EJE
Assessment of prognostic factors in pediatric adrenocortical tumors: the modified pediatric S-GRAS score in an international multicenter cohort-a work from the ENSAT-PACT working group
Maria Riedmeier, 1,20D Shipra Agarwal,3 Sonir Antonini,4 Tatiana El-Jaick B. Costa,5
Orhan Diclehan,6 Martin Fassnacht,7,8[D Bonald C. Figueiredo,9 Tulay Guran, 10[D Christoph Härtel,1 Imme Haubitz,1 Jan Idkowiak, 11,12,13 Michaela Kuhlen,2,14[D Lucia Noronha, 15 Ivy Zortea S. Parise,5 Antje Redlich,2,16[D Soraya Puglisi, 17[D Ekinci Saniye,6 Paul-Gerhardt Schlegel, 1,2,7 Bilgehan Yalcin,6 and Verena Wiegering 1,2,7,18,*
1Department of Pediatrics, Division of Pediatric Hematology, Oncology and Stem Cell Transplantation, University Hospital Würzburg, University of Wuerzburg, Josef-Schneiderstr. 2, 97080 Wuerzburg, Germany
2KIONET, The Phase I/II Pediatric Oncology Network Bavaria, 91054 Erlangen, Germany
3Department of Pathology, All India Institute of Medical Sciences, New Delhi 110029, India
4Department of Pediatrics, Ribeirao Preto Medical School, University of Sao Paulo, Ribeirao Preto, Sao Paulo 14051-200, Brazill 5Hospital Infantil Joana Gusmão, Department of Pediatrics, 152 Rui Barbosa St., Florianópolis, SC 88025-300, Brazil
6Department of Pediatric Oncology, Hacettepe University Faculty of Medicine, 06100 Ankara, Turkey
7Comprehensive Cancer Centre Mainfranken, University of Wuerzburg Medical Centre, Josef-Schneiderstr. 2, 97080 Wuerzburg, Germany 8Department of Medicine, Division of Endocrinology and Diabetes, University of Wuerzbrug Medical Centre, Josef-Schneiderstr. 2, 97080 Wuerzburg, Germany
9Pelé Pequeno Príncipe Research Institute and Pequeno Príncipe Faculty, Silva Jardim Avenue, Água Verde, Curitiba, PR 80.250-200, Brazil
10Department of Pediatric Endocrinology and Diabetes Istanbul, Marmara University School of Medicine, Istanbul 34722, Turkey
11Institute of Metabolism and Systems Research, College of Medical and Dental Sciences, University of Birmingham, Birmingham B152TT, United Kingdom
12Department of Endocrinology and Diabetes, Birmingham Women’s and Children’s NHS Foundation Trust, Birmingham, Birmingham B46NH, United Kingdom
13Centre of Endocrinology, Diabetes and Metabolism, Birmingham Health Partners, University of Birmingham, Birmingham B15 2TT, United Kingdom
14Pediatrics and Adolescents Medicine, Faculty of Augsburg, University of Augsburg, 86156 Augsburg, Germany
15Serviço de Anatomia Patológica, Hospital de Clínicas, Universidade Federal do Paraná, 181 General Carneiro, Alto da Glória, Curitiba, PR 80060-900, Brazil
16Pediatric Oncology, Otto-von-Guericke-University, 39120 Magdeburg, Germany
17Internal Medicine, Department of Clinical and Biological Sciences, S. Luigi Gonzaga Hospital, University of Turin, Orbassano 10043, Italy 18Mildred Scheel Early Career Center, University Hospital Wuerzburg, 97080 Wuerzburg, Germany
*Corresponding author: Department of Pediatrics, Division of Pediatric Hematology, Oncology and Stem Cell Transplantation, University Hospital Wuerzburg, Josef-Schneiderstr. 2, 97080 Würzburg, Germany. Email: Wiegering_v@ukw.de
Abstract
Objective: Pediatric adrenocortical carcinoma (pACC) is rare, and prognostic stratification remains challenging. We aimed to confirm the prognostic value of the previously published pediatric scoring system (pS-GRAS) in an international multicenter cohort.
Design: Analysis of pS-GRAS items of pACC from 6 countries in collaboration of ENSAT-PACT, GPOH-MET, and IC-PACT.
Methods: We received patient data of the pS-GRAS items including survival information from 9 centers. PS-GRAS score was calculated as a sum of tumor stage (1 = 0; 2-3 = 1; 4 = 2 points), grade (Ki67 index: 0%-9% = 0; 10%-19% = 1; ≥20% = 2 points), resection status (R0 = 0; RX/R1/R2 = 1 point), age (<4 years =0; ≥4 years = 1 point), and hormone production (androgen production = 0; glucocorticoid-/mixed-/no-hormone production = 1 point) generating 8 scores and 4 groups (1: 0-2, 2: 3-4, 3: 5, 4: 6-7). Primary endpoint was overall survival (OS).
Results: We included 268 patients with median age of 4 years. The analysis of the pS-GRAS score showed a significantly favorable prognosis in patients with a lower scoring compared to higher scoring groups (5-year OS: Group 1 98%; group 2 87% [hazard ratio {HR} of death 3.6, 95% CI of HR 1.6-8.2]; group 3 43% [HR of death 2.8, 95% CI 1.9-4.4]; group 4: OS 18% [HR of death 2.1, 95% CI 1.7-2.7]). In the multivariable analysis, age (HR of death 3.5, 95% CI 1.8-7.0), resection status (HR of death 5.5, 95% CI 2.7-11.1), tumor stage (HR of death 1.9, 95% CI of HR 1.2-3.0), and
Ki67 index (HR of death 1.7, 95% CI 1.2-2.4) remained strong independent outcome predictors. Especially infants < 4 years showed more often low-risk constellations with a better OS for all tumor stages.
Conclusion: In an international multicenter study, we confirmed that the pS-GRAS score is strongly associated with overall survival among patients with pACC. Age, resection status, stage, and Ki67 index are important parameters for risk stratification.
