Purpose: The cyclin-dependent kinase (CDK) 4/6 inhibitor, ribociclib (LEE011), displayed preclinical activity in neuroblastoma and malignant rhabdoid tumor (MRT) models. In this phase I study, the maximum tolerated dose (MTD) and recommended phase II dose (RP2D), safety, pharmacokinetics (PK), and preliminary activity of single-agent ribociclib were investigated in pediatric patients with neuroblastoma, MRT, or other cyclin D–CDK4/6–INK4–retinoblastoma pathway-altered tumors.

Experimental Design: Patients (aged 1–21 years) received escalating once-daily oral doses of ribociclib (3-weeks-on/1-week-off). Dose escalation was guided by a Bayesian logistic regression model with overdose control and real-time PK.

Results: Thirty-two patients (median age, 5.5 years) received ribociclib 280, 350, or 470 mg/m2. Three patients had dose-limiting toxicities of grade 3 fatigue (280 mg/m2; n = 1) or grade 4 thrombocytopenia (470 mg/m2; n = 2). Most common treatment-related adverse events (AE) were hematologic: neutropenia (72% all-grade/63% grade 3/4), leukopenia (63%/38%), anemia (44%/3%), thrombocytopenia (44%/28%), and lymphopenia (38%/19%), followed by vomiting (38%/0%), fatigue (25%/3%), nausea (25%/0%), and QTc prolongation (22%/0%). Ribociclib exposure was dose-dependent at 350 and 470 mg/m2 [equivalent to 600 (RP2D)–900 mg in adults], with high interpatient variability. Best overall response was stable disease (SD) in nine patients (seven with neuroblastoma, two with primary CNS MRT); five patients achieved SD for more than 6, 6, 8, 12, and 13 cycles, respectively.

Conclusions: Ribociclib demonstrated acceptable safety and PK in pediatric patients. MTD (470 mg/m2) and RP2D (350 mg/m2) were equivalent to those in adults. Observations of prolonged SD support further investigation of ribociclib combined with other agents in neuroblastoma and MRT. Clin Cancer Res; 23(10); 2433–41. ©2017 AACR.

Translational Relevance

High-risk neuroblastoma and malignant rhabdoid tumors (MRT) are pediatric tumors with poor survival outcomes; despite the use of intensive multimodality treatment regimens, novel targeted therapies are urgently required. Both diseases frequently exhibit aberrations in the cyclin D–cyclin-dependent kinase (CDK) 4/6–inhibitor of CDK4 (INK4)–retinoblastoma pathway, and preclinical evidence suggests that targeting this pathway with the CDK4/6 inhibitor ribociclib may be of therapeutic benefit.

This is the first clinical study exploring the use of a CDK4/6 inhibitor in a pediatric population. In this phase I study, we established the maximum tolerated dose (MTD) and recommended phase II dose (RP2D) of single-agent ribociclib in pediatric patients with neuroblastoma, MRTs, or other cancers with documented cyclin D–CDK4/6–INK4–retinoblastoma pathway aberrations. Ribociclib displayed an acceptable safety profile with dose-dependent pharmacokinetic characteristics, as well as preliminary signs of tumor stabilization, supporting further investigation of ribociclib in combination with other anticancer agents in pediatric cancers.

High-risk neuroblastoma and malignant rhabdoid tumors (MRT) are pediatric tumors with poor survival outcomes; despite the development of intensive multimodality therapy regimens, novel effective therapies are urgently needed (1–3). Molecular abnormalities that drive disease progression by promoting cell-cycle progression might present new therapeutic targets in these cancers (4).

Numerous studies have identified genetic aberrations in neuroblastoma and MRTs that increase cyclin-dependent kinase (CDK) 4/6 activity. CCND1, CDK4, and CDK6 have been shown to be highly expressed in neuroblastoma, and CCND1 and CDK6 expression were found to be uniformly high relative to other tumors (5–7). Genomic amplification of CCND1 (cyclin D1 tumors) and CDK4 and deletion of CDKN2A are correlated with poor prognosis and, notably, approximately 30% of neuroblastomas exhibit aberrations in genes which regulate the G1 checkpoint (6). An RNAi kinome screen in neuroblastoma cell lines identified CDK4 as a potential therapeutic target (8). In MRTs, biallelic loss of SMARCB1 (also known as SNF5, INI1, and BAF47), a gene that encodes a core subunit of the SWI/SNF chromatin remodeling complex, was reported in 12 of 13 (92%) MRT cell lines and 50 of 51 (98%) of tumors and appears to be a hallmark of this type of pediatric cancer (9, 10). SMARCB1 limits activity of cyclin D–CDK4/6 through direct transcriptional repression of CCND1 and upregulation of genes encoding the cyclin D–CDK4/6 negative regulators, p21Cip1 and p16INK4A (11, 12). Consequently, loss of SMARCB1 results in altered expression of key cell-cycle regulators and reversal of cell-cycle arrest (13). Genetic ablation experiments have demonstrated that cyclin D1 is a key mediator in the genesis of MRTs, as SMARCB1 heterozygous mice lacking cyclin D1 fail to develop MRTs, and siRNA-mediated knockdown of CCND1 induces G0–G1 arrest and apoptosis in MRT cells (14). Therefore, targeting cyclin D1 or the cyclin D–CDK4/6–inhibitor of CDK4 (INK4)–retinoblastoma (Rb) axis is a biologically rational strategy for inhibiting neuroblastoma and MRT growth.

Ribociclib (LEE011) is an orally bioavailable, highly specific inhibitor of CDK4/6 (15). In vitro screens of a panel of more than 500 cell lines constituting the Novartis Cancer Cell Line Encyclopedia identified neuroblastoma and MRT cell lines to be among the most sensitive to ribociclib treatment (15). In addition, ribociclib caused cell-cycle arrest and senescence in a series of human neuroblastoma-derived cell lines (particularly those harboring MYCN amplification; ref. 16), and treatment of both neuroblastoma and MRT subcutaneous xenograft models resulted in tumor growth delays (15, 16). Collectively, these findings led to the hypothesis that ribociclib may have activity in neuroblastoma and MRTs with CDK4/6 activation through p16 loss or CCND1, CDK4, or CDK6 amplification.

Here, we report results of the first clinical trial of a CDK4/6 inhibitor in pediatric patients, a phase I study of ribociclib in pediatric patients with MRT, neuroblastoma, or cyclin D–CDK4/6–INK4–Rb pathway-activated tumors (NCT01747876).

Study design and objectives

This was a phase I, multicenter, international, open-label study of single-agent ribociclib, composed of a dose-escalation portion in pediatric patients, followed by an expansion portion in pediatric patients with neuroblastoma and MRT. The primary objective was to determine the maximum tolerated dose (MTD) and/or recommended phase II dose (RP2D) of ribociclib. Secondary objectives included assessment of safety, tolerability, pharmacokinetics (PK), and preliminary clinical activity of ribociclib.

