Phase III Randomized Trial of Single vs. Tandem Myeloablative Consolidation Therapy for High-Risk Neuroblastoma

THIS PROTOCOL IS FOR RESEARCH PURPOSES ONLY, AND SHOULD NOT BE COPIED, REDISTRIBUTED OR USED FOR ANY OTHER PURPOSE. MEDICAL AND SCIENTIFIC INFORMATION CONTAINED WITHIN THIS PROTOCOL IS NOT INCLUDED TO AUTHORIZE OR FACILITATE THE PRACTICE OF MEDICINE BY ANY PERSON OR ENTITY. RESEARCH means a systematic investigation, including research development, testing and evaluation, DESIGNED TO DEVELOP OR CONTRIBUTE TO GENERALIZABLE KNOWLEDGE. THIS PROTOCOL IS THE RESEARCH PLAN DEVELOPED BY THE CHILDREN'S ONCOLOGY GROUP TO INVESTIGATE A PARTICULAR STUDY QUESTION OR SET OF STUDY QUESTIONS AND SHOULD NOT BE USED TO DIRECT THE PRACTICE OF MEDICINE BY ANY PERSON OR TO PROVIDE INDIVIDUALIZED MEDICAL CARE, TREATMENT, OR ADVICE TO ANY PATIENT OR STUDY SUBJECT. THE PROCEDURES IN THIS PROTOCOL ARE INTENDED ONLY FOR USE BY CLINICAL ONCOLOGISTS IN CAREFULLY STRUCTURED SETTINGS, AND MAY NOT PROVE TO BE MORE EFFECTIVE THAN STANDARD TREATMENT. ANY PERSON WHO REQUIRES MEDICAL CARE IS URGED TO CONSULT WITH HIS OR HER PERSONAL PHYSICIAN OR TREATING PHYSICIAN OR VISIT THE NEAREST LOCAL HOSPITAL OR HEALTHCARE INSTITUTION.

For Statistics and Data Center Contact Person see: https://members.childrensoncologygroup.org

SECTION PAGE
STUDY COMMITTEE ..... 6
ABSTRACT ..... 10
EXPERIMENTAL DESIGN SCHEMA ..... 12
1.0 GOALS AND OBJECTIVES (SCIENTIFIC AIMS) ..... 13
1.1 Primary Objectives ..... 13
1.2 Secondary Objectives ..... 13
2.0 BACKGROUND ..... 14
2.1 Definition of High-Risk Neuroblastoma ..... 14
2.2 High-risk Neuroblastoma Therapy ..... 15
2.3 Myeloablative regimens ..... 16
2.4 Tandem Myeloablative Regimens ..... 17
2.5 Novel Induction Regimen ..... 19
2.6 Topotecan ..... 20
2.7 Topotecan and Cyclophosphamide ..... 20
2.8 Local Radiation Therapy ..... 22
2.9 Surgery ..... 23
2.10 Neurologic Outcome of Patients with Epidural Neuroblastoma ..... 24
2.11 Maintenance Therapy ..... 24
2.12 Biologic Correlative Studies ..... 25
2.13 Gender and Race Differences ..... 29
2.14 Peripheral Blood Stem Cell Immunocytology (Amendment #2) ..... 30
3.0 STUDY ENROLLMENT AND PATIENT ELIGIBILITY ..... 30
3.1 Study Enrollment ..... 30
3.2 Patient Criteria ..... 32
4.0 TREATMENT PLAN ..... 34
4.1 General Guidelines ..... 34
4.2 Induction Therapy Administration ..... 35
4.3 Local Control - Surgery ..... 42
4.4 Consolidation Therapy ..... 42
4.5 Consolidation Therapy Regimen A: Single HSCT (CEM) ..... 43
4.6 Consolidation Therapy Regimen B: Tandem HSCT #1 (Thiotepa and Cyclophosphamide) ..... 47
4.7 Radiation therapy ..... 54
4.8 Maintenance Phase: Isotretinoin (Accutane) (13-cis-retinoic acid) Therapy ..... 55
5.0 DOSE MODIFICATIONS FOR TOXICITIES ..... 56
5.1 Myelosuppression During Induction ..... 56
5.2 Hematuria During Induction ..... 56
5.3 Renal Toxicity During Induction ..... 57
5.4 Cardiac Toxicity During Induction ..... 57
5.5 Hepatotoxicity During Induction ..... 58
5.6 Gastrointestinal Toxicity During Induction ..... 58
5.7 Ototoxicity During Induction ..... 59
5.8 Neurologic Toxicity During Induction ..... 59
5.9 Allergic Reactions ..... 59
5.10 Other toxicities During Induction ..... 59
5.11 Dose Modifications for 13-cis-Retinoic Acid (cis-RA) Therapy ..... 59
6.0 DRUG INFORMATION ..... 60
7.0 EVALUATIONS/MATERIAL AND DATA TO BE ACCESSIONED ..... 61
7.1 Required Clinical, Laboratory and Disease Evaluations ..... 61
7.2 Required Observations During Follow-up After Completion of 13-cis-Retinoic Acid Therapy (for All Patients) ..... 66
8.0 OFF PROTOCOL THERAPY CRITERIA AND OFF STUDY CRITERIA ..... 67
8.1 Off Protocol Therapy Criteria ..... 67
8.2 Off Study Criteria ..... 67
9.0 STATISTICAL CONSIDERATIONS ..... 68
9.1 Statistical Design ..... 68
9.2 Patient Accrual and Expected Duration of Trial ..... 69
9.3 Statistical Analysis Methods ..... 70
9.4 Gender and Minority Accrual Estimates ..... 77
10.0 EVALUATION CRITERIA ..... 78
10.1 Common Terminology Criteria for Adverse Events (CTCAE) ..... 78
10.2 Response Criteria for Patients with Solid Tumors ..... 78
10.3 International Staging System ..... 78
10.4 International Response Criteria ..... 78
11.0 ADVERSE EVENT REPORTING REQUIREMENTS ..... 79
11.1 Purpose ..... 79
11.2 Determination of Reporting Requirements ..... 79
11.3 Reporting of Adverse Events for Commercial Agents - AdEERS abbreviated pathway ..... 79
11.4 Routine Adverse Event Reporting ..... 80
11.5 Reporting Secondary AML/MDS ..... 80
12.0 RECORDS AND REPORTING ..... 80
12.1 Categories of Research Records ..... 80
12.2 CDUS ..... 81
13.0 SURGICAL GUIDELINES ..... 81
13.1 Surgical rationale ..... 81
13.2 Pre-operative management ..... 81
13.3 Sampling requirements ..... 81
13.4 Operative management ..... 82
13.5 Management of Surgical Complications ..... 83
13.6 Special techniques ..... 84
13.7 Specimen requirements ..... 85
14.0 PATHOLOGY GUIDELINES AND SPECIMEN REQUIREMENTS ..... 85
14.1 Rapid Pathology Review ..... 85
14.2 Specimen Shipping ..... 87
15.0 SPECIAL STUDIES SPECIMEN REQUIREMENTS ..... 87
15.1 Biologic Requirements ..... 87
15.2 Cyclophosphamide Pharmacogenomics ..... 87
15.3 Topotecan Pharmacokinetics ..... 88
15.4 Minimal Residual Disease Studies - Bone Marrow ..... 89
15.5 Minimal Residual Disease - Peripheral Blood Stem Cells ..... 91
15.6 Minimal Residual Disease - Peripheral Blood ..... 91
15.7 13-cis-Retinoic Acid) Pharmacokinetics ..... 92
15.8 Cellular Immunity Against Neuroblastoma ..... 93
16.0 IMAGING STUDIES REQUIRED AND GUIDELINES FOR OBTAINING ..... 94
16.1 CT Scans ..... 94
16.2 MRI Scans ..... 94
16.3 Skeletal Scintigraphy ..... 95
16.4 MIBG Scintigraphy ..... 95
16.5 Plain Film Radiography ..... 95
16.6 Tumor Measurement ..... 95
16.7 Central Review of Imaging ..... 95
17.0 RADIATION THERAPY GUIDELINES ..... 96
17.1 Indications for Radiation Therapy ..... 97
17.2 Emergency Irradiation ..... 98
17.3 Equipment ..... 98
17.4 Target Volumes ..... 98
17.5 Target Dose ..... 101
17.6 Treatment Technique ..... 102
17.7 Organs at Risk ..... 102
17.8 Dose Calculations and Reporting ..... 102
17.9 Quality Assurance Documentation ..... 103
17.10 Definitions of Deviations in Protocol Performance ..... 105
18.0 HEMATOPOIETIC TRANSPLANT GUIDELINES ..... 105
19.0 RECOMMENDED PROCEDURE FOR PBSC MOBILIZATION AND COLLECTION ..... 105
19.1 Catheter Use ..... 105
19.2 PBSC Mobilization ..... 106
19.3 PBSC Collection ..... 106
19.4 PBSC Analyses ..... 107
19.5 Cryopreservation of PBSC Products ..... 107
19.6 PBSC Infusion ..... 107
APPENDIX I: PERFORMANCE STATUS SCALES/SCORES ..... 108
APPENDIX II: SUPPORTIVE CARE GUIDELINES ..... 109
APPENDIX III: NB WITH INTRASPINAL EXTENSION +/- SPINAL CORD COMPRESSION ..... 112
APPENDIX IV: INTERNATIONAL NEUROBLASTOMA STAGING SYSTEM (INSS) ..... 113
APPENDIX V: TUMOR SIZE MEASUREMENTS BY CROSS-SECTIONAL IMAGING ..... 114
APPENDIX VI: NEUROBLASTOMA RESPONSE CRITERIA ..... 115
APPENDIX VII: NEUROBLASTOMA STUDY ASSIGNMENT TABLE ..... 116
APPPENIX VIII: TOPOTECAN PHARMACOKINETICS DATA COLLECTION FORM ..... 117
APPENDIX IX: 13-CIS RA PHARMACOKINETIC DATA COLLECTION FORM ..... 118
APPENDIX X: SPECIMEN SUBMISSION SCHEDULE ..... 119
APPENDIX XI: YOUTH SUMMARIES FOR CHILDREN AND TEENS ..... 120
REFERENCES ..... 124
SAMPLE INFORMED CONSENT / PARENTAL PERMISSION FOR PARTICIPATION IN RESEARCH ..... 131
PART 1: CONSENT FOR INDUCTION THERAPY ..... 131
PART 2 A: CONSENT FOR PATIENTS RANDOMIZED TO CONSOLIDATION THERAPY ..... 152
PART 2 B: CONSENT FOR PATIENTS TO BE NON-RANDOMLY ASSIGNED TO SINGLE MYELOABLATIVE CONSOLIDATION THERAPY ..... 177

宫

Cyclophosphamide NSC# 26271
Topotecan NSC# 609699
Filgrastim NSC# 614629
Cisplatin NSC# 119875
Etoposide NSC# 141540
Mesna NSC# 113891
Doxorubicin NSC# 123127
Vincristine NSC# 67574
Melphalan NSC# 008806
Carboplatin NSC# 241240
Thiotepa NSC# 6396
Isotretinoin NSC# 329481

The primary goal of ANBL0532 is to test whether further intensification of myeloablative therapy will improve cure rate for high risk neuroblastoma.

