Purpose: The aim of this study was to conduct a systematic review, and where possible meta-analyses, of molecular and biological tumor markers described in neuroblastoma, and to establish an evidence-based perspective on their clinical value for the screening, diagnosis, prognosis, and monitoring of patients.

Experimental Design: A well-defined, reproducible search strategy was used to identify the relevant literature from 1966 to February 2000.

Results: A total of 428 papers studying the use of 195 different tumor markers in neuroblastoma were identified. Small sample sizes, poor statistical reporting, large heterogeneity across studies (e.g., in cutoff levels), and publication bias limited meta-analysis to the area of prognosis only; MYCN, chromosome 1p, DNA index, vanillylmandelic acid:homovanillic acid ratio, CD44, Trk-A, neuron-specific enolase, lactate dehydrogenase, ferritin, and multidrug resistance were all identified as potentially important prognostic tools.

Conclusions: This systematic review forms a knowledge base of the tumor markers studied thus far in neuroblastoma, and has identified some of the most important prognostic markers, which should be considered in future research and treatment strategies. Importantly, the review has also highlighted some general problems across primary tumor marker studies, in particular poor and heterogeneous reporting. These need to be addressed to allow better clinical interpretation and enable more appropriate evidence-based reviews in the future. In particular, collaboration of cancer research groups is needed to enable bigger sample sizes, standardize methods of analysis and reporting, and facilitate the pooling of individual patient data.

Neuroblastoma is a neuroblastic tumor of the primordial neural crest and is the most common extracranial solid tumor of childhood, comprising between 8 and 10% of all childhood cancers. It is an enigmatic tumor demonstrating diverse clinical and biological characteristics and behavior (1). Tumors may regress spontaneously, reflecting induction of apoptosis or differentiation, or they may exhibit extremely malignant behavior with very low cure rates. The spectrum of clinical behavior suggests that genetic, biological, and morphological features may be useful markers to stratify children with this disease for the most appropriate management. Knowledge of prognostic markers may also help understand the genesis of this disease.

Neuroblastoma is predominantly a disease of the first decade with ∼80% of children presenting at <4 years old; median age is 22 months. The incidence in the United Kingdom and the United States is ∼1 in 7000 live births, and there is slight sex predominance in most series with a male-to-female ratio of 1.2:1. The disease accounts for 15% of all childhood cancer deaths, indicating the poor prognosis of many of the tumors (2, 3, 4). Children with stage 1, 2, or 4 s disease, or presenting in the first year of life have a good prognosis. In contrast, children (≥1 year of age) with stage 3 and 4 disease have 3-year survival rates of 50% and 15%, respectively. Most children present over the age of 1 year with metastatic (stage 4) disease; this group has an overall survival of 10–20% (3, 4).

A number of genetic and biological features have been investigated in recent years in an effort to improve the understanding of the behavior of neuroblastoma and to identify tumor markers that would improve cure rates by facilitating the screening, diagnosis, prognosis, or monitoring of patients. In particular, many prognostic studies have identified many tumor markers associated with overall or disease-free survival, including MYCN copy number, ploidy, and deletion or loss of heterozygosity of chromosome 1p and gain of chromosome 17q. However, it has proved difficult to identify which prognostic markers are the most useful, reflecting the complex nature of the tumor and the lack of large prospective clinical outcome studies.

The aim of this study was to conduct a systematic review, and where possible meta-analyses, of molecular and biological tumor markers described in neuroblastoma, and to establish an evidence-based perspective on their clinical value for the screening, diagnosis, prognosis, and monitoring of patients. This should facilitate identification of the most useful tumor markers for clinical management and the development of future research strategies by: (a) establishing the importance of the markers studied; and (b) identifying the markers that warrant additional investigation.

The systematic review followed the guidelines contained in NHS Centre for Reviews and Dissemination (1996), and its underlying philosophy was to maintain breadth, synthesize the evidence qualitatively, and then, only where appropriate, use quantitative methods, making procedures explicit and transparent throughout (5).

Search Strategy.

The three on-line bibliographic databases Medline, Embase, and Cancerlit were chosen as a basis for identifying the relevant literature from 1966 to February 2000. An iterative procedure was used to develop an optimal search strategy, which culminated in the use of three important sets of keywords in the strategy (Table 1). The keywords in {Neuroblastoma} related to the family of this disease, whereas those in {Tumor Marker} included the named markers thought a priori to be potentially important. The set {Clinical Area} included more specific terms for the clinical use of markers in children. A paper was included if a word from {Neuroblastoma}, a word from {Tumor Marker}, and a word from {Clinical Area} were included anywhere in the paper.

Three investigators independently performed the assessment of the papers. All three had previous experience of identifying relevant tumor marker literature for a systematic review and subsequent meta-analysis (6, 7). Furthermore, the second investigator is a pediatric oncology consultant, with a special interest in neuroblastoma, and the third investigator is a translational scientific research fellow, with a special interest in small round cell cancers of childhood. Both of these investigators held regular meetings with the first investigator about the review process and the assessment of the literature.

The first investigator read the available abstract to classify each paper into one of three categories: “relevant,” “uncertain,” or “not relevant.” The second and third investigators, who had more background clinical knowledge in the research area, checked all of the abstracts where classification was uncertain, and ∼10% of the papers in each of the relevant and not relevant categories. Copies of all of the papers classified as relevant, together with all of the papers in which relevance remained unclear after assessment of abstracts by the three investigators, were obtained and then read thoroughly to make a final decision as to their inclusion.

Inclusion.

