The Association Between Birth Weight and the Risk of Neuroblastoma: A Meta- Analysis of Observational Studies Involving 4,361,141 Participants

doi:10.21203/rs.3.rs-2483109/v1

Download PDF

Research Article

The Association Between Birth Weight and the Risk of Neuroblastoma: A Meta- Analysis of Observational Studies Involving 4,361,141 Participants

https://doi.org/10.21203/rs.3.rs-2483109/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Purpose

One of the most common extracranial childhood solid tumors is neuroblastoma. In this study, it was aimed to perform a systematic review and meta-analysis to evaluate the risk of neuroblastoma in both high and low birth weight.

Methods

The PRISMA and MOOSE guidelines were followed during the design, analysis, and reporting of this study. A comprehensive literature search was performed for the published papers in Embase, PubMed/Medline, Scopus, and Web of Science databases. The odds ratio (OR) of neuroblastoma in high and low birth weight groups, with 95% confidence intervals (CIs), were calculated using the random-effects and fixed-effects model.

Results

A total of 16 papers and 4,361,141 participants were included in this study. In this meta-analysis, it was observed a U-shaped relationship between birth weight and neuroblastoma. Individuals with both high and low birth weights may face an increased risk of developing neuroblastoma later in life (OR = 1.17, 95% CI: 1.06–1.29, p = 0.002; OR = 1.19; 95% CI: 1.03–1.37, p = 0.017, respectively).

Conclusion

As a result, both high and low birth weight in individuals may be among the important risk factors for neuroblastoma development.

neuroblastoma

meta-analysis

birth weight

One of the most common extracranial childhood solid tumors is neuroblastoma [1]. It develops from embryonic neural crest tissue and contributes to around 15% of all pediatric cancer fatalities. The incidence rate is 10,7 cases per 1,000,000 individuals aged 0–14 years [1, 2]. Because of its unexpected biological activity, which varies from spontaneous remission to highly quick metastatic spread and making detection, therapy, and prognosis difficult, it has been labeled a "clinical enigma" of cancer research [3].

A definitive risk factor for neuroblastoma has yet to be discovered. Since neuroblastoma occurs in the beginning of life, it has been hypothesized that prenatal risk factors may play an important role in the development of this disease. However, in few studies conducted so far, it has been reported that several factors such as maternal smoking [4], maternal medications [5], or supplements taken during pregnancy [6], may affect the development of the disease.

Considering all this, it is considered as an alternative approach to concentrate on exposure indicators, which can be easily collected in large samples with low bias and are well-established markers of the intrauterine environment. Birth weight is one such marker as it is taken immediately after birth by experienced individuals with low systematic error and is usually documented in medical records [7]. Studies have reported that external conditions such as nutrition during pregnancy and maternal diseases are important in terms of birth weight [8, 9]. The effect of birth weight in newborns and children with neuroblastoma has been previously reviewed in various studies [10–13]. In this study, we aimed to evaluate the association between birth weight and neuroblastoma risk with a meta-analysis.

Literature search and search strategy

“Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA)” [14] and “Meta-analysis of Observational Studies in Epidemiology (MOOSE)” [15] reporting guidelines were closely followed during the reporting, design, and analysis of this study. Between February - March 11 (2022), structured and comprehensive literature searches were conducted for the published papers in Web of Science, PubMed / Medline, Scopus, and Embase databases, and an update was performed on April 12 for this search. Related four major factors were considered for the search query lines when choosing the keywords: “birth weight”, “neuroblastoma”, “prenatal”, “birth characteristics”. The related keywords were combined using Medical Subject Headings (MeSH) and text terms, and the Boolean operators (AND / OR) were used to integrate the keywords. Search strategies in the related literature are available in Supplementary Table S1. A flow-chart demonstrating the selection process is also presented in Fig. 1. Relevant studies that could be included in the meta-analysis were downloaded from related databases. Afterward, these studies were transferred to Mendeley data management program for data evaluation and analysis. The study protocol is registered in PROSPERO with ID number CRD42021274163.