Keywords: pediatric adrenocortical cancer, pediatric adrenocortical carcinoma, pediatric adrenocortical tumor, prognostic score, S-GRAS score
Significance
Herein, we present an international, multicenter retrospective study on the prognostic value of the previously published pediatric scoring system (pS-GRAS). We analyzed pS-GRAS data from 268 patients across 9 centers in 6 different countries. Our findings confirm a strong association between pS-GRAS scoring and patient prognosis. The pS-GRAS scoring can be easily assessed shortly after diagnosis to define risk groups for treatment. This is a first step toward using a combined scoring system that includes histopathological, clinical, and molecular data for pACC. The pS-GRAS can now serve as a validated prognostic tool to consult with patients and guide treatment decisions.
Introduction
Pediatric childhood adrenocortical carcinoma (pACC) incidence rate in United States is around 0.3 per million children per year less than 20 years of age.1 The Manchester Children’s Tumor Registry recorded 12 pediatric ACC cases over a 33-year period (1954-1986), with an estimated incidence rate of 0.38 cases per million children per year less than 15 years of age,2 while in France, ACT incidence is 0.2 per million children per year.3 Pediatric ACC is remarkably frequent in southeastern and southern Brazil, where most of the patients carry the germline mutation TP53 R337H.4,5
Pathogenesis of pACC remained widely unclear with excep- tion of a clear association to TP53-related cancer syndromes -such as Li-Fraumeni syndrome-mainly due to a specific in- herited germline mutation at codon 337 in the TP53 suppres- sor gene, especially in young childen.6-8
Even though there is an overlap regarding clinical character- istics and behavior with adult ACCs, especially in infants and younger children, clinical courses differ. Thus, data from the adult setting cannot be uncritically transferred to children.9-13 Recent analyses have revealed age, virilization alone, low pathological tumor score (eg, Wieneke, Weiss, Van Slooten, and Helsinki), and localized tumor stage to be strong outcome predictors.10,12,14-17 In the adult setting, the S-GRAS score has been shown to be superior in terms of prognostic value for disease-free survival and progression-free survival compared with current standard ACC prognostic tools.18-20 Given the distinct features of pediatric ACC,12 we evaluated an adapted S-GRAS score for pACC retrospectively by searching pub- lished pediatric ACC cases21: We demonstrated in a retro- spective heterogeneous patient cohort that the combination of clinical and histopathological criteria has a good predictive value and may therefore serve as a helpful tool for risk strati- fication in future studies. To validate these data, we retro- and prospectively determined pS-GRAS in an international multi- center context. In the current study, we were able to collect a complete dataset for all required pS-GRAS items of 268 pedi- atric patients from 6 countries to establish a valuable risk stratification as a base for future clinical studies.
Methods
A modified, age-dependent pS-GRAS score for pACC patients was calculated using the previous published version of the score21 in a slightly adjusted version: Tumor stage according
to the modified TNM classification (used by around 70% of the partners according to a survey) and the staging system of Children Oncology Group (COG), firstly described by Sandrini and later modified (used by around 30% of the part- ners according to a survey)12,15,22,23 (1 = 0; 2-3 = 1; 4 = 2 points), grade (Ki67 index 0%-9% =0; 10%-19% = 1; ≥20% = 2 points), resection (R) status (R0 complete resection = 0; RX [uncertain resection status], R1 [indicates the removal of all macroscopic disease, but microscopic margins are posi- tive for tumor], R2 [indicates macroscopic residual disease {either locally or due to remaining metastases}] = 1 point), age at diagnosis (<4 years = 0; ≥4 years = 1 point), hormone-related symptoms due to adrenal hormone excess (androgen produc- tion = 0; glucocorticoid-/mixed-/no-hormone production = 1 point), generating 8 pS-GRAS scores, and 4 pS-GRAS groups (1: 0-2, 2: 3-4, 3: 5, 4: 6-7) (see Supplementary Table S1). Due to the consideration of potential international variations within the resection data, we consolidated the resection data previously categorized into 4 groups (R0=0, Rx=1, R1=2, R2=3 points) as outlined in our prior publication21 into only 2 groups (R0=0 and Rx/R1/R2 = 1 point). Endpoints of the study were death and overall survival (OS).
An international cohort was assembled by including 268 pa- tients from various cooperation partners of the international pediatric ACC working groups ENSAT-PACT, IC-PACT, and from the GPOH-MET registry. Patients from 6 countries were included if they met the following inclusion criteria: Proven adrenocortical tumor, age between 0 and 18 years, presence of all pS-GRAS criteria, overall survival, and follow- up time.
Supporting the validation of the scoring system, we also in- corporated data from the previous analysis of pS-GRAS (see Figure 5A). This dataset encompassed 733 pACC patients from 33 available retrospective studies in the literature.21
Continuous variables were presented as median and inter- quartile ranges and categorical variables as counts and percen- tages. Statistical analyses were performed in MEDAS software (Grund EDV, Margetshöchheim). Categorical variables were compared between 2 groups using either chi-squared or, when the values were expected to be small, the Fisher or Mehta and Patel exact test. Continuous measurements were compared be- tween 2 groups by the Mann-Whitney U test. Comparisons of more than 2 groups were performed by rank variance analysis according to Kruskal and Wallis.
For survival endpoints, we performed univariable analyses using Kaplan-Meier (KM) survival curves according to the pS-GRAS score and its components (stage, Ki67 index, resec- tion status, age, hormone activity). The prognostic effect of pS-GRAS score and its individual components was assessed using multivariable survival analysis employing Cox regres- sion. Hazard ratio (HR), 95% CI, and P-values were reported. P-values <. 05 were considered significant.
An ethical approval for the collection of anonymized data in the current project was obtained in Würzburg, Germany (ref- erence number: 2022102002). All research complied with the Declaration of Helsinki.
