Participants in stratum S1 will undergo standard dose craniospinal radiation with boost to the primary tumor site once daily 5 days a week for 6 weeks. Six weeks after completion of radiotherapy, patients receive one cycle of chemotherapy (cisplatin, vincristine, cyclophosphamide) once every 4 weeks for 4 cycles in absence of unacceptable toxicity. After completion of 4 cycles of chemotherapy, participants who are skeletally mature will receive maintenance chemotherapy with vismodegib. Some participants will complete aerobic training and/or neurocognitive remediation.

Participants in stratum S2 will undergo high dose craniospinal radiation with boost to the primary tumor site once daily 5 days a week for 6 weeks. Six weeks after completion of radiotherapy, patients receive one cycle of chemotherapy (cisplatin, vincristine, cyclophosphamide) once every 4 weeks for 4 cycles in absence of unacceptable toxicity. After completion of 4 cycles of chemotherapy, participants who are skeletally mature will receive maintenance chemotherapy with vismodegib. Some participants will complete aerobic training and/or neurocognitive remediation.

Participants in stratum N1 will undergo standard dose craniospinal radiation with boost to the primary tumor site once daily 5 days a week for 6 weeks. Six weeks after completion of radiotherapy, patients receive one cycle of chemotherapy (cisplatin, vincristine, cyclophosphamide) once every 4 weeks for 4 cycles in absence of unacceptable toxicity. Some participants will complete aerobic training and/or neurocognitive remediation.

Participants in stratum N2 will undergo standard dose craniospinal radiation with boost to the primary tumor site once daily 5 days a week for 6 weeks. Six weeks after completion of radiotherapy, patients receive standard chemotherapy (cisplatin, vincristine, cyclophosphamide) for 4 cycles intermixed with an additional 3 cycles of chemotherapy with pemetrexed and gemcitabine in absence of unacceptable toxicity. Some participants will complete aerobic training and/or neurocognitive remediation.

Participants in stratum N3 will undergo high dose craniospinal radiation with boost to the primary tumor site once daily 5 days a week for 6 weeks. Six weeks after completion of radiotherapy, patients receive standard chemotherapy (cisplatin, vincristine, cyclophosphamide) for 4 cycles intermixed with an additional 3 cycles of chemotherapy with pemetrexed and gemcitabine in absence of unacceptable toxicity. Some participants will complete aerobic training and/or neurocognitive remediation.

Route of Administration (ROA): Intravenously (IV)

ROA: IV

ROA: IV

ROA: Orally (PO)

ROA: IV

ROA: IV

Progression-free survival (PFS) will be measured from diagnosis to the earliest of disease progression or death from any cause. Patients who have not experienced one of these events will be censored at their last date of contact. PFS in SJMB12 stratum S1 skeletally immature cohort will be compared to historical controls from SJMB03 as well as from another published cohort (Thompson et al., 2016) using log-rank test.

Progression-free survival (PFS) will be measured from diagnosis to the earliest of disease progression or death from any cause. Patients who have not experienced one of these events will be censored at their last date of contact. PFS in SJMB12 stratum S2 skeletally mature cohort will be compared to historical controls from SJMB03 as well as from another published cohort (Thompson et al., 2016) using log-rank test.

Progression-free survival (PFS) will be measured from diagnosis to the earliest of disease progression or death from any cause. Patients who have not experienced one of these events will be censored at their last date of contact. PFS in SJMB12 stratum S2 skeletally immature cohort will be compared to historical controls from SJMB03 as well as from another published cohort (Thompson et al., 2016) using log-rank test.

Overall survival (OS) will be measured from diagnosis to death from any cause. Patients who survive will be censored at their last date of contact. OS in SJMB12 stratum S1 skeletally mature cohort will be compared to historical controls from SJMB03 as well as from another published cohort (Thompson et al., 2016) using log-rank test.

Overall survival (OS) will be measured from diagnosis to death from any cause. Patients who survive will be censored at their last date of contact. OS in SJMB12 stratum S1 skeletally immature cohort will be compared to historical controls from SJMB03 as well as from another published cohort (Thompson et al., 2016) using log-rank test.

Overall survival (OS) will be measured from diagnosis to death from any cause. Patients who survive will be censored at their last date of contact. OS in SJMB12 stratum S2 skeletally mature cohort will be compared to historical controls from SJMB03 as well as from another published cohort (Thompson et al., 2016) using log-rank test.

Overall survival (OS) will be measured from diagnosis to death from any cause. Patients who survive will be censored at their last date of contact. OS in SJMB12 stratum S2 skeletally immature cohort will be compared to historical controls from SJMB03 as well as from another published cohort (Thompson et al., 2016) using log-rank test.