Study Selection and Inclusion/exclusion Criteria

Original case-control or cohort studies published about the effect of birth weight and risk of neuroblastoma were included in this meta-analysis. Studies that were published in the English language were researched, and no other types of paper were examined. In addition, studies covering the following criteria listed below were included in this research: i) Being an original study examining the relationship between birth weight and neuroblastoma risk, and ii) odds ratio (OR) with 95% CIs for risk of neuroblastoma in at least two strata of birth weight had to be reported. Exclusion criteria: i) reviews, guidelines, opinions, or other non-original data publications; ii) projects and clinical trials that were incomplete; and iii) no clinical evidence from animal and laboratory studies.

Data Extraction and Acquisition

Papers were initially searched based on the title and abstract in related database, and the full text of the appropriate papers were examined. The article titles and abstracts, and any differences amongst co-authors regarding which papers were eligible and which were not were handled using Delphi consensus criteria were examined by two independent investigators (MEA and HE) [16], and the data was extracted into a pre-defined spreadsheet created using Microsoft Excel®. The following study characteristics were extracted: “publication year”, “country”, “region”, “study design”, “year of birth”, “age at diagnosis”, “study size”, “matching ratio (if case–control study)”, “matching variables (if case–control study)”, “source of controls (if case–control study)”, “source of case diagnosis”, “source of data for birth weight”, “effect measures” and “confounders”. All data entries were validated by an independent reviewer. To overcome data limitations - in case of missing data or doubt- the corresponding author(s) of the articles were contacted via email to obtain more details. Prepared data were cross-checked by two investigators via a standard spreadsheet to reach consensus.

Statistical analysis

Forest plots were utilized to compute and graphically illustrate the OR with 95% CI of both high and low birth weight, and to summarize them. Three different meta-analytical methodologies were associated with the study in conducting the meta-analysis: 1) 4000g was taken as the limit to assess neuroblastoma risk below and above 4000g for dichotomous comparison; 2) for low birth weight, 2500g was taken as the cut-off point; and 3) For trend analysis, the regression coefficients obtained from the study were combined with the pool-first method [17]. We estimated the neuroblastoma risk above and below the threshold value, with crude OR and 95% CI, with dual assessments.

We used the “pool-first method” to measure the dose–response association between birth weight and neuroblastoma risk in studies with more than two categories of birth weight. Greenland and Longnecker [17] proposed this approach, which permits meta-analytic pooling of regression data while accounting for the fact that estimates used to generate single regression coefficients are linked within one research. This problem is handled by pooling the data inside each study (“pool first”) to acquire the relevant regression coefficients for each research, and then pooling these regression coefficients. The resultant pooled regression coefficient can be viewed similarly to a single study's regression coefficient. A U-shaped relationship between birth weight and the risk of neuroblastoma was discovered by visually inspecting the plots of 15 of the 16 independent studies. A log-quadratic model was used to construct a study-specific linear and quadratic regression coefficient, as well as 95% CI, for each of these investigations. To use a random-effects model, the resultant linear and quadratic regression coefficients and their 95% CIs were pooled using the pool-first strategy [17]. We replicated the pool-first analysis after doing the sensitivity analyses, this time using a linear model confined to the birth-weight spectrum and studies that used registries to gather birth-weight data.

Cochran's Q test and I² statistics, as developed by Higgins et al. [18], was used to quantified between-study heterogeneity in all meta-analyses. A ratio of more than 50% in I² statistics and a p ≤ 0.05 in Cochran's Q test revealed that significant heterogeneity [18]. The analyses were carried out using both random-effect and fixed-effect models. Sensitivity analyses -“all studies were excluded from analysis separately”- were conducted to test the reliability of the study results. To investigate probable sources of variability between research outcomes, four distinct subgroup analyses were conducted. To begin, we produced separate estimates for all research conducted in North America (The United States and Canada) and those conducted in Europe. Next, we divided all research into two groups based on how they obtained birth weight (registry/certificate vs. interview). Then, subgroup estimates were produced based on how the case diagnosis was obtained (registry vs others). Finally, we divided all the research into two groups based on the source of controls (registry/certificate vs. others). For each significant outcome in our research, Egger's linear regression test and funnel plots were utilized to examine the possibility of publication bias. The statistical significance level was determined as a 2-sided p < 0.05. All analyses were performed using ProMeta3® meta-analysis software [19].

