Received: 1 October 2020
Revised: 27 October 2020
Accepted: 14 November 2020
Check for updates
Pediatric Blood & Cancer
SOCIÉTÉ INTERNATIONALE D’ONCOLOGIE PÉDIATRIQUE
aspho The American Society of Pediatric Hematology/Oncology
INTERNATIONAL SOCIETY OF PAEDIATRIC ONCOLOGY
Clinical prognostic factors in pediatric adrenocortical tumors: A meta-analysis
Elisa Zambaiti1 İD Miriam Duci1 Federica De Corti1 Piergiorgio Gamba1
Patrizia Dall’Igna2 1D
Filippo Ghidini1 Calogero Virgone1 İD
1 Department of Women’s and Children’s Health, University Hospital of Padua, Padua, Italy
2 Department of Emergencies and Organ Transplantation, Azienda Ospedaliero-Universitaria Consorziale Ospedale Pediatrico Giovanni XXIII, Bari, Italy
Correspondence
Calogero Virgone, Department of Women’s and Children’s Health, University Hospital of Padua, Via Giustiniani 3, 35128 Padua, Italy. Email: calogero.virgone@unipd.it
Abstract
Pediatric adrenocortical tumors (ACT) are rare and sometimes aggressive malignan- cies, but there is no consensus on the outcome predictors in children. A systematic search of MEDLINE, SCOPUS, Web of Science, and the Cochrane Library for studies from 1994 to 2020 about pediatric ACT was performed. In 42 studies, 1006 patients, aged 0-18 years, were included. The meta-analyses resulted in the following predictors of better outcome: age <4 years (P <. 00001), nonsecreting tumors (P =. 004), complete surgical resection (P <. 00001), tumor volume (P <. 0001), tumor weight (P <. 00001), tumor maximum diameter (P = . 0009), and Stage I disease (P < . 00001). Moreover, patients affected by Cushing syndrome showed a worse outcome (P < . 0001). Inter- national prospective studies should be implemented to standardize clinical prognostic factors evaluation, together with pathological scores, in the stratification of pediatric ACT.
KEYWORDS
adrenocortical carcinoma, adrenocortical tumors, children, meta-analysis, prognosis study, prog- nostic factors, systematic review
1 INTRODUCTION
Pediatric adrenocortical tumors (ACTs) are rare and sometimes aggres- sive endocrine malignancies, frequently associated with Li-Fraumeni syndrome, a familial cancer predisposition disorder caused by germline mutations in the tumor suppressor gene TP53, and account for approx- imately 0.38% of all childhood cancer cases.1 ACT develop from the adrenal cortex, with an estimated incidence of one per million each year, and comprise benign adrenocortical adenomas (ACA) and highly malignant adrenocortical carcinomas (ACC), whose pathogenesis is incompletely understood. Patients with ACC generally have a poor clinical outcome, because there is no effective therapy for advanced and metastatic forms, accounting for a 5-year overall survival of less than 40%.2 On the other hand, ACA is associated with excellent prog- nosis, but only about 20% of pediatric ACTs are classified as ACA.
As a matter of fact, children with ACC seem to have a better out- come when compared with adults, and some authors attributed this finding to an overestimation of malignancy based on the use of adult scores, which may be not fully applicable to the pediatric population.3,4 The Wieneke index has been reported to have a prognostic value more reliable in comparison with adult scores,5,6 and lately the five- item score7 was described to be useful when stratifying patients with ACT.
In this scenario, it is still difficult to establish which patients may benefit from a more aggressive medical approach after surgery, and the use of pediatric pathology scores could be a valid option if considered together with clinical risk factors, as it happens in the management of other pediatric tumors as neuroblastoma or rhabdomyosarcoma.
In the past, various clinical features were variably reported to asso- ciate with a poor outcome: age >4 years, volume, weight, size, Cushing syndrome, R1 resections, initial biopsy, and stage are the clinical fea- tures more commonly taken into account to define risk stratification.8,9
Unfortunately, there is no consensus on which clinical features may be determinant to predict the outcome in pediatric ACT, as most of the features described above did correlate with prognosis in a univariate analysis but were not confirmed at the multivariate analysis.8
2 METHODS
2.1 Protocol, eligibility criteria, information sources, and search
This review was performed according to an a priori designed protocol and recommended for systematic reviews and meta-analysis.10-12
MEDLINE, SCOPUS, Web of Science, and the Cochrane Library, including the Cochrane Database of Systematic Reviews (CDSR), Database of Abstracts of Reviews of Effects (DARE), and the Cochrane Central Register of Controlled Trials (CENTRAL), were searched elec- tronically in April 2020 utilizing combinations of relevant medical subject heading (MeSH) terms, keywords, and word variants for “adrenocortical tumors,” “adrenocortical adenoma,” “adrenocortical carcinoma,” “prognostic factors,” and “children” (Supporting Informa- tion S1). The search and selection criteria were restricted to English language. Reference lists of relevant articles and reviews were manu- ally searched for additional reports. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA-P) guidelines13 were followed.
The study was registered with the PROSPERO database (registra- tion number CRD42020180970).
2.2 Study selection, data collection, and data items
Studies were assessed according to the following criteria: population, outcome, clinical features, and tumor features. Inclusion criteria were: studies reporting pediatric cases of ACTs with available data on treatment, histology, and follow up; publication span including the last 25 years (1994-2020). Four authors, divided in two groups (Filippo Ghidini, Calogero Virgone and Elisa Zambaiti, Miriam Duci) reviewed all abstracts independently. Agreement about potential relevance was reached by consensus, and full-text copies of those articles were obtained. The same reviewers independently extracted relevant data regarding study characteristics and ACT features and outcomes. Case reports or case series without outcome data and/or clinical data were excluded. Inconsistencies were discussed by the reviewers and consensus reached. If more than one study was published for the same cohort with identical end points, the report containing the most comprehensive information on the population was included to avoid overlapping populations. For those articles in which information was not reported but the methodology was such that this information would have been recorded initially, the authors were contacted. Quality assessment of the included studies was performed using the Newcastle-Ottawa Scale for cohort studies (Supporting Information S2).14
2.3 Summary measures, synthesis of the results, and risk of bias
The clinical prognostic factors analyzed in this review were age, sex, tumor secreting status, tumor volume, tumor weight, tumor maxi- mum diameter, surgical results, and stage. Age greater than or less than 4 years was considered according to previous reports, which showed that children under 4 years of age usually show a favorable outcome. Indeed, distant metastases are observed at a higher rate in adolescents.7,8
Secretion pattern has been sporadically reported to associate with outcome: Cushing syndrome and nonsecreting tumors were reported to have worse prognosis.2
Cut-off values for tumor volume (200 cm3), for tumor weight (100 or 400 g), and for tumor maximum diameter (<5 cm, between 5 and 10 cm, >10 cm) were established in previous literature data.1,8,9 Although size in the ACT literature has been described with different units of mea- surement, the evidence that larger tumors may present a dismal out- come at higher rates is widely recognized.
Surgical results were defined as complete resection or incom- plete resection (microscopic or macroscopic residuals) in patients who underwent surgery on the primary tumor. A complete resection at diagnosis is considered to be a major prognostic factor, since ACTs are localized in greater part. Stage was assessed on the basis of COG staging system for ACT proposed by Sandrini et al15 and later mod- ified by Michalkiewicz et al2 and Ribeiro et al,1 as it represents the most currently used worldwide: studies in which stage assessment was performed with different systems were not included into this analysis.
The outcome analyzed in this systematic review was the reported patient’s status (alive or dead).
2.4 Statistical analysis
Overall, we evaluated separately the association between eight clinical prognostic factors and the outcome for a total of 17 separated meta- analyses.
