Abstract
Background:
The International Neuroblastoma Risk Group (INRG) classifier utilizes a staging system based on pretreatment imaging criteria in which image-defined risk factors (IDRFs) are used to evaluate the extent of locoregional disease. Children’s Oncology Group (COG) study ANBL0531 prospectively examined institutional determination of IDRF status and compared that to a standardized central review.
Methods:
Between 9/2009–6/2011, patients with intermediate-risk neuroblastoma were enrolled on ANBL0531 and had IDRF assessment at treating institutions. Paired COG pediatric surgeons and radiologists performed blinded central review of diagnostic imaging for the presence or absence of IDRFs. Second blinded review was performed in cases of discordance. Comparison of local and central review was performed using the Kappa coefficient to determine concordance in IDRF assessment.
Results:
211 patients enrolled in ANBL0531 underwent IDRF assessment; 3 patients were excluded due to poor image quality. Central reviewer pairs agreed on the presence or absence of any IDRF in 170/208 (81.7%; κ=0.48) cases. Thirteen (6.3%) cases could not be adjudicated after second blinded review. Radiologists were more likely to identify IRDFs as present than surgeons (p<0.001). Local and central reviewers agreed on the presence or absence of any IDRF in only108/208 (51.9%; κ=0.06) cases.
Conclusions:
Among experienced pediatric surgeons and radiologists participating in central review, concordance was moderate, with agreement in 81.7% of cases. On comparison of local and central assessment of IDRFs, concordance was poor. These data indicate that greater standardization, education, technology, and training are needed to improve the assessment of IDRFs in children with neuroblastoma.
Keywords: Neuroblastoma, Image-defined risk factor, IDRF, Concordance
1.0. Introduction
Neuroblastoma is a common childhood malignancy with heterogenous clinical phenotypes [1]. Patients at intermediate risk for recurrence, as determined by both clinical and biologic risk factors, typically undergo both chemotherapy and surgery, and overall survival is excellent at 93–100% [2, 3]. Treatment strategy and outcomes for patients with neuroblastoma largely depend on risk group, yet prior research was hampered by the use of different staging systems. The International Neuroblastoma Staging System (INSS) included completeness of surgical resection as a factor in the stratification of therapy but did not allow for de novo disease staging prior to surgical intervention. In 2005, the International Neuroblastoma Risk Group (INRG) collaborative sought to develop a pre-treatment risk classification system and addressed this challenge by utilizing image-defined risk factors (IDRF) at diagnosis to preoperatively identify tumors that may be challenging to resect [4]. A tumor is designated as having an IDRF if there is radiographic evidence of vascular encasement and invasion or compression of critical structures, with specific criteria for each body compartment. Furthermore, IDRFs are viewed as a proxy for as yet undefined underlying tumor biology that leads to a locally aggressive tumor phenotype. The concordance in identification of IDRFs by local treating sites and a central review panel has not been previously assessed.
The Children’s Oncology Group (COG) study ANBL0531 was designed as a therapeutic clinical trial for children with intermediate-risk neuroblastoma [2]. The primary objective of the study was to use clinical and tumor biologic factors to assign therapy, and in doing so reduce chemotherapy duration and cumulative dose to minimize toxicity in a subset of patients. Importantly, complete response was not required at the end of therapy. The goal of surgical resection in the context of this trial was safe tumor removal without damage to vital organs, blood vessels, or nerves. At the time of inception of this trial, INSS staging was used to assign treatment. A secondary objective of the study was to prospectively validate the prognostic value of the INRG IDRF system and to compare the local institutional determination of IDRFs with that of a COG central review committee.