Myeloblative consolidation therapy is given at the completion of induction therapy and is categorized into Regimen A and Regimen B. Regimen A consists of one myeloablative consolidation with a carboplatin etoposide - melphalan (CEM) preparative regimen. For Regimen B, patients will receive two myeloablative consolidations: Thiotepa - cyclophosphamide (TC) preparative regimen followed by CEM preparative regimen.

Most patients will be randomized into either Regimen A or Regimen B. Randomization will be stratified by stage, MYCN status and response to induction therapy. Patients 365 to 547 days of age ( 12 - 18 months) with Stage 4, MYCN nonamplified tumor with unfavorable histopathology or diploid DNA content or with indeterminant histology or ploidy and patients who are greater than 547 days of age with Stage 3, MYCN nonamplified tumor AND unfavorable histopathology or indeterminant histology will be nonrandomly assigned to single myeloablative transplant (Regimen A).

ANBL0532 will also assess the efficacy of a dose intensive topotecan containing induction regimen, substituting 2 cycles of dose intensive cyclophosphamide and topotecan for the initial 2 cycles of induction utilized on the prior COG A3973 protocol. Patients will receive 6 cycles of induction chemotherapy as outlined in the schema below. Patients will undergo peripheral blood stem cell (PBSC) harvest after 2 cycles with no ex vivo manipulation prior to cryopreservation. PBSC collection will be performed prior to randomization. The goal of apheresis will be collection of divided into 3 aliquots. A minimum collection of , divided into 2 aliquots is required for each patient while an additional aliquot of is strongly recommended to be stored as backup for delayed engraftment or for future use. Tumor response will be assessed after 2 cycles of induction chemotherapy. Patients will undergo surgical resection after Cycle 5 (or 6 if necessary). Tumor response will be assessed after completion of surgical resection and 6 cycles of chemotherapy (end-induction). Patients with progressive disease (PD) at end of induction evaluation will be taken off protocol therapy, all other patients will continue on therapy.

ANBL0532 will also test whether additional radiation therapy delivered to gross residual disease improves local control for those patients with less than a gross total resection. After completion of transplant and recovery from acute toxicities, patients will undergo radiation to primary site of disease
and to MIBG-avid sites seen at pre-transplant (end-induction) evaluation. Patients who achieve end of induction primary site complete response (CR) will receive 21.6 Gy external beam radiation therapy (EBRT) to the primary site (standard dose) while patients achieving ( residual soft tissue) at the primary site will receive an additional boost of 14.4 Gy for a total dose of 36 Gy EBRT delivered to gross residual primary site disease. Pre-surgical and end of induction CT or MRI scans and operative reports will be centrally reviewed in real time for all patients assessed by the treating institution to have a complete response at the primary site at the end of induction. Tumor response will be assessed after completion of radiation therapy.

Patients will be encouraged to enroll onto ANBL0032 or ANBL0931for Maintenance biologic therapy. Alternatively, post-transplant Maintenance therapy with cis-RA daily for 14 days every 28 days repeated for 6 months will be administered.

Pharmacogenomics studies will be performed to evaluate whether host genetic polymorphisms in cyclophosphamide metabolizing enzymes are associated with cyclophosphamide/topotecan - related toxicity and tumor response. Cis-RA pharmacokinetics and pharmacogenomic studies will be performed to evaluate whether the variability of cis-RA continuous steady state plasma levels and/or genetic variations in retinoic acid metabolic enzymes correlate with event-free survival or systemic toxicities in high-risk neuroblastoma patients. Total topotecan levels will be assessed to formulate a population-based dosing model for topotecan in children with high-risk neuroblastoma.

Post-operative complications will be studied prospectively in patients with high risk neuroblastoma to better characterize the incidence and timing of occurrence. Specifically, the occurrence of bowel obstruction, renal atrophy, chylous leak and chronic diarrhea will be evaluated through 12 months from completion of therapy. Rates of these complications will be correlated with the surgical and radiotherapeutic efforts in obtaining local control as they may impact on future management decisions. The neurologic and orthopedic outcome for patients with intraspinal extension of primary tumors (referred to as paraspinal tumors) will be studied prospectively on this protocol, to better characterize the multiple aspects of disability that these patients may be at risk for. It is the intent of this study that patients with clinical signs of spinal cord compression will be treated with primary chemotherapy; however, multidisciplinary discussion among oncology, neurosurgery, and radiation oncology is strongly encouraged at the time of diagnosis.

Clinical outcome data for patients treated on this study will be used in conjunction with ongoing biologic studies on ANBL00B1 to identify new biological surrogate markers for disease relapse and/or disease progression. In addition, immunologic assays will be employed to assess functional cellular immunity against neuroblastoma at diagnosis and after myeloablative therapy and to enumerate T-cell lymphocyte recovery following single versus tandem myeloablative therapy.

Children's Oncology Group

1.1.1

To improve the 3-year event free survival (EFS) rate of high-risk neuroblastoma patients through treatment with a tandem consolidation of Thiotepa/Cyclophosphamide followed by Carboplatin/Etoposide/Melphalan (CEM) as compared to single CEM consolidation.
1.1.2

To improve the rate of end-induction complete response and very good partial response, compared to historical controls, by use of a topotecan-containing induction regimen.
1.1.3

To improve the 3 -year local control rate, compared to historical controls, by increasing the local dose of radiation to the residual primary tumor for patients with less than a gross total resection.

1.2.1

To evaluate the pharmacogenetic relationship of cyclophosphamide-metabolizing enzymes (CYP2B6, CYP2C9 and GSTA1 genotypes) with toxicity and response following dose intensive cyclophosphamide and topotecan induction chemotherapy.
1.2.2

To determine if resection completeness is predictive of a) local control rate; or b) event free survival rate in patients with high-risk neuroblastoma.
1.2.3

To prospectively describe the complications related to efforts at local control (surgery and radiation therapy) in patients with high risk neuroblastoma.
1.2.4

To describe the neurologic outcome of patients with paraspinal primary neuroblastoma tumors.
1.2.5

To determine the variability of 13-cis-retinoic-acid pharmacokinetics and relationship to pharmacogenomic parameters and determine if pharmacokinetics and/or genetic variations correlate with EFS or systemic toxicity as follows:
a) To determine the variability of 13 -cis-retinoic-acid pharmacokinetics and relationship to pharmacogenomic parameters.
b) To determine if 13-cis-retinoic-acid pharmacokinetic levels are predictive of the EFS rate or associated with systemic toxicity following 13-cis-retinoic acid.
c) To determine if pharmacogenomic variations are predictive of the EFS rate or associated with systemic toxicity following 13-cis-retinoic acid.
1.2.6

To evaluate total topotecan pharmacokinetics and correlate with patient specific data for use in an ongoing topotecan population pharmacokinetic analysis.
1.2.7

To evaluate the presence and function of T cells capable of recognizing neuroblastoma by assessing:

Children's Oncology Group
a) If T cells recognizing the neuroblastoma antigen survivin circulate at diagnosis;
b) If these T cells can be expanded using autologous antigen presenting cells (APCs);
c) If these T cells will kill neuroblastoma cells as detected in functional assays; and
d) If the presence and activity of anti-neuroblastoma immunity is decreased by stem cell transplant.
1.2.8

To characterize the recovery of T cell numbers after myeloablative consolidation and hematopoietic stem cell transplant (HSCT) and assess the impact of tandem myeloablative consolidation on T cell recovery.
1.2.9

To characterize minimal residual disease burden using RT-PCR evaluation of a panel of neuroblastomaspecific transcripts in patient bone marrow and peripheral blood following induction chemotherapy and after single versus tandem myeloablative chemotherapy and to evaluate impact on EFS.
1.2.10

To evaluate the EFS and OS rate for patients months with Stage 4, nonamplified tumor with unfavorable histopathology or diploid DNA content or with indeterminant histology or ploidy and patients who are greater than 547 days of age with Stage 3, MYCN nonamplified tumor AND unfavorable histopathology or indeterminant histology following treatment with single myeloablative transplant.

Neuroblastoma, a neoplasm of the sympathetic nervous system, is the second most extracranial malignant solid tumor of childhood. The current survival for a child year at diagnosis with Stage 4 neuroblastoma is only Novel therapies to improve initial disease response and treatment of minimal residual disease are required to improve survival for these children with high-risk neuroblastoma. Standard therapy for patients with high-risk neuroblastoma involves at least 4 components: Induction, Local Control, Consolidation and Treatment of minimal disease with biologic agents. The use of these 4 components has evolved over the last 20 years, based upon work by the POG and CCG, international cooperative groups and smaller cohort studies.

The primary aim of ANBL0532 is to improve event free survival for children with high-risk neuroblastoma. This study is a logical extension of the COG Neuroblastoma committee's strategic plan and will build upon our previous findings to study the use of intensified consolidation (Primary Aim 1), the use of a novel induction regimen (Primary Aim 2), and the use of increased dose local radiation therapy (Primary Aim 3). Specifically, a limited institution pilot study through COG (ANBL00P1) has demonstrated the feasibility and tolerable toxicity of the proposed tandem transplant consolidation regimen. An additional limited institution pilot study through COG (ANBL02P1) has demonstrated feasibility/tolerable toxicity of the dose intensive cyclophosphamide/topotecan administration proposed in the induction phase and ability to harvest PBSC following such chemotherapy. ANBL0532 will also test whether additional radiation therapy will improve local control for those patients with less than a gross total resection, a hypothesis derived directly from retrospective analysis of the prior CCG 3891 high-risk neuroblastoma protocol. Patients completing this study will be eligible for COG ANBL0032 or ANBL0931, a study of immunotherapy as treatment of minimal residual disease.

Patients eligible for ANBL0532 must have high-risk neuroblastoma as defined by risk criteria that include age, stage, and the biologic features of MYCN amplification, histopathology, and DNA ploidy (Appendix VII). The stratification algorithm to be used for ANBL0532 and ANBL0531 (Intermediate Risk Neuroblastoma clinical trial) has been revised from recent clinical trials and is based upon data from legacy CCG and POG clinical trials. These data suggest that patients up to 18 months of age who are diagnosed with biologically favorable stage 4 neuroblastoma share the same excellent prognosis that has previously been demonstrated for infants less than 12 months of age with non-amplification of MYCN

Children's Oncology Group
oncogene (MYCN-NA) stage 4 disease. Data from CCG 3881 and 3891 trials demonstrated a 6-year for infants months, for toddlers months, and for children 18-24 months of age at diagnosis of stage -NA neuroblastoma. In this analysis, Shimada histology also correlated with age at diagnosis and survival, with nearly all patients with favorable Shimada histology (FH) tumors being in the 12-18 month age group. Data from POG clinical trials also suggest a favorable outcome for toddlers ages 12-18 months with stage D, MYCN-NA, hyperdiploid tumors, treated on a variety of POG protocols with a 4 - year EFS of , and for patients age 6-12 months, 12-18 months, and 18-24 months, respectively, at diagnosis. Further analyses of CCG 3891 data reveal an excellent outcome for the rare patient age 12-18 months with stage 3, MYCN-NA but unfavorable Shimada histology (UH) disease. Measurement of tumor DNA index was not performed in this study. The 3 -year EFS/OS was for patients months with Stage 3, MYCN-NA and UH neuroblastoma ( ).