Data were analyzed by using Review Manager (RevMan, Version 5.3, Cochrane Collaboration). For categorical variables risk ratio (RR) with 95% confidence intervals (CIs) were generated. We performed descriptive statistics, and calculated the pooled RR and 95% CI for each prognostic factor that was studied in more than one different study. The RRs were pooled using the fixed-effects model. The hetero- geneity was assessed by the chi-square test and quantified using the /2 statistic. The I 2 statistic represents the percentage of between-study variation that is due to heterogeneity rather than chance: a value of 0% indicates no observed heterogeneity, whereas I 2 values of ≥50% indicate a substantial level of heterogeneity. A fixed-effects model was used if substantial statistical heterogeneity was not present. Publication bias was examined by funnel plot analysis. A P-value of less than .05 was considered statistically significant.
2226 papers found
Scopus: 1219
PubMed: 376
Web of Science: 631
Manual search: 7
Total (minus overlap): 1895
Rejected on Tital/Abstract: 1620
· case reports
· review of the literature
· age > 18 years
· clinical data/outcome not available
Full Text found: 275
Rejected on Full Text: 233
· case reports
· review of the literature
· age > 18 years
· clinical data/outcome not available
· mixed population
· raw data not available
· overlapping series
Accepted studies: 42 (1006 patients)
3 RESULTS
A total of 2233 articles were identified and 275 were assessed with respect to their eligibility for inclusion. Forty-two studies were included in the systematic review (Figure 1), accounting for a total of 1006 pediatric patients with diagnosis of an ACT. The general charac- teristics of the studies included in the systematic review are reported in Table 1.
Quality assessment of the included studies was performed using Newcastle-Ottawa Scale for cohort studies. Almost all the included studies showed an overall good rate with regard to the selection and comparability of the study groups and to the ascertainment outcome of interest (Supporting Information S2). The major weaknesses of these studies were represented by their retrospective design, a lack of homo- geneity of the clinical features evaluated, and the different thresholds adopted to define clinical features (as age, tumor volume, weight, max- imum diameter, and stage).
3.1 Age
Thirty-three papers entered this meta-analysis.2,5,16-45 An age of more than 4 years was found to have RR of 3.07 (P < . 00001; 95% CI 2.35- 4.02; I 2 0%) (Figure 2).
3.2 | S Sex
Thirty-six papers2,5,16-18,20-47,49,50 allowed the evaluation of gender effect on outcome: females were found to have RR of 1.21, without reaching a significance (P = . 14; 95% CI 0.94-1.55; /2 0%).
3.3 Tumor secreting status
Three different meta-analyses were performed in order to deter- mine the effect of secretion pattern on outcome. Firstly, the patients affected by Cushing syndrome versus those without Cushing syndrome were compared (Figure 3A). Twenty-seven studies2,5,16-19,23,24,26,29-34,36-41,43,45,46,48,51 assessed the outcomes between these two groups, and being affected by Cushing syndrome resulted in a worse outcome (P < . 0001; RR 1.88; 95% CI 1.38-2.58; [2 6%). Secondly, 19 studies5,18,20,26,30-33,36-38,40,41,43,45,46,49,51 com- pared the secreting tumors and nonsecreting masses, identifying hormonal secretion as a negative prognostic factor (P = . 004; RR 1.63; 95% CI 1.17-2.27; I 2 0%). Finally, the comparison between patients affected by Cushing syndrome and nonsecreting tumors was performed in 17 studies5,18,20,26,30-33,37,38,40,41,43,45,46,49,51 and no difference in outcome was found (P = . 40; RR 1.19; 95% CI 0.79-1.79; I 2 0%) (Figure 3B).
3.4 Tumor volume
Five papers reported tumor volume.5,18,20,30,31 The meta-analysis (Supporting Information S3) showed a worse outcome for the tumors more than 200 cm3 of volume (P <. 0001; RR 3.37; 95% CI 1.93-5.88; /2 60%).
3.5 Tumor weight
Tumor weight was reported in 15 studies. Two meta-analyses were performed for the two different cutoffs. Either tumors weighing more than 400 g17-21,27,31,33,35,37-40,42,52 or tumors weighing more than 100 g17,18,20,21,31,33-35,37-40,42,45,52 were considered as risk factor for a worse outcome (respectively, P <. 00001; RR 4.29; 95% CI 3.19-5.78; I 2 0% and P < .00001; RR 5.46; 95% CI 3.24-9.20; I 2 0%) (Figure 4A,B).
3.6 Tumor maximum diameter
Three different meta-analyses were performed, considering tumor maximum diameter as a risk factor. The different groups were com- pared as follows. First, tumors <5 cm versus tumors >5 and <10 cm were included in 11 studies,5,25,29,32-35,37,42,45,52 showing a worse outcome for the tumors >5 and <10 cm (P = . 008; RR 2.85; 95% CI 1.31-6.17; I 2 0%). Second, tumors <5 cm versus tumors >5 cm were included in 15 studies,5,19,22,23,25,28,29,32-35,37,42,45,52 resulting
| Author, year (location) | Study design | Sample size | Period | M;F | Mean age at diagnosis (years) | Number of deaths | Mean follow-up years (range) |
|---|---|---|---|---|---|---|---|
| Federici, 1994 (Italy) | Cohort | 12 | 1976-1989 | 5;7 | 5.0 | 2 | 6.4 (1.5-14) |
| Damiani, 1995 (Brazil) | Cohort | 33 | 1975-1993 | 11;22 | 3.7 | 17 | 6.7 (0-20) |