Results
Patient characteristics
In total, we included clinical and pathological characteristics of 268 patients from 9 centers (Brazil,2 Germany,2 Great Britain, India, Italy, and Turkey2). Twenty-four percent (n=65) of patients were reported from Brazil and 76% (n=203) from other international centers. These 2 cohorts are reported separately as Brazilian pACC patients were associ- ated with the specific TP53-variant of Brazil and this subcohort differs slightly from the other international centers in hormone activity, age, and stage distribution as shown in Table 1.
The median age was 4.0 years (ranging from 0.1-18 years). A total of 33% were male, and 67% were female. There was no association between sex and age. The majority of the
patients showed hormone activity (>80%), most of them only androgen (41%) or mixed hormone activity (33%). The stage distribution according to modified TNM classifica- tion/COG classification by Sandrini22 was as follows: 34% stage I, 29% stage II, 13% stage III, and 24% stage IV, respect- ively. The median follow-up was 4.96 years (ranging from 0-27.7 years), 67% showed a complete remission, whereas 25% died due to ACC (details are listed in Table 1).
Evaluation of pS-GRAS criteria in pACC
Overall survival KM curves for pS-GRAS score analysis and its components as univariable analysis are illustrated in Figure 1A-F. A favorable prognosis was noted in patients with lower scores compared to higher scores with an excellent 5-year OS of 98% of group 1, compared to 87% of group 2 (HR of death 3.6, 95% CI 1.6-8.2, P =. 002), 43% of group 3 (HR of death 2.8, 95% CI 1.9-4.4, P <. 00001), and only 18% of group 4 (HR of death 2.1, 95% CI 1.7-2.7, P <. 00001) (for details, see univariable analysis in Table 2). The difference in overall survival was significant for the groups 1-4 (group 1 vs 2 P = . 026, group 2 vs 3 P <. 000001, group 3 vs 4 P =. 002; see Figure 1A). This confirmed the value of this adapted score for the pediatric patient population.
Univariable analysis showed a strong association of all eval- uated variables with overall survival (see Figure 1B-F and Table 2). In the multivariable survival analysis components, age (HR of death 3.5, 95% CI 1.8-7.0; P =. 00028), stage
| Clinical characteristicsª | Entire validation cohort | International | Brazil |
|---|---|---|---|
| n | 268 | 203 | 65 |
| Age (median and range in years) | 4.0 (0.1-18.0) | 5.2 (0.1-18) | 2.1 (0.08-16.01) |
| Sex (f/m) | 179/89 | 128/75 | 51/14 |
| Tumor stage | |||
| 1 | 92 (34%) | 52 (26%) | 40 (62%) |
| 2 | 77 (29%) | 69 (34%) | 8 (12%) |
| 3 | 34 (13%) | 29 (14%) | 5 (8%) |
| 4 | 65 (24%) | 53 (26%) | 12 (18%) |
| Resection state | |||
| R0 | 176 (66%) | 131 (65%) | 45 (69%) |
| R1 | 36 (13%) | 26 (13%) | 10 (15%) |
| R2 | 33 (12%) | 28 (14%) | 5 (8%) |
| Rx | 8 (3%) | 5 (2%) | 3 (5%) |
| Tumor spillage | 13 (5%) | 11 (5%) | 2 (3%) |
| No resection | 2 (1%) | 2 (1%) | 0 (0%) |
| Hormone production | |||
| No hormone | 49 (18%) | 46 (23%) | 3 (5%) |
| Androgen | 109 (41%) | 89 (44%) | 20 (31%) |
| Glucocorticoid | 22 (8%) | 21 (10%) | 1 (2%) |
| Mixed | 88 (33%) | 47 (23%) | 41 (63%) |
| Follow-up | |||
| Follow-up (median and range in years) | 4.96 (0.025-27.68) | 4.27 (0.025-27.64) | 7.21 (0.03-24.0) |
| CR | 180 (67%) | 131 (65%) | 49 (75%) |
| Dead due to second cancer | 3 (1%) | 3 (1%) | 0 (0%) |
| Alive with disease | 13 (5%) | 10 (5%) | 3 (5%) |
| DOD | 68 (25%) | 55 (27%) | 13 (20%) |
| Lost to follow-up | 4 (1%) | 4 (2%) | 0 (0%) |
| Decades of diagnosis | |||
| Before 2000 | 25 (9%) | 22 (11%) | 3 (5%) |
| 2000-2010 | 120 (45%) | 99 (49%) | 21 (32%) |
| >2010 | 123 (46%) | 82 (40%) | 41 (63%) |
Abbreviations: f/m, female, male; n, number; CR, complete remission; DOD, dead of disease.
aClinical characteristics including age (in years), sex (f/m), tumor stage, hormone activity, resection state, hormone production, follow-up (in years), and decades of diagnosis each in total number of patients (n) and percentage (%).
A
pS-GRAS score
B
Age
Overall survival [%]
Overall survival [%]
100
PS-GRAS 1
100
90
PS-GRAS 2
90
age < 4 years
80
80
70
70
60
60
50
pS-GRAS 3
50
age >= 4 years
40
40
30
30
20
pS-GRAS 4
20
10
10
0
2
4
6
8
10
12
14
16
18
20
0
2
4
6
8
10
12
14
16
18
20
Time [years]
Time [years]
Numbers at risk (n = 268)
<4
134
113
99
80
59
50
38
27
22
15
11
≥4
134
76
50
36
27
18
9
7
5
4
2
C
Tumor stage
D
Ki67 index
Overall survival [%]
Overall survival [%]
100
100
90
++
stage
index 0 - 9%
stage I
90
80
80
index 10 - 19%
70
70
60
60
index > 19%
50
stage
50
40
stage IV
40
30
30
20
20
10
10
0
2
4
6
8
10
12
14
16
18
20
Time [years]
E
Resection status
F
Hormone status
Overall survival [%]
Overall survival [%]
100
100
90
RO
90
80
80
androgen production
70
70
60
60
glucocorticoid / mixed / no hormone production
50
R1
50
40
40
30
30
20
R2
20
10
Rx
10
0
2
4
6
8
10
12
14
16
18
20
0
2
4
6
8
10
12
14
16
18
20
Time [years]
Time [years]
Numbers at risk (n = 268)
Numbers at risk (n = 268)
RO
176
147
122
94
73
57
39
29
23
16
10
Andr.