Based upon these data, ANBL0532 will exclude patients age 12-18 months with stage 4, MYCN-NA, hyperdiploid, and FH tumors or stage -NA, unfavorable histopathology. These patients will be treated on the intermediate-risk protocol, ANBL0531, representing a reduction of therapy in both doseintensity and cumulative dose. Of note, this clinically favorable cohort of 12-18 month old patients with stage 4 disease represents only of all stage 4 patients enrolled onto CCG3891 for whom MYCN analyses were available and only of all stage D patients enrolled onto POG trials for whom and DNA ploidy analyses were available.

In contrast, patients 12-18 months of age with stage -NA tumors characterized by either diploid DNA content (4-year EFS 66.6%) or UH tumors (6-year EFS 46%) and patients greater than 18 months of age with stage 3 tumors and unfavorable histology (5-year EFS 53%) have a less favorable outcome and support further attempts to optimize therapy. However, given that the majority of these patients treated on legacy CCG or POG trials did not receive myeloablative therapy, a further escalation to tandem transplantation may not be warranted. Therefore, patients 12-18 months of age with stage 4 MYCN-NA tumors but with diploid DNA content, UH or indeterminant histology or DNA ploidy and children greater than 18 months of age with stage -NA tumors with UH, estimated to be of all high risk stage patients, will be eligible for treatment on ANBL0532 but will be nonrandomly assigned to single myeloablative therapy. A uniformed treatment approach will allow improved analysis of outcome for this relatively rare patient cohort.

High-risk neuroblastoma is largely chemotherapy responsive, but despite improvements in complete response rates, the majority of patients still relapse. To improve event free survival, COG and legacy CCG and POG clinical trials have intensified chemotherapy during induction and consolidation phases of therapy and added biologic agents for treatment of minimal residual disease. CCG-3891 study demonstrated that increased dose intensity of consolidation with purged bone marrow transplant and administration of the biologic agent 13 -cis retinoic acid (cis-RA) following consolidation therapy improved outcome for patients with high-risk neuroblastoma . The 8 -year EFS and OS rates for patients treated with one transplant and cis-RA ( ) are and , respectively (personal communication; K. Matthay and W.London). The recently completed COG Phase III trial, A3973, evaluated whether the use of a more intensive induction chemotherapy regimen (N7), reported by investigators at Memorial Sloan-Kettering to produce a CR/VGPR rate, will result in improved EFS and end-induction response rate when compared to CCG-3891. A3973 also tested whether the use of anti-neuroblastoma purged PBSC rescue improves EFS for patients with high-risk neuroblastoma. The results of COG A3973 indicate no difference in event-free or overall survival with the use of purged PBSC products for transplant. Feasibility and toxicity data from COG A3973, ANBL00P1 and ANBL0P2 cited below support the continued utilization of a dose intensive induction regimen and myeloablative chemotherapy consolidation regimen as the "backbone" schema as proposed in this successor study for high-risk neuroblastoma. The current COG ANBL0032 has demonstrated that immunotherapy (ch14.18 antibody plus cytokines) in addition to cis-RA improves EFS. Patients treated on ANBL0532 will be eligible for participation on ANBL0032 or ANBL0931.

The results of the randomized CCG-3891 protocol demonstrate that a myeloablative therapy with carboplatin, etoposide, melphalan and 1000 cGy total body irradiation (TBI) and purged autologous bone marrow rescue improves EFS for patients with high-risk neuroblastoma. The 3 -year EFS was for all patients on study, of which had stage 4 neuroblastoma. The 3-year EFS from time of assigned randomization was for autologous bone marrow transplantation (ABMT) versus for continuation chemotherapy (CC) ( ). Overall survival was not significantly different for the two regimens, perhaps partly due to use of ABMT salvage in patients who progressed after chemotherapy consolidation. An apparent advantage was seen for ABMT within strata separated either by MYCN amplification, bone marrow immunocytology or stage. More recent evaluation of outcome for those children with high-risk neuroblastoma treated on CCG 3891 revealed an overall 8-year EFS and OS rate +/- SE from the time of enrollment onto CCG 3891 of and , respectively, with a median follow-up time of patients who did not experience an event of 7.7 years ( 130 days -12.8 years) (Table 1). The 8 -year EFS from the time of randomization for patients randomized to ABMT ( ) was vs. for patients randomized to CC . The 8 -year OS for ABMT was not statistically significantly higher than for CC, vs. , (personal communication, K. Matthay and W. London). These data continue to suggest that therapy intensification diminishes risks for neuroblastoma recurrence. Based on the incremental improvement in EFS with ABMT, myeloablative consolidation has become a standard of care for patients with high-risk disease in the United States and abroad.

While TBI was used in the initial studies of autologous transplantation for neuroblastoma , the worldwide trend has been to eliminate this modality due to excessive toxicity including second malignancies. It is now clear that retrospective comparisons of TBI and non-TBI containing regimens showed no difference in patient outcomes. Smedler et al. reported a moderate general developmental delay; with motor, perceptual and cognitive defects in ten children transplanted for leukemia or neuroblastoma with TBI regimens. Children transplanted for neuroblastoma with TBI also demonstrate abnormal growth rates, in some cases associated with growth hormone deficiency. TBI regimens can also cause late effects such as cataracts, other endocrine dysfunction, and secondary malignant neoplasms. There has been no prospective comparative trial to determine whether the addition of TBI improves survival from high-risk neuroblastoma. A retrospective review of 51 patients transplanted for advanced or poorly responding neuroblastoma from the European BMT Solid Tumor Registry found no significant difference in outcome between TBI and non-TBI regimens. A separate retrospective review of 62 concurrently treated stage 4 neuroblastoma patients with and without TBI also demonstrated equivalent survival rates. The CHLA group has shown a 3-year EFS rate of in 73 patients analyzed from the end of induction therapy who were treated on protocol 91-LA6 with a non-TBI regimen of carboplatin, etoposide and melphalan with local irradiation (CEM-LI, Table 1). A preliminary comparison of the EFS of patients with stage 4 neuroblastoma over one year of age at diagnosis transplanted before progression with CEM-LI ( ) with the same cohort of patients treated with CEMTBI on CCG-321P3 ( ) showed no significant difference in outcome ( ). Overall, there were 4/73 (5%) toxic deaths in the patients transplanted before progression with CEM on the 91-LA6 protocol as compared to toxic deaths following CEM-TBI in similar patients transplanted on CCG-321P3. These data suggest that the proposed CEM-LI without TBI is at least as effective as prior TBI regimens with the caveats that sample size limited the ability to detect a statistical difference in EFS and that trend towards improvement in EFS in this historical comparison could also be attributed to general improvements in care and induction regimens used in the 91-LA-6 (personal communication, J. Villablanca).

The COG A3973 recently completed Phase III trial utilized the CEM preparative regimen based upon the results for the 91LA6 clinical trial noted above. Of the 489 eligible patients enrolled onto A3973, 368 had completed preparative regimen of carboplatin, etoposide, melphalan and autologous transplant with either purged or unpurged PBSC and reported on toxicity. Nearly of the patients developed Grade 3 or 4 stomatitis during consolidation. One patient developed Grade 3 cardiac dysfunction, and four had renal
dysfunction. Three patients developed renal failure requiring dialysis. Two of these patients had severe sinusoidal-obstruction syndrome (SOS), formerly VOD, the other had severe capillary leak syndrome with grade 2 elevation of bilirubin, but no hepatic enlargement. All 3 patients proceeded to post transplant therapy on study. In total, 12 of 368 ( ) patients have died of complications during Consolidation phase of therapy, mainly due to infectious complications ( ). There have been 20 patients reported with SOS, which resulted in death in the 3 patients. These toxicities are similar to those reported using alternative myeloablative consolidation regimens and did not meet designed stopping rules. Twoyear EFS for the tranpslanted cohort from time of study enrollment was . The 91LA6 and COG A3973 experience to date suggest tolerable toxicity following CEM-LI consolidation therapy. Comparative analysis of 91LA6, CCG 321-P3 and A3973 indicates that the CEM-LI regimen without TBI is at least as effective as prior TBI regimens. Therefore, CEM-LI will be employed in each arm of this successor Phase III study comparing single versus tandem consolidation therapy.

High-risk neuroblastoma remains incurable in of patients. Further dose intensity may yield further improvements in EFS. To achieve this, tandem, or dual-cycle, transplant approaches have been developed. An initial attempt at tandem transplant using marrow support was piloted by the European Neuroblastoma Group in the LMCE2. Median time to reach the second ABMT was 3 mos. (range 2-6 mos.), and median time to ANC was 43 days (15-120). Overall survival at 2 years was in a very high-risk group of patients, but the toxic death rate was high at .

Grupp et al. explored the use of tandem transplant with PBSC support in children with high-risk solid tumors. The myeloablative regimens employed in the CHP-594/DFCI 34-DAT study were an initial regimen of etoposide, cyclophosphamide and carboplatin followed 4-6 weeks later by a second regimen of melphalan with 12 Gy of TBI. Initial data on 55 enrolled patients (ages 1 to 18 years) demonstrated the feasibility of the approach with 98 cycles of myeloablative therapy with stem cell rescue completed. Four patients who completed the first HSCT course did not complete the second, and there were three toxic deaths. Median time to neutrophil engraftment was 11 days. One patient experienced delayed engraftment but recovered following infusion of back-up PBSC. The 3-year EFS was with a median follow-up of 22 months from diagnosis. These data are substantiated by a more recent analysis of 97 consecutive children (including the 55 previously published) treated with the CHP-594/DFCI 34-DAT that shows a 3 -year PFS and OS of and ), respectively; a 5 -year PFS and OS of ( CI ) and ( CI ), respectively and a 7 -year PFS of Of the 89 patients who received at least one HSCT, the 5 -year and 7 -year PFS was and , respectively (Table 1).

Toxicity and feasibility data on tandem transplantation without TBI have been evaluated in POG 9640 and COG ANBL00P1 studies. POG 9640 enrolled 33 patients and utilized 5 cycles of induction chemotherapy followed by tandem transplantation with unpurged PBSC support. The tandem regimens were: Carboplatin, etoposide, cyclophosphamide (AT#1) followed by thiotepa and cyclophosphamide (AT#2). This regimen was found to be feasible and tolerable. There were no toxic deaths associated with the tandem regimen. A total of 21 patients underwent at least one course myeloablative therapy and 17 patients received tandem regimens. Of the 17 patients who underwent tandem transplant, 12 were alive without evidence of disease at 2 years. The 5 -year EFS and OS rates +/- standard error of the 33 patients enrolled was and , respectively. Twenty-two patients received at least one transplant, and the 5-year EFS and OS rates from time of first transplant were and , respectively (personal communication S. Grupp and W. London, MS in preparation).