2.0. Materials and Methods
The Children’s Oncology Group study ANBL0531 (ClinicalTrials.gov identifier: NCT00499616) was a prospective, single arm, phase III clinical trial conducted at COG sites following approval by institutional human subjects committees. Risk-stratification and therapy assignment utilized age, INSS stage, International Neuroblastoma Pathology Classification (INPC), tumor MYCN status, and tumor ploidy. At the time that ANBL0531 was conducted, neither IDRF status nor INRG stage was used for treatment assignment. Therapy reduction was prescribed for patients with favorable biology tumors, including those without loss of heterozygosity at 1p36 or 11q23. Treatment response was assessed using the 1993 International Neuroblastoma Response Criteria (INRC) after protocol-specific modification [5].
The study was amended on September 21, 2009, after initial study activation on October 8, 2007, to prospectively collect data regarding the presence or absence of IDRFs as assessed at the local (institutional) level and centrally based on the standard list of 20 possible IDRFs (Table 1). IDRFs vary based on the anatomic location of the tumor, and generally assess invasion or compression of local structures and/or encasement of nerves or critical blood vessels. Local institutions indicated on a case report form whether any IDRF was present on imaging performed at study entry. Blinded COG central review of initial cross-sectional imaging (either computed tomography (CT) or magnetic resonance imaging (MRI)) was performed by a panel of two experienced pediatric surgical oncologists and two experienced pediatric radiologists to compare with the local institutional assessment of IDRFs. For central review, baseline imaging studies for each patient were reviewed by one radiologist and one surgeon from the panel and each IDRF identified was scored as present, absent, or indeterminate. Patients were excluded from analysis if image quality was deemed poor by the central review panel. If any IDRF was present, the patient was scored as “IDRF present” and determination was compared between the surgeon and radiologist. When both evaluators agreed, the agreed upon designation was recorded. If there was disagreement between the initial two central reviewers, a blinded review by a second surgeon/radiologist participating in the panel was performed. If three of the four reviewers agreed as to the presence or absence of an IDRF, the consensus assignment was recorded. If two reviewers designated an IDRF as present and two found no IDRF, an unblinded final adjudication was performed among all four reviewers to reach a consensus. The final consensus decision was recorded as the central review assessment. IDRF review agreement was measured using the Kappa coefficient.
Table 1:
INRG Image-Defined Risk Factors (IDRF).
| Image-Define Risk Factors (IDRF) By Anatomic Compartment |
|---|
| Neck: |
| Tumor encasing carotid and/or vertebral artery and/or internal jugular vein |
| Tumor extending to base of skull |
| Tumor compressing the trachea |
| Cervico-thoracic junction: |
| Tumor encasing brachial plexus roots |
| Tumor encasing subclavian vessels and/or vertebral and/or carotid artery |
| Tumor compressing the trachea |
| Thorax: |
| Tumor encasing the aorta and/or major branches |
| Tumor compressing the trachea or principal bronchi |
| Lower left mediastinal tumor, infiltrating the costo-vertebral junction between T9–12 |
| Thoraco-abdominal: |
| Tumor encasing the aorta and/or vena cava |
| Abdomen: |
| Tumor infiltrating the porta hepatic and/or the hepatoduodenal ligament |
| Tumor encasing branches of the superior mesenteric artery at the mesenteric root |
| Tumor encasing the origin of the celiac axis, and/or of the superior mesenteric artery |
| Tumor invading one or both renal pedicles* |
| Tumor encasing the aorta and/or vena cava |
| Tumor encasing the iliac vessels |
| Pelvic tumor crossing the sciatic notch |
| Dumbbell tumors with signs of spinal cord compression (any compartment) |
| Infiltration of adjacent organs/structures: |
| Pericardium, diaphragm, kidney, liver, duodeno-pancreatic block, or mesentery |
Updated guideline in 2011 that any contact with the renal pedicle was to be considered an IDRF [6].
Of note, a guideline was published by the INRG to clarify the interpretation of IDRF involving the renal hilum in October 2011 [6], after ANBL0531 was closed to enrollment. This guideline clarified that any contact with the renal hilum was to be designated as an IDRF and encasement of the renal vessels was not required. This guideline was adhered to for central review but was not utilized for local review.