In the initial search, a total of 595 articles were found in related databases (with 78 in Web of Science, 118 in PubMed, 112 in Embase, and 287 in Scopus). After a preliminary review and the elimination of duplicates, 467 papers were screened and chosen for further evaluation. A total of 18 studies were included in this study after applying the inclusion/exclusion criteria. Only an adjusted OR for neuroblastoma following high birth weight was reported in one research [20] and an unadjusted OR could not be determined from the data. Another study [21] had to be removed due to significant case overlap with a prior study. In one paper [22], since both USA and UK data were reported, it was assumed that this research represents separate studies in the quantitative analysis. In addition, since only low birth weight OR was reported in one study [23] and only high birth weight OR was also reported in one study [24], the relevant studies were added to the analyses separately. As a result, a total of 16 papers (Fig. 1) could be used for meta-analysis [10–13, 22–33] The major parameters of the included studies are available in Table 1.

Table 1

Distribution of included studies by country, design, birth year (s), diagnosis year(s), age at diagnosis (years), and number of cases/controls
Study (First author (year))	Country	Design	Birth year(s)	Diagnosis year(s)	Diagnosis age (years)	Number of cases/controls
Bjorge (2008) (25)	Norway	Cohort	1967–2004	1967–2004	0–15	Cohort: 2,127,452
Bluhm (2008) (26)	Sweden	Case-control	1973-95	1973-95	NA	245/1,225
Buck (2001) (10)	USA	Case-control	1971-87	1976–87	0–5	310/155
Chow (2007) (27)	USA	Case–control	1980–2004	1980–2004	NA	240/2,400
Hamrick (2001) (28)	USA, Canada	Case–control	1974–94	1992–94	0–18	504/504
Johnson and Spitz (1985) (23)	USA	Case-control	1949–78	1964–78	0–14	157/314
Johnson (2008) (29)	USA	Case-control	1976–2004	1988–2004	0–14	155/8.752
McLaughlin (2008) (11)	USA	Case-control	1983–2001	1985–2001	0–14	529/12,010
Munzer (2008) (12)	France	Case-control	1989–2004	2003–04	0–14	191/1,681
Neglia (1988) (24)	USA	Case-control	NA	NA	0–9	97/388
O'Neill (2015) (22)	USA	Case-control	1970–2004	1980–2004	0–14	16,554/53,716
O'Neill (2015) (22)	UK	Case-control	1980–2007	1980–2007	0–14	23,772/33,206
Parodi (2014) (30)	Italy	Case-control	1998–2001	1998–2001	0–10	153/1,044
Rios (2016) (31)	France	Case-control	NA	2003-2004-2010-2011	0–15	357/1,783
Schuz (1999) (32)	Germany	Case-control	1978–94	1992–94	0–14	183/1,785
Uruyama (2007) (13)	USA	Case-control	1983–97	1988–97	0–4	508/1,015
Heck (2020) (33)	Taiwan	Cohort	2004–2014	2004–2014	0–14	Cohort: 2,079,037
NA: not available