109
80
65
50
38
32
25
16
13
9
8
R1
49
33
22
17
11
10
7
4
3
2
2
Oth.
159
109
84
66
48
36
22
18
14
10
5
R2
33
8
4
4
2
1
1
1
1
1
1
Rx
10
1
1
1
0
0
0
0
0
0
0
| Numbers at | risk (n = 268) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 94 | 78 | 65 | 48 | 37 | 31 | 26 | 19 | 16 | 13 | 8 | |
| 2 | 82 | 71 | 59 | 50 | 38 | 30 | 18 | 13 | 9 | 5 | 4 | |
| 3 | 45 | 26 | 18 | 14 | 9 | 6 | 3 | 2 | 2 | 1 | 1 | |
| 4 | 47 | 14 | 7 | 4 | 2 | 1 | 0 | 0 | 0 | 0 | 0 | |
| Numbers at risk (n = 268) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| I | 92 | 77 | 66 | 51 | 40 | 33 | 25 | 22 | 18 | 14 | 8 |
| Il | 77 | 59 | 48 | 36 | 24 | 16 | 11 | 5 | 3 | 2 | 2 |
| III | 34 | 25 | 18 | 14 | 12 | 10 | 7 | 5 | 4 | 3 | 3 |
| IV | 65 | 28 | 17 | 15 | 10 | 9 | 4 | 2 | 2 | 0 | 0 |
| 0 | 2 4 | 6 | 8 | 10 | 12 | 14 | 16 | 18 20 | ||
|---|---|---|---|---|---|---|---|---|---|---|
| Time [years] | ||||||||||
| Numbers at risk (n = 268) | ||||||||||
| 0-9 | 71 | 57 | 45 | 35 | 27 | 22 | 15 | 12 | 10 | 8 4 |
| 10-19 | 73 | 55 | 43 | 34 | 27 | 23 | 19 | 14 | 10 | 7 6 |
| >19 | 124 | 77 | 61 | 47 | 32 | 23 | 13 | 8 | 7 | 4 3 |
Figure 1. pS-GRAS score. Kaplan-Meier curves depicting univariate analysis of overall survival from diagnosis of pACC patients (n=268) according to pS-GRAS scoring: (A) pS-GRAS groups (1: 0-2, 2: 3-4, 3: 5, 4: 6-7) and of each pS-GRAS component: (B) Age at diagnosis (<4 years, ≥4 years), (C) tumor stage (I-IV), (D) Ki67 proliferation index (0%-9%, 10%-19%, >19%), (E) resection status (R0, no resection, RX, R1, R2), (F) hormone status (androgen production, glucocorticoid-/mixed-/no-hormone production).
(HR of death 1.9, 95% CI 1.2-3.0; P = . 009), Ki67 expression (HR of death 1.7, 95% CI 1.2-2.4; P =. 007), and resection stage (HR of death 5.5, 95% CI 2.7-11.1; P <. 00001) seem to have the strongest power to predict overall survival whereas hormone expression (HR of death 1.1, 95% CI 0.6-1.8; P =. 75) had no significant impact (see Table 3). Details are given in the following paragraphs.
Relationship between age and tumor characteristics and/or outcome
Age has a significant impact on hormone activity and the kind of hormones produced, as well as case fatality and stage
distribution: Patients below the age of 4 years showed more often low-risk constellations with a better (superior) OS as shown in Figures 1B and 2A-D. Regarding the distribution of the tumor stages in the subgroups in detail, we observed that the tumor stages 1-4 occur in every age, but pACCs in children < 4 years more often present with lower tumor stages (see Figure 2A). Furthermore, the prognosis for each stage group was age-independent (data not shown). Children older than 4 years more often presented with inactive or glucocorticoid-producing tumors. We could not find any correlation of Ki67 with age.
In the KM curve (see Figure 2B), an age of 3.8 years proved to be the best cutoff for OS probability, well corresponding to the cutoff of 4 years commonly used in the literature.
| Variableª | Two-year-survival | Five-year-survival | HR of death | 95% CI | P-value | |
|---|---|---|---|---|---|---|
| n | n % | n % | ||||
| PS-GRAS | ||||||
| 1 94 | 78 99 | 55 98 | 1 | Reference | ||
| 2 82 | 71 96 | 55 87 | 3.6 | 1.6 | 8.2 | .0022 |
| 3 45 | 26 63 | 15 43 | 2.8 | 1.9 | 4.4 | <. 00001 |
| 4 47 | 14 39 | 7 18 | 2.1 | 1.7 | 2.7 | <. 00001 |
| Age | ||||||
| <4 134 | 113 95 | 89 91 | 1 | Reference | ||
| ≥4 134 | 76 68 | 43 50 | 6.9 | 3.8 | 12.5 | <. 00001 |
| Stage | ||||||
| I 92 | 77 98 | 56 95 | 1 | Reference | ||
| II 77 | 59 93 | 44 88 | 2.5 | 0.8 | 8.3 | .13 |
| III 34 | 25 78 | 16 49 | 3.5 | 2.1 | 6.0 | .00001 |
| IV 65 | 28 48 | 16 34 | 2.8 | 2.0 | 3.9 | <. 00001 |
| Ki67 | ||||||
| 0%-9% 71 | 57 97 | 42 91 | 1 | Reference | ||
| 10%-19% 73 | 55 84 | 37 78 | 3.0 | 1.1 | 8.2 | .028 |
| >19% 124 | 77 72 | 53 57 | 2.6 | 1.6 | 4.0 | .00003 |
| Resection status | ||||||
| R0 176 | 147 97 | 107 92 | 1 | Reference | ||
| Rxb 10 | 1 11 | 1 11 | 31.4 | 11.7 | 84.1 | <. 00001 |
| R1c 49 | 33 79 | 20 52 | 2.7 | 1.9 | 3.9 | <. 00001 |
| R2 33 | 8 26 | 4 13 | d | |||
| Hormone status | ||||||
| Androgen 109 | 80 84 | 53 77 | 1 | Reference | ||
| Glucoc .- /mixed/no hormones 159 | 109 80 | 79 67 | 1.5 | 0.9 | 2.5 | .11 |
aVariables including pS-GRAS score and its single components (age, stage, Ki67 status, resection status, and hormone production) in numbers of patients, two- and five-years overall survival (in numbers and percent), Hazard Ratio of death (HR), 95% Confidence interval (CI), and P-value.