The primary tumor’s general anatomic location was described, and the median number of IDRFs identified per initial central reviewer pair was analyzed for each general anatomic location. The central reviewers’ specialty (surgeon or radiologist) was also noted. Differences between the outcomes of the surgeon and radiologist central reviews by anatomic location were assessed using the Wilcoxon signed-rank test and McNemar’s test. Analyses were performed using SAS® version 9.4. P-values less than 0.05 were considered statistically significant.
3.0. Results
A total of 404 eligible intermediate-risk patients were enrolled on ANBL0531 between October 2007 and June 2011. Of these patients, 211 (52.2%) had assessment of primary tumor IDRFs performed at the treating institution. Three patients were excluded from analysis due to poor quality of the images. Blinded central review of IDRFs was performed for 208 patients. The review panel consisted of two review teams with one radiologist and one surgeon on each team. One team initially reviewed 105 cases, and the second team reviewed 103 cases.
On primary central review, there was concordance between the surgeon and radiologist on the presence or absence of at least one IDRF in 170 out of the 208 patients (81.7%). (Figure 1A). The central reviewers disagreed in 38 cases (18.3%). Radiologists were more likely to identify IRDFs as present than surgeons (p<0.001). Correlation for identifying whether any IDRF was present between the radiologists and surgeons was 0.4837 using the Kappa coefficient, or moderate agreement (Table 2) [7].
Figure 1.
A. Frequencies of IDRF determination by surgeon and radiologist on primary central review. B. Frequencies of IDRF consensus by surgeon and radiologist on secondary central review and unblinded adjudication.
Table 2.
Frequencies of IDRF determination by surgeon and radiologist central reviewers. Kappa Coefficient presented as a measure of agreement between central reviewers.
| Concordance of Surgeon and Radiologist Central Review of IDRF Determination | |||
|---|---|---|---|
| Radiologist | |||
| Surgeon | No | Yes | Unknown |
| No | 27 | 31 | 0 |
| Yes | 5 | 143 | 2 |
| Unknown | 0 | 0 | 0 |
| Kappa Coefficient | 0.4837 (moderate agreement) | ||
The 38 cases in which the initial reviewers did not agree underwent adjudication by the second team of central reviewers. Consensus was reached in 25/38 cases while consensus was not reached in 13 cases (6.3% of all total cases). The 13 unresolved cases were not included in further analyses; however, consensus was achieved in 12/13 cases after unblinded review by all 4 central reviewers. For this final case, which involved a tumor with possible extension to the base of the skull, the imaging review committee chair made the final determination (Figure 1B). Surgeons changed their determination to match the radiologists in 8/13 cases. The IDRFs in question for these cases were related to the renal pedicle (n=5), neck region (n=4), thoracic or thoraco-abdominal cavity (n=3), and a dumbbell tumor (n=1). For the 5 cases involving the renal pedicle, the tumor was abutting but not encasing the hilum. An IDRF was identified as present by both radiologists and deemed absent by both surgeons in all 5 cases. After joint review, the central panel agreed that an IDRF was present in all 5 cases.
Concordance between local and central review was assessed in three ways. First, if either member of the central reviewer pair (surgeon or radiologist) detected the presence of any IDRF, then IDRF was considered present per central review. Among these 208 cases, the local institution and central reviewers agreed in 108 (51.9%) cases (Table 3). Among the 100 cases in which there was discordance between local and central review, at least one central reviewer identified an IDRF as present while the local institution indicated no IDRF present (n=49) or ‘Unknown’ (n=37). Central review did not identify an IDRF in 14 cases for which local review indicated that an IDRF was present (n=12) or ‘Unknown’ (n=2). Concordance between the central reviewers and institutions was 0.0640 using the Kappa coefficient, which is considered poor agreement (N=208).
Table 3.
