2) HLA-compatible unrelated donors or (3) HLA haplo-type mismatched related donors using a new chemotherapeutic cytoreductive regimen. Candidates for transplant will be stratified according to their donor types.

Candidates for this trial will include patients with high risk forms of ALL, AML or CML, non-Hodgkin's lymphoma, or myelodysplastic syndrome for whom an allogeneic marrow transplant is clearly indicated, and patients with aplastic anemia refractory to ATG or cyclosporine treatment and who are transfusion dependent. All research participants will be conditioned for transplantation with intravenous busulfan (busulfex®) (0.8- 1.0 mg/Kg/dose Q6H x 10 doses), melphalan (70 mg/m2/dose x 2 doses) and fludarabine (25mg/m2/day x 5 doses). Doses of busulfan will be adjusted according to plasma levels. All research participants will also receive ATG (Thymoglobulin®) prior to transplant to promote engraftment. No drug prophylaxis againstGvHD will be administered post transplant. All research participants will also receive G-CSF post-transplant to foster engraftment.

The preferred source of stem cells will be peripheral blood stem cells (PBSC) induced and mobilized by treatment of the donor with G-CSF for 5-6 days. PBSC obtained through 2-3 leukaphereses will be Isolex® 300i separated CD34+ stem cell column selected and E-rosette depleted (E-). The CD34 + E peripheral blood progenitors will then be administered to the research participants after they have completed cytoreduction. If the use of CD34 + E- PBSC is not possible, the alternative graft will consist of bone marrow derived stem cells T-cell depleted by soybean agglutinin and E-rosetting (SBA-E-).

Research participants will be carefully monitored for engraftment, chimerism, incidence and severity of acute and chronic GvHD, regimen-related toxicity, characteristics of hematopoietic and immune reconstitution and ultimate survival and disease-free survival. This phase II trial is designed to investigate the feasibility and safety of a chemotherapy-based cytoreductive regimen plus a T-cell depleted peripheral blood stem cell (PBSC) or bone marrow stem cell transplant (BMT) for the treatment of high risk patients with advanced stages of hematologic malignancies. The majority of research participants will receive grafts derived from PBSC and will be the focus of the trial. The study population will be segmented into three research participant groups based on the type of donors used for the hematopoietic stem cell graft: (1) HLA-identical sibling or related donor, (2) HLA-compatible unrelated donor, and (3) HLA-mismatched related donor.

A maximum of 25 PBSC research participants in both related groups will be accrued onto the study; a maximum of 70 PBSC research participants in the unrelated group will be accrued. In order to reduce patient risk, the study design includes early termination of any trial group in the event of excessive graft failure, grade 3-4 acute graft-versus-host disease, or early transplant related mortality during the accrual period. Excessive failure is defined differently in the three donor groups. Stopping rules for the three research participant populations will be utilized. In addition to the 120 PBSC research participants, we anticipate approximately 25 BMT research participants treated across the three donor groups. These research participants will be followed and at the conclusion of the trial descriptive statistics on this subgroup will be recorded.

Inclusion Criteria:

• Diagnosis: Acute myelogenous leukemia (AML) Acute undifferentiated leukemia (AUL); Acute lymphoblastic leukemia (ALL); Acute biphenotypic leukemia (AbiL); Chronic myelogenous leukemia (CML); Non Hodgkin's lymphoma (NHL); Myelodysplastic syndrome (MDS) Aplastic anemia (AA)
• Status: CR1 first remission; CR2 second remission; CP chronic phase
• HLA-matched Related donors: Research participants who have an HLA-matched related donor are eligible for entry on this protocol. This will include a healthy related donor who is genotypically or phenotypically matched at all A, B and DRB1 loci, as tested by DNA analysis.
• HLA-compatible Unrelated donors: Research participants who do not have a related HLA-matched donor but have an unrelated donor who is either matched at all A, B and DRB1 loci or who is mismatched at 1/6 loci (A, B, or DRB1) as tested by DNA analysis, will be eligible for entry on this protocol.
• HLA-mismatched Related donors: Research participants who do not have a related or unrelated HLA-compatible donor must have a healthy family member who is at least HLA-haplotype identical to the recipient.
• Research participants must have a healthy HLA compatible related or unrelated donor who is willing to receive G-CSF injections and undergo apheresis for PBSC collection, or undergo a marrow harvesting procedure.
• Research participants should be < 55 years. There is no lower age limit. Research participants > 55 years will be accrued on a case by case basis after discussion and approval by the BMT Service.
• Research participants may be of either gender or any ethnic background.
• Research participants must have a Karnofsky (adult) or Lansky (pediatric) Performance Status > 70%
• Research participants must have adequate physical function

Exclusion Criteria:

• Active CNS or skin leukemic/lymphomatous involvement
• Female research participants who are pregnant or breast-feeding
• Active viral, bacterial or fungal infection
• Research participant seropositive for HIV-I/II; HTLV -I/II
• Research participants who have undergone a prior allogeneic or autologous stem cell transplant within the previous six months.