Table 2

Descriptive and clinical features of included studies in the meta-analysis
Study (First author)	Source of controls	Source of case diagnosis	Source of data for birth weight	Original study results
Bjorge (2008) (25)	NA (cohort: population)	Cancer registry	Birth registry	“Unadjusted risk ratio (95% CI) for birth weight 4000–4499 g (vs 3000–3499 g): 1.4 (0.9–2.1); for birth weight < 2500 g: 0.6 (0.2–1.9)”
Bluhm (2008) (26)	Birth registry	Swedish Cancer Register or Death Register.	National Cancer Register	“OR (95% CI), adjusted for birth year, for birth weight > 4500 g (vs 2500–4499 g): 1.07 (0.53–2.18); for birth weight < 2500 g: 1.05 (0.51–2.19)”
Buck (2001) (10)	Birth registry	Cancer registry	Birth certificates	“Unadjusted OR (95% CI) for birth weight > 4000 g (vs 3000–3499 g): 1.2 (0.6–2.2); for birth weight < 2500 g: 0.9 (0.4–2.2)”
Chow (2007) (27)	Birth certificates	Cancer registry	Birth certificates	“OR (95% CI), adjusted for birth year, for birth weight > 4000 g (vs 2500–3999 g): 1.25 (0.87–1.79); for birth weight < 2500 g: 0.75 (0.38–1.51)”
Hamrick (2001) (28)	Random digit dialing	Clinical records^a	Interview	“OR (95% CI), adjusted for gender, race, maternal education, and household income, for birth weight 4001–4499 g (vs 2501–4000 g): 1.1 (0.7–1.7); for birth weight 1500–2500 g: 1.1 (0.6–2.0)”
Johnson and Spitz (1985) (23)	Birth certificates	Death certificates	Birth certificates	“Unadjusted OR (95% CI) for birth weight < 2500 g (vs 4380 g): 3.22 (1.13–9.20)”
Johnson (2008) (29)	Birth registry	Cancer surveillance	Birth records	“Hazard ratio (95% CI), adjusted for sex and birth year, for birth weight 4000 g (vs 2500–4000 g): 1.10 (0.70–1.73); for birth weight < 2500 g: 1.17 (0.60–2.28)”
McLaughlin (2008) (11)	Birth certificates	Cancer registry	Birth certificates	“Risk ratio (95% CI), adjusted for birth year, region, gender, and race, for birth weight 44500 g (vs 2500–3499 g): 1.4 (0.7–2.5); for birth weight < 2500 g: 1.5 (1.0–2.1)”
Munzer (2008) (12)	Random digit dialing	Cancer registry	Interview	“OR (95% CI), adjusted for age and gender, for birth weight 4000 g (vs 3000–3499 g): 1.6 (0.9–2.8); for birth weight < 2500 g: 1.8 (0.8–3.8)”
Neglia (1988) (24)	Birth certificates	Clinical records	Birth certificates	“Unadjusted OR (95% CI) for birth weight 4000 g (vs < 4000 g): 0.96 (0.47–1.73)”
O'Neill (2015) (22)	Birth records	Cancer registry	Birth certificates	“OR adjusted for gestational age, birth order, plurality, maternal age, and race/ethnicity for birth weight > 4000 g (vs 3000–3999) 1.22 (1.02–1.45); for birth weight < 2500 1.36 (0.96–1.93)- 1.14 (0.87–1.51)”
O'Neill (2015) (22)	Birth records	National Cancer Register	Birth records	“OR adjusted sex, period, and region of birth for birth weight > 4000 g (vs 3000–3999) 1.27 (0.94–1.71); for birth weight < 2500 1.31 (0.61–2.78)-0.98 (0.58–1.64)”
Parodi (2014) (30)	Birth certificates	Clinical records	Interview	“OR (95% CI), adjusted for birth year, for birth weight > 4000 g 1.1 (0.57–2.00); for birth weight < 2500 g: 0.59 (0.22–1.6)”
Rios (2016) (31)	Random digit dialing	Clinical records	Birth records	“OR (95% CI), adjusted for age and sex, birth-order, maternal age, urban status of the area of residence and study birth weight 4000 g (vs 3000–3499 g): 1.4 (0.9–2.2); for birth weight < 2500 g: 1.2 (0.9–1.7)”
Schuz (1999) (32)	Population	Cancer registry	Interview	“OR (95% CI), adjusted for socio-economic status, for birth weight 4000 g (vs 2500–4000 g): 1.3 (0.8–2.1); for birth weight < 2500 g: 2.4 (1.2–4.8)”
Uruyama (2007) (13)	Birth registry	Cancer registry	Birth certificates	“Unadjusted OR (95% CI) for birth weight (term) 4000 g (vs 2500–4000 g): 1.25 (0.88–1.78); for birth weight < 2500 g (term): 1.40 (0.65–3.04)”
Heck (2020) (33)	NA (cohort: population)	National Cancer Register	Birth registry	“HR (95% CI) Adjusted mother’s age, father’s age, family income, for birth weight > 4000 g (vs 2,500–3,999g): 0.77 (0.25–2.42); for birth weight < 2500 g: 0.79 (0.42–1.45)”
NA: not available
^a Cases came from two collaborative clinical trial groups.