Frequencies of IDRF determination by the institution and central review, using three methods. Kappa Coefficient presented as a measure of agreement between central review and institution.
| Concordance of Institutional Reading of IDRF with Central Review | |||
|---|---|---|---|
| Agreement between Central Review and local institution | # Cases (%) | Kappa coefficient | Agreement |
| Is IDRF present, yes or no? (n=208, all cases) |
108/208 (51.9%) | 0.0640 | Poor |
| Excluding initial Central Review mismatches | 96/170 (56.5%) | 0.1026 | Poor |
| Determination of No IDRF present (n=195, adjudicated cases) |
16/36 (44.4%) | ||
| Determination of + IDRF present | 90/159 (56.6%) | 0.1160 | Poor |
| ‘False positive’ reading of IDRF by local institution | 13/36 (36.1%) | ||
| ‘False negative’ reading of IDRF by local institution | 40/159 (25.2%) | ||
In the second approach, cases where the initial central reviewers disagreed (n=38) were excluded. Local and central reviewers agreed in 96/170 (56.5%) cases (Table 3). The central reviewers identified an IDRF as being present in 60 cases that were not designated as having an IDRF based on local institution review (no IDRF present in 35 cases and ‘Unknown’ in 25). In 14 cases, central reviewers did not identify the presence of an IDRF while local institutions designated an IDRF as present (n=12) or ‘Unknown’ (n=2). The Kappa coefficient for identifying an IDRF between the central reviewers and institutions was 0.1026, which is considered poor agreement.
Using the third approach, the 195 cases in which consensus was reached with either initial or secondary central review were evaluated. Among these cases, there were 76 (39.0%) cases identified with no IDRF present per either central or institutional review (Table 3). In 16 of these 76 cases (21.0%), the institution and central reviewers agreed that there was no IDRF present. The presence of an IDRF was designated based on either institutional or central review in 172 of 195 cases (88.2%). Local and central reviewers agreed that an IDRF was present in 90/172 cases (52.3%). The institution detected an IDRF in 13 cases where central reviewers did not, while central reviewers identified an IDRF in 69 cases where the institution indicated that no IDRF was present or where the institution called the IDRF status ‘Unknown’. Concordance between local and central review was 0.1160, which is considered poor agreement.
Among the entire cohort of 208 patients, 95 patients had localized tumors with at least one IDRF present. Central reviewers often selected multiple IDRFs for a single patient and all IDRFs were included in this analysis. IDRFs were most often located in the abdomen/pelvis (n=55) and least often were designated as thoraco-abdominal (n=5) (Table 4). The number of IDRFs identified in each general anatomic location on initial central review were assessed and the number identified by surgeons compared to radiologists were compared (Table 5). Radiologists were more likely to identify IDRFs in the neck compared to surgeons (p=0.03). There was no statistically significant difference in the number of IDRFs identified in the remaining general anatomic locations.
Table 4.
Frequency and percentage of the presence of an IDRF in each general anatomic location, by any central reviewer (at least two and up to four), for the INRG L2 cohort (N=95). A single patient can have multiple IDRF sites in multiple general anatomic locations.
| General Anatomic Location | Frequency | Percentage |
|---|---|---|
| Neck tumor | 11 | 11.6 |
| Cervico-thoracic tumor | 22 | 23.2 |
| Thoracic tumor | 32 | 33.7 |
| Thoraco-abdominal tumor | 5 | 5.3 |
| Abdomen/pelvis tumor | 55 | 57.9 |
| Dumbbell tumor | 32 | 33.7 |
| Involvement/infiltration of adjacent organs/structures | 36 | 37.9 |
Table 5.