To be included in the systematic review a paper had to provide a quantitative result or give tabulated individual patient data (IPD) evaluating the use of a tumor marker in neuroblastoma. The paper had to be based on a primary research study of humans relevant to the clinical area of screening, diagnosis, prognosis, or monitoring. There was no restriction on age of patients in the study, although ∼90% of papers included just 0–18-year-olds. The criteria for classifying the four clinical areas was that the paper had to present data in the form of summary statistics or IPD for: (a) screening: the use of tumor markers to screen an apparent healthy population; (b) diagnosis: tumor marker levels considered of diagnostic value; (c) prognosis: tumor marker levels at a measured point in time with relation to the outcome of patients at the end of a specific follow-up period; and (d) monitoring: tumor marker levels taken repeatedly during a follow-up period with relation to disease status over that period.

Exclusion.

Papers that reported only laboratory work, methodology for identifying new markers, or results from animal studies were excluded. Review articles and foreign language papers were also excluded. Histological characteristics of tumors [such as the presence of differentiated ganglia in neuroblastoma (Shimada index)] were not included in the markers reviewed.

Information Extracted.

From the included papers, information was extracted on the tumor marker used and to which clinical area it related: screening, diagnosis, prognosis, or monitoring. Among the covariate information extracted from each paper on prognosis was whether survival was overall (OS) or disease-free (DFS), the marker cutoff level if applicable and, if so, the total number of patients and deaths within each high and low subgroup. The age range and stages of neuroblastoma disease represented by the patients in each study were also recorded, as these were known a priori to be important prognostic clinical features (8).

Meta-Analysis and Assessment of Publication Bias.

Meta-analysis was performed, where possible, to combine all of the relevant results found from the literature search (9). For each of the areas of screening, diagnosis, and monitoring, only those tumor markers on which 3 or more papers provided data were considered. For the area of prognosis, due to the many prognostic markers and prognostic studies identified, meta-analysis was limited to those reported in ≥10 papers. Both fixed and random effects meta-analyses were used, with the latter preferred if there was evidence of heterogeneity. Meta-analysis for clinically relevant subgroups of patients (e.g., age <1 and stage 4 disease) was also considered, but was only performed where sufficient data were available.

For the meta-analysis of data from the prognosis papers, the extraction of the loge (hazard ratio; HR) and its variance was the desired target. These statistics were chosen because they provide an important comparative estimate of the risk of death/disease recurrence between two groups of patients.

It was common for a paper to report more than one prognostic result by relating one or more markers to OS and/or DFS, and also by providing unadjusted and/or adjusted results (e.g., adjusted for age and stage of disease). Estimates of the loge (HR) and its variance comparing two groups defined by a single marker level were sought from all of the OS and DFS reports using the methods described by Parmar et al.(10). An unadjusted estimate was preferred for each report, as adjusted results are likely to be highly inconsistent in the factors for which adjustment are made (11). An adjusted estimate was sought in the absence of an unadjusted result.

An assessment of the publication bias in the prognosis literature was made by application of two appropriate statistical tests to the MYCN results (12, 13). The Trim and Fill method was also used to assess the likely impact of publication bias on the pooled MYCN results (14).

Economic and Psychosocial Effects of Tumor Markers.

A set of keywords was used to screen the abstracts of the papers classified as relevant for any economic or psychosocial results relating to the use of a tumor marker in neuroblastoma for any of the clinical areas (Table 1).

Literature Search Results.

We identified 3415 papers from the searches; 1536 were first identified in Medline, an additional 473 in Embase, and then an additional 1406 from Cancerlit. These were classified by the three investigators (Fig. 1). The second and third investigator agreed that 85.7% of a sample of the first investigator’s relevant papers were indeed relevant or uncertain (42 of 49). They also agreed that 193 (86.9%) of a sample of 222 not-relevant papers checked were indeed not relevant, and of the 29 others classified 8 papers as relevant and 21 papers as uncertain. After obtaining and reading the entire articles, 15 of these 21 uncertain papers were ultimately classified as not relevant. Thus, 208 of the first investigator’s 222 not-relevant papers sampled by the other investigators were correctly classified (93.7%). Overall, 428 papers were considered relevant and included in our review (Fig. 1; complete list of these references available on the internet).6

Tumor Markers Identified Overall and Within Each Clinical Area.

A total of 195 different tumor markers were studied in these 428 papers in relation to the screening, diagnosis, prognosis, or monitoring of neuroblastoma (Table 2). There were 49 different papers on screening, 288 on diagnosis, 260 on prognosis, and 51 on monitoring; 201 of the 428 papers covered two or more clinical areas.

Screening.

The review identified 49 papers that gave quantitative data relating to the use of tumor markers in screening and potentially considered the evaluation of a population-based screening program for neuroblastoma. These papers covered programs established in geographical regions of Austria, Canada, France, Germany, Japan, and the United Kingdom The studies considered a variety of outcomes including: (a) feasibility/uptake rate; (b) the number of false-positive and false-negative cases; (c) incidence; (d) stage distribution; and (e) mortality. In terms of outcomes (c), (d), and (e), some studies have undertaken, or are designed to enable in the future, a comparison between screened and control (nonscreened) populations. Unfortunately, a quantitative synthesis of results was not feasible given the heterogeneity in how and which outcomes were reported. However, a qualitative assessment suggested that considerable uncertainty still surrounds whether population-based screening for neuroblastoma is cost-effective overall, and, if so, the optimal age at which to screen, and also the optimal screening strategy, i.e., one-stage or multistage. Recent studies have shown that early screening (before 6 months of age) is not informative (see “Discussion”).

Diagnosis.

It was not possible to perform a meta-analysis of the data from the diagnosis papers, because the results mostly only compared the number of neuroblastoma patients with high/positive marker levels to those with low/negative levels respectively. Marker levels from patients with neuroblastoma were rarely compared with those from a sample of healthy controls in the diagnosis papers, e.g., none of the 22 papers reporting levels of serum lactate dehydrogenase at diagnosis compared patients with neuroblastoma to healthy controls. Current clinical practice suggests that urinary catecholamines are important for the differential diagnosis of neuroblastoma from other small round cell tumors; however, the poor quality of the published literature prevented a quantitative evaluation of this practice.