A total of 4,361,141 participants were involved in the research, Two of the 16 studies were cohort, and the other 14 studies used a case–control strategy with matching ratios ranging from 1:1 to 1:10 in case–control studies. The initial research came out in 1985, and the most current one came out in 2020. The research was carried out in the “United States, Canada, France, Norway, Sweden, Italy, Taiwan and Germany”. Participants varied in age from 0 to 18. The number of participants in the study ranged from 471 to 2,127,452. Cases came from cancer registries in ten of the investigations, whereas they came from other sources in the other six.

High Birth Weight And Neuroblastoma Risk

There were 15 studies [10–13, 22, 24–33] that provided data for calculating the OR (95% CI) of risk of neuroblastoma in patients with a high birth weight (> 4000 g) compared to those with lower birth weight. The forest plot with ORs with 95% CIs, as well as the pooled estimate for the risk of neuroblastoma in high birth weight participants, are presented in Fig. 2. High birth weight was associated with an increased risk of neuroblastoma. This impact measure was the same when the random-effects model and the fixed-effects model were used (OR = 1.17; 95% CI: 1.06–1.29, p = 0.002; heterogeneity: Chi² = 2.33, df = 15, I² = 0%, p > 0.05). Sensitivity analyses were performed by extracting each study separately. No significant change was observed among the studies (df = 15, I² = 0%, p > 0.05) (Fig. 3). The pooled estimate was quite robust, according to sensitivity analysis (fixed-effects model): excluding individual study values resulted in pooled ORs ranging from 1.17 (95% CI: 1.05–1.29) to 1.18 (95% CI: 1.07–1.30). Visual inspection of the funnel plots (Supplementary Fig. S1), as well as Begg's test (Z value: -1.14; p = 0.255) and Egger's linear regression test revealed no evidence of publication bias (Intercept: -0.14; p = 0.609).

Low Birth Weight And Neuroblastoma Risk

15 studies [10–13, 21–26, 28–31, 35] provided data for the computation of the OR (95% CI) of risk of neuroblastoma in patients with low birth weight (< 2500 g) compared to those with birth weights more than this threshold value. Figure 4 depicts a forest plot with ORs and 95% CIs, as well as the pooled estimate of the risk of neuroblastoma following low birth weight. Low birth weight was associated with an elevated risk of neuroblastoma in both the random-effects model (OR = 1.19; 95% CI: 1.03–1.37, p = 0.017) and the fixed-effects model (OR = 1.19; 95% CI: 1.05–1.36, p = 0.007; heterogeneity: Chi² = 16.93, df = 15, I² = 0%, p = 0.323). Sensitivity analyses were also performed by extracting each study separately for low birth weight (df = 15, I² = 11.42%, p > 0.05). No noticeable change was observed in the analysis results. Thus, the robustness of the analysis results was confirmed by sensitivity analysis (Fig. 5). No noticeable publication bias was observed among the included studies according to the symmetry of the funnel plot (Supplementary Fig S2), Begg's test (Z-value: -1.93; p = 0.054), and Egger's linear regression test (Intercept: -0.71; p = 0.341).

Subgroup Analysis

Subgroup analysis was used to further analyze the connection with both high and low birth weight (as region). In terms of the connection with high birth weight, region-specific pooled estimates revealed no significant differences (Table 3). The relationship with low birth weight and neuroblastoma was stronger in European research than in those conducted in the United States or Canada. The technique used to collect birth weight data had no effect on the strength of the link with high birth weight. In contrast, it had a significant impact on the link with low birth weight: studies that employed interview-based data revealed a > 45% higher risk, but those that relied on registry data had a significantly lower estimate. As seen in Table 3, although the manner of acquiring the case diagnosis had no significant effect on either the high or low birth weight estimates, the low birth weight estimate was greatly influenced by the source of controls, with registry-based studies having lesser effects.