Median (minimum, maximum) number of IDRF sites identified in each general anatomic location by the initial central review pair, for the INRG L2 cohort (N=95). A single patient can have multiple IDRF sites in multiple general anatomic locations. Statistical significance assessed by differences in the paired scores by surgeons versus radiologists for each patient.
| General Anatomic Location | Surgeon | Radiologist | P-value |
|---|---|---|---|
| Neck tumor | 0 (0, 2) | 0 (0, 2) | 0.031 |
| Cervico-thoracic tumor | 0 (0, 3) | 0 (0, 3) | 0.27 |
| Thoracic tumor | 0 (0, 4) | 0 (0, 4) | 0.075 |
| Thoraco-abdominal tumor | 0 (0, 1) | 0 (0, 1) | 1.00 |
| Abdomen/pelvis tumor | 0 (0, 5) | 1 (0, 6) | 0.07 |
| Dumbbell tumor | 0 (0, 1) | 0 (0, 1) | 0.22 |
| Involvement/infiltration of adjacent organs/structures | 0 (0, 1) | 0 (0, 1) | 0.81 |
4.0. Discussion
ANBL0531 was the first prospective cooperative group study to include both institutional and central IDRF assessments and to evaluate concordance between these assessments. In addition, concordance between IDRF assessment by surgeons and radiologists was evaluated using imaging from a large cohort of intermediate-risk neuroblastoma patients. Concordance between surgeons and radiologists during central review was only moderate, with agreement among experienced pediatric surgeons and radiologists in 81.7% of cases. The current study also demonstrated that concordance between local and central review was poor, although this may be related to the fact that the IDRF system was new at COG sites during the study period and local IDRF assessment may not have always been reported by expert surgeons and radiologists. Systematic improvement is required to improve accuracy of IDRF assessment.
During central review, study pairs consisting of a pediatric surgeon and pediatric radiologist agreed 81.7% of the time as to whether at least one IDRF was present or not. In cases with disagreement on initial review, secondary review achieved consensus in most cases. However, a small subset of cases (6.3%) could not be adjudicated even with secondary review. The most debated IDRFs involved the renal pedicle and neck, where radiologists were more likely than surgeons to identify an IDRF as present. Clinical judgement of the surgeon relating to surgical risk and resectability may impact their determination of whether an IDRF is present. If surgeons are less concerned about resection of a tumor that partially encases vessels, they may be less likely to consider this an IDRF, or conversely be more concerned about other IDRFs such as involvement of the celiac or superior mesenteric artery at its origin. The 13 cases with discordant central review highlight the notion that clinical judgement of the surgeon regarding resectability may influence their determination of whether an IDRF is present. All five cases involving the renal hilum were tumors that abutted the hilum and did not encase renal pedicle structures, leading surgeons to initially deem IDRFs as absent due to an estimation that these tumors could be treated with up front resection. While the IDRF guidelines were changed based on evidence that any renal involvement increased risk of nephrectomy [8], in these instances, surgeon perception of risk with renal hilum involvement was less and impacted their IDRF assessment. Enhanced understanding of the relationship between the presence of specific IDRFs and incidence of surgical complications may improve utilization of IDRFs for preoperative treatment decisions relating to up front resection versus neoadjuvant chemotherapy, particularly in children with intermediate-risk disease who may potentially avoid chemotherapy if offered surgery.
While the presence of IDRFs has been associated with increased risk of surgical complications as well as inferior long-term outcomes [9–12], it is unclear whether this is due to a direct relationship between IDRFs and more aggressive tumor biology, or whether specific IDRFs impart a worse prognosis compared to others. Avanzini, et al, reported that involvement of the superior mesenteric artery (SMA) at the root of the mesentery was the only IDRF associated with incomplete resection, while celiac artery and SMA IDRFs were associated with worse event-free and overall survival [9]. The total number of IDRFs has also been used to predict outcomes, with 6 or more IDRFs significantly increasing the risk of surgical complications as well as inferior survival [11]. For intermediate-risk patients, the surgical goal is a safe resection, and significant questions remain regarding the utility of IDRFs in predicting surgical outcomes and/or patient event-free or overall survival. An in-depth analysis of the association between specific IDRFs and surgical complications may allow for a better understanding of the prognostic value of IDRFs and their potential role in surgical decision-making. Comprehensive and accurate documentation of IDRFs in both operative reports and radiology reports may facilitate this research.