“Including data of 2 patients without resection.
“Including data of 13 patients with tumor spillage.
HR of death for R2 couldn’t be calculated because statistical variation was too big.
| Variablesª | n | HR of death | 95% CI | P-value | |
|---|---|---|---|---|---|
| Ageb | 268 | 3.5 | 1.8 | 7.0 | .00028 |
| Hormone status | 268 | 1.1 | 0.6 | 1.8 | .75 |
| Tumor stage | 268 | 1.9 | 1.2 | 3.0 | .009 |
| Ki67 | 268 | 1.7 | 1.2 | 2.4 | .007 |
| Resection status | 268 | 5.5 | 2.7 | 11.1 | <. 00001 |
ªCalculation of variables of pS-GRAS score of 268 patients using multivariable survival statistics employing Cox regression including report of Hazard Ratio of death (HR), 95% Confidence interval (CI), and P-value. Parameters of the pS-GRAS score: Age: = 4 years (1), hormone status:
androgen excess (0), glucocorticoid / mixed/ no hormone excess (1), tumor stage: 1 (0), 2-3 (1), 4 (2), Ki67 expression: = 20% (3) and resection status: RO (0), Rx/ R1/ R2 (1).
Clinical characteristics and survival
The OS of the entire cohort was 91.9% (95% CI 90.2-92.6) after 1 year, 81.5% (95% CI 79.1-84.0) after 2 years, and 71.3% (95% CI 68.3-74.3) after 5 years (Figure 3A). Due to the higher propor- tion of low-risk patients, the Brazilian cohort tended to have a bet- ter survival than the non-Brazilian cohort (P =. 054). Patients treated before the year 2000 had a remarkably poorer survival than treatment in the 21st century (Figure 3B, P <. 001). Excluding patients diagnosed and treated before 2000, the valid- ity of the pS-GRAS scoring remains consistent (data not shown).
We confirmed a strong correlation of survival with less advanced tumor stage, R0 resection, and low Ki67 index (P <. 001; Figure 1C-E).
A
Age and stage distribution
B
Age distribution
Overall survival [%]
100
age ⇐ 2.5 years
50%
29%
90
2.5 < age ⇐ 3.8 years
19%
80
70
10%
13%
3.8 <age ⇐ 4.8 years
60
28%
12%
4.8 < age ⇐ 6 years
39%
50
6 < age ⇐ 18 years
40
30
< 4 years
>= 4 years
20
Stage
I,
II,
III,
N
10
0
2
4
6
8
10
12
14
16
18
20
Time [years]
C
Age of the Brazilian cohort
D
Age of the non-Brazilian cohort
Overall survival [%]
Overall survival [%]
100
100
age < 4 years
90
90
age < 4 years
80
80
70
70
60
60
50
50
age >= 4 years
age >= 4 years
40
40
30
30
20
20
10
10
0
2
4
6
8
10
12
14
16
18
20
0
2
4
6
8
10
12
14
16
18
20
Time [years]
Time [years]
| Numbers at risk (n = 268) | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| <= 2.5 | 105 | 90 | 81 | 65 | 47 | 40 | 32 | 22 | 19 | 13 | 9 |
| >2.5-3.8 | 22 | 18 | 14 | 11 | 8 | 7 | 4 | 3 | 1 | 1 | 1 |
| >3.8-4.8 | 18 | 10 | 6 | 6 | 5 | 4 | 2 | 2 | 2 | 1 | 1 |
| >4.8-6 | 23 | 13 | 10 | 8 | 6 | 5 | 3 | 2 | 1 | 1 | 1 |
| >6-18 | 100 | 58 | 38 | 26 | 20 | 12 | 6 | 5 | 4 | 3 | 1 |
| Numbers at risk (n | = 65) | Numbers at risk (n = 203) | |||||||||||||||||||||
| <4 | 48 | 42 | 37 | 32 | 26 | 23 | 18 | 13 | 11 | 7 | 6 | <4 | 86 | 71 | 62 | 48 | 33 | 27 | 20 | 14 | 11 | 8 | 5 |
| ≥4 | 17 | 5 | 5 | 5 | 3 | 2 | 1 | 1 | 0 | 0 | 0 | ≥4 | 117 | 71 | 45 | 31 | 24 | 16 | 8 | 6 | 5 | 4 | 2 |
Figure 2. (A-D) Age at diagnosis and overall survival from diagnosis. A: Pie chart illustrating age at diagnosis (<4 years and >4 years) and stage distribution (stages I-IV in %). B: Kaplan-Meier curves depicting distribution of age at diagnosis for all patients (<2.5 years, 2.5-3.8 years, 3.8-4.8 years, 4.8-5 years, >5 years) and overall survival; C and D: Kaplan-Meier curve depicting age at diagnosis (<4 years and >4 years) and overall survival in the Brazilian cohort (C) and the non-Brazilian cohort (D), P= . 000024 and P= . 000011.