The promising results of the CHOP/DFCI experience, the POG 9640 study and the 91LA6 CEM-LI study have led the COG Neuroblastoma Committee to consider intensification of CEM-LI in a tandem regimen. The feasibility and toxicity of incorporating CEM-LI in a tandem regimen has been studied in COG ANBL00P1. This limited-institution pilot study utilized five cycles of induction chemotherapy followed by tandem transplantation. The tandem regimens were Thiotepa/Cyclophosphamide followed by CEM-LI. Thiotepa/Cyclophosphamide regimen was chosen as the initial transplant based upon the known toxicity

Children's Oncology Group
profile of agents alone and in combination when delivered at myeloablative doses, the acceptable toxicity profile of the combination demonstrated in the POG 9640 tandem transplant study and the known anti-neuroblastoma activity of each agent including their ability to cross the blood brain barrier. Data obtained from a CHOP pilot tandem regimen of Thiotepa/Cyclophosphamide followed by full dose CEM (equivalent to A3973 dosing) revealed excessive toxicity following CEM. Therefore, a dose reduction of CEM agents was utilized in the ANBL00P1 protocol and will be utilized in ANBL0532. The observed toxicity following CEM in the CHOP pilot and in the COG A3973 trial suggest that CEMrelated toxicity would interfere with subsequent delivery of myeloablative chemotherapy and support the sequence of thiotepa/cyclophosphamide followed by CEM in a tandem myeloablative regimen.

Of the 41 eligible patients enrolled onto ANBL00P1, 8 patients were removed from protocol prior to consolidation. Of the remaining 33 patients, 7 received only a single transplant while 26 patients received tandem transplants. Reasons for not receiving the 2 nd transplant included progressive disease ( 1 patient), family preference (2 patients), mortality following transplant #1 ( ) and prolonged toxicity following transplant #1 ( , delayed engraftment, elevated liver transaminases, renal insufficiency). Therefore, of all patients and of those patients eligible for consolidation therapy received the tandem consolidation regimen. This compares favorably to the legacy CCG3891 trial where of patients assigned to ABMT received the intended therapy. Overall, 3 patients have experienced treatment related mortality. Twenty-two events occurred in 41 eligible patients enrolled at diagnosis resulting in 2 -year EFS from time of diagnosis of . These data support the feasibility of this tandem myeloablative chemotherapy regimen, although small numbers of accrued patients limits the ability to assess efficacy.

The summary data presented in Table 1 support the investigation of tandem transplant consolidation to improve EFS from high risk neuroblastoma. A comparison of the CCG3891 and CHP/DFCI trials, both TBI-containing trials, reveals an improvement in EFS following tandem consolidation; (CCG 3891 from time randomized to ABMT: 8-year EFS 27+/- 5% versus CHP/DFCI from diagnosis: 7-year EFS 45+/ versus CHP/DFCI from time on initial myeloablative regimen: 7 -year EFS ). A trend toward improvement in EFS following tandem transplant is even observed when comparing the highly selected cohort of CCG3891 patients who received ABMT and were randomized to receive cis-RA ( of those patients patients randomized to ABMT) versus the CHP/DFCI patients who received the intended tandem transplant ( of patients enrolled); CCG 3891: 8-year EFS versus CHP/DFCI 7-year EFS 52 +/- 11%. Furthermore, the 91LA6 data suggests that CEM provides similar EFS to the single TBI-CEM containing regimen used in CCG3891, supporting the elimination of TBI from a tandem regimen. The EFS and OS results of the various non-TBI containing tandem regimens (ANBL00P1 and POG 9640) vary between studies, reflecting the difficulty in evaluating efficacy of therapy regimens in small, non-randomized clinical trials. However, the combined data from the CHP/DFCI, the POG 9640 and the ANBL00P1 trials demonstrate that intensification of consolidation therapy with PBSC support is feasible, does not increase treatment related mortality and may decrease recurrence for patients with high-risk neuroblastoma.

Children's Oncology Group

1 = Data represent 8yr EFS data for CCG 3891; EFS calculated from time of enrollment at initial diagnosis; EFS calculated from time of initial HSCT for those patients who received ABMT and were randomized to receive cis-RA (CCG3891 ); 4= EFS calculated from time of initial HSCT (91LA6 ), or from time of initial HSCT who received at least one HSCT regimen (CHP594/DFCI34 , POG9640 N=22, ANBL00P1 ); calculated from time of initial HSCT for those patients who received tandem HSCT per CHP 594/DFCI34 ( ); 6= protocol only required a minimum collection of cells ; median collection cells (range ). Not available

The emergence of chemotherapy-resistant tumor cells is the major obstacle to improved initial tumor response and to the cure of high-risk neuroblastoma. A marked escalation in chemotherapy dose intensity has resulted in improved initial tumor response rates. A highly dose intensive combination of cyclophosphamide, vincristine and doxorubicin alternating with cisplatin and etoposide has been studied in small numbers of patients at Memorial Sloan Kettering Cancer Center with a reported CR plus very good partial response rate (VGPR) of Unfortunately, this excellent induction response rate was not reproduced in a multi-center trial performed by the French Society of Pediatric Oncology (SFOP) where 21 of 47 patients who received the MSKCC induction regimen achieved CR at metastatic sites of disease, results that were similar to prior less dose intensive regimens. This regimen was further evaluated in over 400 patients with high risk neuroblastoma enrolled on the recently completed COG Phase III trial (A3973), demonstrating a CR plus VGPR of . Data from the A3973 and SFOP trials suggest that chemotherapy resistance remains an obstacle to the cure of high risk neuroblastoma. Further dose intensification of induction chemotherapy is limited by hematopoietic and mucosal toxicity arguing for the addition of new non-cross resistant agents to further improve tumor response rates.

Topotecan, 9-dimethylaminomethyl-10-hydroxycamptothecin, is a semi-synthetic derivative of camptothecin that inhibits topoisomerase I enzyme resulting in double stranded DNA breaks and cellular apoptosis. Preclinical investigations in murine xenograft models have demonstrated antitumor activity against neuroblastoma, synergistic activity when combined with alkylator and platinum anti-tumor compounds and both dosage and schedule dependency of anti-tumor activity. Anti-tumor activity demonstrated in xenograft models is maximal following daily x 5 days administration or daily x 5 days x 2 weeks at single day systemic exposure of , a systemic exposure proven to be feasible in subsequent clinical trials. Although topotecan demonstrates a non-cross resistant mechanism of action cross-resistance to etoposide, a topoisomerase II inhibitor commonly used in neuroblastoma therapy, has been demonstrated by in vitro analyses of patient-derived neuroblastoma cell lines resistant to etoposide. Patient-derived neuroblastoma cell lines established early in the clinical course (diagnosis or after induction therapy) were more likely to be sensitive to topotecan, supporting the incorporation of topotecan into Induction therapy, prior to significant etoposide exposure. In vivo, topotecan is eliminated through hepatic metabolism and renal excretion. Topotecan penetrates the CNS, a site of recurrence in neuroblastoma. Topotecan has a limited toxicity profile notable for dose-limiting myelosuppression and mild to moderate nausea, vomiting and mucositis, regardless of administration schedule. Antineuroblastoma activity has been demonstrated in both Phase I and II trials. A Phase I trial administered topotecan at dosages of day to day for 5 days to pediatric patients with recurrent tumors. Of nine patients with neuroblastoma, a partial response was demonstrated in 3 patients while stable disease occurred in one patient. A Phase II upfront window trial in patients with newly diagnosed high-risk neuroblastoma (POG-9341) identified topotecan ( day for 5 days) as an active antineuroblastoma agent with of patients achieving a complete ( ) or partial tumor response ( ) following 2 cycles of single agent therapy. In a more traditional topotecan single agent Phase II trial (topotecan day for 5 days) enrolling 37 patients with refractory neuroblastoma, complete response (CR) was achieved in 2 patients, mixed response (MR) in 5 patients and stable disease (SD) in 8 patients. These data are consistent with topotecan's anti- neuroblastoma efficacy documented in preclinical studies and support the incorporation of topotecan into upfront therapy for neuroblastoma.

Topotecan and cyclophosphamide have been successfully combined without excessive toxicity and have anti-neuroblastoma activity. A phase I trial in pediatric patients with refractory or recurrent solid tumors demonstrated the topotecan MTD as day administered over 30 minutes daily for 5 days in combination with cyclophosphamide day administered daily for 5 days. Myelosuppression, defined as Grade 4 neutropenia, was the dose limiting toxicity without additional significant toxicity observed. Anti-tumor activity was further evaluated in a Phase II trial of topotecan day x 5 days combined with cyclophosphamide day for 5 days administered to 83 pediatric patients with refractory solid tumors. Six of thirteen patients with refractory/recurrent neuroblastoma ( 5 received prior HSCT) achieved a PR ( 3 had received a prior BMT), while 2 patients achieved mixed response or stable disease. Preliminary data from POG-9642, a randomized study of topotecan (2 day for 5 days) versus topotecan/cyclophosphamide (at Phase II doses noted above) in patients with recurrent neuroblastoma demonstrate a trend toward improved CR/PR rate following topotecan/cyclophosphamide therapy ( versus 20%), decreased rate of grade 3 or 4 infectious complications and increased time to tumor progression (Frantz C., COG 9462 Public Report, 2/11/2003).

Xenograft models predict that higher systemic exposure ( ) to topotecan will improve tumor response. A topotecan systemic exposure of is predicted to be achieved by doses of 1.0 to day. Limited pediatric trials, both non-myeloablative and myeloablative, have incorporated such doses of topotecan. Repetitive delivery of a dose-intensive topotecan and cyclophosphamide regimen (MSKCC regimen), continuous infusion high-dose cyclophosphamide ( 4200 hours) combined with continuous infusion topotecan ( hours), has been administered to pediatric patients with recurrent solid tumors. Severe neutropenia was observed following all courses of therapy, with of courses complicated by fever and complicated by
bacteremia. Median duration of neutropenia was 7 days (range days); however, all patients recovered neutrophil count to and platelet to by 28 days. Nonhematologic toxicity was limited to Grade 4 hyperbilirubinemia and Grade 3 cardiac dysfunction in one patient during documented sepsis and three episodes of Grade 3 mucositis, two associated with herpes simplex infection. There were no toxic deaths. Cyclophosphamide and topotecan doses of day for 3 days and 1.25 to 4 day for 5 days have been successfully combined with melphalan as a myeloablative regimen for adult patients with advanced ovarian cancer. Non-hematopoietic toxicity was limited to mucositis with maximum of Grade 2 toxicity occurring in 16 of 53 patients treated.

The limited toxicity profile following dose-intensive administration of topotecan and cyclophosphamide support the potential feasibility for its incorporation into an upfront induction regimen. To this end, COG ANBL02P1 examined the feasibility of adding dose intensive topotecan and cyclophosphamide onto the N7 backbone chemotherapy administered in COG A3973 study. Escalated doses of topotecan (1.2 day for 5 days) and cyclophosphamide ( day for 5 days) (T/C) replace the initial 2 cycles of A3973 induction chemotherapy allowing incorporation of topotecan while maintaining dose intensity of known active agents in the treatment of neuroblastoma (see Table 2). The topotecan dose of day days was chosen based upon a predicted topotecan AUC ( *hr) that is within the range known to produce tumor response in murine xenograft models and the ability to administer with cyclophosphamide. Chemotherapy cycles were scheduled every 21 days, PBSC harvest occurred after T/C cycles and surgical resection of residual primary tumor after cycle 5. Thirty-one patients, 3 with INSS Stage 3 and 28 with Stage 4 were enrolled. Pharmacokinetically guided topotecan dosing (target systemic exposure of AUC determined by single day topotecan lactone levels) demonstrated the ability to achieve the target AUC in of patients during T/C Cycle 1 and in of patients during T/C Cycle 2. Furthermore, the median topotecan dose required to achieve the targeted AUC during Cycle 1 and Cycle 2 was day and day , respectively and support the use of day dosing in the ANBL0532 induction. Successful PBSC harvest as defined by a minimum of CD34 cells/kg was achieved in patients for whom data are available; median harvest of cells/kg (range 2.24-548) collected over a median of 1 day (range 1-3 days). A PBSC harvest of CD34 cells was achieved in 23 patients ( ). One patient was harvested after Cycle 5 due to physician discretion and achieved a PBSC collection of CD34 cells/kg. Only one patient required 3 days of apheresis but achieved a total PBSC collection of CD34 cells/kg. All PBSC collections were free of tumor contamination using an immunocytochemical detection assay. There were no dose limiting toxicities during Cycles 1 and 2 of induction and no induction related moralities. The majority of patients ( ) experienced Grade 3 or 4 hematopoietic toxicity. Documented infection occurred in patients during T/C cycles compared to of patients during subsequent induction cycles on ANBL02P1 or patients during initial 3 cycles of the MSKCC induction on COG A3973. The intended dose intensity of all chemotherapy agents was maintained in 30 of 31 patients. One patient required a dose reduction of chemotherapy doses during Cycle 6 induction due to prolonged myelosuppression. Complete response or partial response was achieved in of patients following 2 cycles of T/C. End-induction response rate (CR/VGPR) was ( ), 1 patient developed progressive disease following completion of induction but prior to initiating consolidation therapy.