| Bergadà, 1996 (Argentina) | Cohort | 20 | 1970-1991 | 5;15 | 7.1 | 2 | 10 (0-23) |
| Mendonca, 1996 (Brazil) | Cohort | 18 | 1980-1992 | 5;13 | 3.1 | 2 | 5.1 (0.5-9.5) |
| Michalkiewicz, 1997 (USA, Brazil, Canada, Uruguay, Norway, Chile) | Cohort | 20 | 1988-1994 | 18;1 | 2.0 | 1 | 2.5 (0.1-6.2) |
| Mayer, 1997 (Canada) | Cohort | 11 | 1972-1996 | 3;8 | 7.0 | 1 | 5.6 (0.8-15) |
| Driver, 1998 (UK) | Cohort | 18 | 1954-1995 | 3;15 | 5.6 | 12 | 4.3 (0.1-26) |
| Teinturier, 1999 (France) | Cohort | 54 | 1973-1993 | 27;27 | 4.0 | 28 | 3.2 (0.1-20) |
| Wolthers, 1999 (UK) | Cohort | 30 | 1976-1996 | 7;23 | 4.9 | 4 | 3.8 (0.4-75) |
| Wilkin, 2000 (Canada) | Cohort | 10 | - | 2;8 | 5.7 | 1 | 7.5 (0.5-18) |
| Mishra, 2001 (India) | Cohort | 10 | 1990-1999 | 4;6 | 7.4 | 5 | 2.2 (0-7.6) |
| Misu, 2001 (Japan) | Cohort | 4 | - | 3;1 | 2.5 | 1 | 5.9 (3.0-12) |
| Ciftici, 2001 (Turkey) | Cohort | 30 | 1970-1999 | 11;19 | 6.7 | 11 | 2.5 |
| Latronico, 2001 (Brazil) | Cohort | 18 | - | 5;13 | 2.2 | 1 | 5.3 (0.2-13) |
| McDonnell, 2003 (Australia) | Cohort | 12 | 1976-2001 | 6;6 | 2.5 | 3 | 10 (1.0-25.8) |
| Narasismhan, 2003 (India) | Cohort | 8 | 1989-2000 | 2;6 | 2.5 | 2 | 2.3 (0.5-5.0) |
| Sandrini, 2005 (Brazil) | Cohort | 21 | - | 7;14 | 3.7 | 4 | 8.6 (0.3-19) |
| Figuereido, 2005 (Brazil) | Cohort | 9 | - | 3;6 | 2.6 | 2 | - |
| Barbosa, 2004 (Brazil) | Cohort | 7 | - | 4;3 | 5.1 | 1 | 5.8 (1.2-8.2) |
| Pinto, 2005 (Brazil) | Cohort | 16 | - | 6;10 | 4.4 | 1 | 4.4 (1.0-10) |
| Rosati, 2008 (Brazil) | Cohort | 12 | - | 3;9 | 3.6 | 5 | 6.6 (2.4-15.8) |
| Loncarevic, 2008 (Germany) | Cohort | 14 | 1998-2007 | 6;8 | 3.8 | 6 | 4.4 (4.3-9.8) |
| Lorea, 2012 (Brazil) | Cohort | 60 | 1991-2009 | 15;45 | 3.4 | 10 | 5.7 (0.7-14) |
| Waldmann, 2012 (Germany) | Cohort | 2 | - | 1;1 | 16 | 1 | 4.2 (1.9-6.5) |
| Fragoso, 2012 (Brazil) | Cohort | 24 | 1990-2010 | 7;17 | 3.8 | 5 | 6.7 (0.6-14) |
| Custodio, 2013 (Brazil) | Cohort | 19 | 2005-2010 | 4;15 | 1.5 | 4 | 3.8 (2.0-5.8) |
| Borges, 2013 (Brazil) | Cohort | 57 | 1991-2009 | 14;43 | 3.4 | 10 | 5.2 (0-14) |
| Wendt, 2014 (USA) | Cohort | 5 | 1975-2011 | 3;2 | 3.1 | 3 | 7.6 (0.5-32) |
| Nazli Gonç, 2014 (Turkey) | Cohort | 18 | 1999-2013 | 8;10 | 5.7 | 6 | 6.0 (1.0-11) |
| Kerkhofs, 2014 (The Netherlands) | Cohort | 12 | 1989-2013 | 3;9 | 4.1 | 5 | 6.4 (0.1-14.8) |
| Dall'Igna, 2014 (Italy) | Cohort | 58 | 1982-2011 | 19;39 | 5.1 | 12 | 5.9 (0.2-9.9) |
| Sakoda, 2014 (UK) | Cohort | 29 | 1987-2011 | 14;15 | 2.2 | 10 | 2.1 (0.1-15) |
| Chatterjee, 2015 (India) | Cohort | 13 | 2005-2014 | 4;9 | 2.9 | 3 | 2.2 |
| Das, 2016 (India) | Cohort | 14 | 2005-2015 | 5;9 | 2.1 | 3 | 4.1 (0.5-5) |
| Bulzico, 2016 (Brazil) | Cohort | 27 | 1997-2015 | 8;19 | 5.1 | 10 | 2.3 (0.1-12.8) |
| Pinto, 2017 (USA) | Cohort | 59 | 2006-2013 | 17;42 | 3.3 | 14 | 4.0 (0-23) |
| Picard, 2018 (France) | Cohort | 95 | 2000-2018 | 28;67 | 5.0 | 16 | 5.3 (0.2-17) |
| Jehangir, 2019 (India, Australia) | Cohort | 22 | 2006-2016 | 12;10 | 6.3 | 2 | 6.0 (2.1-6.7) |
| Monteiro, 2019 (Brazil) | Cohort | 12 | 2004-2015 | 2;10 | 2.0 | 1 | 4.5 (0.4-11) |
| Parise, 2019 (Brazil) | Cohort | 48 | - | 18;30 | 2.8 | 15 | 7.0 (0.2-21) |
| Zekri, 2020 (Egypt) | Cohort | 18 | 2007-2016 | 6;12 | 4.0 | 8 | 3.2 |
| Pinto, 2020 (USA) | Cohort | 11 | - | 3;8 | 4.5 | 3 | - |
| Study or Subgroup | log[Risk Ratio] | SE | Weight | Risk Ratio IV, Fixed, 95% CI | Risk Ratio IV, Fixed, 95% CI | |
|---|---|---|---|---|---|---|
| Barbosa 2004 | 1.3218 | 1.4888 | 0.8% | 3.75 [0.20, 69.39] | ||
| Bergada 1996 | -0.4055 | 1.3385 | 1.0% | 0.67 [0.05, 9.19] | ||
| Borges 2013 | 1.6247 | 0.563 | 5.9% | 5.08 [1.68, 15.30] | ||
| Chatterjee 2015 | 1.1939 | 0.6898 | 3.9% | 3.30 [0.85, 12.75] | ||
| Custodio 2013 | 1.404 | 0.6323 | 4.7% | 4.07 [1.18, 14.06] | ||
| Dall'Igna 2014 | 1.4018 | 0.7282 | 3.5% | 4.06 [0.97, 16.93] | ||
| Damiani 1995 | 1.0986 | 0.3761 | 13.3% | 3.00 [1.44, 6.27] | ||
| Das 2016 | 0 | 1.3093 | 1.1% | 1.00 [0.08, 13.02] | ||
| Federici 1994 | 2.1972 | 1.3916 | 1.0% | 9.00 [0.59, 137.65] | ||
| Figuereido 2004 | -0.539 | 0.7237 | 3.6% | 0.58 [0.14, 2.41] | ||
| Fragoso 2012 | 1.8971 | 1.0355 | 1.8% | 6.67 [0.88, 50.74] | ||
| Gonc 2014 | 0.6931 | 0.7265 | 3.6% | 2.00 [0.48, 8.31] | ||
| Jehangir 2019 | -0.3677 | 1.3461 | 1.0% | 0.69 [0.05, 9.69] | ||
| Kerkhofs 2014 | 2.6856 | 1.3748 | 1.0% | 14.67 [0.99, 217.06] | ||
| Latronico 2001 | 3.2958 | 1.453 | 0.9% | 27.00 [1.57, 465.74] | ||
| Lonarevic 2008 | 0.2877 | 0.6124 | 5.0% | 1.33 [0.40, 4.43] | ||
| Lorea 2012 | 1.6247 | 0.563 | 5.9% | 5.08 [1.68, 15.30] | ||
| Mayer 1997 | -0.6286 | 1.5303 | 0.8% | 0.53 [0.03, 10.71] | ||
| McDonnell 2003 | 1.0296 | 1.0757 | 1.6% | 2.80 [0.34, 23.06] | ||
| Mendonca 1995 | 1.2528 | 1.2956 | 1.1% | 3.50 [0.28, 44.35] | ||
| Michalkiewicz 1997 | 1.204 | 1.4549 | 0.9% | 3.33 [0.19, 57.72] | ||
| Mishra 2001 | 2.0614 | 1.3561 | 1.0% | 7.86 [0.55, 112.09] | ||
| Monteiro 2019 | 1.6864 | 1.5348 | 0.8% | 5.40 [0.27, 109.35] | ||
| Narasismhan 2003 | 1.0986 | 1.1547 | 1.4% | 3.00 [0.31, 28.84] | ||
| Parise 2019 | 1.2321 | 0.3959 | 12.0% | 3.43 [1.58, 7.45] | ||
| Pinto 2005 | 2.0541 | 1.5459 | 0.8% | 7.80 [0.38, 161.42] | ||
| Pinto 2017 | 3.5316 | 1.4153 | 0.9% | 34.18 [2.13, 547.60] | ||