Study treatment

Patients received escalating doses of once-daily ribociclib on a 3-weeks-on/1-week-off schedule in 28-day treatment cycles. An intermittent dosing schedule was chosen on the basis of the expected bone marrow suppression associated with CDK4/6 inhibition which was confirmed in the adult phase I study with ribociclib (17). Ribociclib was administered orally in a fasted state, either as whole capsules or opened and poured onto semi-solid food or dissolved in water in case of swallowing difficulties. The starting dose was 280 mg/m2/d, which was 80% of the RP2D in adults (600 mg) adjusted for body surface area.

Patient population

Patients aged 1–21 years with a confirmed diagnosis of neuroblastoma, MRT, or other tumors with documented evidence of cyclin D–CDK4/6–INK4–Rb pathway abnormalities (assessed by DNA sequencing or immunohistochemistry) that had progressed despite standard therapy were included in this study. MRT included diagnoses of atypical teratoid rhabdoid tumor [ATRT; referred to here as primary central nervous system (CNS) MRT] and rhabdoid tumor of the kidney and other soft tissues, for which the morphology and immunophenotypic panel were consistent with rhabdoid tumors. Loss of SMARCB1 was confirmed by immunohistochemistry or molecular confirmation of biallelic SMARCB1 loss/mutation. Patients were required to have a Lansky play performance (≤16 years) or Karnofsky (>16 years) score of at least 50% and measurable and/or evaluable disease. Archival tumor biopsies for molecular analysis were required from patients at diagnosis or relapse prior to study entry.

Patients who had received myeloablative therapy with autologous stem cell transplant within 3 months, or any prior allogenic stem cell transplant, were excluded. Other key exclusion criteria included uncontrolled cardiovascular conditions, prior history of QTc prolongation or QT corrected using Fridericia's formula (QTcF) > 450 ms on screening, left ventricular ejection fraction < 45%, impairment of gastrointestinal (GI) function or GI disease that may affect the absorption of ribociclib, and prior exposure to CDK4/6 inhibitors.

Safety and efficacy assessments

Dose-limiting toxicities (DLTs) were evaluated during the first treatment cycle as part of the dose-escalation procedure and were defined as adverse events (AEs) or clinically significant abnormal laboratory values assessed as unrelated to disease, disease progression, intercurrent illness, or concomitant medications, occurring within the first 28 days of treatment with ribociclib. AEs were assessed continuously according to Common Terminology Criteria for AEs v4.03. Efficacy assessments were conducted at screening; day 22 of cycles 2, 4, and 6; day 22 of cycle 10, and every 4 cycles thereafter. Tumor response was assessed locally according to RECIST 1.1 for all patients. Where possible, patients with neuroblastoma were also assessed by modified International Neuroblastoma Response Criteria (INRC), with stable disease (SD) defined as neither sufficient tumor shrinkage to qualify for partial response nor sufficient growth to qualify for progressive disease (18), and those with primary CNS tumors were assessed by Revised Assessment in Neuro-Oncology (RANO) criteria (19).

PK assessments

Samples for PK evaluation were collected on cycle 1 day 1 (at pre-dose, and 1, 2, 4, and 8 hours post-dose), day 2 pre-dose (at 24 hours post-day 1 dose), day 8 (at pre-dose and 2 hours post-dose), day 15 (at pre-dose, and 1, 2, 4, and 8 hours post-dose), and day 22 (at 24 hours post-day 21 dose), cycle 2 day 1 (at pre-dose and 4 hours post-dose), and cycle 3 day 1 (at pre-dose). Steady-state values were calculated by combining data collected on days 15 and 22. Plasma samples were assayed using a validated liquid chromatography-tandem mass spectrometry assay with a lower limit of quantification of 1.0 ng/mL.

Biomarker assessments

Next-generation sequencing (NGS) of archival tumor samples was conducted to evaluate the relationship between antitumor activity and molecular aberrations in cancer-related genes. Tumor samples were sequenced on the Foundation Medicine T5 panel of about 300 genes known to be altered in solid tumors (20). Known and likely mutations were defined by their allelic fraction within the tumor or by mutations in genes known to be tumor suppressors or oncogenes.

Statistical analyses

Dose escalation was guided by an adaptive Bayesian logistic regression model using the Escalation With Overdose Control principle (21). The dose-determining set, which included all patients from the safety set who met the minimum exposure criterion (≥16 of the 21 planned daily doses during the first cycle) and had scheduled safety evaluations or experienced a DLT, was used for DLT/MTD determination. Patients who did not experience a DLT during the first cycle were considered to have sufficient safety evaluations if observed for at least 28 days following the first dose with enough safety data to conclude that a DLT did not occur.

Initial enrollment at each dose level was in cohorts of 3 to 6 patients. The MTD was defined as the highest dose level not expected to result in DLTs in more than 33% of patients; the RP2D was subsequently selected, taking into account the overall safety profile and PK characteristics of ribociclib at each dose level. The Bayesian design allowed for enrollment of additional patients at lower dose levels that were considered safe, while higher dose levels were still being evaluated for DLTs, thereby maximizing recruitment opportunities in this rare patient population. Approximately 28 patients were required to be treated with ribociclib to ensure the model had reasonable operating characteristics to recommend an MTD.

Baseline patient characteristics, NGS, and efficacy analyses were performed on the full analysis set, comprising all patients who received at least one dose of ribociclib, whereas safety analyses were performed on the safety analysis set, consisting of all patients who received at least one dose of ribociclib and had at least one post-baseline safety assessment. PK analyses were conducted on the PK analysis set, which included all patients who had at least one blood sample providing evaluable drug concentration data. PK parameters were determined from individual concentration–time profiles in plasma by noncompartmental analysis using Phoenix software (Pharsight Corporation; Version 6.4). The parameters are summarized by dose group using descriptive statistics. SAS version 9.3 was used in all analyses, with the exception of Bayesian modeling, for which programs were compiled using R (version 2.15.1) and WinBUGS (version 1.4.3).

The cutoff date for safety and efficacy analyses was December 25, 2014, and for PK and NGS data was April 9, 2015. A further update on patient disease status was provided at the time of writing this article (February 8, 2016).

Ethics

The study was designed and conducted in accordance with the ICH Harmonized Tripartite Guidelines for Good Clinical Practice, with applicable local regulations (including European Directive 2001/20/EC and US Code of Federal Regulations Title 21), and with the ethical principles laid down in the Declaration of Helsinki. The study protocol and informed consent form were approved by an institutional review board or ethics committee at each site, and all patients (or their guardians) provided written informed consent prior to study entry.

Patient characteristics

Between May 2013 and June 2014, 32 patients were enrolled. Median age was 5.5 years (range, 1.0–20.0 years) and 22 (69%) patients were aged ≤12 years (Table 1). Fifteen patients (47%) had neuroblastoma, 15 had MRT (of which 13 had primary CNS MRT), one (3%) had rhabdomyosarcoma with CDK4 amplification, and one had anaplastic meningioma with rhabdoid differentiation and homozygous loss of CDKN2A/B. All patients underwent prior surgery and received prior antineoplastic therapy; the median number of prior treatment regimens was 3. Most patients (88%) also had prior radiotherapy.

Table 1.