From the staging standpoint, it is important to have a consistent definition of IDRFs that is used throughout the world. Concordance at the central review level was moderate in this study, yet experienced pediatric surgeons and radiologists still disagreed in almost 20% of cases. However, incidence of disagreement amongst trained radiologists for any radiographic interpretation is reported as high as 22% [13], and disagreement is highest for radiographic studies for pediatric patients who undergo neuro or body imaging [14], thereby highlighting the challenge of accurate radiographic interpretation even amongst the most experienced clinicians. It is important to note that this study evaluated IDRF assessment shortly after the INRG approach was adopted, and ANBL0531 was the first COG study to prospectively collect IDRF data. Radiologists were more likely to identify IDRFs as present than surgeons and may have followed IDRF guidelines more strictly than surgeons, who may have used IDRF in a more general sense when considering surgical risk. Further analysis of concordance in IDRF assessment and utility over time will be important to monitor as both radiologists and surgeons have become more accustomed to formalized IDRF assessment. Additionally, techniques to improve concordance should be explored, such as novel 3-D imaging modalities which may improve the accuracy of imaging used for tumor staging and IDRF assessment [15,16].
Concordance between local and central IDRF assessment was poor in all comparisons. Across analyses, local institutions were less likely to identify the presence of an IDRF compared to central review. It is notable that the INRG IDRF guidelines were updated after ANBL0531 was closed to enrollment to reflect that any contact with the renal vasculature was considered an IDRF rather than renal vasculature encasement alone being considered an IDRF. This guideline change was not in place at the time of local institutional review, but was incorporated into central review, and may explain some of the discrepancies in IDRF classification. Another key factor was that IDRF designation may not have been performed by a consistently designated person such as a trained radiologist or surgeon at the local level, and local assessment may have varied based on the specialty and experience of individual performing imaging assessment for data collection purposes. Standardized radiology templates for cross-sectional imaging reporting may facilitate more accurate IDRF identification and reporting at the local level for future studies. Surgeon and radiologist involvement in the development and implementation of such templates to ensure data validity and reliability is essential to ensure proper reporting of IDRFs for assignment of stage in the INRG classification system.
Several important limitations should be considered when interpreting the results of this study. First, local sites were only asked to identify whether any IDRF was present; therefore, a comprehensive comparison of IDRF assessment was not possible at the local level. Additionally, local assessment was not performed by a designated individual. This may limit interpretation of the poor concordance identified but does reflect the reality of clinical practice in which the experience level of those involved in imaging assessment varies. Additionally, institutions that selected ‘Unknown’ were analyzed as discordant, but this may reflect the experience of the individual entering data rather than the true IDRF assessment by the treatment team. Furthermore, the INRG definition change regarding renal pedicle IDRF may have impacted concordance between local and central review. Finally, the current study captures the time period shortly after the inception of IDRF assessment in North America; it is possible that both local and central IDRF assessment has improved over time as IDRFs have since become an integral component of neuroblastoma evaluation over the last decade. Therefore, a more contemporaneous assessment of IDRF concordance will be incorporated into current COG studies.
This current study of children with intermediate-risk neuroblastoma is the first prospective, cooperative group study to evaluate concordance of the INRG IDRF system at the institutional and central level. Central concordance was moderate, suggesting a need for improvement to ensure accurate assessment of pre-operative risk stratification and INRG staging. Thorough and accurate documentation of IDRFs in both surgeon operative reports and radiology reports is critical to facilitate this research. Furthermore, optimized collection of IDRF data is required for accurate INRG staging which is now an integral part of COG and most international risk classification systems [4, 17]. Future studies focused on the association of IDRFs and tumor biology with outcomes are needed to better understand the prognostic value of the current IDRF system and potentially revise the current system. Additionally, monitoring of IDRF concordance over time and novel methods to improve concordance, such as 3-D modeling,
Acknowledgments
The authors would like to recognize the contributions of the late Dr. SL Wootton-Gorges as a dedicated member of the central review panel for ANBL0531.