In summary, this dose intensive cyclophosphamide and topotecan containing induction regimen was well tolerated with expected and reversible toxicities. Dose intensity of standard induction chemotherapy agents was not limited by the addition of dose-intensive topotecan. The ability to achieve sufficient PBSC harvest with in vivo tumor purging was demonstrated. Together these data support the investigation of efficacy of this novel induction regimen in ANBL0532.

Children's Oncology Group

Chemotherapy Dose Intensity - Comparison of Therapeutic Regimens

Dose intensities are calculated as week administered over the entire induction regimen.

For stage III and IV neuroblastoma, local relapse is a major component of treatment failure in several published series, all of which included radiation delivered to the site of primary disease. As systemic therapy for high-risk neuroblastoma becomes more aggressive, response rates improve, and survival increases, local control becomes a formidable problem. This is evident in a recent analysis of CCG 3891, consisting of chemotherapy, primary surgery, 10 Gy of external beam radiation therapy (EBRT) to gross residual disease, followed by randomized assignment to continuation chemotherapy (CC) or autologous bone marrow transplantation (ABMT). ABMT patients received 10 Gy of total body irradiation (TBI). Estimated 5-year locoregional recurrence rates were and for CC and ABMT patients, respectively . Results of this trial revealed that for patients who received 10 Gy of EBRT to the primary, the addition of 10 Gy of TBI and ABMT decreased local recurrence compared with and . This indicates that in combination with EBRT to the primary tumors site, the addition of 10 Gy of TBI as a component of high dose chemotherapy with ABMT improved local control compared with CC without TBI. Several single institution reports further support the administration of pre-chemotherapy or pre-surgery primary tumor volume and regional lymph nodes primary site radiation therapy ( 21 Gy) results in improved local tumor control. Based upon these data, the currently completed COG A3973 delivered 21.6 Gy EBRT to pre-surgical primary tumor volume following recovery from consolidation myeloablative therapy.

Several analyses of outcome for high-risk neuroblastoma suggest that current local therapy for incompletely resected patients is inadequate. Wolden et al. reported on a series of patients with stage 4 neuroblastoma receiving 1.5 Gy twice a day to 21 Gy to pre chemotherapy, pre-surgery primary tumor volume and regional lymph nodes. The actuarial locoregional control rate at 5 years was . A recent update of this experience reported a probability of primary-site failure among 99 patients, most of whom ( 92 patients) had no evidence of disease in the primary site at the time of irradiation. Among seven patients with disease at the primary site at the time of irradiation, three had disease that recurred locally. EBRT did not increase acute toxicity, except for increased total parenteral nutrition administration.

Published studies have sought to assess the appropriate radiation dose to achieve local control in patients with neuroblastoma. A dose response to radiation is supported by several publications; however, these publications describe results in stage III disease or do not incorporate biological prognostic markers.

In the analysis of radiation therapy in CCG 3891, in combination with EBRT to the primary tumors site, the addition of 10 Gy total body irradiation (TBI) as a component of autologous bone marrow transplant (ABMT) improved local control compared with continuation chemotherapy (CC) without TBI. In addition, a dose response to radiation administered to the primary tumor site is supported by the patients in this study that received 20 Gy for extra-abdominal primary tumors. Of 36 patients with extraabdominal primaries, 6 patients received 20 Gy EBRT (2 also received TBI) while 30 patients received no EBRT (10 of these received TBI). Local relapse rates at 5 years were and for patients with and without EBRT, respectively Jacobson et al. reported on the experience at the University of Florida, using radiation doses up to 45 Gy. The authors noted no evaluable complications related to radiation alone, but rather related more commonly to the tumor itself or surgical resection. Investigators at the University of Utah performed a similar analysis of radiation dose response for neuroblastoma. The authors observed local failures after doses up to 45 Gy in children older than three years. Although authors of both studies state that doses lower than 45 Gy may be adequate for some patients with neuroblastoma, the vast minority of patients in these studies had high-risk disease and therefore no firm conclusions can be drawn regarding the appropriate dose to the primary site in patients with high-risk disease. In fact, a recent abstract from the University Children's Hospital in Cologne, Germany, reported that 36 Gy of external beam radiation therapy (EBRT) administered to patients with incompletely resected primary tumors could compensate for "the disadvantage of incomplete response to induction chemotherapy." In this report of the NB97 trial for patients one year of age or older with stage 4 neuroblastoma, patients with isolated localized residual disease had improved outcome following EBRT ( 3 -y-EFS 100%, 3-y-OS 100%) compared those not receiving EBRT despite residual tumor tissue ( 3 -yEFS , p<0.001; 3-y-OS , p<0.001).

Given the high rates of local recurrence, particularly following an incomplete surgical resection, and the tolerable toxicity of EBRT evident in multi-institutional trials, patients enrolled onto ANBL0532 who have an incomplete surgical resection of the primary tumor will receive 21.6 Gy EBRT to the postinduction chemotherapy, pre-operative primary tumor volume and an additional boost of 14.4 Gy EBRT to the gross residual volume (total dose 36 Gy to gross residual tumor volume). The additional dose of 14.4 Gy to gross residual disease is unlikely to result in significantly increased toxicities. In CCG 3891, the administration of EBRT was not associated with enhanced toxicities with the exception of parenteral nutrition requirement, which occurred in and of patients who did and did not receive EBRT, respectively . Furthermore, 36 Gy is well within normal tissue tolerance for most organs within the field of radiation. Kidneys and liver will likely prove to be the dose-limiting structures and the same guidelines for maximal doses to normal structures will be followed, as is standard in all current neuroblastoma protocols. It should also be noted that the boost for patients with incomplete resections will deliver an additional 14.4 Gy (from 21.6 Gy to 36.0 Gy) only to regions of gross residual disease, a target that generally is substantially smaller than the original pre-operative tumor volume that receives 21.6 Gy.

Biopsy at initial diagnosis is necessary to obtain adequate tissue for biologic studies and enrollment on the Neuroblastoma Biology Protocol. Resection of the primary tumor and bulky metastatic disease is usually necessary to achieve CR or VGPR after induction chemotherapy. A retrospective analysis of high risk neuroblastoma patients enrolled onto CCG 3891 demonstrated improved resectability of primary tumor after initial chemotherapy and revealed a trend toward improved survival for those patients who underwent gross total resection of primary tumor, 5-year EFS and , respectively Acute surgical complication rates and characteristics were independent of timing of primary tumor resection. Based upon these data, patients enrolled onto the recently completed A3973 underwent delayed surgical resection of the primary tumor after 5 cycles of chemotherapy. Tumor biopsy at diagnosis was required to obtain adequate tissue for assessment of biologic characteristics. ANBL0532 will utilize this treatment strategy and prospectively evaluate if complete primary tumor resection is predictive of local control and EFS in high risk neuroblastoma patients.

Early and late postoperative complications are known to occur following extensive tumor resection for neuroblastoma. These complications may include adhesive bowel obstruction, chylous leak, ipsilateral renal injury, atrophy or loss, and chronic diarrhea. Group wide studies have regularly sought data on intraoperative and 30 day postoperative complications. Adkins et al. reported normal organ removal in of high risk neuroblastoma patients who underwent complete resection of primary tumor. There was no statistical difference in acute intra-operative or peri-operative complications, including risk for hemorrhage, renal injury, wound complication or acute bowel obstruction, regardless of extent of surgical resection. There is potential for bias in this study, given the retrospective collection of data. Extended observations are necessary to detect late postoperative complications, as they may relate to the extent of resection and subsequent surgical and radiotherapeutic efforts to achieve local control. Specifically, extended follow up will be necessary to detect the occurrence of adhesive bowel obstruction and renal atrophy related to surgically induced ischemic events while the risk for chylous leak or diarrhea may be related to extensive retroperitoneal dissection. Rates of these complications should be correlated with the surgical and radiotherapeutic efforts in obtaining local control as they may impact on future management decisions.

Epidural or intradural extension of tumor occurs in approximately of patients diagnosed with neuroblastoma, and may occur with or without neurologic impairment. Epidural neuroblastoma with spinal cord compression is an oncologic emergency that warrants immediate intervention. However, the choice of emergent therapy remains somewhat controversial, with pediatric oncologists favoring chemotherapy and neurosurgeons favoring laminectomy with surgical decompression. There are several retrospective reports of similar neurologic outcome for patients treated with chemotherapy versus laminectomy, and therefore emergent chemotherapy has become a preferable alternative to laminectomy in most cases, in an effort to diminish the devastating late orthopedic effects of laminectomy in infants or young children. Radiation therapy has generally been reserved for situations of progressive or persistent neurologic dysfunction despite chemotherapy and laminectomy.

This study will prospectively characterize multiple aspects of neurologic disability at diagnosis and during the follow-up period in an effort to systematically study the neurologic outcome of patients with primary neuroblastoma tumors with intraspinal extension (see Appendix III). A multidisciplinary evaluation at the time of diagnosis, including oncology, neurosurgery, and radiation oncology, is strongly encouraged. It is the intent of this study that patients with clinical signs of spinal cord compression will be treated with primary chemotherapy, unless the multidisciplinary evaluation determines that surgical decompression would be more appropriate. Radiation therapy should be reserved for patients whose neurologic status deteriorates despite chemotherapy and surgical decompression.