| Rosati 2008 | 1.1756 | 0.5578 | 6.0% | 3.24 [1.09, 9.67] | ||
| Sakoda 2014 | 2.0281 | 0.6873 | 4.0% | 7.60 [1.98, 29.23] | ||
| Sandrini 2004 | 2.6391 | 1.4282 | 0.9% | 14.00 [0.85, 230.06] | ||
| Wendt 2014 | 0.2877 | 0.8165 | 2.8% | 1.33 [0.27,6.61] | ||
| Wilkin 2000 | -1.0986 | 1.5275 | 0.8% | 0.33 [0.02, 6.65] | ||
| Zekri 2019 | 0.5108 | 0.5578 | 6.0% | 1.67 [0.56, 4.97] | ||
| Total (95% CI) | 100.0% | 3.07 [2.35, 4.02] | ||||
| Heterogeneity: Chiz= 29.91, df = 32 (P = 0.57); 12= 0% Test for overall effect: Z = 8.20 (P < 0.00001) | 0.002 | 0.1 1 10 | ||||
| Favours | [> 4 years] Favours [< 4 years] | |||||
500
| a | Weight | Risk Ratio IV, Fixed, 95% CI | Risk Ratio IV, Fixed, 95% CI | ||
|---|---|---|---|---|---|
| Study or Subgroup | log[Risk Ratio) | SE | |||
| Barbosa 2004 | -0.6061 | 0.5903 | 7.4% | 0.55 [0.17. 1.73] | |
| Bergada 1996 | 1.4271 | 1.4888 | 1.2% | 4.17 [0.23. 77.10] | |
| Borges 2013 | 1.0986 | 0.5774 | 7.7% | 3.00 [0.97. 9.30] | |
| Ciftici 2001 | 0.3677 | 0.6504 | 6.1% | 1.44 [0.40, 5.17] | |
| Custodio 2013 | -0.5596 | 1.3671 | 1.4% | 0.57 [0.04. 8.33] | |
| Dalligna 2014 | 1.3218 | 0.6357 | 6.4% | 3.75 [1.08. 13.04] | |
| Federici 1994 | 0.6931 | 1.2748 | 1.6% | 2.00 [0.16. 24.33] | |
| Figuereido 2004 | 0.2231 | 1.245 | 1.7% | 1.25 [0.11, 14.34] | |
| Gonc 2014 | -0.4055 | 0.7638 | 4.4% | 0.67 [0.15, 2.98] | |
| Latronico 2001 | 1.0986 | 1.5706 | 1.0% | 3.00 [0.14. 65.16] | |
| Lonarevic 2008 | -1.5041 | 1.3591 | 1.4% | 0.22 [0.02. 3.19] | |
| Lorea 2012 | 1.6479 | 0.4068 | 15.6% | 5.20 [2.34. 11.53] | |
| Mayer 1997 | 1.6582 | 1.5079 | 1.1% | 5.25 [0.27, 100.85] | |
| McDonnell 2003 | -0.4055 | 1.472 | 1.2% | 0.67 [0.04. 11.94] | |
| Mendonca 1995 | 2.4567 | 1.4703 | 1.2% | 11.67 [0.65, 208.19] | |
| Michalkiewicz 1997 | 1.6582 | 1.5717 | 1.0% | 5.25 [0.24, 114.28] | |
| Misu 2001 | 1.0986 | 1.4142 | 1.3% | 3.00 [0.19. 47.96] | |
| Monteiro 2019 | 0.6286 | 1.5303 | 1.1% | 0.53 [0.03. 10.71] | |
| Narasismhan 2003 | -0.2877 | 1.1902 | 1.8% | 0.75 [0.07. 7.73] | |
| Parise 2019 | .0.2346 | 0.5549 | 8.4% | 0.79 [0.27. 2.35] | |
| Pinto 2005 | 0.4055 | 1.5546 | 1.1% | 1.50 [0.07. 31.58] | |
| Pinto 2017 | 0.821 | 0.6439 | 6.2% | 2.27 [0.64. 8.03] | |
| Pinto 2020 | 1.3863 | 1.1726 | 1.9% | 4.00 [0.40. 39.83] | |
| Rosati 2008 | 0.1823 | 0.5477 | 8.6% | 1.20 [0.41, 3.51] | |
| Sandrini 2004 | 0.3365 | 1.1084 | 2.1% | 1.40 (0.16. 12.29] | |
| Wilkin 2000 | 2.0794 | 1.486 | 1.2% | 8.00 [0.43, 147.22] | |
| Zekri 2019 | 1.0498 | 0.6574 | 6.0% | 2.86 [0.79. 10.36] | |
| Total (95% CI) | 100.0% | 1.88 [1.38, 2.58] | |||
| Heterogeneity: Chi= 27.56, df= 26 (P = 0.38); P | = 6% | 0.005 | 0.1 1 10 | ||
| Test for overall effect | Z = 3.95 (P < 0.0001) | Favours [Cushing] Favours [No Cushing] | |||
| b | log[Risk Ratio] | SE | Weight | Risk Ratio IV, Fixed, 95% CI | Risk Ratio IV, Fixed, 95% CI |
|---|---|---|---|---|---|
| Study or Subgroup | |||||
| Borges 2013 | -0.4235 | 1.3517 | 1.6% | 0.65 [0.05. 9.26] | |
| Ciftici 2001 | 1.0498 | 0.391 | 18.7% | 2.86 [1.33. 6.15] | |
| Custodio 2013 | 0.08 | 1.0032 | 2.8% | 1.08 [0.15. 7.74] | |
| Dalligna 2014 | 0.3321 | 0.576 | 8.6% | 1.39 [0.45, 4.31] | |
| Damiani 1995 | 0.2911 | 0.4364 | 15.0% | 1.34 [0.57. 3.15] | |
| Gonc 2014 | 0.47 | 0.7984 | 4.5% | 1.60 [0.33. 7.65] | |
| Lonarevic 2008 | 0.2231 | 0.8062 | 4.4% | 1.25 [0.26, 6.07] | |
| Lorea 2012 | -0.4235 | 1.3517 | 1.6% | 0.65 [0.05. 9.26] | |
| Mayer 1997 | 0.1054 | 1.4944 | 1.3% | 1.11 [0.06. 20.79] | |
| McDonnell 2003 | 1.4351 | 0.9952 | 2.9% | 4.20 [0.60, 29.54] | |
| Monteiro 2019 | 0.2877 | 1.5 | 1.3% | 1.33 [0.07. 25.22] | |
| Narasismhan 2003 | -0.2231 | 1.3323 | 1.6% | 0.80 [0.06. 10.89] | |
| Parise 2019 | -0.2624 | 0.6534 | 6.7% | 0.77 [0.21. 2.77] | |
| Pinto 2017 | 0.5543 | 0.5065 | 11.2% | 1.74 [0.65. 4.70] | |
| Pinto 2020 | 1.1939 | 0.6898 | 6.0% | 3.30 [0.85. 12.75] | |
| Rosati 2008 | -0.7732 | 1.2532 | 1.8% | 0.46 [0.04. 5.38] | |
| Sandrini 2004 | 1.1527 | 0.8836 | 3.7% | 3.17 [0.56, 17.90] | |
| Wilkin 2000 | 0.5108 | 1.4376 | 1.4% | 1.67 [0.10. 27.89] | |
| Zekri 2019 | 0.069 | 0.7591 | 5.0% | 1.07 [0.24. 4.74] | |
| Total (95% CI) | 100.0% | 1.63 [1.17, 2.27] | |||
| Heterogeneity: Chiª = 9.07, df = 18 (P = 0.96); P= Test for overall effect: Z = 2.88 (P = 0.004) | 0% | 0.05 | 0.2 1 5 20 | ||
| Favours | [non secreting] Favours [secreting] | ||||
200
FIGURE 3 Forest plots showing risk ratio (RR) (A) according to Cushing’s syndrome (Cushing vs non-Cushing); and (B) secretion status (secreting vs nonsecreting tumors), for each study and pooled for all studies. Only first author of each study is given
| a | SE | Risk Ratio | Risk | Ratio 95% CI | ||
|---|---|---|---|---|---|---|
| Study or Subgroup | log[Risk Ratio] | Weight | IV, Fixed, 95% CI | IV, Fixed, | ||
| Bergada 1996 | 1.3863 | 1.299 | 4.2% | 4.00 [0.31, 51.02] | ||
| Borges 2013 | 2.8637 | 1.0156 | 6.9% | 17.53 [2.39, 128.28] | ||
| Custodio 2013 | 1.8718 | 1.0831 | 6.1% | 6.50 [0.78, 54.31] | ||
| Damiani 1995 | 1.6895 | 0.6743 | 15.6% | 5.42 [1.44, 20.31] | ||
| Lorea 2012 | 2.9444 | 1.0145 | 6.9% | 19.00 [2.60, 138.77] | ||
| McDonnell 2003 | 2.1282 | 1.3853 | 3.7% | 8.40 [0.56, 126.89] | ||
| Mendonca 1995 | 1.6094 | 1.2649 | 4,4% | 5.00 [0.42, 59.65] | ||