Patient and disease characteristics

Ribociclib 280 mg/m2Ribociclib 350 mg/m2Ribociclib 470 mg/m2All patients
Characteristicn = 5n = 15n = 12N = 32
Median age, y (range) 2.0 (1.0–17.0) 6.0 (2.0–20.0) 7.0 (1.0–20.0) 5.5 (1.0–20.0) 
 1–<2, n (%) 2 (40) 1 (8) 3 (9) 
 2–<6, n (%) 2 (40) 7 (47) 4 (33) 13 (41) 
 6–<12, n (%) 3 (20) 3 (25) 6 (19) 
 12–<18, n (%) 1 (20) 4 (27) 2 (17) 7 (22) 
 ≥18, n (%) 1 (7) 2 (17) 3 (9) 
Male, n (%) 4 (80) 9 (60) 8 (67) 21 (66) 
Median BSA at screening, m2 (range) 0.6 (0.4–1.7) 0.8 (0.5–1.8) 0.9 (0.6–1.7) 0.8 (0.4–1.8) 
Diagnosis, n (%) 
 Neuroblastoma 1 (20) 7 (47) 7 (58) 15 (47) 
 Primary CNS MRT 3 (60) 7 (47) 3 (25) 13 (41) 
 Extra-CNS MRT 1 (20) 1 (8) 2 (6) 
 Rhabdomyosarcomaa 1 (7) 1 (3) 
 Anaplastic meningioma with rhabdoid differentiationb 1 (8) 1 (3) 
Performance status [Lansky scale (≤ 16 y) or Karnofsky score (> 16 y)], n (%) 
 100 1 (20) 9 (60) 7 (58) 17 (53) 
 90 4 (80) 2 (13) 3 (25) 9 (28) 
 80 2 (13) 1 (8) 3 (9) 
 70 1 (7) 1 (3) 
 50 1 (7) 1 (3) 
 40c 1 (8) 1 (3) 
Prior surgery, n (%) 5 (100) 15 (100) 12 (100) 32 (100) 
Prior radiotherapy, n (%) 4 (80) 15 (100) 9 (75) 28 (88) 
Number of prior antineoplastic regimens, n (%) 
 0 or 1 2 (40) 3 (20) 4 (33) 9 (28) 
 2 or 3 2 (40) 4 (27) 3 (25) 9 (28) 
 ≥4 1 (20) 8 (53) 5 (42) 14 (44) 
Ribociclib 280 mg/m2Ribociclib 350 mg/m2Ribociclib 470 mg/m2All patients
Characteristicn = 5n = 15n = 12N = 32
Median age, y (range) 2.0 (1.0–17.0) 6.0 (2.0–20.0) 7.0 (1.0–20.0) 5.5 (1.0–20.0) 
 1–<2, n (%) 2 (40) 1 (8) 3 (9) 
 2–<6, n (%) 2 (40) 7 (47) 4 (33) 13 (41) 
 6–<12, n (%) 3 (20) 3 (25) 6 (19) 
 12–<18, n (%) 1 (20) 4 (27) 2 (17) 7 (22) 
 ≥18, n (%) 1 (7) 2 (17) 3 (9) 
Male, n (%) 4 (80) 9 (60) 8 (67) 21 (66) 
Median BSA at screening, m2 (range) 0.6 (0.4–1.7) 0.8 (0.5–1.8) 0.9 (0.6–1.7) 0.8 (0.4–1.8) 
Diagnosis, n (%) 
 Neuroblastoma 1 (20) 7 (47) 7 (58) 15 (47) 
 Primary CNS MRT 3 (60) 7 (47) 3 (25) 13 (41) 
 Extra-CNS MRT 1 (20) 1 (8) 2 (6) 
 Rhabdomyosarcomaa 1 (7) 1 (3) 
 Anaplastic meningioma with rhabdoid differentiationb 1 (8) 1 (3) 
Performance status [Lansky scale (≤ 16 y) or Karnofsky score (> 16 y)], n (%) 
 100 1 (20) 9 (60) 7 (58) 17 (53) 
 90 4 (80) 2 (13) 3 (25) 9 (28) 
 80 2 (13) 1 (8) 3 (9) 
 70 1 (7) 1 (3) 
 50 1 (7) 1 (3) 
 40c 1 (8) 1 (3) 
Prior surgery, n (%) 5 (100) 15 (100) 12 (100) 32 (100) 
Prior radiotherapy, n (%) 4 (80) 15 (100) 9 (75) 28 (88) 
Number of prior antineoplastic regimens, n (%) 
 0 or 1 2 (40) 3 (20) 4 (33) 9 (28) 
 2 or 3 2 (40) 4 (27) 3 (25) 9 (28) 
 ≥4 1 (20) 8 (53) 5 (42) 14 (44) 

NOTE: Data cutoff: December 25, 2014.

Abbreviation: BSA, body surface area.

aRhabdomyosarcoma with CDK4 amplification.

bAnaplastic meningioma with rhabdoid differentiation, with CDKN2A/B loss.

cProtocol waiver approved.

Study treatment

Five patients received ribociclib 280 mg/m2, 15 received 350 mg/m2, and 12 received 470 mg/m2. The median exposure to treatment was 2 (range, 0–13) cycles and 7 patients received ribociclib for ≥4 cycles. At data cutoff (December 25, 2014), 27 patients had discontinued treatment and 5 patients were still on treatment. The primary reasons for treatment discontinuation were disease progression (n = 23, 72%), AEs (n = 2, 6%), withdrawal of consent (n = 1, 3%), and death due to disease progression (n = 1, 3%).

DLTs

Thirty patients were evaluable for the dose-determining set. Two patients were excluded as they had not received 75% (≥16 days) of the planned daily doses of ribociclib during cycle 1 and did not experience DLTs. Of the 30 evaluable patients, 3 patients reported DLTs during cycle 1: 1 case of grade 3 fatigue at 280 mg/m2 and 2 of grade 4 thrombocytopenia at 470 mg/m2. In addition, 33% of patients at the 470 mg/m2 dose level experienced hematologic toxicities that despite not meeting the criteria for DLT, led to dose delays and modifications in subsequent cycles; no hematologic DLTs were reported at the 350 mg/m2 dose. The safety and tolerability data did not support further dose escalation of ribociclib and the MTD was determined as 470 mg/m2/d on a 3-weeks-on/1-week-off dosing schedule. The 350 mg/m2/d dose, which is equivalent to the adult RP2D (600 mg; refs. 17, 22), was selected as the RP2D based on the overall safety profile and PK characteristics (discussed below).

Safety and tolerability

All patients were evaluable for safety analysis. AEs of any grade suspected to be study drug-related occurred in 31 (97%) patients and grade 3/4 AEs occurred in 25 (78%) patients (Table 2).

Table 2.