Funding
Supported in part by the National Institutes of Health- National Cancer Institute Grant U10 CA180899 (Children’s Oncology Group Statistics and Data Center), National Clinical Trials Network Operations Center Grant No. U10 CA180886, and the St. Baldrick’s Foundation, and NCI R35 CA220500. Author AN supported by NCI COG Statistics & Data Center grant U10 CA180899.
List of Abbreviations:
- INRG
International Neuroblastoma Risk Group
- IDRF
image-defined risk factor
- COG
Children’s Oncology Group
- INSS
International Neuroblastoma Staging System
- INPC
International Neuroblastoma Pathology Classification
- INRC
International Neuroblastoma Response Criteria
- CT
computed tomography
- MRI
magnetic resonance imaging
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Competing Interests: Author AN serves on a data safety monitoring board for Novartis.
Level of Evidence: Treatment Study, Level III
Disclosure
The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Contributor Information
Erin G Brown, Division of Pediatric Surgery, Department of Surgery, University of California Davis Children’s Hospital, Sacramento, CA.
E Stanton Adkins, Department of Pediatrics, University of South Carolina, Columbia, SC.
Peter Mattei, Division of Pediatic Surgery, Department of Surgery, Children’s Hospital of Philadelphia, Philadelphia, PA.
Fredric A Hoffer, Department of Radiology, University of Washington, Seattle, WA.
Sandra L Wootton-Gorges, Department of Radiology, University of California Davis Children’s Hospital, Sacramento, CA.
Wendy B London, Department of Pediatrics, Dana-Farber Cancer Institute and Boston Children’s Hospital, Boston, MA.
Arlene Naranjo, University of Florida Children’s Oncology Group Statistics and Data Center, Gainesville, FL.
Mary L Schmidt, Division of Pediatric Hematology and Oncology, Department of Pediatrics, University of Illinois Cancer Center, Chicago, IL.
Michael D Hogarty, Division of Pediatric Hematology and Oncology, Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, PA.
Meredith S Irwin, Division of Pediatric Hematology and Oncology, Department of Pediatrics, The Hospital for Sick Children, Toronto, ON.
Susan L Cohn, Division of Pediatric Hematology and Oncology, Department of Pediatrics, University of Chicago, Chicago, IL.
Julie R Park, Division of Pediatric Hematology and Oncology, Department of Pediatrics, St. Jude Children’s Research Center, Memphis, TN.
John M Maris, Division of Oncology, Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, PA.
Rochelle Bagatell, Division of Pediatric Hematology and Oncology, Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, PA.
Clare J Twist, Division of Pediatric Hematology and Oncology, Department of Pediatrics, Roswell Park Comprehensive Cancer Center, Buffalo, NY.
Jed G Nuchtern, Division of Pediatric Surgery, Department of Surgery, Texas Children’s Hospital, Houston, TX.
Andrew M Davidoff, Department of Surgery, St. Jude Children’s Research Hospital, Memphis, TN.
Erika A Newman, Division of Pediatric Surgery, Department of Surgery, CS Mott Children’s Hospital, Ann Arbor, MI.
Dave R Lal, Division of Pediatric Surgery, Department of Surgery, Children’s Wisconsin, Milwaukee, WI.