Table 3

The association between birth weight and the neuroblastoma risk: moderator analysis
Characteristics of study	Category	High birth weight OR (95% CI)	Low birth weight OR (95% CI)
Geographic area	North America	1.16 (1.03–1.32)	1.18 (1.00-1.40
Geographic area	Europe	1.19 (1.01–1.41	1.27 (1.03–1.57)
Source of data for birth weight	Registry	1.17 (1.05–1.31)	1.17 (1.01–1.35)
Source of data for birth weight	Interview	1.21 (1.00-1.45)	1.45 (1.07–1.96)
Source of case diagnosis	Registry	1.20 (1.07–1.34)	1.20 (1.03–1.42)
Source of case diagnosis	Other^a	1.09 (0.88–1.35)	1.24 (0.99–1.56)
Source of controls	Registry/certificate	1.18 (1.05–1.32)	1.14 (0.97–1.35)
Source of controls	Other^b	1.17 (0.97–1.41)	1.36 (1.09–1.69)
^a Included: death certificates, records, surveillance.
^b Included: population, random digit dialing.

Several prior reports have shown that birth weight is significantly associated with different types of cancer in infants and children [34, 35]. However, it is a well-known fact that there is insufficient evidence in studies on neuroblastoma. A total of 16 papers were included in this systematic review and meta-analysis. 4,361,141 participants worldwide were systematically analyzed, and a meta-analysis was conducted. A U-shaped association between birth weight and neuroblastoma was discovered in this study. Individuals with both high and low birth weights may face an increased risk of developing neuroblastoma later in life (OR = 1.17, 95% CI: 1.06–1.29, p = 0.002; OR = 1.19; 95% CI: 1.03–1.37, p = 0.017, respectively). In addition to all these, our sensitivity analyses showed that there may be a bias in relation to low birth weight. Studies using interview method as data source had stronger relationships than studies using recordings. Importantly, the results of our meta-analysis differ in some respects from the results reached by the authors of the individual studies. Namely, Urayama et al. clearly concluded that high birth weight is a risk factor for neuroblastoma [13]. In the studies conducted by Schuz et al. [21], Hemrick et al. [28], and Johnson and Spitz [23], it was concluded that low birth weight was strongly associated with neuroblastoma risk. Similar to our study, four studies [11, 12, 22, 31] reported a U-shaped relationship between birth weight and neuroblastoma. Five study reports [24, 25, 27, 29, 33] stated that there was no significant association between birth weight and neuroblastoma, while the authors of one study did not comment on this issue [10]. It is thought that the authors of the related studies may have overlooked an association with high birth weight, either because it was not statistically significant or because it may have been “masked” because of the selection of the reference layer for birth weight. Very importantly, the birth weight measurement error must be considered at this point. In addition, studies on the relationships between birth weight and disease risks can be confusing in many ways. Correction and sensitivity analyses for confounders were done by matching in the included studies, but the adjusted estimates did not differ significantly from the unadjusted results.

When the relationship between high birth weight and neuroblastoma was examined, there was a significant level of homogeneity between the study results (according to I² statistics which a method for assessing study heterogeneity [18]. In addition, sensitivity analyses, a method that excludes each study separately, were performed. There was no noticeable difference in the analysis results. These findings showed that the study was stable and very robust. In addition, studies reporting > 4500 (extremely high birth weight) birth weights have reported an increased risk of neuroblastoma [11, 25, 28]. In our subgroup and sensitivity analyses, the relationship between low birth weight and neuroblastoma was also evaluated. A stronger association was reported in the analyses in interview-based studies than in objective sources such as birth records/registry. This suggested that mothers may remember their children born with low birth weight lower than they actually are. To be clear, it is anticipated that there may be a bias in interview-based studies.