| Mishra 2001 | 0.6931 | 0.7071 | 14.2% | 2.00 [0.50. 8.00] | ||
| Narasismhan 2003 | -0.7621 | 1.387 | 3.7% | 0.47 [0.03. 7.07] | ||
| Parise 2019 | 2.0794 | 0.9954 | 7.2% | 8.00 [1.14, 56.28] | ||
| Pinto 2005 | 1.7918 | 1.5546 | 2.9% | 6.00 [0.29. 126.31] | ||
| Pinto 2017 | 3.3839 | 1.4058 | 3.6% | 29.49 [1.87, 463.69] | ||
| Sakoda 2014 | 1.5404 | 0.705 | 14.3% | 4.67 [1.17, 18.58] | ||
| Wilkin 2000 | 0.9163 | 1.5166 | 3.1% | 2.50 [0.13, 48.85] | ||
| Wolthers 1999 | 2.1484 | 1.4744 | 3.3% | 8.57 [0.48, 154.19] | ||
| Total (95% CI) | 100.0% | 5.46 [3.24, 9.20] | ||||
| Heterogeneity: Chiª = 10.17, df= 14 (P = 0.75); P= 0% Test for overall effect: Z = 6.37 (P < 0.00001) | 0.002 | 0.1 1 10 | ||||
| Favours [>100g] | Favours [<100g] | |||||
200
500
| b | log[Risk Ratio] | SE | Weight | Risk Ratio IV, Fixed, 95% CI | Risk Ratio IV. Fixed, 95% CI | |
|---|---|---|---|---|---|---|
| Study or Subgroup | ||||||
| Bergada 1996 | 1.7346 | 1.268 | 1.4% | 5.67 [0.47, 68.02] | ||
| Borges 2013 | 1.5404 | 0.5034 | 9.0% | 4.67 [1.74, 12.52] | ||
| Chatterjee 2015 | 2.1001 | 1.4058 | 1.2% | 8.17 [0.52, 128.43] | ||
| Custodio 2013 | 1.3218 | 0.9661 | 2.5% | 3.75 [0.56, 24.91] | ||
| Damiani 1995 | 1.2958 | 0.3451 | 19.2% | 3.65 [1.86, 7.19] | ||
| Jehangir 2019 | 2.6027 | 1.0138 | 2.2% | 13.50 [1.85, 98.47] | ||
| Lorea 2012 | 1.5999 | 0.5046 | 9.0% | 4.95 [1.84, 13.32] | ||
| McDonnell 2003 | 1.4917 | 0.78 | 3.8% | 4.44 [0.96. 20.50] | ||
| Mishra 2001 | 0.9808 | 0.5845 | 6.7% | 2.67 [0.85, 8.38] | ||
| Narasismhan 2003 | -0.2231 | 1.3323 | 1.3% | 0.80 [0.06. 10.89] | ||
| Parise 2019 | 1.4018 | 0.3643 | 17.2% | 4.06 [1.99, 8.30] | ||
| Pinto 2005 | 2.7081 | 1.5055 | 1.0% | 15.00 [0.78, 286.81] | ||
| Pinto 2017 | 2.0072 | 0.5108 | 8.8% | 7.44 [2.73, 20.25] | ||
| Sakoda 2014 | 1.2528 | 0.3887 | 15.1% | 3.50 [1.63, 7.50] | ||
| Wolthers 1999 | 2.0149 | 1.1972 | 1.6% | 7.50 [0.72, 78.36] | ||
| Total (95% CI) | 100.0% | 4.29 [3.19, 5.78] | ||||
| Heterogeneity: Chi? = 6.50, df= 14 (P = 0.95); P= 0% Test for overall effect: Z = 9.63 (P < 0.00001) | 0.005 | 0.1 1 10 | ||||
| Favours [>400 g] Favours [<400 g] | ||||||
| a | log[Risk Ratio) | SE | Weight | Risk Ratio IV, Fixed, 95% CI | Risk Ratio IV, Fixed, 95% CI | |
|---|---|---|---|---|---|---|
| Study or Subgroup | ||||||
| Chatterjee 2015 | 0.7419 | 1.361 | 5.8% | 2.10 [0.15, 30.25] | ||
| Dalligna 2014 | 2.3885 | 1.4126 | 5.4% | 10.90 [0.68, 173.67] | ||
| Das 2016 | 1.1575 | 1.4124 | 5.4% | 3.18 [0.20, 50.69] | ||
| Federici 1994 | 1.3218 | 1.452 | 5.1% | 3.75 [0.22, 64.56] | ||
| Fragoso 2012 | 2.3979 | 1.4241 | 5.3% | 11.00 [0.67, 179.30] | ||
| Kerkhofs 2014 | 1.0498 | 1.4167 | 5.4% | 2.86 [0.18, 45.90] | ||
| Latronico 2001 | 1.3218 | 1.5626 | 4.4% | 3.75 [0.18, 80.19] | ||
| Mayer 1997 | 0.6286 | 1.5303 | 4.6% | 1.87 [0.09, 37.63] | ||
| McDonnell 2003 | 1.9459 | 1.4142 | 5.4% | 7.00 [0.44, 111.91] | ||
| Mendonca 1995 | 1.9279 | 1.4779 | 5.0% | 6.88 [0.38, 124.53] | ||
| Mishra 2001 | 0.2231 | 0.7416 | 19.7% | 1.25 [0.29, 5.35] | ||
| Narasismhan 2003 | 0.2231 | 1.245 | 7.0% | 1.25 [0.11, 14.34] | ||
| Sakoda 2014 | 1.1632 | 0.9747 | 11.4% | 3.20 [0.47, 21.62] | ||
| Wilkin 2000 | 0.4055 | 1.5138 | 4.7% | 1.50 [0.08. 29.15] | ||
| Wolthers 1999 | 1.9741 | 1.4395 | 5.2% | 7.20 [0.43, 120.96] | ||
| Total (95% CI) | 100.0% | 2.98 [1.56, 5.67] | ||||
| Heterogeneity: ChP = 5.02, df= 14 (P = 0.99); P= 0% | 0.005 | 0.1 1 10 | ||||
| Test for overall effect: Z= 3.32 (P = 0.0009) | Favours [>5 cm] | Favours [<5 cm] | ||||
200
200
| Risk Ratio Risk Ratio IV, Fixed, 95% CI IV, Fixed, 95% CI | ||||||
|---|---|---|---|---|---|---|
| b | log[Risk Ratio) | SE | Weight | |||
| Study or Subgroup | ||||||
| >10 cm | ||||||
| Chatterjee 2015 | 2.4159 | 1.4003 | 2.5% | 11.20 [0.72, 174.24] | ||
| Dall'igna 2014 | 1.3863 | 0.4964 | 19.6% | 4.00 [1.51, 10.58] | ||
| Das 2016 | 2.4567 | 1.4209 | 2.4% | 11.67 [0.72, 188.97] | ||
| Federici 1994 | 2.9087 | 1.4056 | 2.4% | 18.33 [1.17, 288.19] | ||
| Fragoso 2012 | 1.7004 | 0.5558 | 15.7% | 5.48 [1.84, 16.28] | ||
| Jehangir 2019 | 1.2238 | 1.3195 | 2.8% | 3.40 [0.26, 45.15] | ||
| Kerkhofs 2014 | 1.7918 | 1.4259 | 2.4% | 6.00 [0.37, 98.15] | ||
| Mayer 1997 | 0.1054 | 1.4944 | 2.2% | 1.11 [0.06, 20.79] | ||
| McDonnell 2003 | 1.8101 | 0.8015 | 7.5% | 6.11 [1.27, 29.40] | ||
| Mendonca 1995 | 2.0149 | 1.1972 | 3.4% | 7.50 [0.72, 78.36] | ||
| Mishra 2001 | 0.1542 | 0.8309 | 7.0% | 1.17 [0.23. 5.95] | ||
| Narasismhan 2003 | -0.1054 | 1.3375 | 2.7% | 0.90 [0.07, 12.38] | ||
| Sakoda 2014 | 1.0761 | 0.4877 | 20.3% | 2.93 [1.13. 7.63] | ||
| Wilkin 2000 | 0 | 1.4907 | 2.2% | 1.00 [0.05, 18.57] | ||
| Wolthers 1999 | 1.6582 | 0.8381 | 6.9% | 5.25 [1.02, 27.14] | ||
| Total (95% CI) | 100.0% | 3.87 [2.51, 5.96] | ||||
| Heterogeneity: ChP = 8.78, df = 14 (P = 0.84); P= | 0% | 0.005 | 0.1 1 10 | |||
| Test for overall effect: Z = 6.15 (P < | 0.00001) | Favours [<10cm] | Favours [>10cm] | |||
in a worse outcome for those >5 cm (P = . 0009; RR 2.98; 95% CI 1.56-5.67; I 2 0%) (Figure 5A). Finally, the last meta-analysis among 15 studies5,19,22,23,25,28,29,32-35,37,42,45,52 showed that tumors larger than 10 cm had worse outcome (P < .00001; RR 3.87; 95% CI 2.51-5.96; I 2 0%) (Figure 5B).