References
- [1].Davidoff AM. Neuroblastoma. Semin Pediatr Surg 2012;21(1):2–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Twist C, Schmidt M, Naranjo A, London W, et al. Maintaining Outstanding Outcomes Using Response- and Biology-Based Therapy for Intermediate-Risk Neuroblastoma: A Report From the Children’s Oncology Group Study ANBL0531. J Clin Oncol 2019;37(34):3243–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Baker DL, Schmidt ML, Cohn SL, et al. Outcome after reduced chemotherapy for intermediate-risk neuroblastoma. N Engl J Med 2010;363(14):1313–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Monclair T, Mosseri V, Cecchetto G, et al. Influence of image-defined risk factors on the outcome of patients with localised neuroblastoma. A report from the LNESG1 study of the European International Society of Paediatric Oncology Neuroblastoma Group. Pediatr Blood Cancer 2015;62(9):1536–42. [DOI] [PubMed] [Google Scholar]
- [5].Brodeur GP J; Berthold F; Carlsen NL, et al. Revisions of the International Criteria for Neuroblastoma Diagnosis, Staging, and Response to Treatment. J Clin Oncol 1993;11(8):1466–77. [DOI] [PubMed] [Google Scholar]
- [6].Brisse H, McCarville MG,C; Krug B, et al. Guidelines for Imaging and Staging of Neuroblastic Tumors: Consensus Report from the International Neuroblastoma Risk Group Project. Radiology 2011;261(1):243–57. [DOI] [PubMed] [Google Scholar]
- [7].Landis JR and Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33(1):159–74. [PubMed] [Google Scholar]
- [8].Shamberger RC, Smith EI, Joshi VV, et al. The Risk of Nephrectomy During Local Control in Abdominal Neuroblastoma. J Pediatr Surg 1998;33(2):161–4. [DOI] [PubMed] [Google Scholar]
- [9].Avanzini S, Pio L, Erminio G, et al. Image-defined risk factors in unresectable neuroblastoma: SIOPEN study on incidence, chemotherapy-induced variation, and impact on surgical outcomes. Pediatr Blood Cancer 2017;64(11). [DOI] [PubMed] [Google Scholar]
- [10].Phelps HM, Ndolo JM, Van Arendonk KJ, et al. Association between image-defined risk factors and neuroblastoma outcomes. J Pediatr Surg 2019;54(6):1184–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Saksiri P, Sripornsawan P, Kritsaneepaiboon S, et al. Correlation between image-defined risk factors and surgical complications in patients with neuroblastoma: a retrospective study. Pediatr Surg Int 2023;39(1):70. [DOI] [PubMed] [Google Scholar]
- [12].Pohl A, Erichsen M, Stehr M, et al. Image-defined Risk Factors Correlate with Surgical Radicality and Local Recurrence in Patients with Neuroblastoma. Klin Padiatr 2016;228(3):118–23. [DOI] [PubMed] [Google Scholar]
- [13].Geijer H, Geijer M. Added value of double reading in diagnostic radiology,a systematic review. Insights Imaging 2018;9(3):287–301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Eakins C, Ellis W, Pruthi S, et al. Second opinion inter- pretations by specialty radiologists at a pediatric hospital: rate of disagreement and clinical implications. AJR Am J Roentgenol 2012;199:916–20. [DOI] [PubMed] [Google Scholar]
- [15].Valls-Esteve A, Adell-Gomez N, Pasten A, et al. Exploring the Potential of Three-Dimensional Imaging, Printing, and Modeling in Pediatric Surgical Oncology: A New Era of Precision Surgery. Children (Basel) 2023;10(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Wellens LM, Meulstee J, van de Ven CP, et al. Comparison of 3-Dimensional and Augmented Reality Kidney Models With Conventional Imaging Data in the Preoperative Assessment of Children With Wilms Tumors. JAMA Netw Open 2019;2(4):e192633. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Irwin M, Naranjo A, Zhang FF, et al. Revised Neuroblastoma Risk Classification System: A Report From the Children’s Oncology Group. J Clin Oncol 2021; 39(29): 3229–3241. [DOI] [PMC free article] [PubMed] [Google Scholar]