Progression-free survival (PFS) will be measured from diagnosis to the earliest of disease progression or death from any cause. Patients who have not experienced one of these events will be censored at their last date of contact. PFS in SJMB12 stratum N2 will be compared to historical controls from SJMB03 using log-rank test.

Overall survival (OS) will be measured from diagnosis to death from any cause. Patients who survive will be censored at their last date of contact. OS in SJMB12 stratum N2 will be compared to historical controls from SJMB03 using log-rank test.

Progression-free survival (PFS) will be measured from diagnosis to the earliest of disease progression or death from any cause. Patients who have not experienced one of these events will be censored at their last date of contact. PFS in SJMB12 stratum N3 will be compared to historical controls from SJMB03 using log-rank test.

Overall survival (OS) will be measured from diagnosis to death from any cause. Patients who survive will be censored at their last date of contact. OS in SJMB12 stratum N3 will be compared to historical controls from SJMB03 using log-rank test.

Progression-free survival (PFS) in stratum N1 will be compared to historical controls from SJMB03 using log-rank test.

Overall survival (OS) will be measured from diagnosis to death from any cause. Patients who survive will be censored at their last date of contact. OS in SJMB12 stratum N1 will be compared to historical controls from SJMB03 using log-rank test.

Intermediate and high risk Non-WNT Non-SHH medulloblastoma patients treated on strata N2 and N3 will receive pemetrexed and gemcitabine during adjuvant chemotherapy. Percentage of participants who complete treatment will be estimated and reported with an exact 2-sided 95% confidence interval.

Patients with SHH tumor treated on strata S1 and S2 will receive 12 courses of vismodegib during maintenance therapy. For each risk-based stratum separately, percentage of participants who complete treatment will be estimated and reported with an exact 2-sided 95% confidence interval.

To estimate stratum specific cumulative incidence of local disease failure at 2 and 5 years.

To evaluate the effects of an aerobic training intervention, delivered during the radiation therapy period and at home, prior to the start of chemotherapy, on hand grip strength as measured using a hand-held dynamometer among children treated for medulloblastoma. Participants will be seated with the shoulder at 0-10 degrees and the elbow in 90 degrees of flexion. The forearm will be positioned in neutral. Each participant will complete three trials, the average used for analysis.

To evaluate the effects of an aerobic training intervention, delivered during the radiation therapy period and at home, prior to the start of chemotherapy, on range of motion as measured with a goniometer among children treated for medulloblastoma. The goniometer is a reliable and valid measure of active and passive range of motion with standard procedures.

To evaluate the effects of an aerobic training intervention, delivered during the radiation therapy period and at home, prior to the start of chemotherapy, on overall flexibility among children treated for medulloblastoma. Flexibility will be measured by having participants perform a "sit and reach test." A yardstick is placed on a firm flat surface and tape is placed across it at a right angle to the 15 inch mark. The participant sits with the yardstick between the legs, with legs extended at right angles to the taped line on the floor. The heels of the feet touch the edge of the taped line and are 10-12 inches apart. The participant reaches forward with both hands as far as possible, and the best value for three trials, in centimeters, at the fingertips is recorded.

Measured by the Bruininks-Oseretsky Test of Motor Proficiency, Version 2 (BOT-2). To evaluate the effects of an aerobic training intervention, delivered during the radiation therapy period and at home, prior to the start of chemotherapy, on motor proficiency among children treated for medulloblastoma.

To evaluate the effects of an aerobic training intervention, delivered during the radiation therapy period and at home, prior to the start of chemotherapy on health related quality of life (QoL) among children treated for medulloblastoma. QoL will be assessed using the 23-item PedsQL(TM) 4.0 Generic Core Scales which encompasses four subscales (1) physical functioning (eight items), (2) emotional functioning (five items), (3) social functioning (five items), and (4) school functioning (five items) and the 24-item PedsQL(TM) Brain Tumor Module which encompasses six scales: (1) cognitive problems (seven items), (2) pain and hurt (three items), (3) movement and balance (three items), (4) procedural anxiety (three items), (5) nausea (five items), and (6) worry (three items).

To evaluate the effects of an aerobic training intervention, delivered during the radiation therapy period and at home, prior to the start of chemotherapy, on fatigue among children treated for medulloblastoma. Fatigue will be assessed using the 18-item PedsQL(TM) Multidimensional Fatigue Scale which encompasses three subscales: (1) general fatigue (six items), (2) sleep/rest fatigue (six items), and (3) cognitive fatigue (six items).