AEs (all grade ≥ 10%) suspected to be study drug-related

Ribociclib 280 mg/m2Ribociclib 350 mg/m2Ribociclib 470 mg/m2All patients
n = 5n = 15n = 12N = 32
AEs, n (%)All gradeGrade 3/4All gradeGrade 3/4All gradeGrade 3/4All gradeGrade 3/4
All AEs 4 (80) 2 (40) 15 (100) 13 (87) 12 (100) 10 (83) 31 (97) 25 (78) 
Hematologic AEs 
 Neutropenia 2 (40) 1 (20) 13 (87) 11 (73) 8 (67) 8 (67) 23 (72) 20 (63) 
 Leukopenia 1 (20) 11 (73) 7 (47) 8 (67) 5 (42) 20 (63) 12 (38) 
 Anemia 1 (20) 7 (47) 1 (7) 6 (50) 14 (44) 1 (3) 
 Thrombocytopenia 8 (53) 4 (27) 6 (50) 5 (42) 14 (44) 9 (28) 
 Lymphopenia 6 (40) 2 (13) 6 (50) 4 (33) 12 (38) 6 (19) 
Nonhematologic AEs 
 Vomiting 1 (20) 6 (40) 5 (42) 12 (38) 
 Fatigue 1 (20) 1 (20) 3 (20) 4 (33) 8 (25) 1 (3) 
 Nausea 1 (20) 4 (27) 3 (25) 8 (25) 
 QTc prolongation 4 (27) 3 (25) 7 (22) 
 Decreased appetite 1 (20) 1 (20) 3 (20) 2 (17) 6 (19) 1 (3) 
 AST increased 2 (13) 1 (7) 3 (25) 5 (16) 1 (3) 
 Asthenia 3 (20) 1 (8) 4 (13) 
 Increased creatinine 3 (20) 1 (8) 4 (13) 
 Hypophosphatemia 1 (20) 3 (20) 4 (13) 
Ribociclib 280 mg/m2Ribociclib 350 mg/m2Ribociclib 470 mg/m2All patients
n = 5n = 15n = 12N = 32
AEs, n (%)All gradeGrade 3/4All gradeGrade 3/4All gradeGrade 3/4All gradeGrade 3/4
All AEs 4 (80) 2 (40) 15 (100) 13 (87) 12 (100) 10 (83) 31 (97) 25 (78) 
Hematologic AEs 
 Neutropenia 2 (40) 1 (20) 13 (87) 11 (73) 8 (67) 8 (67) 23 (72) 20 (63) 
 Leukopenia 1 (20) 11 (73) 7 (47) 8 (67) 5 (42) 20 (63) 12 (38) 
 Anemia 1 (20) 7 (47) 1 (7) 6 (50) 14 (44) 1 (3) 
 Thrombocytopenia 8 (53) 4 (27) 6 (50) 5 (42) 14 (44) 9 (28) 
 Lymphopenia 6 (40) 2 (13) 6 (50) 4 (33) 12 (38) 6 (19) 
Nonhematologic AEs 
 Vomiting 1 (20) 6 (40) 5 (42) 12 (38) 
 Fatigue 1 (20) 1 (20) 3 (20) 4 (33) 8 (25) 1 (3) 
 Nausea 1 (20) 4 (27) 3 (25) 8 (25) 
 QTc prolongation 4 (27) 3 (25) 7 (22) 
 Decreased appetite 1 (20) 1 (20) 3 (20) 2 (17) 6 (19) 1 (3) 
 AST increased 2 (13) 1 (7) 3 (25) 5 (16) 1 (3) 
 Asthenia 3 (20) 1 (8) 4 (13) 
 Increased creatinine 3 (20) 1 (8) 4 (13) 
 Hypophosphatemia 1 (20) 3 (20) 4 (13) 

NOTE: Data cutoff: December 25, 2014.

Abbreviation: AST, aspartate aminotransferase.

The most common all-grade (≥35%) AEs suspected to be study drug-related were hematologic including neutropenia (72% of patients; 63% grade 3/4), leukopenia (63%; 38%), anemia (44%; 3%), thrombocytopenia (44%; 28%), and lymphopenia (38%; 19%). The most common all-grade (≥20%) nonhematologic AEs suspected to be study drug-related were vomiting (38%; 0% grade 3/4), fatigue (25%; 3%), nausea (25%; 0%), and QTc prolongation (22%; 0% grade 2–4). Newly occurring post-baseline values of QTcF ≥ 450 ms were observed in 3 patients (9%): 2 (13%) in the 350 mg/m2 group at cycle 1 day 15 and 1 (8%) in the 470 mg/m2 group at cycle 1 day 8. An increase in baseline of >30 ms was reported in 12 patients (38%): 6 (40%) in the 350 mg/m2 group and 6 (50%) in the 470 mg/m2 group.

Clinical activity

All 32 patients were evaluable for efficacy (Fig. 1). At data cutoff, the best overall response per RECIST 1.1 or INRC was SD in 9 of the 32 (28%) patients (7 with neuroblastoma and 2 with primary CNS MRT). Five of these patients achieved prolonged SD for more than 6, 6, 8, 12, and 13 cycles, respectively. Analysis of disease history, metaiodobenzylguanidine (MIBG) status, and prior treatment experience revealed no obvious differences between the clinical characteristics of patients with neuroblastoma who achieved SD and those who did not.

Figure 1.

Duration of exposure to ribociclib. Graph illustrates treatment responses, per RECIST 1.1, and duration of exposure following ribociclib administration in pediatric patients (N = 32) with primary CNS MRT, NB, ANM, MRT, and RMS. The different colors correspond to the different doses of ribociclib administered. Symbols refer to treatment responses (SD, progressive disease, neither complete nor progressive disease). Primary tumor type is indicated for each patient. At data cutoff, 4 patients had SD for more than 24 weeks; 5 patients were ongoing (>). ANM, anaplastic meningioma; NB, neuroblastoma; RMS, rhabdomyosarcoma. Data cutoff: December 25, 2014.

Figure 1.

Duration of exposure to ribociclib. Graph illustrates treatment responses, per RECIST 1.1, and duration of exposure following ribociclib administration in pediatric patients (N = 32) with primary CNS MRT, NB, ANM, MRT, and RMS. The different colors correspond to the different doses of ribociclib administered. Symbols refer to treatment responses (SD, progressive disease, neither complete nor progressive disease). Primary tumor type is indicated for each patient. At data cutoff, 4 patients had SD for more than 24 weeks; 5 patients were ongoing (>). ANM, anaplastic meningioma; NB, neuroblastoma; RMS, rhabdomyosarcoma. Data cutoff: December 25, 2014.

Close modal

On a subsequent follow-up for clinical response in February 2016, 5 patients had SD for >6 months. Two patients with heavily pretreated neuroblastoma upon study entry progressed after achieving SD for 15 and 8 months, respectively; the patient who previously achieved prolonged SD for 13 cycles (who had MIBG-positive neuroblastoma and bone marrow metastases) remained on treatment with a best overall response of SD after 26 months; the 2 patients with primary CNS MRT remained on treatment with SD after 24 and 20 months, respectively.