Prognosis.

The 12 most commonly studied prognostic markers were each selected for an in-depth study to establish their individual value as a prognostic tool; each marker was studied in ≥10 prognosis papers (Table 3). The prognostic value of CD44 expression was also evaluated, because all of its 8 prognostic studies were contained within those papers of the other 12 markers to be evaluated. Hence, 13 markers overall were evaluated as a prognostic tool. These 13 markers were studied in 211 (81.2%) of all of the prognosis papers. Within these there were 575 reports of prognostic power assessment, where levels of 1 of these 13 tumor markers were related to OS or DFS by summary statistics or IPD.

Weakness of reporting, analysis, and presentation of results meant that only 204 (35.5%) estimates of both the loge (HR) and its variance could be extracted. Meta-analyses were additionally restricted by large variability in both clinical and statistical factors relating to the 204 estimates. For example, for the marker MYCN, 94 estimates of the loge (HR) and variance were obtained but these involved 9 different cutoff points to dichotomize the marker, 9 different stage groups, 4 different age groups, 17 adjusted/77 unadjusted estimates, and 2 different outcomes (OS and DFS). Type of treatment and method of marker measurement were not recorded but both would have added additional heterogeneity to that already noted. A more in-depth evaluation of these reporting problems, together with recommendations for improvement, are provided elsewhere (15).

Whereas acknowledging the problems implied by this large degree of heterogeneity, it remained important to use the data extracted for each marker and determine which were potentially the most important markers for future research. Therefore, meta-analysis was performed for each of the 13 markers separately for OS and DFS. The meta-analysis results for each marker are presented in Table 3 and have been classified into three marker groups: DNA/chromosome abnormalities, biological markers, and urinary catecholamines. For the vast majority of markers, there is a statistically significant difference (in terms of the HR) between the groups defined by the markers. For example, there was strong statistically significant evidence that amplification of the MYCN gene was associated with a worse OS and DFS. The risk of death was 5.48 times greater for patients with MYCN amplification compared with those that did not have amplification [HR = 5.48; 95% confidence interval (CI), 4.30–6.97], and similarly for risk of disease recurrence (HR = 4.28; 95% CI, 3.34–5.49). All of the papers that could be included in the meta-analyses were published from 1985 onwards, with the majority after 1989. For example, of the 70 articles used for the MYCN OS and DFS meta-analyses none were published before 1985, only 8 were published from 1985 to 1989, and 62 were published from 1990 onwards. Hence, the prognostic studies we included have been reported after the improved method for staging and treatment of neuroblastoma that has improved survival for children with this disease over the last 20 years.

Monitoring.

The review identified 51 papers that provided quantitative data evaluating the serial use of tumor markers to aid the clinical management of patients with neuroblastoma. However, there was considerable heterogeneity between the studies in terms of: (a) tumor markers considered (e.g., VMA, HVA, MYCN, ferritin, NSE, and lactate dehydrogenase); (b) outcome (OS and DFS); (c) statistical analyses undertaken and reporting of results; (d) length of follow-up, e.g., treatment phase, long-term follow-up; (e) number of patients; (f) age/stage distribution; and (g) number of serial measurements.

The combination of these problems with the relatively few papers identified meant that any meta-analysis using these studies was not worthwhile, statistically or clinically.

Publication Bias.

The estimates of the loge (HR) and its variance obtained for MYCN were used to assess the possibility of publication bias. Both Begg and Egger tests produced Ps < 0.001, and therefore publication bias was strongly suspected to be a problem (12, 13). In fact, using the Trim and Fill method, 17 studies with smaller HRs were estimated as missing from our results, in addition to the 45 studies included in the analysis. Hence, it appears likely that the effect size from the original meta-analyses may be biased upwards, i.e., they may overestimate the true underlying loge (HR) for MYCN. It is highly likely that this problem is also true for the other markers.

Economic and Psychosocial Results.

No papers made an economic evaluation of the use of tumor markers in neuroblastoma, but 2 papers reported cost data in relation to screening (16, 17). They are both somewhat dated and contain few details about cost calculations, which made it difficult to assess the accuracy of the claims made or the relevance of the findings to current practice. Furthermore, none of the 428 relevant papers reported any data on the psychosocial consequences for children and their families of using tumor markers clinically in neuroblastoma, even in the monitoring papers.

This is the first systematic review of tumor markers for neuroblastoma that has been reported, and it forms a knowledge base for future research. A systematic review is the preferred means of identifying and combining existing evidence (18), and it is particularly important for evidence-based evaluations of tumor markers in neuroblastoma, and indeed other rare diseases, because (sometimes conflicting) evidence relating to markers is published across a number of studies, many of which involve small numbers of patients. Systematic reviews are also important because they can highlight underlying problems across individual studies and help identify future research needs (19). Indeed, both of these aspects are demonstrated throughout our review and form the most important messages of this paper, which we hope will help improve tumor marker studies in the future.

Appraisal of the Systematic Review.

During the systematic review we classified 3415 papers overall and ultimately identified 428 relevant papers, which showed diversity in primary interest, methodology, analysis of data, and quality of reporting. The search strategy used is likely to have identified the majority of the available literature, targeting in particular the databases specializing in scientific and clinical reporting, although we acknowledge the possibility that the review may not be fully comprehensive, reflecting publication and reporting bias. Our initial search strategy included the names of those markers known a priori to be potentially important but, in light of the many markers identified during the review, this list was certainly not exhaustive, although we did include more general terms such as “marker” that will have limited the number of markers missed.