To evaluate the effects of an aerobic training intervention, delivered during the radiation therapy period and at home, prior to the start of chemotherapy, on memory function at the end of the intervention and at the end of adjuvant chemotherapy, among children treated for medulloblastoma. Memory will be measured using different instruments as age appropriate: CogStage for age 5 and older (Continuous Paired Associate Learning task), California Verbal Learning Test, Children's Version (CVLT-C) for age 6 to < 17 years or California Verbal Learning Test, Second Edition (CVLT-II) for age ≥17 years.

To evaluate the effects of an aerobic training intervention, delivered during the radiation therapy period and at home, prior to the start of chemotherapy, on attention at the end of the intervention and at the end of adjuvant chemotherapy, among children treated for medulloblastoma. Attention will be measured using different instruments as age appropriate: Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition (WPPSI-IV) for ages 3 to < 6 years (Digit Span Forward subtest), Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV) for ages 6 to < 17 years (Digit Span Forward subtest) or Wechsler Adult Intelligence Scale, Fourth Edition (WAIS-IV) for ages ≥17 years (Digit Span Forward subtest). We will also use Conners' Continuous Performance Test, Kiddie Version V.5 for ages 4 to < 6 years and Conner's Continuous Performance Test, Second Edition (CPT-II) for ages ≥6 years, in addition to Cog Stage for children ≥5 years (Detection and Identification tasks).

To evaluate the effects of an aerobic training intervention, delivered during the radiation therapy period and at home, prior to the start of chemotherapy, on executive function at the end of the intervention and at the end of adjuvant chemotherapy, among children treated for medulloblastoma. Executive function is measured as age-appropriate by the following: BRIEF-P (age 3 to <6 years), BRIEF (age 6 to <19 years) and BREIF-A (age ≥19 years). Connor's Parent Rating Scale, Third Edition (CPRS-III) will also be used for age 6 to <19 years, as well as CogState One-Back and Groton Maze tasks for age ≥5 years.

To evaluate the impact of an aerobic training intervention on sleep quality and quantity in children with medulloblastoma. Sleep quality and quantity will be measured by Actigraph accelerometer as well as a sleep diary.

To evaluate the relation between baseline cognitive performance and sleep quality in children with medulloblastoma. Multivariable general linear mixed models will be used to look for associations between sleep quality as measured by Actigraph accelerometer and a sleep diary with measures of cognitive performance (memory, attention and executive function (detailed above) and internalizing and externalizing behavior. Internalizing and externalizing behaviors will be measured using the Behavior Assessment System for Children, 2nd Edition (BASC-2). There are Preschool, Child and Adolescent versions.

To evaluate the relation between baseline cognitive performance and sleep quantity in children with medulloblastoma. Multivariable general linear mixed models will be used to look for associations between sleep quantity as measured by a sleep diary with measures of cognitive performance (memory, attention and executive function and internalizing and externalizing behavior (detailed above).

To evaluate the relation between baseline cognitive performance and fatigue in children with medulloblastoma. Multivariable general linear mixed models will be used to look for associations between fatigue score as measured by the Multidimensional Fatigue Scale with measures of cognitive performance (memory, attention and executive function and internalizing and externalizing behavior (detailed above).

Intellectual function will be measured as age-appropriate using the Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition (WPPSI-IV) for age 3 to < 6 years, Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV) for age 6 to < 17 years, and Wechsler Adult Intelligence Scale, Fourth Edition (WAIS-IV) for age ≥17 years. Linear mixed-effects models will be used to model change in intellectual function over time.

Intellectual function will be measured as age-appropriate using the Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition (WPPSI-IV) for age 3 to < 6 years, Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV) for age 6 to < 17 years, and Wechsler Adult Intelligence Scale, Fourth Edition (WAIS-IV) for age ≥17 years. Linear mixed-effects models and multivariate linear mixed-effects models will be used to examine change in intellectual function with demographic and clinical factors.

Academic ability will be measured using the Woodcock Johnson Tests of Academic Achievement-Third Edition (WJ-III-ACH) which has six subsets: Letter-Word Identification, Passage Comprehension, Reading Fluency, Calculation, Applied Problems, and Math Fluency. Academic ability will also be measured using the Comprehensive Test of Phonological Processing (CTOPP) which has three subsets: Elision, Blending Words, and Rapid Naming. Linear mixed-effects models will be used to model change in academic ability over time.

Linear mixed-effects models and multivariate linear mixed-effects models will be used to examine change in academic ability (measured as described above) with demographic and clinical factors.