PK analysis

Following oral dosing, ribociclib was rapidly absorbed with the median time to reach peak plasma concentrations (Tmax) between 2 and 4 hours across dose levels, and exposure demonstrated dose dependency with high interpatient variability (Table 3). On the basis of trough concentrations, steady state appeared to have been reached by approximately day 8 of repeat dosing. At the RP2D, the geometric mean (CV%) area under the curve from time 0 to 24 hours (AUC0–24) was 25,200 h*ng/mL (46.1%), and the maximum plasma concentration (Cmax) was 1,790 ng/mL (55.2%) at steady state. The overall accumulation of ribociclib from day 1 to steady state at day 15 was 2- to 3-fold and did not differ between dose levels. The median effective half-life at steady state ranged from 30 to 41 hours across dose levels.

Table 3.

PK parameters

DoseDayStatisticsAUC0–24, h*ng/mLCmax, ng/mLTmax, hCLss/F, L/h
Ribociclib 280 mg/m2, n = 5 C1D1 n – 
  Mean (SD) 10,700 (7,000) 1,040 (542) – – 
  Geometric mean (CV%) 9,120 (70) 944 (53) – – 
  Median (range) 9,250 (4,580–22,000) 937 (542–1,910) 2.0 (1.1–4.1) – 
 C1D15 n 
  Mean (SD) 27,600 (28,200) 2,070 (2,100) – 11 (12.9) 
  Geometric mean (CV%) 19,500 (131) 1,530 (99) – 6.1 (408) 
  Median (range) 13,800 (8,910–60,100) 1,110 (860–5,210) 2.1 (1.1–3.8) 11 (1.8–20.1) 
Ribociclib 350 mg/m2, n = 15 C1D1 n 12 12 12 – 
  Mean (SD) 14,500 (11,900) 1,480 (1,140) – – 
  Geometric mean (CV%) 11,500 (74) 1,210 (68) – – 
  Median (range) 10,000 (5,320–43,600) 1,130 (514–4,140) 2.0 (1.0–4.1) – 
 C1D15 n 12 12 
  Mean (SD) 27,700 (15,000) 2,010 (958) – 10.1 (3.3) 
  Geometric mean (CV%) 25,200 (46) 1,790 (55) – 9.7 (31) 
  Median (range) 24,500 (15,100–60,100) 2,010 (633–4,270) 2.1 (1.1–23.8) 9.4 (6.8–15.6) 
Ribociclib 470 mg/m2, n = 12 C1D1 n – 
  Mean (SD) 18,700 (6,730) 2,060 (1,140) – – 
  Geometric mean (CV%) 17,500 (43) 1,830 (57) – – 
  Median (range) 17,600 (8,010–28,500) 1,960 (796–4,360) 4.0 (2.0–4.1) – 
 C1D15 n 10 10 
  Mean (SD) 30,300 (14,700) 2,540 (1,590) – 9.5 (3.1) 
  Geometric mean (CV%) 27,300 (57) 2,010 (92) – 9.2 (34) 
  Median (range) 29,100 (13,300–50,500) 2,500 (562–5,050) 3.9 (2.0–23.6) 9.4 (6.5–12.6) 
DoseDayStatisticsAUC0–24, h*ng/mLCmax, ng/mLTmax, hCLss/F, L/h
Ribociclib 280 mg/m2, n = 5 C1D1 n – 
  Mean (SD) 10,700 (7,000) 1,040 (542) – – 
  Geometric mean (CV%) 9,120 (70) 944 (53) – – 
  Median (range) 9,250 (4,580–22,000) 937 (542–1,910) 2.0 (1.1–4.1) – 
 C1D15 n 
  Mean (SD) 27,600 (28,200) 2,070 (2,100) – 11 (12.9) 
  Geometric mean (CV%) 19,500 (131) 1,530 (99) – 6.1 (408) 
  Median (range) 13,800 (8,910–60,100) 1,110 (860–5,210) 2.1 (1.1–3.8) 11 (1.8–20.1) 
Ribociclib 350 mg/m2, n = 15 C1D1 n 12 12 12 – 
  Mean (SD) 14,500 (11,900) 1,480 (1,140) – – 
  Geometric mean (CV%) 11,500 (74) 1,210 (68) – – 
  Median (range) 10,000 (5,320–43,600) 1,130 (514–4,140) 2.0 (1.0–4.1) – 
 C1D15 n 12 12 
  Mean (SD) 27,700 (15,000) 2,010 (958) – 10.1 (3.3) 
  Geometric mean (CV%) 25,200 (46) 1,790 (55) – 9.7 (31) 
  Median (range) 24,500 (15,100–60,100) 2,010 (633–4,270) 2.1 (1.1–23.8) 9.4 (6.8–15.6) 
Ribociclib 470 mg/m2, n = 12 C1D1 n – 
  Mean (SD) 18,700 (6,730) 2,060 (1,140) – – 
  Geometric mean (CV%) 17,500 (43) 1,830 (57) – – 
  Median (range) 17,600 (8,010–28,500) 1,960 (796–4,360) 4.0 (2.0–4.1) – 
 C1D15 n 10 10 
  Mean (SD) 30,300 (14,700) 2,540 (1,590) – 9.5 (3.1) 
  Geometric mean (CV%) 27,300 (57) 2,010 (92) – 9.2 (34) 
  Median (range) 29,100 (13,300–50,500) 2,500 (562–5,050) 3.9 (2.0–23.6) 9.4 (6.5–12.6) 

NOTE: Data cutoff: April 9, 2015.

Abbreviations: AUC0–24, area under the curve from time zero to 24 hours post-dose; C1D1, cycle 1 day 1; C1D15, cycle 1 day 15; CLss/F, apparent total body clearance of drug from the plasma at steady state; Cmax, maximum plasma concentration; CV, coefficient of variation; Tmax, time to reach maximum plasma concentration.

Genetic alterations

Twenty-five patients had samples available for mutation analysis, of which 19 samples had mutations identified by T5 panel analysis (Fig. 2). All patients with MRT for which samples were available (n = 11) were confirmed to have SMARCB1 alterations including one of the patients with SD; no alterations related to the cyclin D–CDK4/6–INK4–Rb pathway were identified in the patients with MRT. CDK4 amplification was identified in 3 patients with neuroblastoma; 2 of these patients had disease progression and 1 had SD and had ongoing treatment (>6 cycles) at data cutoff. Only one patient with neuroblastoma and progressive disease exhibited an amplification in MYCN, a gene previously linked to ribociclib sensitivity in neuroblastoma-derived cell lines (16). There were no discernible differences between the genetic profile of patients who had SD and those who had disease progression in patients who had neuroblastoma or MRT. Retrospective genetic and immunohistochemical analyses indicated that the RB gene was intact and expression of the protein was unaltered in all enrolled patients.

Figure 2.

Genetic alterations in tumor samples. Results of NGS analyses on 19 archival tumor samples to evaluate the relationship between antitumor activity and molecular aberrations. Tumor samples were sequenced on the Foundation Medicine T5 panel of about 300 genes known to be altered in solid tumors. Gain- or loss-of-function mutations in oncogenes and tumor suppressors and other alteration types are indicated for each patient (according to the legend). Treatment response (SD, PD, UNK), best percentage change in the sum of longest diameters (RECIST 1.1), duration of treatment exposure (days), primary tumor type, and genetic alteration are also displayed. NB, neuroblastoma; PD, progressive disease; UNK, unknown. Data cutoff: April 9, 2015. *, Best percentage change in sum of longest diameters as per RECIST 1.1; †, Best percentage change was not calculable, as patients had either non-target lesions at baseline or had different evaluation methods between baseline and post-baseline assessments; ‡, Patient did not have post-baseline efficacy evaluation.