Initially our strategy had been to check the references quoted in all of the selected papers, to identify any other papers missing from the three databases we searched. This was not feasible given the large literature base, nor was it possible to systematically check for duplicates of patients across papers. We did not include foreign language papers because of the difficulties in translation, and this may have introduced bias if statistically or clinically significant studies were more likely to be (re)written for publication in an English language journal (20).

Clinical Interpretation.

Neuroblastoma is a multifaceted disease. The expansion in biological and cytogenetic markers is an indication that the cancer process is complex with multiple changes taking place with a neuroblastoma cell. Research studies have tried to identify which of these markers are initiating factors contributing to the cancer phenotype and which may be regarded as secondary. Initial studies for a particular marker have inevitably sought to link that one marker with survival, to establish its importance. However, it is now becoming clear that a number of key biological markers are likely to be interlinked and must be evaluated in combination with one another and not in isolation.

The poor and heterogeneous reporting restricted any quantitative synthesis in the areas of screening, diagnosis, and monitoring, and therefore any overall clinical evaluation. Recent papers, published since the start of our review, suggest that there is currently insufficient evidence to support a screening program for infants up to 6 months of age, and the majority of authors conclude that it should be discontinued (21). Clinical and histopathological features, which have not been evaluated in our review, are the most informative for the diagnosis of neuroblastoma, and these include age at diagnosis, tumor histology, and primary tumor site. However, the detection of catecholamine metabolites in urine is also used for the differential diagnosis of neuroblastoma from other small round cell tumors of childhood. For the use of molecular and biological markers in diagnosis, the few studies comparing a healthy control group to patients with neuroblastoma was particularly disappointing, and future studies need to address this. Similarly, monitoring studies need to report the differences in serial marker measurements between those who develop a recurrence of disease and those who remain disease-free, preferably for a many patients over a long follow-up period. Where possible, research groups need to collaborate and pool together resources to enable bigger sample sizes and achieve consistency across studies, which should be targeted to address the important issues. Only then will the benefits of using tumor markers for screening, diagnosis, and monitoring be properly ascertainable.

This systematic review did produce an evaluation of the most commonly reported individual markers for prognosis; MYCN, chromosome 1p, DNA index, VMA:HVA ratio, CD44, Trk-A, NSE, lactate dehydrogenase, ferritin, and multidrug resistance were all identified as potentially important prognostic tools. However, the pooled results must be treated with caution given the reporting problems and large heterogeneity of clinical/statistical factors across studies. Recent studies have also indicated that chromosome 17q gains also have important prognostic significance (22, 23, 24, 25, 26, 27, 28). Unfortunately, many of these studies were published after the start of our review, and consequently chromosome 17q was not among the prognostic markers we selected for the in-depth evaluation above. However, in light of this current knowledge, we have subsequently extracted, wherever possible, HR results from each of the 8 prognosis papers our review identified for this marker. Meta-analysis of these suggests that patients with gain of chromosome 17q have a significantly worse DFS (from 3 studies: HR = 4.16; 95% CI, 2.56–6.77) and OS (from 3 studies: HR = 4.30; 95% CI, 2.70–6.86) compared with those who did not. However, these results are again subject to the problems of poor reporting and heterogeneity.

It was not possible to compare subgroups of patients, individual prognostic markers, or assess the benefits of using any of the markers in combination. Availability of full IPD, including all of the exact values of all markers assessed and subgroup information (e.g., age and stage of disease), from each paper would facilitate such assessments in the future (15). Ideally, large multicenter studies should also be initiated to assess the benefits of using chromosome 17q and the other important prognostic markers, both individually and in combination, to improve strategies for the stratification of children with neuroblastoma for therapy. Results from such studies would enable clinicians to clearly see which are the most appropriate individual and combinations of prognostic markers to use.

Reporting Problems.

Importantly, our review has highlighted problems in how individual prognostic marker studies are designed and reported. Many study reports do not allow for adequate extraction of data in order for comparisons to be made, and the large heterogeneity in cutoff levels and other measures limit evaluations of markers across studies and within specific subgroups of patients (e.g., age <1). There is also the threat of publication bias across the literature. For those researchers studying prognostic tumor markers, Altman and Lyman (29) have proposed important guidelines for both conducting and evaluating prognostic factor studies that should be considered. We have also provided guidelines for improved statistical reporting of primary studies to facilitate the evaluation and comparison of individual markers, identify the additional benefits of using combinations of markers, and ultimately allow important evidence-based reviews to be made (15). In particular, presentation of the HR with some measure of precision (e.g., 95% CI) and availability of IPD, either in the paper or on the internet (30), are both highly desirable. The availability of IPD has allowed important evidence-based reviews to be made in other cancer settings (31, 32).

Psychosocial and Economic Issues.

The clinical implications of the results must also be considered together with psychosocial and economic aspects of tumor markers, for example in monitoring and screening, respectively. Our search found no published evidence on the psychosocial consequences for children and their families of using tumor markers clinically in neuroblastoma, but this probably reflects the few papers studying the use of tumor markers for monitoring patients. Psychosocial evaluations of tumor markers are clearly important and have been performed for other disease settings (33). Future research on the use of tumor markers in pediatric oncology should seek to include an assessment of the psychosocial outcomes of using markers, particularly for prognostic and monitoring purposes.

We also only identified two studies that included an economic evaluation of using tumor markers, and so the cost-effectiveness of individual markers could not be evaluated. A clear prescription for future trials and studies is therefore to include an economic evaluation element wherever possible, either in terms of a cost-effectiveness study or the identification of resource use that would permit decision models to be developed.

Summary.