3.7 Surgery results
A meta-analysis (Supporting Information S3) including 13 studies5,20,23,36,27,32,33,37,38,42,51,53,54 compared the outcome between patients undergone complete resection and those undergone incom- plete resection. The first ones showed a better outcome (P < . 00001; RR 3.50; 95% CI 2.51-4.88; /2 16%).
3.8 Stage
Five different meta-analyses assessed the impact on the outcome of the different stage of disease at diagnosis (Supporting Information S4). The stages were compared as follows. Stage I versus Stage II, III, and IV included eight studies.7,18,20,31,36,38,40,46 The meta-analysis showed a better outcome for Stage I (P < . 00001; RR 6.94; 95% CI 3.35- 14.4; I 2 12%). Then, Stage I versus Stage II and III included seven studies.7,18,20,31,38,40,46 This resulted in a better outcome for patients with Stage I disease (P < .00001; RR 4.69; 95% CI 2.11-10.4; I 2 28%). Stage I versus Stage II, including six studies,7,18,20,31,38,40 showed a bet- ter outcome once again for Stage I (P <. 002; RR 5.02; 95% CI 1.82-13.9;
I 2 53%). Stage II was compared to Stage III in six studies,18,20,31,38,40,46 and no difference was found (P < .12; RR 1.93; 95% CI 0.84-4.47; I 2 39%). Lower stages (Stage I and II) versus advanced disease (Stage III and IV) showed RR of 4.99 for advanced tumors (95% CI 3.57-6.98; P <. 00001; /2 0%).7,18,20,31,36,38,40,46
4 DISCUSSION
4.1 Main findings
The findings from this systematic review partially confirmed the previ- ous reports. Age <4 years, R0 resection, tumor diameter <5 cm, tumor volume <200 cm3, tumor weight <100 g and <400 g, and lower stage at diagnosis strongly correlated with good outcome. An association with worse outcome was also found when patients presenting with Cushing syndrome were compared with patients without Cushingoid features (secreting and nonsecreting together), but not when com- pared with nonsecreting tumors only. Secreting tumors showed a slight increase in the RR when compared with nonsecreting tumors. Lower stages are associated with a better outcome independently from the inclusion of metastatic patients into the analysis.
4.2 Limitations
Limitations and bias derived from the features of the studies are included in this review. The main weaknesses of these series were
represented by their retrospective design, small sample size, dif- ferent thresholds or classifications used (such as to describe vol- ume and weight, or stage), and by the fact that most of the stud- ies did not explore all the clinical features taken into account by this systematic review. The variability in the thresholds limited the evaluation of some clinical features, as it lowered the number of patients who entered the single analysis, decreasing the rate of events. In this scenario, it is plausible that the relationship between a given clinical feature and the outcome may have been under- or overestimated.
4.3 Implications for clinical practice
The current pathological scores may not be sufficient to determine the malignant behavior of localized ACTs, especially in case of a Stage II-III disease. The Wieneke index seems to be the most reliable score in pedi- atric age so far,5,6,27 but because it addresses a group of tumors with indeterminate malignancy, it still leaves partially unsolved the deci- sion on which may be the best postoperative treatment (chemotherapy associated to mitotane or mitotane alone) for this subset of patients.
In the past, various clinical features have been considered to be associated with poor outcome, but they varied depending on the study: age >4 years, size, Cushing syndrome, incomplete resections, and stage have been more commonly reported. Additionally, initial biopsy is con- sidered to be an unfavorable prognostic factor and it should be avoided as much as possible.2,8,47,55,56 Moreover, few attempts have been made to overcome this variability of evidence before the present study, but they retrospectively took into account a limited number of patients and without reaching a wider consensus on a possible risk stratification. A paper published by the European group (EXPERT) tried to define a subset of “high-risk” tumors based on the presence of one of the fol- lowing features: age at diagnosis, volume more than 200 cm3, Cushing syndrome, initial biopsy (open or tru-cut), surgical excision with resid- uals or spillage, regional lymph node involvement, histologic vascular invasion, and distant metastases at diagnosis.8 An analysis performed on 111 patients of the American National Cancer Data Base suggested including in the risk stratification age, extension into adjacent organs, metastases, and incomplete resection.55
Identifying some clinical features, in order to stratify patients who are at major risk of relapse, could lead to identifying a subgroup of ACT which may be elected to a systemic treatment after surgery, but also in which a mini-invasive approach at diagnosis should be avoided.
The use of adjuvant mitotane for 24-36 months in some Stage II and III patients, alone or in combination with standard chemotherapy, has been found to be determinant56-59 to improve overall survival and event-free survival. In these cases, a distinction between benign and malignant forms may in turn avoid important and unnecessary side effects, or represent a possible improvement in terms of survival.
Indeed, the use of a mini-invasive approach (laparoscopic or robotic- assisted) in ACT is also the object of debate: some authors suggested that it could be associated with a higher rate of incomplete resection
or spillage60 and discouraged this technique.61,62 The definition of pre- operative clinical risk factors (eg, age >4 years, Cushing syndrome, or nonsecreting tumor), aside from size and suspicious nodes or exten- sion into adjacent organ at imaging, could drive the surgeon’s decision in order to avoid unnecessary hazards with potentially fatal outcomes.
4.4 Implications for research
Despite the presence of national registries and large studies, and a considerable number of ACT patients reported in the literature, the prognostic value to be attributed to a given clinical feature is currently extremely variable. For this reason, international cooperative studies or registries prospectively aimed at defining a risk stratification for pediatric ACT, ideally based both on clinical and histological features, are needed. These studies should consider prospectively the predictive accuracy of the different features, alone and in combination. The opti- mal cutoffs of quantitative variables, such as size and volume, should be investigated in order to provide a precise estimation of the pre- and postsurgical risk of localized ACT. Furthermore, such studies should provide standardization for qualitative variables in order to homoge- nize data collected in different countries (eg, staging or tumor scores).
