This phase II trial studies the effect of P140K MGMT hematopoietic stem cells, O6-benzylguanine, temozolomide, and carmustine in treating participants with supratentorial glioblastoma or gliosarcoma who have recently had surgery to remove most or all of the brain tumor (resected). Chemotherapy drugs, such as 6-benzylguanine, temozolomide, and carmustine, work in different ways to stop the growth of tumor cells, either by killing the cells, by stopping them from dividing. Placing P140K MGMT, a gene that has been created in the laboratory into bone marrow making the bone more resistant to chemotherapy, allowing intra-patient dose escalation which kills more tumor cells while allowing bone marrow to survive.

Inclusion Criteria:

• Patients with histologically confirmed, newly diagnosed, supratentorial glioblastoma or gliosarcoma who have undergone gross total tumor resection or near gross total resection (resection of >85% of enhancing tumor demonstrated by MRI) are eligible up to 35 days post-operatively. Patients with primarily infratentorial disease, or with multifocal,or leptomeningeal dissemination of disease will be excluded. In general, patients will not have > 1 cm residual measurable or evaluable disease after surgical tumor resection.
• Patient must have unmethylated MGMT
• Absence Of IDH1 or IDH2mutation on tumor tissue by a CLIA-approved immunohistochemistry or DNA sequencing test on local testing
• Patients aged 18-75 years.
• ECOG performance status 0-1or Karnofsky ≥ 70.
• No myelosuppressive chemotherapy or hematopoietic cell transplantation prior to the diagnosis of GBM and no prior chemotherapy (including Gliadel BCNU wafers) for GBM
• Life expectancy of at least 12 weeks.
• No plan for hypofractionated radiation therapy
• Adequate hematologic (absolute neutrophil count (ANC)≥ 1000/mm3, platelets ≥ 100,000/mm3, Hgb ≥ 9.5, hepatic (Bilirubin ≤ 2.0 mg/dl, AST and ALT less than or equal to 3 times institutional upper limit of normal, prothrombin time <1.2 times normal), and renal (serum creatinine ≤ 2.0 mg/dl or Creatinine Clearance ≥ 60mL/min/1.73 m2for subjects with serum creatinine levels above institutional normal). These tests will be repeated within 2 weeks of treatment with BG and TMZ, and must meet the same criteria. -Post-operative steroids are i) tapered to ≤ 8mg dexamethasone/day(or equivalent)and ii) patient has been on a stable or decreasing steroid dose for the 7 days prior to enrollment
• Patients of child-bearing potential must agree to using single barrier contraception.
• Must be willing and able to understand provide informed consent.
• Patient must have all sutures removed prior to registration
• Patient must be considered to be clinically stable.
• The subject will be identified as a candidate for an autologous transplant via an evaluation by a transplant physician per standard of care. Participants will be screened by their transplant physician and social work for a history of substance abuse per screening tool such as SIPAT. Any participant with positive screen for significant substance abuse will undergo evaluation and must have a treatment, management plan in place and must have formal review of medical team prior to initiation of transplant procedures.
• No evidence of active infection.
• Availability of 10unstained slides or FFPE sample of tumor for molecular or histopathological studies.
• Negative screening for Hepatitis B, C and HIV

Exclusion Criteria:

• Any known medical or hereditary condition associated with immunosuppression;orothermedical illness which may jeopardize patient safety.
• Known history of HIV seropositivity. This exclusion is included for two reasons. First, there is evidence of decreased marrow reserve in HIV+ patients and antiviral treatment is associated with myelosuppression. Thus, drug treatment designed to be myelosuppressive may bemore toxic in this patient population. Second, extensive laboratory culturing of the bone marrow and peripheral blood progenitor cells is required. No preclinical samples which are HIV+ have been evaluated with the gene transfer modality proposed and thus the feasibility and safety of gene transfer and selection in HIV+ samples cannot yet be advocated. Such studies are planned so as to not preclude HIV+ patients in later studies.
• Pregnant or lactating women. There is data to indicate that BCNU and TMZ is teratogenic and carcinogenic. Thus, its use in pregnant women would confer unnecessary risk to the fetus.
• Patients with symptomatic pulmonary disease and other severe co-morbid respiratory conditions, including patients with active pulmonary infection and/or pulse oximetry < 90% and a corrected DLCO < 50% of predicted. However, subjects with a corrected DLCO in the range of 50-70% should have Pulmonologyclearance prior to intervention.
• Patients with known diagnosis heart failure or cardiac insufficiency and an LVEF of < 40%. History of acute coronary event including MI within 6 months prior to study enrollment.
• Known history of cardiac arrhythmias including atrial fibrillation, tachyarrhythmiaor bradycardia.Inability to undergo repeated MRI evaluation; or allergy or intolerance of Gadolinium-containing contrast agent.
• Active illicit drug use or diagnosis of alcoholism.
• Prior diagnosis of any malignant disease with the exception of non-melanomatous skin cancer, or carcinoma in situof the cervix, bladder, prostate, or breast, unless patient has been disease-free/in remission for ≥2 years prior to date of study enrollment.
• Mental incapacity or psychiatric illness preventing informed consent.
• History of Hepatitis B or C or Hepatitis grade ≥3 are excluded due to the potential for additional hepatotixicity

Individual participant data that underlie or influence the results observed from the study.

Beginning 3 months and ending 5 years following article publication.

Investigators who provide a methodologically sound proposal for use of requested data.