This systematic review has emphasized the uncertainty on the clinical value of the studied tumor markers in neuroblastoma, reflecting the small size of many studies and poor statistical reporting. We have managed to assess the significance of the most commonly reported prognostic markers; MYCN, chromosome 1p, DNA index, VMA:HVA ratio, CD44, Trk-A, NSE, lactate dehydrogenase, ferritin, and multidrug resistance were all identified as potentially important prognostic tools. Each of these, and the more recently identified changes in chromosome 17q, should be considered in the development of future research strategies and for the stratification of children with neuroblastoma for different treatment strategies.

The multiplicity and complexity of tumor markers in neuroblastoma underlines the need for studies to be coordinated, through the cooperation of cancer research groups, using multiple laboratories and standardizing methods of analysis and reporting. In particular, collaboration is needed to consider prospectively planned pooled analyses and facilitate the pooling of IPD, a strategy that would adequately ensure a quantitative synthesis (rather than merely a qualitative synthesis) to address the questions of interest, such as which combinations of markers provide the best prognostic tool, or whether monitoring patients with neuroblastoma using markers is cost-effective. Large, multicenter collaborative studies would require agreement as to which markers to measure, and we have provided a base that highlights the ones reported in greatest detail thus far.

Grant support: National Health Service Health and Technology Assessment (HTA) Program (Grant 97/15/03).

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.

Requests for reprints: Richard David Riley, Department of Health Sciences, University of Leicester, 22-28 Princess Road West, Leicester, LE1 6TP, United Kingdom. Phone: 44-116-2525427; Fax: 44-116-2523272; E-mail: rdr3@leicester.ac.uk

6

Internet address: http://www.prw.le.ac.uk/epidemio/personal/rdr3/paed2.html.

Fig. 1.

Flow chart showing the results at each stage of the process used to identify the final set of relevant papers in the neuroblastoma review.

Fig. 1.

Flow chart showing the results at each stage of the process used to identify the final set of relevant papers in the neuroblastoma review.

Close modal
Table 1

Set of keywords used in the neuroblastoma literature search of Medline, Embase, and Cancerlit

A.
NeuroblastomaClinical Area
Neuroblastoma Patient(s) Prognostic 
Ganglioneuroblastoma Child Diagnostic 
Ganglioneuroma Children Pediatric 
 Prognosis Paediatric 
 Diagnosis Screening 
 Monitoring Infant(s) 
 Follow-up  
A.
NeuroblastomaClinical Area
Neuroblastoma Patient(s) Prognostic 
Ganglioneuroblastoma Child Diagnostic 
Ganglioneuroma Children Pediatric 
 Prognosis Paediatric 
 Diagnosis Screening 
 Monitoring Infant(s) 
 Follow-up  
B. Tumor marker
Tumour marker(s) DOPA Homovanillic acid 
Tumor marker(s) Neuron-specific enolase HVA 
Marker(s) NSE Normetanephrine 
N-MYC Ferritin NM 
NMYC Lactate dehydrogenase Metanephrine 
MYCN LDH MN 
Tyrosine hydroxylase Ganglioside(s) 3-methoxy tyramine 
TH Monosialganglioside 3-MT 
Dopa-decarboxylase Disialoganglioside Vanillacetic acid 
DDC C-neu VPA 
Phenylethanolamine-N-methyl transferase C-myc Vanillglycol 
PNMT Neuropeptide(s) VG 
PGP9.5 Somatostatin receptors Vanillglycol acid 
Dopamine-β-hydroxylase Telomerase VGA 
DBH CD44 Catechol acetic acid 
Phenylalanine Mitotic index CAA 
Drug resistance RT-PCR VAA 
MRP Dopamine Norepinephrine 
Tyrosine NB84 Vanilalamine 
3,4-dihydroxyl phenyl alanine Noradrenaline VA 
1p deletion Adrenaline Trk A 
DNA diploidy Vanillylmandelic acid Trk B 
17q VMA Trk C 
14q Epinephrine  
B. Tumor marker
Tumour marker(s) DOPA Homovanillic acid 
Tumor marker(s) Neuron-specific enolase HVA 
Marker(s) NSE Normetanephrine 
N-MYC Ferritin NM 
NMYC Lactate dehydrogenase Metanephrine 
MYCN LDH MN 
Tyrosine hydroxylase Ganglioside(s) 3-methoxy tyramine 
TH Monosialganglioside 3-MT 
Dopa-decarboxylase Disialoganglioside Vanillacetic acid 
DDC C-neu VPA 
Phenylethanolamine-N-methyl transferase C-myc Vanillglycol 
PNMT Neuropeptide(s) VG 
PGP9.5 Somatostatin receptors Vanillglycol acid 
Dopamine-β-hydroxylase Telomerase VGA 
DBH CD44 Catechol acetic acid 
Phenylalanine Mitotic index CAA 
Drug resistance RT-PCR VAA 
MRP Dopamine Norepinephrine 
Tyrosine NB84 Vanilalamine 
3,4-dihydroxyl phenyl alanine Noradrenaline VA 
1p deletion Adrenaline Trk A 
DNA diploidy Vanillylmandelic acid Trk B 
17q VMA Trk C 
14q Epinephrine  
C.
PsychosocialEconomic
Quality of life Cost Cea 
Anxiety Cost-effectiveness Cba 
Psychosocial Econ Cua 
Adjustment   
C.
PsychosocialEconomic
Quality of life Cost Cea 
Anxiety Cost-effectiveness Cba 
Psychosocial Econ Cua 
Adjustment   
Table 2

List of tumor markers in neuroblastoma that were identified by the systematic review together with the number of papers overall and within each clinical area