The presence of minimal residual disease despite achievement of maximal response to induction and consolidation therapy results in continued risk for neuroblastoma relapse. This was demonstrated in the CCG-3891 study where the benefit of additional non-cross resistant therapy, 13-cis Retinoid acid (cisRA) administered following myeloablative consolidation therapy was observed. Patients with newly diagnosed high-risk neuroblastoma enrolled in CCG3891 were randomized at week 34 of therapy (following either chemotherapy or ABMT consolidation) to no further therapy versus cis-RA for six months. Cis-RA or isotretinoin is a synthetic retinoid derived from the naturally occurring all transretinoic acid by modification of the terminal carboxyl group. When neuroblastoma cell lines are exposed to all-trans-retinoic acid (trans-RA) or cis-RA in vitro, they exhibit decreased proliferation, decreased expression of the MYCN oncogene, and morphological differentiation. Growth arrest and differentiation in response to cis-RA have been observed in neuroblastoma cell lines initiated from tumors at the time of progression after chemoradiotherapy, suggesting that resistance to cytotoxic chemotherapy does not induce resistance to cis-RA. cis-RA was well tolerated in a pediatric Phase I trial using an intermittent administration schedule of twice daily administration for 14 days followed by 14 days with no therapy. cis-RA toxicities are generally mild; consisting primarily of chelitis, dry skin, and hypertriglyceridemia, with hypercalcemia seen at higher doses. A significant difference in the three
year EFS from the time of this randomization in the patients receiving cis-RA ( ) versus those with no further therapy ( ), with p value of 0.027 in a test of proportions. This advantage for EFS was seen in subgroups by prior randomization, such that the best overall EFS was for patients treated with ABMT with cis-RA ( EFS), second was ABMT without cis-RA ( ), third was chemotherapy consolidation with cis-RA ( ), and fourth was chemotherapy without cis-RA ( ).

Despite the use of cis-RA, greater than of children will develop recurrent neuroblastoma. To further improve outcome, the efficacy of novel, anti-neuroblastoma targeted immunotherapy to eliminate minimal residual disease has been evaluated. Anti-gangliosidase (GD2) monoclonal antibody ch14.18 has shown preclinical and early clinical activity against neuroblastoma, which was enhanced when combined with GM-CSF or IL2. COG ANBL0032 was designed to determine if adding ch14.18 + GM-CSF + IL2 to standard therapy of isotretinoin after intensive multimodality therapy improved outcome for high-risk neuroblastoma patients. High-risk neuroblastoma patients who responded to induction therapy and stem cell transplant were randomized to 6 cycles of isotretinoin (standard therapy) or 6 cycles of isotretinoin with 5 concomitant cycles of ch14.18 combined with GM-CSF or IL2 in alternating cycles (immunotherapy). An intent-to-treat randomized comparison (Lan-DeMets interim monitoring, cumulative alpha ) was performed for event free survival (EFS). A total of 226 eligible patients were evenly randomized to standard or immunotherapy. Immunotherapy was associated with grade pain, vascular leak syndrome and hypersensitivity reactions in and of patients, respectively. With of expected events observed, the study met criteria for early stopping for efficacy. Median follow-up was 2.1 years. Two-year EFS estimates were for patients randomized to immunotherapy versus for standard therapy ( ). OS was also superior ( , unadjusted for interim looks) for immunotherapy: versus isotretinoin alone . Moreover, for the major subgroup of patients year of age with stage 4 disease, EFS was significantly greater ( 0.02 ) for the immunotherapy group ( ), as compared to isotretinoin alone ( ). (*2-year estimates) (Alice Yu; personal communication, COG ANBL0032 study report).

Immunotherapy consisting of ch14.18 with GM-CSF and IL2 significantly improves outcome for highrisk neuroblastoma patients. Ch14.18 remains under clinical investigation through COG ANBL0032 and ANBL0931 protocols to further evaluate efficacy and toxicity. Given the significant improvement in survival from high risk neuroblastoma following antiGD2 antibody therapy, patients enrolled onto ANBL0532 should be encouraged to participate in actively enrolling clinical trials of ch14.18 antibody, (i.e. ANBL0032 or ANBL0931). Patients who are not eligible to receive ch14.18 antibody therapy or decline immunotherapy will remain on ANBL0532 and receive six months of cis-RA post ASCT based on the significant improvement in EFS with cis-RA documented by CCG3891 trial.

Enrollment on COG ANBL00B1 (Neuroblastoma Biology Study) is an eligibility requirement for ANBL0532 as assignment to clinical risks groups and different COG neuroblastoma therapeutic clinical trials is dependent upon accurate histologic and biologic diagnosis. Although submission of paraffinembedded tissue is acceptable for enrollment onto ANBL00B1, the COG Neuroblastoma committee encourages the submission of fresh or snap-frozen tissue, peripheral blood, or bone marrow specimen. The acquisition of these primary tumor specimens will enable additional correlative biologic studies. Enrollment on both COG ANBL0532 and ANBL00B1 has the significant advantage of allowing the correlation of novel biologic parameters with a clinically well-characterized (including treatment and outcome) cohort of patients.

Although the efficacy of cyclophosphamide is well established in NB patients, considerable inter-patient variability exists in response and toxicity. Variability in cyclophosphamide metabolism may result in such clinical variability. Cyclophosphamide dosing based on body weight ( ) or BSA ( )
results in a wide area under the concentration-time curve (AUC) range of the parent drug, and an even greater variability in the AUC of its metabolites. Children have a faster clearance of cyclophosphamide relative to adults, although the clinical significance of this difference in disposition is unclear. Recent studies have revealed an association between the AUC of cyclophosphamide and its metabolites to the risk of recurrence in children receiving standard dose cyclophosphamide (i.e., ) for treatment of B-cell non-Hodgkin lymphoma.

The pharmacokinetics of cyclophosphamide and its metabolites HCY and CEPM were evaluated in 18 of the 31 patients with newly diagnosed high-risk neuroblastoma enrolled in COG ANBL02P1. Median patient age was 2.5 years (range: 0.9-9.4).

The median clearance was (range: ) after dose 1 and 4.35 (1.55-8.54) after dose 4, consistent with previous reports Topotecan did not appear to effect cyclophosphamide clearance. Similar to our observations in patients receiving Cyclophosphamide/TBI, there was considerable interpatient variability in the AUC of cyclophosphamide with even greater interpatient variability in exposure to the metabolites. After dose 1, variability of the AUC of cyclophosphamide ( 3.1 fold), HCY ( 8.4 fold) and CEPM ( 8.3 fold) was comparable to that measured in cyclophosphamide/TBI patients (i.e., 3.3, 7.8 and 16.1 fold, personal communication J. McCune). Auto-induction of cyclophosphamide clearance was observed in 14 of the 15 children for whom cyclophosphamide AUC was available after doses 1 and 4, again similar to other reports. The and increased in four and three children, respectively.

Multivariate analysis was conducted to evaluate for determinants of AUC of cyclophosphamide, HCY and CEPM. Two of the 21 children received cyclophosphamide ; their AUCs were normalized to cyclophosphamide to allow for inclusion in this multivariate analysis. Multivariate regression analysis indicated that age, weight and body surface area explained only of the variability in CY clearance after dose 1. Furthermore, the AUC of cyclophosphamide and its metabolites were not consistently correlated.

Cyclophosphamide pharmacokinetic assessments are unlikely to be used clinically given the variability in cyclophosphamide metabolism and the resource intensity of cyclophosphamide pharmacokinetic evaluations. Alternatively, genetic polymorphisms in genes involved in cyclophosphamide metabolism may correlate with cyclophosphamide pharmacokinetics and may predict toxicity and anti-tumor activity of the ANBL0532 cyclophosphamide containing induction regimen. Cyclophosphamide is activated by several cytochrome P450 enzymes which have genetic polymorphisms, specifically CYP3A4/5, CYP2C19, CYP2C9 and CYP2B6. CYP2B6 and alleles have been associated with either altered activation of HCY or clinical outcome. Their frequencies in Caucasians are and , respectively. Several genetic polymorphisms in CYP2C9 have functional consequences on the enzyme's activity; CYP2C92 and CYP2C93 are the most common variants. In Caucasians, of individuals are homozygotes for CYP2C91, 22% are heterozygous or homozygous for CYP2C92 and 13% carry CYP2C9*3. The expression of these alleles is less common in African-Americans or Chinese (<2% -

Children's Oncology Group
, respectively). The CYP2C92 allele is characterized by an arginine to cysteine exchange at position 144 (Cys144 - Ile 359), while in CYP2C93 isoleucine 359 is changed to leucine (Arg144 Leu359). Both CYP2C92 and CYP2C93 are non-null alleles which code for functional protein with reduced levels of enzyme activity. CYP2C92 carriers have a moderate reduction in activity, while the reduction in activity is most marked for CYP2C93. GSTA1, the predominant GST involved in the metabolism of cyclophosphamide, exhibits a polymorphism that influences its hepatic expression. GSTA1B, which is several linked single nucleotide polymorphisms in the proximal promoter region of the GSTA1 gene, has reduced levels of GSTA1 enzyme. The expression of GSTA1B does not differ by ethnic group; GSTA1*A/B occurs in and GSTA1B/*B occurs in of a population of breast cancer patients.

ANBL0532 will assess whether host genetic polymorphisms in CYP2B6, CYP2C9 and GSTA1 genotypes involved in the metabolic activation and clearance of cyclophosphamide are correlated with Cyclophosphamide/Topo associated toxicity and tumor response. Prior studies have established genotype-phenotype relationships for the CYP2B6, CYP2C9 and GSTA1 genes for both in vitro analyses and in metastatic breast cancer patients receiving high dose cyclophosphamide as a component of a HSCT preparative regimen. The pharmacology laboratory at the Fred Hutchinson Cancer Research Center is currently conducting an analysis of CYP2B6, CYP2C9, CYP3A5, ABCC2, GSTA1, GSTM1, GSTT1 genetic polymorphisms in 147 patients who received CY in combination with TBI in preparation for HSCT. Preliminary analysis suggests that polymorphisms in CYP2B6, CYP2C9, and GSTA1 are key enzymes associated with toxicity (personal communication, J. McCune). An exploratory analysis of covariance (ANCOVA) was conducted using limited ANBL02P1 data. Age was used as a co-variate for three genes (GST, CYP2C9, CYP2C19) upon the following endpoints after the first cyclophosphamide dose (i.e., dose 1): AUCCY, CY clearance, AUCHCY, AUCCEPM. CYP2C9 was the only gene that trended towards statistical significance with AUCCY, cyclophosphamide clearance ( ), AUCHCY , AUCCEPM ( ).

The variability of cis-RA continuous steady state (CSS) plasma levels and/or genetic variations in retinoic acid metabolic enzymes may correlate with event-free survival or systemic toxicity (CTC Grade skin, hypercalcemia and hepatic toxicity) in high-risk neuroblastoma patients. Pharmacokinetics studies performed on patients in the Phase I trial of cis-RA showed significant interpatient and intrapatient variability in peak serum levels (mean peak serum cis-RA concentration: ; trough concentration: ). It was noted that peak serum concentrations above were associated with a higher incidence of grade 3 and 4 toxicities in patients. cis-RA may be subject to first-pass metabolism and subsequent plasma (and tumor) concentrations will depend on the rate of metabolism to its inactive 4-oxo metabolite. Preliminary data from a pilot study in the UK (UKCCSG Study PK 2000 08 ) has shown that plasma concentrations of 4-oxo-13-cis-RA can accumulate to exceed those of the parent compound. A - fold variation in 4-oxo-13-cis-RA peak plasma concentrations was found. It is proposed that increased levels of this metabolite in vivo may lead to a diminished efficacy of cis-RA.