Figure 2.

Genetic alterations in tumor samples. Results of NGS analyses on 19 archival tumor samples to evaluate the relationship between antitumor activity and molecular aberrations. Tumor samples were sequenced on the Foundation Medicine T5 panel of about 300 genes known to be altered in solid tumors. Gain- or loss-of-function mutations in oncogenes and tumor suppressors and other alteration types are indicated for each patient (according to the legend). Treatment response (SD, PD, UNK), best percentage change in the sum of longest diameters (RECIST 1.1), duration of treatment exposure (days), primary tumor type, and genetic alteration are also displayed. NB, neuroblastoma; PD, progressive disease; UNK, unknown. Data cutoff: April 9, 2015. *, Best percentage change in sum of longest diameters as per RECIST 1.1; †, Best percentage change was not calculable, as patients had either non-target lesions at baseline or had different evaluation methods between baseline and post-baseline assessments; ‡, Patient did not have post-baseline efficacy evaluation.

Close modal

This was the first study to evaluate the safety and show preliminary activity of the highly selective CDK4/6 inhibitor ribociclib in pediatric patients with neuroblastoma, MRT, or other cancers with documented cyclin D–CDK4/6–INK4–Rb pathway aberrations. Initiation of this first-in-class pediatric study occurred early in development due to the biological rationale (supported by preclinical data), and the urgent need for effective targeted therapies in these poor prognosis tumor types, where disease stabilization can be considered a meaningful outcome.

The first-in-human phase I study, in which adult patients with advanced solid tumors or lymphomas were treated with single-agent ribociclib, demonstrated preliminary antitumor activity. Among 132 evaluable patients, 3 had partial responses (one with head and neck acinar carcinoma and CDK2NA loss, one with PIK3CA-mutant, CCND1-amplified, estrogen receptor–positive breast cancer, and one with BRAF/NRAS wild-type, CCND1-amplified melanoma); 43 (32.6%) patients had SD (17). The most common study drug-related grade 3/4 adverse events in ≥10% of adult patients at the RP2D of 600 mg/d (3-weeks-on/1-week-off schedule) were neutropenia (28%), leukopenia (19%), and lymphopenia (18%; ref. 17).

In the current study, an intermittent 3-weeks-on/1-week-off schedule for ribociclib was investigated in pediatric patients. Although greater efficacy might be obtained with continuous daily dosing to prevent any potential tumor regrowth during the resting period, continuous dosing of 400 and 600 mg dose levels in adult patients was deemed infeasible due to the high level of hematologic toxicities requiring dose interruptions (17). Therefore, an intermittent schedule was selected in the current study, at doses of ≥280 mg/m2 (considered to be therapeutic; equivalent to adult doses of ≥400 mg).

Single-agent ribociclib was well-tolerated in pediatric patients, with a similar safety profile to that observed in adult patients (17). The majority of AEs were hematologic, and there was no worsening of toxicities following cycle 1, which was also consistent with the experience in adult patients (17). Grade 3/4 hematologic AEs were more frequent at the pediatric 350 mg/m2 dose level in this heavily pretreated patient population than the equivalent 600 mg/d intermittent dose in adult patients; neutropenia (73% vs. 28%), leukopenia (47% vs. 19%), thrombocytopenia (27% vs. 9%), and lymphopenia (13% vs. 18%; ref. 17). Hematologic AEs have also been associated with the CDK4/6 inhibitor palbociclib, suggesting an on-target effect for this class of drugs (23). At the 470 mg/m2 dose of ribociclib, hematologic toxicities were predominant and thrombocytopenia was the DLT in pediatric patients. In addition, nausea, fatigue, and vomiting were frequently observed at this dose, preventing further escalation—the MTD was therefore established as 470 mg/m2/d on a 3-weeks-on/1-week-off schedule.

Ribociclib exposure was dose-dependent, although PK parameters in pediatric patients exhibited high interpatient variability, as also reported in adult patients. A robust analysis of PK parameters and age or body surface area was not possible due to the limited number of patients treated; however, ribociclib exposures at 350 and 470 mg/m2 were found to be equivalent to those observed at 600 to 900 mg fixed doses in adults in the first-in-human study (17). On the basis of all available safety and PK data, ribociclib 350 mg/m2/d on a 3-weeks-on/1-week-off dosing schedule was selected as the RP2D in pediatric patients.

During this study, the best overall response was SD per RECIST 1.1 or INRC, achieved by 7 patients with neuroblastoma and 2 patients with primary CNS MRT. All the patients with SD received ribociclib for more than 4 cycles. There were no objective responses at data cutoff; therefore, the expansion portion of the study did not commence. Nevertheless, on a subsequent follow-up for clinical response, 1 patient with neuroblastoma and 2 patients with primary CNS MRT remained with SD on treatment for ≥20 months. A review of these patients' demographic profiles, prior medical history, and tumor genetics did not reveal any unique features that were predictive of clinical response. However, the observations are noteworthy and warrant further evaluation of ribociclib in these tumors in the clinic, particularly in light of recent preclinical evidence suggesting that ribociclib penetrates into the mouse CNS (24).

NGS data were limited and few mutations were identified other than those that were expected, such as SMARCB1 mutations in patients with MRT, precluding definitive conclusions about the functions of genomic alterations (mutations, deletions, amplifications) in relation to CDK4/6 biology. None of these pediatric tumors exhibited CCND1 gene amplification, an alteration found in 2 of 3 cancers in adult patients who experienced an objective tumor response to ribociclib. In addition, only one patient with progressive disease exhibited MYCN amplification, an alteration associated with ribociclib sensitivity in preclinical studies of neuroblastoma (16). It must be noted that the NGS analysis was performed on a limited number of patient samples, and the association between CCND1 or MYCN amplification and treatment response could therefore not be evaluated. Further analyses in a larger patient population are required to determine the impact of genetic alterations, including CCND1 and MYCN amplifications, on ribociclib sensitivity. The identification of reliable biomarkers for response to CDK4/6 inhibitors has met with significant challenges, particularly with the use of NGS. Alternative pharmacodynamic analyses using biomarkers of on-target effects, such as the impact of treatment on Rb phosphorylation, may provide further insights into the activity of ribociclib in certain patient populations.