Tumor MarkerOverallScreeningDiagnosisPrognosisMonitoring
MYCN 201 148 151 
VMA 125 44 78 45 18 
HVA 105 38 64 35 16 
DNA index/ploidy/diploidy/triploidy/aneuploid/hyperdiploidy 56 37 44 
Chromosome 1p or chromosome 1p36 47 34 40 
Ferritin or isoferritin 49 36 33 
NSE 45 33 28 
LDH 32 22 26 
Dopamine 24 22 10 
TrkA (nerve growth factor receptor) 25 16 16 
Adrenaline/epinephrine 15 15 
Multidrug resistance/associated protein/p-glycoprotein 16 16 
Nonadrenaline/noradrenaline/norepinephrine 13 13 
CD44 10 
Neuropeptide Y 12 10 
Tyrosine hydroxylase 12 11 
Chromosome 17q 11 
Ha-ras/P21/H-ras/c-ha-ras 11 
Telomerase/Telomeric repeats 11 
Chromosome 14q 
GD2 ganglioside 
S100 
Chromosome 11q 
Low affinity nerve growth receptor (LNGFR) 
Metanephrine 
TrkC 
3-methoxy-4-hydroxyphenyl glycol 
4-hydroxy-3-methoxymandelic acid 
Dihydroxyphenylalanine 
Dopamine β hydroxylase 
Proliferating cell nuclear antigen/proliferation index/Ki67/KiS5 protein 
Tumor MarkerOverallScreeningDiagnosisPrognosisMonitoring
MYCN 201 148 151 
VMA 125 44 78 45 18 
HVA 105 38 64 35 16 
DNA index/ploidy/diploidy/triploidy/aneuploid/hyperdiploidy 56 37 44 
Chromosome 1p or chromosome 1p36 47 34 40 
Ferritin or isoferritin 49 36 33 
NSE 45 33 28 
LDH 32 22 26 
Dopamine 24 22 10 
TrkA (nerve growth factor receptor) 25 16 16 
Adrenaline/epinephrine 15 15 
Multidrug resistance/associated protein/p-glycoprotein 16 16 
Nonadrenaline/noradrenaline/norepinephrine 13 13 
CD44 10 
Neuropeptide Y 12 10 
Tyrosine hydroxylase 12 11 
Chromosome 17q 11 
Ha-ras/P21/H-ras/c-ha-ras 11 
Telomerase/Telomeric repeats 11 
Chromosome 14q 
GD2 ganglioside 
S100 
Chromosome 11q 
Low affinity nerve growth receptor (LNGFR) 
Metanephrine 
TrkC 
3-methoxy-4-hydroxyphenyl glycol 
4-hydroxy-3-methoxymandelic acid 
Dihydroxyphenylalanine 
Dopamine β hydroxylase 
Proliferating cell nuclear antigen/proliferation index/Ki67/KiS5 protein 
Table 3

Meta-analysis results for the 13 prognostic markers, grouped by tumor marker class, for overall survival and disease-free survival together with the number of prognosis papers identified overall and the number of estimates of the log (hazard ratio) and variance obtained for each outcome

Marker typeTumor marker (relationship with prognosis)No. prognosis papersOutcomeNo. estimates obtainedPooled hazard ratio95% Confidence intervalP
DNA or chromosome abnormalities MYCN (amplification poor outcome)  DFSa 46 4.28 3.34 to 5.49 <0.0001 
  151 OS 48 5.48 4.30 to 6.97 <0.0001 
 DNA index (diploidy poor outcome)  DFS 0.33 0.20 to 0.56 <0.0001 
  44 OS 11 0.31 0.20 to 0.48 <0.0001 
 Chromosome 1p (deletion poor outcome)  DFS 3.93 2.31 to 6.68 <0.0001 
  40 OS 11 3.12 1.95 to 4.98 <0.0001 
Urinary catecholamines VMA (elevated poor outcome)  DFS Not possible Not possible Not possible 
  36 OS 0.50 0.19 to 1.29 0.15 
 HVA (elevated poor outcome)  DFS Not possible Not possible Not possible 
  26 OS 1.14 0.65 to 1.98 0.65 
 VMA:HVA (small ratio (e.g. <1) poor outcome)  DFS 0.35 0.17 to 0.72 0.0043 
  20 OS 0.44 0.18 to 1.06 0.068 
 Dopamine (elevated poor outcome)  DFS Not possible Not possible Not possible 
  10 OS Not possible Not possible Not possible 
Biological markers CD44 (high expression good outcome)  DFS 0.06 0.02 to 0.21 <0.0001 
  OS Not possible Not possible Not possible 
 TrkA (high expression good outcome)  DFS 0.26 0.16 to 0.42 <0.0001 
  16 OS 0.09 0.05 to 0.16 <0.0001 
 NSE (high serum levels poor outcome)  DFS 5.56 2.11 to 14.7 0.0005 
  28 OS 5.22 3.12 to 8.73 <0.0001 
 LDH (high serum levels poor outcome)  DFS 3.20 2.06 to 4.98 <0.0001 
  26 OS 3.36 1.72 to 6.57 0.0004 
 Ferritin (high serum levels poor outcome)  DFS 4.26 2.42 to 7.53 <0.0001 
  33 OS 2.74 1.92 to 3.91 <0.0001 
 MRP (high expression poor outcome)  DFS 6.37 3.71 to 10.9 <0.0001 
  16 OS 3.52 1.19 to 10.5 0.023 
Marker typeTumor marker (relationship with prognosis)No. prognosis papersOutcomeNo. estimates obtainedPooled hazard ratio95% Confidence intervalP
DNA or chromosome abnormalities MYCN (amplification poor outcome)  DFSa 46 4.28 3.34 to 5.49 <0.0001 
  151 OS 48 5.48 4.30 to 6.97 <0.0001 
 DNA index (diploidy poor outcome)  DFS 0.33 0.20 to 0.56 <0.0001 
  44 OS 11 0.31 0.20 to 0.48 <0.0001 
 Chromosome 1p (deletion poor outcome)  DFS 3.93 2.31 to 6.68 <0.0001 
  40 OS 11 3.12 1.95 to 4.98 <0.0001 
Urinary catecholamines VMA (elevated poor outcome)  DFS Not possible Not possible Not possible 
  36 OS 0.50 0.19 to 1.29 0.15 
 HVA (elevated poor outcome)  DFS Not possible Not possible Not possible 
  26 OS 1.14 0.65 to 1.98 0.65 
 VMA:HVA (small ratio (e.g. <1) poor outcome)  DFS 0.35 0.17 to 0.72 0.0043 
  20 OS 0.44 0.18 to 1.06 0.068 
 Dopamine (elevated poor outcome)  DFS Not possible Not possible Not possible 
  10 OS Not possible Not possible Not possible 
Biological markers CD44 (high expression good outcome)  DFS 0.06 0.02 to 0.21 <0.0001 
  OS Not possible Not possible Not possible 
 TrkA (high expression good outcome)  DFS 0.26 0.16 to 0.42 <0.0001 
  16 OS 0.09 0.05 to 0.16 <0.0001 
 NSE (high serum levels poor outcome)  DFS 5.56 2.11 to 14.7 0.0005 
  28 OS 5.22 3.12 to 8.73 <0.0001 
 LDH (high serum levels poor outcome)  DFS 3.20 2.06 to 4.98 <0.0001 
  26 OS 3.36 1.72 to 6.57 0.0004 
 Ferritin (high serum levels poor outcome)  DFS 4.26 2.42 to 7.53 <0.0001 
  33 OS 2.74 1.92 to 3.91 <0.0001 
 MRP (high expression poor outcome)  DFS 6.37 3.71 to 10.9 <0.0001 
  16 OS 3.52 1.19 to 10.5 0.023 
a