5 CONCLUSIONS
As expected, localized small tumors did show a better prognosis: a low stage, size <10 cm, and weight under 100 g seem to have lesser risk of death. Some clinical factors (age >4 years, Cushing syndrome) corre- lated with a worse outcome and, despite the biological reasons being far from clear, they could be considered when stratifying patients or before planning the surgical approach at diagnosis. Notwithstanding its limitations, this systematic review strengthened in part the evidence coming from the previous literature, and it could be considered a step forward for risk stratification for pediatric ACT.
ACKNOWLEDGMENTS
We thank Carlos Alberto Scrideli (Pediatrics Department, Ribeira õ Preto Medical School, University of Sao Paulo, Brazil) for his contribu- tions to this systematic review in terms of additional explanations on their published data and unpublished data supplied.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author.
ORCID
Elisa Zambaiti (D https://orcid.org/0000-0002-9610-1708 Patrizia Dall’Igna D https://orcid.org/0000-0002-3822-3272 Calogero Virgone (D https://orcid.org/0000-0002-3651-9416
REFERENCES
1. Ribeiro RC, Pinto EM, Zambetti GP, Rodriguez-Galindo C. The Inter- national Pediatric Adrenocortical Tumor Registry initiative: contri- butions to clinical, biological, and treatment advances in pediatric adrenocortical tumors. Mol Cell Endocrinol. 2012;351:37-43.
2. Michalkiewicz E, Sandrini R, Figueiredo B, et al. Clinical and outcome characteristics of children with adrenocortical tumors: a report from the International Pediatric Adrenocortical Tumor Registry. J Clin Oncol. 2004;22:838-845.
3. Dehner LP. Pediatric adrenocortical neoplasms: on the road to some clarity. Am J Surg Pathol. 2003;27(7):1005-1007.
4. Dehner LP, Hill DA. Adrenal cortical neoplasms in children: why so many carcinomas and yet so many survivors. Pediatr Dev Pathol. 2009;12:284-291.
5. Dall’Igna P, Virgone C, De Salvo GL, et al. Adrenocortical tumors in Italian children: analysis of clinical characteristics and P53 sta- tus. Data from the national registries. J Pediatr Surg. 2014;49:1367- 1371.
6. Wieneke JA, Thompson LD, Heffess CS. Adrenal cortical neoplasms in the pediatric population: a clinicopathologic and immunophenotypic analysis of 83 patients. Am J Surg Pathol. 2003;27:867-881.
7. Picard C, Orbach D, Carton M, et al. Revisiting the role of the patholog- ical grading in pediatric adrenal cortical tumors: results from a national cohort study with pathological review. Mod Pathol. 2019;32:546-559.
8. Cecchetto G, Ganarin A, Bien E, et al. Outcome and prognostic fac- tors in high-risk childhood adrenocortical carcinomas: a report from the European Cooperative Study Group on Pediatric Rare Tumors (EXPERT). Pediatr Blood Cancer. 2017;64(6):e26368.
9. Gulack BC, Rialon KL, Englum BR, et al. Factors associated with sur- vival in pediatric adrenocortical carcinoma: an analysis of the National Cancer Data Base (NCDB). J Pediatr Surg. 2016;51:172-177.
10. Henderson LK, Craig JC, Willis NS, Tovey D, Webster AC. How to write a Cochrane systematic review. Nephrology (Carlton). 2010;15(6):617- 624.
11. NHS Centre for Reviews and Dissemination. Systematic Reviews. CRD’s Guidance for Undertaking Reviews in Health Care. York, UK: Centre for Reviews and Dissemination, University of York; 2009
12. Leeflang MM, Deeks JJ, Gatsonis C, Bossuyt PM. Systematic reviews of diagnostic test accuracy. Ann Intern Med. 2008;149(12):889-897.
13. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that eval- uate health care interventions: explanation and elaboration. PLoS Med. 2009;6(7):e1000100.
14. Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for Assessing the Quality if Nonrandomized Studies in Meta-Analyses. Ottawa, Canada: Ottawa Health Research Institute; 1999. http: //www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed June 28, 2020.
15. Sandrini R, Ribeiro RC, DeLacerda L. Childhood adrenocortical tumors. J Clin Endocrinol Metab. 1997;82:2027-2031.
16. Barbosa AS, Giacaglia LR, Martin RM, Mendonca BB, Lin CJ. Assess- ment of the role of transcript for GATA-4 as a marker of unfavor- able outcome in human adrenocortical neoplasms. BMC Endocr Disord. 2004;4(1):3.
17. Bergadá I, Venara M, Maglio S, et al. Functional adrenal cortical tumors in pediatric patients: a clinicopathologic and immunohisto- chemical study of a long term follow-up series. Cancer. 1996;77 (4):771-777.
18. Silva Borges K, Antunes Moreno D. Spindle assembly checkpoint gene expression in childhood adrenocortical tumors (ACT): overexpression of Aurora kinases A and B is associated with a poor prognosis. Pediatr Blood Cancer. 2013;60:1809-1816.
19. Chatterjee G, DasGupta S, Mukherjee G, et al. Usefulness of Wieneke criteria in assessing morphologic characteristics of adrenocortical tumors in children. Pediatr Surg Int. 2015;31(6):563-571.
20. Custodio G, Parise GA, Filho NK, et al. Impact of neonatal screen- ing and surveillance for the TP53 R337H mutation on early detection of childhood adrenocortical tumors. J Clin Oncol. 2012;31(20):2619- 2626.
21. Damiani D, Della Manna T, Aquino LGW, et al. Proliferating cell nuclear antigen immunoreaction in adrenal tumors. Tumori. 1995;81:273-277.
22. Das S, Sengupta M, Islam N, et al. Weineke criteria, Ki-67 index and p53 status to study pediatric adrenocortical tumors: is there a correlation? J Pediatr Surg. 2016;51(11):1795-1800.
23. Federici S, Galli G, Ceccarelli PL, et al. Adrenocortical tumors in chil- dren: a report of 12 cases. Eur J Pediatr Surg. 1994;4(1):21-25.
24. Figueiredo BC, Cavalli LR, Pianovski MA, et al. Amplification of the steroidogenic factor 1 gene in childhood adrenocortical tumors. J Clin Endocrinol Metab. 2005;90(2):615-619.
25. Fragoso MC, Almeida MQ, Mazzuco TL, et al. Combined expression of BUB1B, DLGAP5, and PINK1 as predictors of poor outcome in adreno- cortical tumors: validation in a Brazilian cohort of adult and pediatric patients. Eur J Endocrinol. 2012;166(1):61-67.
26. NazliGönç E, Özön ZA, Çakir MD, Alikasifoglu A, Kandemir N. Need for comprehensive hormonal workup in the management of adrenocorti- cal tumors in children. J Clin Res Pediatr Endocrinol. 2014;6(2):68-73.
27. Jehangir S, Nanjundaiah P, Sigamani E, et al. Pathological prognostica- tion of paediatric adrenocortical tumours: is a gold standard emerging. Pediatr Blood Cancer. 2019;66:e27567.
28. Kerkhofs TM, Ettaieb MH, Verhoeven RH, et al. Adrenocortical carci- noma in children: first population-based clinicopathological study with long-term follow-up. Oncol Rep. 2014;32(6):2836-2844.
29. Latronico AC, Pinto EM, Domenice S, et al. An inherited mutation out- side the highly conserved DNA-binding domain of the p53 tumor sup- pressor protein in children and adults with sporadic adrenocortical tumors. J Clin Endocrinol Metab. 2001;86(10):4970-4973.