A number of cytochrome P450 (CYP) enzymes have been identified as playing a role in the metabolism of cis-RA. CYP2C8 is most important in terms of activity and level of expression in the liver. Genetic polymorphisms in the CYP2C8 gene have been described which result in a lower rate of paclitaxel metabolism and are commonly seen in a Caucasian population. Thus, genetic variation in CYP2C8 activity could underlie individual differences in cis-RA metabolism and bioavailability. The fetal isoform CYP3A7, which is expressed post-natally in a significant number of individuals, also metabolizes cis-RA . The expression of CYP3A7, and thus the contribution of this isoform to cis-RA metabolism, may be predicted by genotyping in a pediatric population. A further aspect of cis-RA metabolism is glucuronidation, both of the parent drug and of 4-hydroxy-metabolites. This conjugation may be mediated by UGT1A1 or UGT2B7, both enzymes that are subject to genetic polymorphisms. Results from this study will provide an insight into whether modulation of cis-RA dosing according to blood levels achieved and/or genotype, could be used in future studies to optimize the treatment of high-

risk neuroblastoma. This study will be conducted in collaboration with the United Kingdom Children's Cancer Study Group (UKCCSG).

Many individual pharmacokinetic studies of topotecan in children with cancer have been conducted, and results have shown wide interindividual variability in topotecan clearance. Due to the inherent variability in topotecan disposition, investigators have used pharmacokinetically guided dosing to individualize the topotecan dosage for each patient based upon their systemic clearance and the goals of the clinical trial. The response rate for this approach has been promising. Although pharmacokinetically guided topotecan is a promising approach to dosing topotecan, it has limitations and it is important to find ways to simplify the current pharmacokinetic dosing strategy.

One approach would be to use a dosing model based upon patient specific covariates identified in a population pharmacokinetic analysis. Such an analysis of topotecan lactone pharmacokinetics has been conducted in a large pediatric patient population, which includes patients with many different diagnoses. The covariates identified in this analysis included BSA, age, phenytoin coadministration, calculated glomerular filtration rate, and serum creatinine. However, for the current application the patient population will be limited to children with high-risk neuroblastoma and only total topotecan (combination of lactone and carboxylate forms) will be available for pharmacokinetic analysis. Thus, a more limited population pharmacokinetic analysis was undertaken of the ANBL02P1 total topotecan plasma concentration-time data. In this analysis, the only patient covariate identified as significantly related to topotecan total clearance was sex with females having a lower clearance. Clearly this covariate will not be very informative in a dosing model. Thus, it would be helpful to gain additional covariate data in this patient population, and relate that to total topotecan clearance determined using a limited sampling model. Ultimately, data collected in this Phase III clinical trial will be used to derive a dosing algorithm for topotecan in children with high-risk neuroblastoma that will either not need plasma topotecan concentrations or at most use only one plasma concentration. By using this dosing model, the interpatient variability in topotecan systemic exposure can be minimized, and the full therapeutic efficacy of this compound can be realized.

Despite low or absent expression of MHC class I, T cells capable of recognizing neuroblastoma cells and acting as effector cells have been described. Survivin has the potential to be a relevant tumor antigen in neuroblastoma, as it is widely expressed and expression is inversely correlated with prognosis. Moreover, as a key inhibitor of apoptosis, survivin may be required for tumor growth and development, making its potential loss as a means of escape of the tumor from immune surveillance difficult. We hypothesize that cytolytic T lymphocytes (CTL) directed against survivin+ targets will be detectable in the blood of neuroblastoma patients. Additionally, these CTL will be expanded by exposure to two different preparations of antigen presenting cells (APCs). One APC type will be transfected with survivin mRNA, providing surviving peptides to the MHC of the APC, while the second APC type will be loaded with neuroblastoma RNA. The APC used in our studies has been an activated B cell, referred to as a CD40-B. Relevant prior work from the Grupp and Vonderheide labs show that: 1) CD40-B cells serve as efficient APCs, inducing antigen-specific and anti-tumor CTL. 2) Large numbers of CD40-B APCs can be grown from small volumes of peripheral blood, as opposed to dendritic cells, which require pheresis to collect in adequate numbers. 3) RNA transfection efficiently transduces CD40-B APC, allowing expansion of tumor-specific CTL from patient blood samples even in a tumor like neuroblastoma in which tumor antigens are not known. 4) Neuroblastoma expresses survivin. 5) Survivin-specific CTL can be detected at diagnosis in up to of neuroblastoma patients and expanded using either survivintransfected CD40-B or CD40-B transfected with neuroblastoma RNA. Iinitial studies demonstrate that T cells active against survivin+ targets emerge whether survivin RNA or whole tumor RNA is used as the antigenic stimulus. 6) Survivin-specific CTL secretes IFN-gamma in response to tumor and effectively lyse survivin+ neuroblastoma targets. 7) CD107a is a marker of CTL function and can also be used to demonstrate that CTL specific for the immunodominant epitope of survivin are responsible for antineuroblastoma killing activity.

These data point to survivin as an important tumor antigen in neuroblastoma, show feasibility of using tumor RNA to provide an antigenic payload to APCs and confirm that neuroblastoma is a tumor susceptible to immune recognition. Data revealing tumor RNA-loaded APC produce tetramer+ CTL directed against survivin that are capable lysing autologous tumor provides key evidence of the importance of survivin in neuroblastoma. These results need to be confirmed in the larger number of samples available from the patients on this clinical trial. These experiments will allow us to further characterize survivin as a potential tumor antigen in neuroblastoma.

Increasing intensity of therapy has the potential to increase extent and duration of immune suppression, especially of adaptive immunity. This was apparent in the CHOP/DFCI tandem HSCT experience where a combination of tandem HSCT, use of TBI and, potentially, use of CD34 selection of PBSC produced significant T cell immune suppression. We hypothesize that patients on the tandem HSCT arm will experience slower T cell recovery, specifically of CD4+ T cells. Enumeration of CD3, CD4 and CD8 cells post HSCT is available at all FACT-approved HSCT centers, and is part of post-HSCT routine follow up at many centers. There is also value in following CD4 recovery to guide the level of viral surveillance and isolation precautions that may or may not be required for an individual patient. Thus, we propose to perform these assays at 1,3 and 6 months post-HSCT, comparing T cell recovery across arms of the study. These assessments will provide a valuable baseline for studies that may utilize either of the ANBL0532 regimens, especially in the context of immunotherapies that may require functional T cell recovery.

Despite initial response to therapy, the majority of children with high-risk neuroblastoma will develop recurrent or progressive disease. More sensitive methods for the detection of minimal residual disease (MRD) are needed. Several reports demonstrated the high sensitivity ( to normal cells) of immunocytological assays in bone marrow, bone marrow autograft and peripheral blood samples. Quantitative levels of bone marrow infiltration using immunocytological assays have been predictive of outcome for patients with stage 3 and 4 disease. . More recently amplification of tissuespecific mRNA by reverse transcriptase polymerase chain reaction (RT-PCR) has been used to detect minimal disease burden (reviewed in ). Investigators have employed an RT-PCR assay, using both tyrosine hydroxylase (TH) and the neuronal gene (PGP 9.5) which has a sensitivity of 1 tumor cell per million hematopoietic cells. Although IC and RT-PCR appear to be ideal tools for monitoring MRD, reaching a sensitivity of one tumor cell detected in to normal cells, their introduction into clinical management of children with neuroblastoma has been slow. ANBL0532 will evaluate the clinical relevance of tumor cells detected by these techniques. Immunocytology and RT-PCR studies will be used on bone marrow, peripheral blood, and PBSC collections to examine the question of tumor contamination. The incidence of RT-PCR positivity in stem cell products infused will be prospectively monitored. Furthermore, the incidence of RT-PCR positivity in bone marrow and peripheral blood will be compared following consolidation therapy between single verus tandem myeloablative therapy.

In CCG-3891, stage at diagnosis and EFS were similar between genders, however there was a trend for males to have better EFS than females on the ABMT arm of the study. There was a significant interactive effect, in which the benefit of ABMT was restricted to males. No differences in outcome by gender were observed in the earlier 321P3 transplant study. There were no differences on CCG-3891 with respect to gender, regarding the effect of 13-cis-RA on EFS. Data from CCG-3891 also showed a lower overall EFS for African Americans versus Whites or Hispanics. African Americans had equivalent outcome on the ABMT arm, but lower EFS if treated on the chemotherapy arm in comparison with Whites. No ethnic differences were seen in the outcomes of patients treated with cis-RA.

Children's Oncology Group

Patients will undergo collection of peripheral blood stem cell product following 2 cycles of induction chemotherapy. The ANBL0532 study has previously mandated immunocytology assessment of peripheral blood stem cell (PBSC) product for tumor contamination. The laboratory at Children's Hospital of Los Angeles (CHLA) that has previously performed the immunocytology assay for CCG 3891, COG A3973 and COG ANBL0532 no longer has CLIA certification and no validated CLIAapproved alternative laboratory testing exists.

ANBL0532 Amendment #2 will NOT REQUIRE immunocytology assessment of PBSC for tumor contamination. The amendment to this study will also remove mandatory PBSC specimen shipment to CHLA for immunocytology testing. Data from the most recent generation of COG high-risk neuroblastoma trials show that tumor contamination is rare and is not an independent prognostic determinate of outcome. COG A3973 determined tumor cell contamination of PBSC by immunocytologic assessment in only 5 of 487 ( ) of PBSC products collected after 2 cycles of induction chemotherapy. More importantly, COG A3973 demonstrated no benefit from anti-tumor purging of PBSC product using the same antibodies used for immunocytologic detection and support the conclusion that tumor recurrence is not directly related to tumor cell re-infusion from the PBSC product. Taken together, the COG neuroblastoma committee has concluded that there are no data to support the clinical utility of immunocytological testing of PBSC products for patients with high-risk neuroblastoma.

Local IRB/REB approval of this study must be obtained by a site prior to enrolling patients. Sites must submit IRB/REB approvals to the NCI's Cancer Trials Support Unit (CTSU) Regulatory Office and allow 3 business days for processing. The submission must include a fax coversheet (or optional CTSU IRB Transmittal Sheet) and the IRB approval document(s). The CTSU IRB Certification Form may be submitted in lieu of the signed IRB approval letter. All CTSU forms can be located on the CTSU web page (https://www.ctsu.org). Any other regulatory documents needed for access to the study enrollment screens will be listed for the study on the CTSU Member's Website under the RSS Tab.

IRB/REB approval documents may be faxed (1-215-569-0206), emailed (CTSURegulatory@ctsu.coccg.org) or mailed to the CTSU Regulatory office.

When a site has a pending patient enrollment within the next 24 hours, this is considered a "Time of Need" registration. For Time of Need registrations, in addition to marking your submissions as 'URGENT' and faxing the regulatory documents, call the CTSU Regulatory Helpdesk at: 1-866-651CTSU. For general (non-regulatory) questions, call the CTSU General Helpdesk at: 1-888-823-5923.

Prior to enrollment on this study, patients must be assigned a COG patient ID number. This number is obtained via the eRDE system once authorization for the release of protected health information (PHI) has been obtained. The assigned COG patient identification number will be used to identify the patient in all future interactions with the COG. If you have problems with registration, please refer to the online help in the eRDE area of the COG website.

In order for an institution to maintain COG membership requirements, every newly diagnosed patient needs to be offered participation in ACCRN07, Protocol for the Enrollment on the Official COG Registry, The Childhood Cancer Research Network (CCRN).

A Biopathology Center (BPC) number will be assigned as part of the registration process. Each patient will be assigned only one BPC number per COG Patient ID. Each patient will be assigned only one PBC number per COG Patient ID. For additional information about the labeling of specimens, please refer to the pathology and/or Biology guidelines in this protocol.