It may not be correct to assume that birth weight or fetal growth alone is a causal factor leading to disease risks [36]. At this point, it is important to focus on the main causes and mechanisms that cause the increase or decrease in birth weight. It is a well-known fact that the incidence of several diseases risk continues to increase rapidly worldwide, with lifestyle changes, genetic factors, ethnicity, and environmental factors such as poor nutrition, polluted air, obesity, and sedentary lifestyle. For example, maternal diabetes mellitus, which is one of the factors that stimulate prenatal weight gain, may be a remarkable factor in this respect. It has been clearly shown in some studies that it leads to an increase in birth weight [37, 38]. Similarly, the fact that the mother is overweight during pregnancy can also cause this. However, in terms of neuroblastoma, not only these factors, but also all factors should be considered.

As a result, both high and low birth weight in individuals may be among the important risk factors for neuroblastoma development. At this point, there is an increasing need to investigate, particularly in randomized controlled trials, several other causative factors that may lead to the development of neuroblastoma and high or low birth weight in infants.

Data availability: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Funding statement: None.

Conflict of interest: None.

Authors’ Contributions: MEA, REC, YB, HE, and HNO contributed significantly to the work's idea, design, and data collection. MEA and HE contributed to data analysis. MEA, REC, YB, HE, and HNO all contributed to the work's data interpretation. All authors contributed to the preparation and critical revision of the paper. The final manuscript was reviewed and approved by all authors.