Attention will be measured using different instruments as age appropriate: Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition (WPPSI-IV) for ages 3 to < 6 years old (Digit Span Forward subtest), Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV) for ages 6 to < 17 years (Digit Span Forward subtest) or Wechsler Adult Intelligence Scale, Fourth Edition (WAIS-IV) for ages ≥17 years (Digit Span Forward subtest). We will also use Conners' Continuous Performance Test, Kiddie Version V.5 for ages 4 to < 6 years and Conner's Continuous Performance Test, Second Edition (CPT-II) for ages ≥6 years, in addition to Cog Stage for children ≥5 years (Detection and Identification tasks). Linear mixed-effects models will be used to model change in the measure of attention over time.

Linear mixed-effects models and multivariate linear mixed-effects models will be used to examine change in attention (measured as described above) with demographic and clinical factors.

Memory will be measured using different instruments as age appropriate: CogStage for age 5 and older (Continuous Paired Associate Learning task), California Verbal Learning Test, Children's Version (CVLT-C) for age 6 to < 17 years or California Verbal Learning Test, Second Edition (CVLT-II) for age ≥17 years. Linear mixed-effects models will be used to model change in memory over time.

Linear mixed-effects models and multivariate linear mixed-effects models will be used to examine change in memory (measured as described above) with demographic and clinical factors.

Cognitive processing speed function will be measured using the age-appropriate processing speed indices: WPPSI-IV which has two subsets: Bug Search and Cancellation, for age 3 to < 6 years old; WISC-IV which has two subsets: Coding and Symbol Search subtests, for 6 to < 17 years, and WAIS-IV which has two subsets: Coding and Symbol Search subtests, for age ≥17 years. Linear mixed-effects models will be used to model change in cognitive processing speed over time.

Linear mixed-effects models and multivariate linear mixed-effects models will be used to examine change in cognitive processing speed (measured as described above) with demographic and clinical factors.

Neurocognitive executive function will be measured using the Behavioral rating inventory of executive function (BRIEF). There are age-appropriate versions: BRIEF-P for age 3 to < 6 years, BRIEF for age 6 to <19 years, and BRIEF-A for age ≥19 years. Executive function will also be measured using Conner's Parent Rating Scale, Third Edition (CPRS-III), for age 6 to < 19 years and also CogState, for age ≥5 years, One-Back and Groton Maze tasks. Linear mixed-effects models will be used to model change in executive function over time.

Linear mixed-effects models and multivariate linear mixed-effects models will be used to examine change in executive function (measured as described above) with demographic and clinical factors.

To assess the impact of a computer-based working memory intervention, relative to standard of care, on additional performance- and rater-based measure of attention. Attention will be measured using different instruments as age appropriate: Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition (WPPSI-IV) for ages 3 to < 6 years (Digit Span Forward subtest), Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV) for ages 6 to < 17 years (Digit Span Forward subtest) or Wechsler Adult Intelligence Scale, Fourth Edition (WAIS-IV) for ages ≥17 years (Digit Span Forward subtest). We will also use Conners' Continuous Performance Test, Kiddie Version V.5 for ages 4 to < 6 years and Conner's Continuous Performance Test, Second Edition (CPT-II) for ages ≥6 years, in addition to Cog Stage for children ≥5 years (Detection and Identification tasks). T tests will be used to compare changes in attention between the intervention and standard of care groups.

To assess the impact of a computer-based working memory intervention, relative to standard of care, on additional performance- and rater-based measure of processing speed. Cognitive processing speed function will be measured using the age-appropriate processing speed indices: WPPSI-IV which has two subsets: Bug Search and Cancellation, for age 3 to < 6 years; WISC-IV which has two subsets: Coding and Symbol Search subtests, for 6 to < 17 years, and WAIS-IV which has two subsets: Coding and Symbol Search subtests, for age ≥17 years. T tests will be used to compare changes in processing speed between the intervention and standard of care groups.

To assess the impact of a computer-based working memory intervention, relative to standard of care, on additional performance- and rater-based measure of executive function. Executive function is measured as age-appropriate by the following: BRIEF-P (age 3 to <6 years), BRIEF (age 6 to <19 years) and BREIF-A (age ≥19 years). Connor's Parent Rating Scale, Third Edition (CPRS-III) will also be used for age 6 to <19 years, as well as CogState One-Back and Groton Maze tasks for age ≥5 years. T tests will be used to compare changes in executive function between the intervention and standard of care groups.