Enrollment on ANBL00B1 is required for all newly diagnosed patients within 21 days of diagnosis. Tissue procurement is mandatory for biology study registration. For patients with stage 4S disease who are very ill and in whom an open biopsy to obtain tissue for diagnosis and biologic studies is considered medically contraindicated, every effort should be made to obtain some tumor tissue by either fine needle aspiration of a metastatic site of disease and/or sampling of involved bone marrow, so that this tumor sample can be submitted for MYCN determination. Consent for ANBL00B1 must be obtained at the time of tissue submission and should be within one week of surgery. Needle biopsies are not sufficient for histologic classification. Investigators are strongly encouraged to obtain adequate tissue (see Sections 13.0, 14.0, and 15.0) via open biopsy techniques. For patients with Stage 4 disease who are days of age with unequivocal neuroblasts in the bone marrow, in whom a diagnostic biopsy is not obtained, a minimum of of involved bone marrow and a blood specimen must be sent to the reference lab per the requirements of ANBL00B1 to be eligible for this study. (See ANBL00B1 protocol for specifics)

Patients may be enrolled on the study once all eligibility requirements for the study have been met. Study enrollment is accomplished by going to the Enrollment application in the eRDE system. If you have problems with enrollment, refer to online help in the Applications area of the COG website.

Patients must be enrolled on ANBL00B1 prior to the time of enrollment on ANBL0532. In emergency situations (or if in the opinion of the treating physician, it is in the patient's best interest) consent can be obtained and patient enrolled on ANBL00B1 and subsequently enrolled on ANBL0532 as soon as the assignment of "High-Risk" has been made in the eRDE system. When ANBL0532 enrollment is done prior to start of beginning protocol therapy, the date protocol therapy is projected to start must be no later than five (5) calendar days after enrollment. In a case of a need for emergency therapy, protocol therapy may start before enrollment on ANBL0532; however, ANBL0532 consent part 1 must be obtained prior to start of therapy AND enrollment must take place within fourteen (14) calendar days of beginning protocol therapy. Study enrollment must occur within 4 weeks of diagnosis or after only one cycle of chemotherapy on the low/intermediate risk neuroblastoma studies, or within 4 weeks of progression to stage 4 for INSS stage .

To allow non-English speaking patients to participate in the study, bilingual health care services will be provided in the appropriate language.

Randomization will take place at completion of Induction phase of therapy via RDE. Patients will be assigned to either single myeloablative therapy or tandem myeloablative therapy. Randomization will be stratified by initial stage of disease, biologic characteristics and response to induction chemotherapy.

Patients 365 to 547 days of age ( months) with Stage 4, MYCN nonamplified tumor with unfavorable histopathology or diploid DNA content or with indeterminant histology or ploidy and patients who are greater than 547 days of age with Stage 3, MYCN nonamplified tumor AND unfavorable histopathology or indeterminant histology will be nonrandomly assigned to single myeloablative transplant arm.

Important note: The eligibility criteria listed below are interpreted literally and cannot be waived (per COG policy posted ). All clinical and laboratory data required for determining eligibility of a patient enrolled on this trial must be available in the patient's medical/research record which will serve as the source document for verification at the time of audit.

Patients must be years of age at the time of initial diagnosis.

Patients must have a diagnosis of neuroblastoma (ICD-O morphology 9500/3) or ganglioneuroblastoma verified by histology or demonstration of clumps of tumor cells in bone marrow with elevated urinary catecholamine metabolites. Patients with the following disease stages at diagnosis are eligible, if they meet the other specified criteria. See Appendix VII.

Patients with newly diagnosed neuroblastoma with INSS Stage 4 are eligible with the following:
a. amplification (greater than four-fold increase in signals as compared to reference signals), regardless of age or additional biologic features.
b. Age > 18 months (>547 days) regardless of biologic features.
c. Age 12-18 months (365-547 days) with any of the following three unfavorable biologic features (MYCN amplification, unfavorable pathology and/or DNA index = 1) or any biologic feature that is indeterminant/unsatisfactory/unknown.

Patients with newly diagnosed neuroblastoma with INSS Stage 3 are eligible with the following:
a. amplification (greater than four-fold increase in signals as compared to reference signals), regardless of age or additional biologic features
b. Age > 18 months (> 547 days) with unfavorable pathology, regardless of status.

Patients with newly diagnosed INSS Stage 2a/2b with amplification (greater than four-fold increase in signals as compared to reference signals), regardless of age or additional biologic features.

Patients with newly diagnosed INSS Stage 4s with MYCN amplification (greater than four-fold increase in signals as compared to reference signals), regardless of additional biologic features.

Patients days initially diagnosed with: INSS stage who progressed to a stage 4 without interval chemotherapy. These patients must have been enrolled on ANBL00B1. It is to be noted that study enrollment must occur within 4 weeks of progression to Stage 4 for INSS Stage 1, 2, 4S.

Patients must have had no prior systemic therapy except for localized emergency radiation to sites of lifethreatening or function-threatening disease and/or no more than one cycle of chemotherapy per low or intermediate risk neuroblastoma therapy (P9641, A3961, ANBL0531) prior to determination of MYCN amplification and histology.

3.2.4.1 Adequate renal function defined as:

The threshold creatinine values in this Table were derived from the Schwartz formula for estimating GFR (Schwartz et al. J. Peds, 106:522, 1985) utilizing child length and stature data published by the CDC.
3.2.4.2 Adequate liver function defined as:

No known contraindication to PBSC collection. Examples of contraindications might be a weight or size less than the collecting institution finds feasible, or a physical condition that would limit the ability of the child to undergo apheresis catheter placement (if necessary) and/or the apheresis procedure.

All patients and/or their parents or legal guardians must sign a written informed consent.

All institutional, FDA, and NCI requirements for human studies must be met.
Identification of COG Approved Transplant Center: Patients entered on study must have their stem cell transplant performed at a COG approved transplant center. A COG-approved transplant center must be identified at the time of study registration. This center may be changed to another COG-approved transplant center after registration if required due to personal or financial patient issues.

3.2.6.1 Females of childbearing potential must have a negative pregnancy test. Patients of childbearing potential must agree to use an effective birth control method.
3.2.6.2 Female patients who are lactating must agree to stop breast-feeding.
3.2.6.3 Patients that are 12-18 months of age with INSS Stage 4 and all favorable biologic features (ie, non-amplified , favorable pathology, and DNA index ) are not eligible.

Timing of protocol therapy administration, response assessment studies, and surgical interventions are based on schedules derived from the experimental design or on established standards of care. Minor unavoidable departures (up to 72 hours) from protocol directed therapy and/or disease evaluations (and up to 1 week for surgery) for valid clinical, patient and family logistical, or facility, procedure and/or anesthesia scheduling issues are acceptable per COG administrative Policy 5.14 (except where explicitlv prohibited within the protocol).

This study will be a Phase III trial that analyses the use of intensified consolidation in high-risk neuroblastoma.

Myeloblative consolidation therapy is given at the completion of induction therapy and is categorized into Regimen A and Regimen B. Regimen A consists of one myeloablative consolidation with a CARBOplatin - etoposide - melphalan (CEM) preparative regimen. For Regimen B, patients will receive two myeloablative consolidations: Thiotepa - cyclophosphamide (TC) preparative regimen followed by CEM preparative regimen.

Most patients will be randomized into either Regimen A or Regimen B. Randomization will be stratified by stage, MYCN status and response to induction therapy.

The following patients will be nonrandomly assigned to single myeloablative transplant (Regimen A):

Patients who develop progressive disease (PD) at the end of induction, those who are unable to obtain adequate PBSC for transplant or who have GFR will be taken off protocol therapy. Refer to protocol section 8.1.1 for complete definition of off protocol therapy criteria.

All patients will have a double lumen central venous line placed prior to beginning induction chemotherapy.

Chemotherapy doses for all drugs EXCEPT topotecan will be adjusted for patients who weigh ; these patients will have chemotherapy dosed per kg rather than per body surface area (BSA). Dosing will be adjusted only during cycles when patient's weight is . Topotecan dosing will be based on BSA regardless of age or weight.

Patients will receive 6 cycles of induction chemotherapy. Chemotherapy cycles (except Cycle 1) may begin when the ANC and platelets after post-chemotherapy nadir. There are no hematologic criteria to begin Cycle 1.

Children's Oncology Group
Patients who received one cycle of chemotherapy per low or intermediate risk neuroblastoma therapy (P9641, A3961, ANBL0531) prior to determination of MYCN amplification and histology will receive all 6 cycles of ANBL0532 induction therapy.
4.1.3.1

Patients will undergo surgical resection after Cycle 5 (or Cycle 6, if medically necessary). Begin next chemotherapy cycle as soon as possible following surgery. Therapy should not be delayed by more than one week unless complications arise. If surgery cannot be scheduled after Cycle 5, it should occur after Cycle 6.
4.1.3.2

Primary and metastatic tumor response will be assessed after 2 cycles and at completion of induction chemotherapy. Primary site tumor assessment prior to surgical resection (Cycle 5 or Cycle 6, if medically necessary) is mandatory as the pre-surgical primary tumor volume will be used to establish radiation fields.

Patients with progressive disease (PD) at the end of induction evaluation will be taken off protocol therapy.
4.1.3.4 Peripheral Blood Stem Cell Harvest (PBSC) (See Section 19.0 for complete details)

Patients will undergo PBSC harvest after 2 cycles REGARDLESS of persistent bone marrow metastatic disease. There will not be any no ex vivo manipulation of PBSC prior to cryopreservation as results from the randomized question in A3973 did not demonstrate a benefit in survival following ex vivo purging of PBSC product.

PBSC collection will be performed prior to randomization. Therefore, all patients must have a PBSC collection that is sufficient for tandem transplant with a goal for collection of divided into 3 separate aliquots. A minimum number of frozen PBSC of CD34 cells divided into 2 aliquots is required for all patients ( CD34 cells for each transplant) to remain eligible for transplant while an additional aliquot of CD34 cells is strongly recommended to be stored for back-up.

Myeloid Growth Factors: During induction cycles 1 and 3-6 (all induction cycles except cycle 2 or other cycles in which PBSC collections are planned), cytokine support need not be limited to G-CSF i.e: other growth factors are permitted according to an institution's standard guidelines. Choice of myeloid growth factor must be recorded appropriately in the therapy delivery maps. Myeloid growth factors (including GCSF) should be administered hours after the last dose of chemotherapy. Daily myeloid growth factor therapy should be continued until the ANC . Discontinue daily myeloid growth factor support a minimum of 24 hours prior to administration of the next chemotherapy cycle. See Appendix II for supportive care guidelines.

Institutions are permitted to use their standard guidelines for hydration/monitoring parameters pertaining to chemotherapy agents utilized in this protocol. In the absence of institutional standards, suggested guidelines are included in the therapy delivery maps that follow.

No hematologic criteria to begin Cycle 1. This course lasts 21 days. This Therapy Delivery Map is on one page.

NOTE: For eRDEs reporting: use Day 22 (or treatment start date for next cycle if > Day 22) as the end date for the current cycle.

Page 1 of 1

COG #

Patients should begin therapy on day 22 of the prior cycle or as soon thereafter as and platelets . This course lasts 21 days. This Therapy Delivery Map is on one page.