Brazilian vs. non-Brazilian cohort
B Decades of diagnosis
Overall survival [%]
Overall survival [%]
100
100
90
90
80
Brazilian
80
2000-2010
70
both
70
>2010
60
non-Brazilian
60
50
50
<2000
40
40
30
30
20
20
10
10
0
2
4
6
8
10
12
14
16
18
20
0
2
4
6
8
10
12
14
16
18
20
Time [years]
Time [years]
| Numbers at risk (n = 268) | Numbers at risk (n = 268) | ||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Bra | 65 | 47 | 42 | 37 | 29 | 25 | 19 | 14 | 11 | 7 | 6 | 00-10 | 120 | 97 | 83 | 70 | 57 | 46 | 36 | 26 | 22 | 15 | 9 |
| Both | 203 | 142 | 107 | 79 | 57 | 43 | 28 | 20 | 16 | 12 | 7 | >2010 | 123 | 78 | 54 | 34 | 18 | 11 | 3 | 1 | 0 | 0 | 0 |
| n.Bra | 268 | 189 | 149 | 116 | 86 | 68 | 47 | 34 | 27 | 19 | 13 | <2000 | 25 | 14 | 12 | 12 | 11 | 11 | 8 | 7 | 5 | 4 | 4 |
Figure 3. (A and B) Influencing factors of survival. Kaplan-Meier curves depicting overall survival (patients n=268) subdividing cohorts by (A) origin of cohort (all patients, Brazilian-, non-Brazilian cohort) and (B) decades of diagnosis (<2000, 2000-2010, >2010).
Regarding Ki67 expression level, we could demonstrate a good differentiation in terms of survival rates (Figure 1D). Analyzing OS depending on Ki67 expression in a stage- dependent manner, we could show that in stage 1, higher expression of Ki67 had no impact on overall survival (how- ever, with a relatively small number of cases with high Ki67 expression at stage 1), but that from stage 2 onwards, Ki67 ex- pression > 10% is associated with significantly worse survival (Figure 4).
Resection status was a main factor for overall survival (Figure 1E, Tables 2 and 3). Even in advanced stages, a com- plete resection of the primary tumor was associated with a sig- nificantly better OS (stage 4 alive in CR: 7/14 with complete resection vs 10/51 with incomplete resection, P = . 02).
There was no significant difference in survival of tumors with androgen tumors compared to tumors with mixed-, glucocorticoid-, or no-hormone production, which is in contrast to our previous findings21 (see Figure 5A and B).
Discussion
Pediatric ACCs are rare and, therefore, management is not internationally standardized. In the adult ACC community, treatment protocols, prognostic scores, and molecular under- standing are much more advanced, however it could be shown that these results cannot be uncritically used for the pediatric ACC patients. For example, it has been shown that histological scores to diagnose ACC in adults (eg, Weiss, Hough, and Van Slooten) have no reliable diagnostic value especially in younger children as it leads to an overestimation of malignancy.
Risk stratification using combined scoring system including histopathological, clinical, and molecular data has been suc- cessful for certain (solid) pediatric tumors such as neuroblast- oma. Therefore, we established a modified pS-GRAS score in our previous work,12 which classifies pACC into 3 prognostic groups (low, intermediate, and high risks) on the basis of pathological and clinical criteria. We could show a
significantly favorable prognosis in patients with a lower scor- ing compared to higher scoring groups. To the best of our knowledge, this is the first risk score taking clinical factors as age and hormone activity at diagnosis into consideration. In comparison to genetic investigations, the data of pS-GRAS are easily available for all patients with pACC.18
However, our previous study-review of patient data of the literature-had several limitations as the majority of the re- ported studies included a small number of patients and there was a heterogeneity of patient cohorts, therapy approaches, and a diversity of reporting accuracy. We also know that because of the low incidence of pACC, we had to select patients over a large time period (1986-2021) for recruiting a sufficiently large cohort. For the retrospective analysis of the pS-GRAS score, only few patients with full datasets were eligible-mostly because data on Ki67/mitosis rate were frequently missing. Nevertheless, we showed in this heteroge- neous patient cohort that the combination of clinical and histopathological criteria seems to have a good predictive value and may therefore serve as a helpful tool for risk stratification in future studies.21 The next step was the valid- ation of pS-GRAS in an international context-which we have done in the current study-in order to establish a valu- able risk stratification as a base for future clinical studies. Due to networking between the pediatric working groups ENSAT-PACT, GPOH-MET, and IC-PACT, we were able to obtain complete dataset for all required pS-GRAS items from 9 centers and 6 countries.