Neuroblastoma has a complex genetic profile that can include MYCN amplification, somatic activating mutations in the ALK oncogene, and other aberrations that activate CDK4/6 activity (6, 25). It has also recently been shown that high-risk neuroblastoma may be defined by activation of telomere maintenance mechanisms caused by MYCN amplification, TERT rearrangements, or ATRX mutations (26), and relapsed neuroblastomas are frequently associated with RAS/mitogen-activated protein kinase (MAPK) pathway mutations that could also influence ribociclib sensitivity (27). In MRT, disruption of the SMARCB1 transcriptional regulator can alter the expression of multiple genes upstream and downstream of CDK4/6 (28, 29). In addition, the LIN28/RAS/MAPK pathway has also recently been identified as a potential driver of MRT tumorigenesis (29). Collectively, these findings suggest that cell-cycle alterations in pediatric tumors such as neuroblastoma and MRT may not be the sole events driving tumorigenesis. Therefore, these pediatric tumors may require targeted, combinatorial therapeutic approaches. However, a key question remains: whether these alternative approaches should select patients with cyclin D–CDK4/6–INK4–Rb pathway alterations or include patients regardless of their pathway status.

Clinical studies of ribociclib in combination with other therapies are currently ongoing in various cancer types. Combination strategies may overcome prior treatment resistance or enhance treatment efficacy where dysregulation of more than one biological pathway is responsible for driving the disease; however, the mechanism of action and cumulative toxicities of additional agents must be carefully considered when designing treatment regimens.

The results from this study demonstrate that ribociclib has an acceptable safety profile and dose-dependent PK in pediatric patients. Several heavily pretreated patients had prolonged SD. Further studies of ribociclib in combination with other active agents in patients with MRT and neuroblastoma are ongoing, such as the AcSé-ESMART multi-arm trial (NCT02813135), which includes ribociclib in combination with everolimus (a PI3K/AKT/mTOR pathway inhibitor) or chemotherapy in patients prescreened for specific molecular alterations. The combination with everolimus was based on the observation that the PI3K/AKT/mTOR pathway, for example, via MYCN, ALK, or RAS activation, may be implicated in the pathogenesis of neuroblastoma and MRT (30, 31). Other ribociclib-based combinations with other targeted therapies, such as MAPK/ERK inhibitors, may be considered in future trials.

S.G. DuBois is a consultant/advisory board member for Loxo. M. Fischer is a consultant/advisory board member for Novartis. A. Marabelle reports receiving speakers bureau honoraria from Bristol-Myers Squibb, GlaxoSmithKline, Merck MSD, Merck Serono, and Roche/Genentech and is a consultant/advisory board member for Genmab, Genticel, Lytix Pharma, Merck Serono, Novartis, Pierre Fabre, and Roche/Genentech. A. Matano reports receiving commercial research grants from Novartis. No potential conflicts of interest were disclosed by the other authors.

Conception and design: B. Geoerger, F. Bourdeaut, J.I. Geller, A.D.J. Pearson, T. Cash, A. Matano, S. Parasuraman, S. Chi

Development of methodology: B. Geoerger, J.I. Geller, A.D.J. Pearson, A. Matano, S. Chi

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): B. Geoerger, F. Bourdeaut, S.G. DuBois, M. Fischer, J.I. Geller, N.G. Gottardo, A. Marabelle, A.D.J. Pearson, S. Modak, G. Robinson, A. Matano, S. Parasuraman, S. Chi

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): B. Geoerger, J.I. Geller, A.D.J. Pearson, T. Cash, A. Matano, S.G. Bhansali, J.R. Dobson, S. Parasuraman

Writing, review, and/or revision of the manuscript: B. Geoerger, F. Bourdeaut, M. Fischer, J.I. Geller, N.G. Gottardo, A. Marabelle, A.D.J. Pearson, S. Modak, T. Cash, G. Robinson, A. Matano, S.G. Bhansali, J.R. Dobson, S. Parasuraman, S. Chi

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): M. Fischer, J.I. Geller, A. Matano, S. Parasuraman

Study supervision: J.I. Geller, A.D.J. Pearson, M. Motta, S. Parasuraman

The authors wish to acknowledge Charles W. M. Roberts, MD, PhD, for generating the preliminary data that served as the basis for the rhabdoid trial and for initially proposing the testing of LEE011 in rhabdoid tumor patients. We thank all patients and their parents and the medical teams of the institutions that participated in the trial. Ribociclib was discovered by Novartis Institutes for BioMedical Research in collaboration with Astex Pharmaceuticals. We thank Eisha Comar PhD for medical editorial assistance with this article.