Working memory intervention baseline=7-10 months after on treatment. Physical exercise intervention baseline=at patient enrollment. Attention will be measured using different instruments as age appropriate: Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition (WPPSI-IV) for 3 to < 6 years (Digit Span Forward subtest), Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV) for 6 to < 17 years (Digit Span Forward subtest) or Wechsler Adult Intelligence Scale, Fourth Edition (WAIS-IV) for ≥17 years (Digit Span Forward subtest). We will also use Conners' Continuous Performance Test, Kiddie Version V.5 for 4 to < 6 years and Conner's Continuous Performance Test, Second Edition (CPT-II) for ≥6 years, in addition to Cog Stage for children ≥5 years (Detection and Identification tasks). ANOVA will be used to compare changes in memory among the groups. In addition, T-tests will be used to compare the combined intervention group to each of the other two groups.

Working memory intervention baseline=7-10 months after on treatment. Physical exercise intervention baseline=at patient enrollment. Cognitive processing speed function will be measured using the age-appropriate processing speed indices: WPPSI-IV which has two subsets: Bug Search and Cancellation, for 3 to < 6 years; WISC-IV which has two subsets: Coding and Symbol Search subtests, for 6 to < 17 years, and WAIS-IV which has two subsets: Coding and Symbol Search subtests, for ≥17 years. ANOVA will be used to compare changes in processing speed among the groups. In addition, T-tests will be used to compare the combined intervention group to each of the other two groups.

Working memory intervention baseline=7-10 months after on treatment. Physical exercise intervention baseline=at patient enrollment. Executive function is measured as age-appropriate by the following: BRIEF-P (3 to <6 years), BRIEF (6 to <19 years) and BREIF-A (≥19 years). Connor's Parent Rating Scale, Third Edition (CPRS-III) will also be used for 6 to <19 years, as well as CogState One-Back and Groton Maze tasks for ≥5 years. ANOVA will be used to compare changes in executive function among the groups. In addition, T-tests will be used to compare the combined intervention group to each of the other two groups.

Attention will be measured using different instruments as age appropriate: Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition (WPPSI-IV) for 3 to < 6 years (Digit Span Forward subtest), Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV) for 6 to < 17 years (Digit Span Forward subtest) or Wechsler Adult Intelligence Scale, Fourth Edition (WAIS-IV) for ≥17 years (Digit Span Forward subtest). We will also use Conners' Continuous Performance Test, Kiddie Version V.5 for 4 to < 6 years and Conner's Continuous Performance Test, Second Edition (CPT-II) for ≥6 years, in addition to Cog Stage for ≥5 years (Detection and Identification tasks). Differences between 6 months following assessment and immediate post-intervention testing, and difference between 6 months following assessment and pre-intervention testing, will be calculated for each subject and then group difference (intervention group vs. control group) will be examined using t-tests.

To evaluate the maintenance of improvements in working memory three months following participation in the computer-based working memory intervention program. Memory will be measured using different instruments as age appropriate: CogStage for ≥5 years (Continuous Paired Associate Learning task), California Verbal Learning Test, Children's Version (CVLT-C) for 6 to < 17 years or California Verbal Learning Test, Second Edition (CVLT-II) for ≥17 years. The differences between six months following assessment and immediate post-intervention testing, as well as the difference between six months following assessment and pre-intervention testing, will be calculated for each subject and then the group difference (intervention group vs. control group) will be examined using t-tests.

To evaluate the maintenance of improvements on processing speed three months following participation in the computer-based working memory intervention program. Cognitive processing speed function will be measured using the age-appropriate processing speed indices: WPPSI-IV which has two subsets: Bug Search and Cancellation, for 3 to < 6 years; WISC-IV which has two subsets: Coding and Symbol Search subtests, for 6 to < 17 years, and WAIS-IV which has two subsets: Coding and Symbol Search subtests, for ≥17 years. The differences between six months following assessment and immediate post-intervention testing, as well as the difference between six months following assessment and pre-intervention testing, will be calculated for each subject and then the group difference (intervention group vs. control group) will be examined using t-tests.

To evaluate the maintenance of improvements in executive function three months following participation in the computer-based working memory intervention program. Executive function is measured as age-appropriate by the following: BRIEF-P (3 to <6 years), BRIEF (6 to <19 years) and BREIF-A (≥19 years). Connor's Parent Rating Scale, Third Edition (CPRS-III) will also be used for 6 to <19 years, as well as CogState One-Back and Groton Maze tasks for ≥5 years. The differences between six months following assessment and immediate post-intervention testing, as well as the difference between six months following assessment and pre-intervention testing, will be calculated for each subject and then the group difference (intervention group vs. control group) will be examined using t-tests.