The results mainly confirm the findings of the previous study of the literature. In the univariable analysis, we could validate age as an important factor for overall survival. The cutoff of 4 years of age for both cohorts, Brazilian and non-Brazilian, is pragmatic as previously described.4,7 We have shown that children <4 years of age have superior overall survival-regardless of tumor stage. Not surprisingly, patients with stage 4 generally have a poor prognosis- regardless of age. The hypothesis of a different pathogenesis of early childhood ACC and a biological similarity of tumors
A
Stage I
B
Stage II
Overall survival [%]
Overall survival [%]
100
Ki67 10-19%
100
Ki67 0-9%
90
KI67 0-9%
KI67 > 19%
90
Ki67 10-19%
80
80
70
70
Ki67 >19%
60
60
50
50
40
40
30
30
20
20
10
10
0
0
2
4
6
8
10
12
14
16
18
20
2
4
6
8
10
12
14
16
18
20
Time [years]
C
Stage III
D
Stage IV
Overall survival [%]
Overall survival [%]
100
Ki67 0-9%
100
90
90
80
80
70
70
60
Ki67 10-19%
Ki67 0-9%
60
50
50
40
Ki67 > 19%
40
30
30
Ki67 10-19%
20
20
Ki67 >19%
10
10
0
2
4
6
8
10
12
14
16
18
20
0
2
4
6
8
10
12
14
16
18
20
| Numbers at risk (n = 92) | ||||||||||||
| 0-9 | 33 | 27 | 23 | 17 | 16 | 13 | 10 | 10 | 8 | 8 | 4 | |
| 10-19 | 31 | 28 | 22 | 16 | 12 | 10 | 9 | 8 | 6 | 4 | 3 | |
| >19 | 28 | 22 | 21 | 18 | 12 | 10 | 6 | 4 | 4 | 2 | 1 | |
| Time [years] | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Numbers at risk (n = 77) | ||||||||||||
| 0-9 | 24 | 19 14 | 10 | 6 | 5 | 4 | 1 | 1 | 0 | 0 | ||
| 10-19 | 19 | 15 | 13 | 11 | 9 | 7 | 5 | 4 | 2 | 2 | 2 | |
| >19 | 34 | 25 | 21 | 15 | 9 | 4 | 2 | 0 | 0 | 0 | 0 | |
| Numbers at risk (n = 34) | Time [vears1 | Numbers at risk (n = 65) | Time Ivears] | ||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0-9 | 3 | 2 | 2 | 2 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0-9 | 11 | 9 | 6 | 6 | 4 | 4 | 1 | 1 | 1 | 0 | 0 |
| 10-19 | 8 | 6 | 4 | 4 | 3 | 3 | 2 | 1 | 1 | 1 | 1 | 10-19 | 15 | 6 | 4 | 3 | 3 | 3 | 3 | 1 | 1 | 0 | 0 |
| >19 | 23 | 17 | 12 | 8 | 8 | 7 | 5 | 4 | 3 | 2 | 2 | >19 | 39 | 13 | 7 | 6 | 3 | 2 | 0 | 0 | 0 | 0 | 0 |
Figure 4. (A-D) Impact of Ki67 expression in the different stage groups on overall survival. Kaplan-Meier curves depicting the influence of Ki67 expression (0%-9%, 10%-19%, >19%) in the particular tumor stages (A: Stage I, B: Stage 2, C: Stage 3; D: Stage 4) on overall survival.
in older patients seems to be supported by ours and other data. 16,27
In addition to age, several other predictive factors such as re- section status, tumor stage, hormonal activity, and Ki67 rate could be confirmed to correlate strongly with overall survival. Surgical tumor resection is the essential therapeutic approach for pACC, even in advanced setting, and the predictive value of resection status is well established, as microscopic or macroscopic tumor residuals significantly worsen the clinical outcome.28,29 R1, R2, and intraoperative spillage are associ- ated with poor outcome and additional treatment measures should be considered.
Tumor staging is a big topic, which has to be discussed and optimized. The tumor staging system-established for pediatric ACC-has high predictive value23 as we could confirm in our cohorts. However, there are still some lim- itations, especially for tumor stages 2 and 3 that do not
correctly discriminate all patients, and at least the thera- peutic implications correlating with the stages need to be improved. 28,30
The importance of hormone activity as prognostic and tu- mor marker has been widely discussed. Our data are in line with previously published data (over 70% of the cases from Brazil) showing that most of the pediatric ACC are hormone active and androgen production seems to be associated with better overall survival. This prognostic “androgen effect” has bias, including younger age. In the opposite direction, iso- lated Cushing’s syndrome tended to occur in the older children (P <. 001).15 In addition, the effect of hormone production was not as pronounced as in the retrospective cohort and not significant after adjustment. As there was no standardized methodology through the reporting centers for the measure- ment of hormone activity, there might be space for improve- ment and further efforts should be done to evaluate the
A
Cohort of the literature
B
Validation cohort
Overall survival [%]
Overall survival [%]
100
100
90
90
80
androgens
80
androgens
70
70
mixed
mixed
glucocorticoids
60
no hormones glucocorticoids
60
50
50
no hormones
40
40
30
30
20
20
10
10
0
2
4
6
8
10
12
14
16
18
20
0
2
4
6
8
10
12
14
16
18
20
Time [years]
Time [years]
| Numbers at risk (n = 424) | Numbers at risk (n = 268) | ||||||||||||||||||||||
| No | 48 | 34 | 24 | 18 | 16 | 12 | 7 | 6 | 5 | 4 | 3 | No | 48 | 31 | 23 | 17 | 12 | 11 | 7 | 6 | 5 | 4 | 1 |
| Mixed | 115 | 71 | 50 | 38 | 27 | 23 | 18 | 8 | 5 | 4 | 3 | Mixed | 89 | 62 | 50 | 42 | 34 | 24 | 14 | 11 | 9 | 6 | 4 |
| Andr. | 218 | 162 | 119 | 89 | 70 | 54 | 42 | 37 | 30 | 20 | 15 | Andr. | 109 | 80 | 65 | 50 | 38 | 32 | 25 | 16 | 13 | 9 | 8 |
| Gluc. | 43 | 29 | 23 | 20 | 15 | 12 | 10 | 10 | 9 | 5 | 4 | Gluc. | 22 | 16 | 11 | 7 | 2 | 1 | 1 | 1 | 0 | 0 | 0 |
Figure 5. (A and B) Impact of hormonal expression on overall survival. Kaplan-Meier curves depicting the influence of hormonal expression (no hormones, mixed, androgens, glucocorticoids) on overall survival of (A) the cohort of the literature (n= 424; see Riedmeier et al. “Assessment of prognostic factors in pediatric adrenocortical tumors: a systematic review and evaluation of a modified S-GRAS score.” European Journal of Endocrinology vol. 187,6 751-763. 26 Oct. 2022, doi:10.1530/EJE-22-0173) and (B) the validation cohort of the current multicenter study (n= 268).
prognostic and diagnostic role of hormone production more detailed.