Financial support for medical editorial assistance was provided by Novartis Pharmaceuticals Corporation.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Pinto
NR
,
Applebaum
MA
,
Volchenboum
SL
,
Matthay
KK
,
London
WB
,
Ambros
PF
, et al
Advances in risk classification and treatment strategies for neuroblastoma
.
J Clin Oncol
2015
;
33
:
3008
17
.
2.
Morgenstern
DA
,
Gibson
S
,
Brown
T
,
Sebire
NJ
,
Anderson
J
. 
Clinical and pathological features of paediatric malignant rhabdoid tumours
.
Pediatr Blood Cancer
2010
;
54
:
29
34
.
3.
Brennan
C
,
Verhaak
R
,
McKenna
A
,
Campos
B
,
Noushmehr
H
,
Salama
S
, et al
The somatic genomic landscape of glioblastoma
.
Cell
2013
;
155
;
462
77
.
4.
Fruhwald
MC
,
Biegel
JA
,
Bourdeaut
F
,
Roberts
CW
,
Chi
SN
. 
Atypical teratoid/rhabdoid tumors-current concepts, advances in biology, and potential future therapies
.
Neuro Oncol
2016
;
18
:
764
78
.
5.
Molenaar
JJ
,
Ebus
ME
,
Koster
J
,
van Sluis
P
,
van Noesel
CJ
,
Versteeg
R
, et al
Cyclin D1 and CDK4 activity contribute to the undifferentiated phenotype in neuroblastoma
.
Cancer Res
2008
;
68
:
2599
609
.
6.
Molenaar
JJ
,
Koster
J
,
Ebus
ME
,
van Sluis
P
,
Westerhout
EM
,
de Preter
K
, et al
Copy number defects of G1-cell cycle genes in neuroblastoma are frequent and correlate with high expression of E2F target genes and a poor prognosis
.
Genes Chromosomes Cancer
2012
;
51
:
10
9
.
7.
Guimier
A
,
Ferrand
S
,
Pierron
G
,
Couturier
J
,
Janoueix-Lerosey
I
,
Combaret
V
, et al
Clinical characteristics and outcome of patients with neuroblastoma presenting genomic amplification of loci other than MYCN
.
PLoS One
2014
;
9
:
e101990
.
8.
Cole
KA
,
Huggins
J
,
Laquaglia
M
,
Hulderman
CE
,
Russell
MR
,
Bosse
K
, et al
RNAi screen of the protein kinome identifies checkpoint kinase 1 (CHK1) as a therapeutic target in neuroblastoma
.
Proc Natl Acad Sci U S A
2011
;
108
:
3336
41
.
9.
Versteege
I
,
Sevenet
N
,
Lange
J
,
Rousseau-Merck
MF
,
Ambros
P
,
Handgretinger
R
, et al
Truncating mutations of hSNF5/INI1 in aggressive paediatric cancer
.
Nature
1998
;
394
:
203
6
.
10.
Jackson
EM
,
Sievert
AJ
,
Gai
X
,
Hakonarson
H
,
Judkins
AR
,
Tooke
L
, et al
Genomic analysis using high-density single nucleotide polymorphism-based oligonucleotide arrays and multiplex ligation-dependent probe amplification provides a comprehensive analysis of INI1/SMARCB1 in malignant rhabdoid tumors
.
Clin Cancer Res
2009
;
15
:
1923
30
.
11.
Kuwahara
Y
,
Charboneau
A
,
Knudsen
ES
,
Weissman
BE
. 
Reexpression of hSNF5 in malignant rhabdoid tumor cell lines causes cell cycle arrest through a p21(CIP1/WAF1)-dependent mechanism
.
Cancer Res
2010
;
70
:
1854
65
.
12.
Zhang
ZK
,
Davies
KP
,
Allen
J
,
Zhu
L
,
Pestell
RG
,
Zagzag
D
, et al
Cell cycle arrest and repression of cyclin D1 transcription by INI1/hSNF5
.
Mol Cell Biol
2002
;
22
:
5975
88
.
13.
Venneti
S
,
Le
P
,
Martinez
D
,
Eaton
KW
,
Shyam
N
,
Jordan-Sciutto
KL
, et al
p16INK4A and p14ARF tumor suppressor pathways are deregulated in malignant rhabdoid tumors
.
J Neuropathol Exp Neurol
2011
;
70
:
596
609
.
14.
Tsikitis
M
,
Zhang
Z
,
Edelman
W
,
Zagzag
D
,
Kalpana
GV
. 
Genetic ablation of cyclin D1 abrogates genesis of rhabdoid tumors resulting from Ini1 loss
.
Proc Natl Acad Sci U S A
2005
;
102
:
12129
34
.
15.
Kim
S
,
Loo
A
,
Chopra
R
,
Caponigro
G
,
Huang
A
,
Vora
S
, et al
LEE011: an orally bioavailable, selective small molecule inhibitor of CDK4/6–Reactivating rb in cancer [abstract]
.
Mol Cancer Ther
2013
;
12
.
pp. PR02
.
16.
Rader
J
,
Russell
MR
,
Hart
LS
,
Nakazawa
MS
,
Belcastro
LT
,
Martinez
D
, et al
Dual CDK4/CDK6 inhibition induces cell-cycle arrest and senescence in neuroblastoma
.
Clin Cancer Res
2013
;
19
:
6173
82
.
17.
Infante
J
,
Cassier
P
,
Gerecitano
J
,
Witteveen
P
,
Chugh
R
,
Ribrag
V
, et al
A phase I study of the cyclin-dependent kinase 4/6 inhibitor ribociclib (LEE011) in patients with advanced solid tumors and lymphomas
.
Clin Cancer Res
2016
;
22
:
5696
705
.
18.
Brodeur
GM
,
Pritchard
J
,
Berthold
F
,
Carlsen
NL
,
Castel
V
,
Castelberry
RP
, et al
Revisions of the international criteria for neuroblastoma diagnosis, staging, and response to treatment
.
J Clin Oncol
1993
;
11
:
1466
77
.
19.
Wen
PY
,
Macdonald
DR
,
Reardon
DA
,
Cloughesy
TF
,
Sorensen
AG
,
Galanis
E
, et al
Updated response assessment criteria for high-grade gliomas: response assessment in neuro-oncology working group
.
J Clin Oncol
2010
;
28
:
1963
72
.
20.
Frampton
GM
,
Fichtenholtz
A
,
Otto
GA
,
Wang
K
,
Downing
SR
,
He
J
, et al
Development and validation of a clinical cancer genomic profiling test based on massively parallel DNA sequencing
.
Nat Biotechnol
2013
;
31
:
1023
31
.
21.
Neuenschwander
B
,
Branson
M
,
Gsponer
T
. 
Critical aspects of the Bayesian approach to phase I cancer trials
.
Stat Med
2008
;
27
:
2420
39
.
22.
Bhansali
S
,
Parasuraman
S
,
Hewes
B
,
Einolf
H
,
Umehara
K
,
Matano
A
, et al
Application of SimCyp® and real-time pharmacokinetic assessments in defining dose in Novel–Novel combinations with potential Drug–Drug interactions in cancer patients [abstract]
.
In
:
Proceedings of the Annual Meeting of the American College of Clinical Pharmacology
; 
2014
Sep 14–16;
Atlanta GA
.
23.
Asghar
U
,
Witkiewicz
AK
,
Turner
NC
,
Knudsen
ES
. 
The history and future of targeting cyclin-dependent kinases in cancer therapy
.
Nat Rev Drug Discov
2015
;
14
:
130
46
.
24.
Patel
YT
,
Davis
A
,
Kala
A
,
Baker
S
,
Roussel
M
,
Stewart
CF
. 
CNS penetration of the CDK4/6 inhibitor ribociclib (LEE011) in non-tumor bearing mice and mice bearing orthotopic pediatric brain tumors
.
Neuro Oncol
2016
;
18
Suppl 6. Abstract nr PDTB-12
.
25.
Pugh
TJ
,
Morozova
O
,
Attiyeh
EF
,
Asgharzadeh
S
,
Wei
JS
,
Auclair
D
, et al
The genetic landscape of high-risk neuroblastoma
.
Nat Genet
2013
;
45
:
279
84
.
26.
Peifer
M
,
Hertwig
F
,
Roels
F
,
Dreidax
D
,
Gartlgruber
M
,
Menon
R
, et al
Telomerase activation by genomic rearrangements in high-risk neuroblastoma
.
Nature
2015
;
526
:
700
4
.
27.
Eleveld
TF
,
Oldridge
DA
,
Bernard
V
,
Koster
J
,
Daage
LC
,
Diskin
SJ
, et al
Relapsed neuroblastomas show frequent RAS-MAPK pathway mutations
.
Nat Genet
2015
;
47
:
864
71
.
28.
Ichimura
K
,
Nishikawa
R
,
Matsutani
M
. 
Molecular markers in pediatric neuro-oncology
.
Neuro Oncol
2012
;
14
Suppl 4
:
iv90
9
.
29.
Weingart
MF
,
Roth
JJ
,
Hutt-Cabezas
M
,
Busse
TM
,
Kaur
H
,
Price
A
, et al
Disrupting LIN28 in atypical teratoid rhabdoid tumors reveals the importance of the mitogen activated protein kinase pathway as a therapeutic target
.
Oncotarget
2015
;
6
:
3165
77
.
30.
Kiessling
MK
,
Curioni-Fontecedro
A
,
Samaras
P
,
Lang
S
,
Scharl
M
,
Aquzzi
A
, et al
Targeting the mTOR complex by everolimus in NRAS mutant neuroblastoma
.
PLoS One
2016
;
11
:
e0147682
.
31.
Westhoff
M-A
,
Karpel-Massler
G
,
Brühl
O
,
Enzenmüller
S
,
Ferla-Brühl
KL
,
Siegelin
MD
, et al
A critical evaluation of PI3K inhibition in glioblastoma and neuroblastoma therapy
.
Mol Cancer Ther
2014
;
27
:
32
.

Supplementary data