DFS, disease-free survival; OS, overall survival.

We thank Suzy Paisley at the School of Health and Related Research (ScHARR) in Sheffield for advice on systematic reviews and literature searching.

1
Brodeur G. M., Castleberry R. P. Neuroblastoma Pizzo P. A. Poplack D. G. eds. .
Principles and Practice of Paediatric Oncology
,
761
-797, J. B. Lippincott Co. Philadelphia  
1997
.
2
Maris M., Matthay K. K. Molecular biology of neuroblastoma.
J. Clin. Oncol.
,
17
:
2264
-2279,  
1999
.
3
Grovas A., Fremgen A., Rauck A., Ruymann F. B., Hutchinson C. L., Winchester D. P., Menck H. R. The National Cancer Data Base Report on patterns of childhood cancer in the United States.
Cancer (Phila.)
,
80
:
2321
-2332,  
1997
.
4
Cotterill S. J., Pearson A. D., Pritchard J., Foot A. B., Roald B., Kohler J. A., Imeson J. Clinical prognostic factors in 1277 patients with neuroblastoma: results of The European Neuroblastoma Study Group ’Survey’ 1982–1992.
Eur. J. Cancer
,
36
:
901
-908,  
2000
.
5
NHS Centre for Reviews and Dissemination .
Undertaking Systematic Reviews of Research on Effectiveness: CRD Guidelines for Those Carrying Out or Commissioning Reviews. CRD Report 4
, NHS Centre for Reviews and Dissemination University of York  
1996
.
6
Riley R. D., Burchill S. A., Abrams K. R., Heney D., Sutton A. J., Jones D. R., Lambert P. C., Young B., Wailoo A. J., Lewis I. J. A systematic review of molecular and biological markers in tumours of the Ewing’s sarcoma family.
Eur. J. Cancer
,
39
:
19
-30,  
2003
.
7
Riley R. D., Burchill S. A., Abrams K. R., Heney D., Lambert P. C., Jones D. R., Sutton A. J., Young B., Wailoo A. J., Lewis I. J. A systematic review and evaluation of the use of tumour markers in paediatric oncology: Ewing’s sarcoma and neuroblastoma.
Health Technol. Assess.
,
7
:
1
-162,  
2003
.
8
Brodeur G. M. Molecular basis for heterogeneity in human neuroblastomas.
Eur. J. Cancer
,
31
:
505
-510,  
1995
.
9
Sutton A. J. Abrams K. R. Jones D. R. Sheldon T. A. Song F. eds. .
Methods for Meta-analysis in Medical Research
, John Wiley London  
2000
.
10
Parmar M. K. B., Torri V., Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints.
Stat. Med.
,
17
:
2815
-2834,  
1998
.
11
Altman D. G. Systematic reviews of studies of prognostic variables Egger M. Smith G. Davey Altman D. G. eds. .
Systematic Reviews in Health Care: Meta-Analysis in Context, Chapter 13
,
228
-247, BMJ Publishing Group London  
2001
.
12
Begg C. B., Mazumdar M. Operating characteristics of a rank correlation test for publication bias.
Biometrics
,
50
:
1088
-101,  
1994
.
13
Egger M., Smith G. D., Schneider M., Minder C. Bias in meta-analysis detected by a simple, graphical test.
Br. Med. J.
,
315
:
629
-634,  
1997
.
14
Duval S., Tweedie R. A non-parametric “trim and fill” method of assessing publication bias in meta-analysis.
Biometrics
,
56
:
455
-463,  
2000
.
15
Riley, R. D., Abrams, K. R., Sutton, A. J., Lambert, P. C., Jones, D. R., Heney, D., and Burchill, S. A. Reporting of prognostic markers: current problems and development of guidelines for evidence-based practice in the future. Br. J. Cancer, in press.
16
Sawada T., Todo S., Fujita K., Iino S., Imashuku S., Kusunok T. Mass screening of neuroblastoma in infancy.
Am. J. Dis. Child.
,
136
:
710
-712,  
1982
.
17
Scriver C. R., Gregory D., Bernstein M., Clow C. L., Weisdorf T., Dougherty G. E., Auray-Blais C., Giguere R., Lemieux B., Laberge C. Feasibility of chemical screening of urine for neuroblastoma case finding in infancy in Quebec.
CMAJ
,
136
:
952
-956,  
1987
.
18
Egger M. Davey Smith G. Altman D. G. eds. .
Systematic Reviews in Health Care: Meta-Analysis in Context
, BMJ Publishing Group London  
2001
.
19
Egger M., Davey Smith G., O’Rourke K. Rational, potentials and promise of systematic reviews Egger M. Smith G. Davey Altman D. G. eds. .
Systematic Reviews in Health Care: Meta-Analysis in Context
,
3
-19, BMJ Publishing Group London  
2001
.
20