Takita J. Molecular Basis and Clinical Features of Neuroblastoma. JMA J. 2021;4(4):321–331.
Matthay KK, Maris JM, Schleiermacher G, et al. Neuroblastoma. Nat Rev Dis Primers. 2016;10(2):16078.
Brodeur GM. Neuroblastoma: biological insights into a clinical enigma. Nat Rev Cancer. 2003;3:203–16.
Kramer S, Ward E, Meadows AT, et al. Medical and drug risk factors associated with neuroblastoma: a case–control study. J Natl Cancer Inst. 1987;78:797–804.
Cook MN, Olshan AF, Guess HA, et al. Maternal medication use and neuroblastoma in offspring. Am J Epidemiol. 2004;159:721–31.
Olshan AF, Smith JC, Bondy ML, et al. Maternal vitamin use and reduced risk of neuroblastoma. Epidemiology. 2002;13:575–80.
Reichman NE, Hade EM. Validation of birth certificate data. A study of women in New Jersey’s HealthStart program. Ann Epidemiol. 2001;11:186–93.
Lunde A, Melve KK, Gjessing HK, et al. Genetic and environmental influenecs on birth weight, birth length, head circumference, and gestational age by use of population-based parent–offspring data. Am J Epidemiol. 2007;165:734–41.
Brooks AA, Johnson MR, Steer PJ, et al. Birth weight: nature or nurture? Early Hum Dev. 1995;42:29–35.
Buck GM, Michalek AM, Chen CJ, et al. Perinatal factors and risk of neuroblastoma. Paediatr Perinat Epidemiol. 2001;15:47–53.
McLaughlin CC, Baptiste MS, Schymura MJ, et al. Perinatal risk factors for neuroblastoma. Cancer Causes Contr. 2009;20:289–301.
Munzer C, Menegaux F, Lacour B, et al. Birth-related characteristics, congenital malformation, maternal reproductive history and neuroblastoma: The ESCALE study (SFCE). Int J Cancer. 2008;122:2315–21.
Uruyama KY, Von Behren J, Reynolds P. Birth characteristics and risk of neuroblastoma in young children. Am J Epidemiol. 2007;165:486–95.
Liberati A, Altman DG, Tetzla J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: Explanation and elaboration. PLoS Med. 2009;6:e1000100.
Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. JAMA. 2000;283 2008–12.
Verhagen AP, de Vet HC, de Bie RA, et al. The Delphi list: a criteria list for quality assessment of randomized clinical trials for conducting systematic reviews developed by Delphi consensus. J Clin Epidemiol. 1998;51(12):1235–41.
Greenland S, Longnecker MP. Methods for trend estimation from summarized dose–response data, with applications to meta-analysis. Am J Epidemiol. 1992;135:1301–9.
Higgins JPT, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–60.
ProMeta-3. 2015. “Professional statistical software for conducting meta-analysis”. It is based on ProMeta 2.1 deployed by Internovi in 2015. https://idostatistics.com/prometa3/.
Yeazel MW, Ross JA, Buckley JD, et al. High birth weight and risk of specific childhood cancers: a report from the Children’s Cancer Group. J Pediatr. 1997;131(5):671–677.
Schuz J, Kaletsch U, Meinert R, et al. Risk factors for neuroblastoma at different stages of disease. Results from a population-based case– control study in Germany. J Clin Epidemiol. 2001;54:702–9.
O’Neill KA, Murphy MF, Bunch KJ, et al. Infant birthweight and risk of childhood cancer: international population-based case control studies of 40 000 cases. Int J Epidemiol. 2015;44(1):153–168.
Johnson CC, Spitz MR. Neuroblastoma: case–control analysis of birth characteristics. J Natl Cancer Inst. 1985;74:789–92.
Neglia JP, Smithson WA, Gunderson P, et al. Prenatal and perinatal risk factors for neuroblastoma. A case–control study. Cancer. 1988;61:2202–6.
Bjorge T, Engeland A, Tretli S, et al. Birth and parental characteristics and risk of neuroblastoma in a population-based Norwegian cohort study. Br J Cancer. 2008;99:1165–69.
Bluhm E, McNeil DE, Cnattingius S, et al. Prenatal and perinatal risk factors for neuroblastoma. Int J Cancer Res. 2008;123(12):2885–2890.
Chow EJ, Friedman DL, Mueller BA. Maternal and perinatal characteristics in relation to neuroblastoma. Cancer. 2007;109:983–92.
Hamrick SEG, Olshan AF, Neglia JP, et al. Association of pregnancy history and birth characteristics with neuroblastoma: a report from the Children’s Cancer Group and the Pediatric Oncology Group. Paediatr Perinat Epidemiol. 2001;15:328–37.
Johnson KJ, Puumala SE, Soler JT, et al. Perinatal characteristics and risk of neuroblastoma. Int J Cancer, 2008;123:1166–72.
Parodi S, Merlo DF, Ranucci A, et al. Risk of neuroblastoma, maternal characteristics and perinatal exposures: the SETIL study. Cancer Epidemiol. 2014;38(6):686–694.
Rios P, Bailey HD, Orsi L, et al. Risk of neuroblastoma, birth-related characteristics, congenital malformations and perinatal exposures: A pooled analysis of the ESCALE and ESTELLE French studies (SFCE). Int J Cancer. 2016;139(9):1936–1948.
Schuz J, Kaatsch P, Kaletsch U, et al. Association of childhood cancer with factors related to pregnancy and birth. In J Epidemiol. 1999;28:631–39.
Heck JE, Lee PC, Wu CK, et al. Gestational risk factors and childhood cancers: A cohort study in Taiwan. Int J Cancer. 2020;147(5):1343–1353.
Hjalgrim LL, Westergard T, Rostgaard K, et al. Birth weight as a risk factor for childhood leukemia: a meta-analysis of 18 epidemiologic studies. Am J Epidemiol. 2003;158:724–35.
Harder T, Plagemann A, Harder A. Birth weight and subsequent risk of childhood primary brain tumours: a metaanalysis. Am J Epidemiol. 2008;168:366–73.
Basso O. Birth weight is forever. Epidemiology. 2008;19:204–5.
Hyperglycemia and adverse pregnancy outcomes. N Engl J Med. 2008;358:1991–2002.
Crowther NJ, Hiller JE, Moss JR, et al. Effect of treatment of gestational diabetes mellitus on pregnancy outcomes. N Engl J Med. 2005;352:2477–86

No competing interests reported.

SUPPLEMENTARYMATERIALS.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

The Association Between Birth Weight and the Risk of Neuroblastoma: A Meta- Analysis of Observational Studies Involving 4,361,141 Participants

Status:

Version 1

Abstract

Purpose

Methods

Results

Conclusion

Figures

Introduction

Methods

Literature search and search strategy

Study Selection and Inclusion/exclusion Criteria

Data Extraction and Acquisition

Statistical analysis

Results

High Birth Weight And Neuroblastoma Risk

Low Birth Weight And Neuroblastoma Risk

Subgroup Analysis

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1