30. 0. Loncarevic IF, Hering A, Posorski N, Linden T, Hoyer H, Bucky P. Num- ber of genomic imbalances correlates with the overall survival for adrenocortical cancer in childhood. Pediatr Blood Cancer. 2008;51:356- 362.
31. Lorea CF, Moreno DA, Borges KS, et al. Expression profile of apoptosis- related genes in childhood adrenocortical tumors: low level of expres- sion of BCL2 and TNF genes suggests a poor prognosis. Eur J Endocrinol. 2012;167(2):199-208.
32. Mayer SK, Oligny LL, Deal C, Yazbeck S, Gagné N, Blanchard H. Childhood adrenocortical tumors: case series and reevalua- tion of prognosis-a 24-year experience. J Pediatr Surg. 1997;32 (6):911-915.
33. McDonnell CM, Zacharin MR. Adrenal cortical tumours: 25 years’ experience at the Royal Children’s Hospital, Melbourne. J Paediatr Child Health. 2003;39:682-685.
34. Mendonca BB, Lucon AM, Menezes CA, et al. Clinical, hormonal and pathological findings in a comparative study of adrenocortical neo- plasms in childhood and adulthood. J Urol. 1995;154(6):2004-2009.
35. Mishra A, Agarwal G, Misra AK, Agarwal A, Mishra SK. Functioning adrenal tumours in children and adolescents: an institutional experi- ence. ANZ J Surg. 2001;71:103-107.
36. Monteiro NML, Rodrigues KES, Vidigal PVT, Oliveira BM. Adrenal car- cinoma in children: longitudinal study in Minas Gerais, Brazil. Rev Paul Pediatr. 2019;37(1):20-26.
37. Narasimhan KL, Samujh R, Bhansali A, et al. Adrenocortical tumors in childhood. Pediatr Surg Int. 2003;19:432-435.
38. Parise IZS, Parise GA, Noronha L, et al. The prognostic role of CD8(+) T lymphocytes in childhood adrenocortical carcinomas com- pared to Ki-67, PD-1, PD-L1, and the Weiss score. Cancers (Basel). 2019;11(11):1730.
39. Pinto EM, Billerbeck AEC, Villares Fragoso MCB, Mendonca B, Latron- ico AC. J Clin Endocrinol Metab. 2005;90(5):2976-2981.
40. Pinto EM, Rodriguez-Galindo C, Pounds SB, et al. Identification of clin- ical and biologic correlates associated with outcome in children with
adrenocortical tumors without germline TP53 mutations: a St Jude Adrenocortical Tumor Registry and Children’s Oncology Group study. J Clin Oncol. 2017;35(35):3956-3963.
41. Rosati R, Cerrato F, Doghman M, et al. High frequency of loss of het- erozygosity at 11p15 and IGF2 overexpression are not related to clini- cal outcome in childhood adrenocortical tutors positive for the R337H TP53 mutation. Cancer Genet Cytogenet. 2008;186:19-24.
42. Sakoda A, Mushtaq I, Levitt G, Sebire NJ. Clinical and histopathological features of adrenocortical neoplasms in children: retrospective review from a single specialist center. J Pediatr Surg. 2014;49(3):410-415.
43. Sandrini F, Villani DP, Tucci S, Moreira AC, de Castro M, Elias LL. Inher- itance of R337H p53 gene mutation in children with sporadic adreno- cortical tumor. Horm Metab Res. 2005;37(4):231-235.
44. Wendt S, Shelso J, Wright K, Furman W. Neoplastic causes of abnormal puberty. Pediatr Blood Cancer. 2014;61:664-671.
45. Wilkin F, Gagné N, Paquette J, Oligny LL, Cheri D. Pediatric adrenocor- tical tumors: molecular events leading to insulin-like growth factor II gene overexpression. J Clin Endocrinol Metab. 2000;85(5):2048-2056.
46. Zekri W, Hammad M, Rashed WM, et al. The outcome of childhood adrenocortical carcinoma in Egypt: a model from developing countries. Pediatr Hematol Oncol. 2020;37(3):198-210.
47. Bulzico D, de Faria PA, de Paula MP, et al. Recurrence and mortality prognostic factors in childhood adrenocortical tumors: analysis from the Brazilian National Institute of Cancer experience. Pediatr Hematol Oncol. 2016;33(4):248-258.
48. Misu Y, Jiang S, Yukfumi Y, et al. Clinicopathological features of pedi- atric functional adrenocortical carcinoma diagnosed by Weiss criteria; an analysis of four cases. Clin Pediatr Endocrinol. 2001;10(2):141-146.
49. Pinto EM, Faucz FR, Paza LZ, et al. Germline variants in phosphodi- esterase genes and genetic predisposition to pediatric adrenocortical tumors. Cancers (Basel). 2020;12(2):506.
50. Waldmann J, Patsalis N, Fendrich V, et al. Clinical impact of TP53 alterations in adrenocortical carcinomas. Langenbecks Arch Surg. 2012;397(2):209-216.
51. Ciftci AO, Senocak ME, Tanyel FC, Büyükpamukçu N. Adrenocortical tumors in children. J Pediatr Surg. 2001;36(4):549-554.
52. Wolthers OD, Cameron FJ, Scheimberg I, et al. Androgen secreting adrenocortical tumours. Arch Dis Child. 1999;80(1):46-50.
53. Driver CP, Birch J, Gough DC, Bruce J. Adrenal cortical tumors in child- hood. Pediatr Hematol Oncol. 1998;15(6):527-532.
54. Teinturier C, Pauchard MS, Brugières L, Landais P, Chaussain JL, Bougnères PF. Clinical and prognostic aspects of adrenocortical neo- plasms in childhood. Med Pediatr Oncol. 1999;32(2):106-111.
55. McAteer JP, Huaco JA, Gow KW. Predictors of survival in pedi- atric adrenocortical carcinoma: a Surveillance, Epidemiology, and End Results (SEER) program study. J Pediatr Surg. 2013;48(5): 1025-1031.
56. Picard C, Faure-Conter C, Leblond P, et al. Exploring heterogene- ity of adrenal cortical tumors in children: the French pediatric rare tumor group (Fracture) experience. Pediatr Blood Cancer. 2020;67: e28086.
57. Redlich A, Boxberger N, Strugala D, et al. Systemic treatment of adrenocortical carcinoma in children: data from the German GPOH- MET 97 trial. Klin Padiatr. 2012;224:366-371.
58. Fassnacht M, Terzolo M, Allolio B, et al. Combination chemotherapy in advanced adrenocortical carcinoma. N Engl J Med. 2012;366:2189- 2197.
59. Terzolo M, Angeli A, Fassnacht M, et al. Adjuvant mitotane treat- ment for adrenocortical carcinoma. N Engl J Med. 2007;356:2372- 2380.
60. Miller BS, Gauger PG, Hammer GD, Doherty GM. Resection of adreno- cortical carcinoma is less complete and local recurrence occurs sooner and more often after laparoscopic adrenalectomy than after open adrenalectomy. Surgery. 2012;152(6):1150-1157.
61. Hubertus J, Boxberger N, Redlich A, von Schweinitz D, Vorwerk P. Surgical aspects in the treatment of adrenocortical carcinomas in children: data of the GPOH-MET 97 trial. Klin Padiatr. 2012;224: 143-147.
62. Miller BS. Changes in the evaluation and management of adrenocorti- cal carcinoma. Ann Surg Oncol. 2018;25(12):3413-3415.
SUPPORTING INFORMATION
Additional supporting information may be found online in the Support- ing Information section at the end of the article.
How to cite this article: Zambaiti E, Duci M, De Corti F, et al. Clinical prognostic factors in pediatric adrenocortical tumors: A meta-analysis. Pediatr Blood Cancer. 2020;e28836. https://doi.org/10.1002/pbc.28836