Egger M., Zellweger-Zähner T., Schneider M., Junker C., Lengeler C., Antes G. Language bias in randomised controlled trials published in English and German.
Lancet
,
350
:
326
-329,  
1997
.
21
Brodeur G. M., Look A. T., Shimada H., Hamilton V. M., Maris J. M., Hann H. W., Leclerc J. M., Bernstein M., Brisson L. C., Brossard J., Lemieux B., Tuchman M., Woods W. G. Biological aspects of neuroblastomas identified by mass screening in Quebec.
Med. Pediatr. Oncol.
,
36
:
157
-159,  
2001
.
22
Abel F., Ejeskar K., Kogner P., Martinsson T. Gain of chromosome arm 17q is associated with unfavourable prognosis in neuroblastoma, but does not involve mutations in the somatostatin receptor 2(SSTR2) gene at 17q24.
Br. J. Cancer
,
81
:
1402
-1409,  
1999
.
23
Bown N., Cotterill S., Lastowska M., O’Neill S., Pearson A. D., Plantaz D., Meddeb M., Danglot G., Brinkschmidt C., Christiansen H., Laureys G., Speleman F. Gain of chromosome arm 17q and adverse outcome in patients with neuroblastoma.
N. Engl. J. Med.
,
340
:
1954
-1961,  
1999
.
24
Cunsolo C. L, Bicocchi M. P., Petti A. R., Tonini G. P. Numerical and structural aberrations in advanced neuroblastoma tumours by CGH analysis; survival correlates with chromosome 17 status.
Br. J. Cancer
,
83
:
1295
-1300,  
2000
.
25
Lastowska M., Cotterill S., Pearson A. D., Roberts P., McGuckin A., Lewis I., Bown N. Gain of chromosome arm 17q predicts unfavourable outcome in neuroblastoma patients. U. K. Children’s Cancer Study Group and the U. K. Cancer Cytogenetics Group.
Eur. J. Cancer
,
33
:
1627
-1633,  
1997
.
26
Plantaz D., Mohapatra G., Matthay K. K., Pellarin M., Seeger R. C., Feuerstein B. G. Gain of chromosome 17 is the most frequent abnormality detected in neuroblastoma by comparative genomic hybridization.
Am. J. Pathol.
,
150
:
81
-89,  
1997
.
27
Vandesompele J., Speleman F., Van Roy N., Laureys G., Brinskchmidt C., Christiansen H., Lampert F., Lastowska M., Bown N., Pearson A., Nicholson J. C., Ross F., Combaret V., Delattre O., Feuerstein B. G., Plantaz D. Multicentre analysis of patterns of DNA gains and losses in 204 neuroblastoma tumors: how many genetic subgroups are there?.
Med. Pediatr. Oncol.
,
36
:
5
-10,  
2001
.
28
Bown N., Lastowska M., Cotterill S., O’Neill S., Ellershaw C., Roberts P., Lewis I., Pearson A. D. 17q gain in neuroblastoma predicts adverse clinical outcome.
Med. Pediatr. Oncol.
,
36(1)
:
14
-19,  
2001
.
29
Altman D. G., Lyman G. H. Methodological challenges in the evaluation of prognostic factors in breast cancer.
Breast Cancer Res. Treat.
,
52
:
289
-303,  
1998
.
30
Hutchon D. J. R. Publishing raw data and real time statistical analysis on e-journals.
Br. Med. J.
,
322
:
530
2001
.
31
Sakamoto J., Teramukai S., Koike A., Saji S., Ohashi Y., Nakazato H. Prognostic value of preoperative immunosuppressive acidic protein in patients with gastric carcinoma. Findings from three independent clinical trials. Tumor Marker Committee for the Study Group of Immunochemotherapy with PSK for Gastric Cancer.
Cancer (Phila.)
,
77
:
2206
-2212,  
1996
.
32
Look M. P., van Putten W. L., Duffy M. J., Harbeck N., Christensen I. J., Thomssen C., Kates R., Spyratos F., Ferno M., Eppenberger-Castori S., Sweep C.G., Ulm K., Peyrat J. P., Martin P. M., Magdelenat H., Brunner N., Duggan C., Lisboa B. W., Bendahl P. O., Quillien V., Daver A., Ricolleau G., Meijer-van Gelder M. E., Manders P., Fiets W. E., Blankenstein M.A., Broet P., Romain S., Daxenbichler G., Windbichler G., Cufer T., Borstnar S., Kueng W., Beex L. V., Klijn J. G., O’Higgins N., Eppenberger U., Janicke F., Schmitt M., Foekens J. A. Pooled analysis of prognostic impact of urokinase-type plasminogen activator and its inhibitor PAI-1 in 8377 breast cancer patients.
J. Natl. Cancer Inst.
,
94
:
116
-128,  
2002
.
33
Park K. C., Schwimmer J., Shepherd J. E., Phelps M. E., Czernin J. R., Schiepers C., Gambhir S. S. Decision analysis for the cost-effective management of recurrent colorectal cancer.
Ann. Surg.
,
233
:
310
-319,  
2001
.