Ki67 expression rate is an established prognostic factor in adult ACC,31,32 and we were able to confirm its importance also for pACC.17,33,34 However, 2 other independent studies performed similar analysis in younger children and found no association between Ki67 and prognosis,35,36 suggesting that age dependency needs to be further evaluated. Interestingly, in stage 1 patients, high Ki67 expression levels were not asso- ciated with an unfavorable outcome, whereas in all other stages, expression levels correlated with outcome. In addition to Ki67 expression levels, necrosis and capsular invasion are microscopic features that have a strong impact on malignancy, as shown by Picard at al.37
By comparing the multinational with the Brazilian cohort, we could confirm already well-described differences: The Brazilian cohort had a lower frequency of non-hormone active and only glucocorticoid-producing tumors, patients were younger, showed a higher proportion of stage I, a slightly better overall survival, and more TP53-related cancer syndromes-associated tumors.12 However, this better outcome relates likely due to the fact that most patients in Brazil are diagnosed as part of a screening program in affected CPS families with the possibility of earlier detection of pACC.4,5 However, prognostic differenti- ation using pS-GRAS scoring works in non-Brazilian and Brazilian pACC patients.
In summary, we were able to confirm the strong associ- ation between our pS-GRAS scoring in an international retro- and prospective multicenter cohorts. It shows a strong correlation with prognosis and can be easily assessed shortly after diagnosis to define risk groups for treatment. PS-GRAS scoring will be a first step using combined scoring system including histopathological, clinical, and molecular data for pACC. However, the role of hormone production
and molecular risk factors needs to be investigated in fur- ther studies to improve this approach. Risk stratification and therapy adjustment very early after diagnosis and add- itional monitoring of minimal residual disease by specific molecular markers (eg, liquid biopsies) should be the long- term goal for pACC. However, achieving this goal is a long- term endeavor. In the interim, pS-GRAS can serve as a now validated prognostic tool to consult patients and guide treatment decisions.
Acknowledgments
This work was supported by a research grant from the Tour of Hope Foundation. We would like to thank the Parents Initiative Group for Children with Leukemia and Solid Tumors Würzburg e.V. for their continuous support.
Supplementary material
Supplementary material is available at European Journal of Endocrinology online.
Funding
This work was supported by a research grant Interdisziplinäres Zentrum für klinische Forschung Würzburg (IZKF) training grant awarded to M.R. (project number: Z-02CSP/23), the Dr. Mildred Scheel Stiftung für Krebsforschung (project num- ber: 70113303-6) awarded to V.W., and by the Deutsche Forschungsgemeinschaft (DFG) German Research Foundation Project 314061271-TRR 205 to M.F. The funder had no role in study design, data collection and analysis, decision to pub- lish, or preparation of the manuscript.
Conflict of interest: None declared.
Authors’ contributions
Maria Riedmeier (Formal analysis [equal], Funding acquisition [equal], Supervision [equal], Writing-original draft [equal]), Shipra Agarwal (Data curation [equal], Writing-review & ed- iting [equal]), Sonir Antonini (Data curation [equal]), Martin Fassnacht (Data curation [equal], Formal analysis [equal], Supervision [equal], Validation [equal], Writing-review & ed- iting [equal]), Bonald Figueiredo (Data curation [equal], Writing -review & editing [equal]), Soraya Puglisi (Data curation [equal], Writing-review & editing [equal]), Verena Wiegering (Data curation [equal], Methodology [equal], Project adminis- tration [equal], Supervision [equal], Writing-original draft [equal]), Christoph Härtel (Writing-review & editing [equal]), Paul-Gerhardt Schlegel (Writing-review & editing [equal]), Tulay Guran (Data curation [equal], Writing-review & editing [equal]), Imme Haubitz (Conceptualization [equal], Formal ana- lysis [equal], Methodology [equal], Software [equal], Validation [equal]), Jan Idkowiak (Data curation [equal], Writing-review & editing [equal]), Michaela Kuhlen (Data curation [equal], Writing-review & editing [equal]), Antje Redlich (Data cur- ation [equal]), Bilgehan Yalcin (Data curation [equal], Writing -review & editing [equal]), Lúcia Noronha (Data curation [supporting]), Ivy Zortéa Parise (Data curation [equal]), Tatiana EI-Jaick B. Costa (Data curation [equal]), Orhan Diclehan (Data curation [equal]), and Ekinci Saniye (Data cur- ation [equal])
References
1. Ries S, Gurney L, Tamra Y. Cancer Incidence and Survival among Children and Adolescents: United States SEER Program 1975-1995, National Cancer Institute, SEER Program. NIH Pub. No. 99-4649. Bethesda.
2. Birch JM, Blair V. Increase in childhood carcinomas in north-west England. Lancet. 1988;1(8589):833. https://doi.org/10.1016/S0140- 6736(88)91699-6
3. Desandes E, Clavel J, Berger C, et al. Cancer incidence among chil- dren in France, 1990-1999. Pediatr Blood Cancer. 2004;43(7): 749-757. https://doi.org/10.1002/pbc.20148
4. Ribeiro RC, Sandrini F, Figueiredo B, et al. An inherited p53 muta- tion that contributes in a tissue-specific manner to pediatric adrenal cortical carcinoma. Proc Natl Acad Sci U S A. 2001;98(16): 9330-9335. https://doi.org/10.1073/pnas.161479898
5. Custódio G, Parise GA, Kiesel Filho N, et al. Impact of neonatal screening and surveillance for the TP53 R337H mutation on early detection of childhood adrenocortical tumors. J Clin Oncol. 2013; 31(20):2619-2626. https://doi.org/10.1200/JCO.2012.46.3711
6. Wasserman JD, Novokmet A, Eichler-Jonsson C, et al. Prevalence and functional consequence of TP53 mutations in pediatric adreno- cortical carcinoma: a children’s oncology group study. J Clin Oncol. 2015;33(6):602-609. https://doi.org/10.1200/JCO.2013. 52.6863
7. Wagner J, Portwine C, Rabin K, Leclerc JM, Narod SA, Malkin D. High frequency of germline p53 mutations in childhood adrenocort- ical cancer. J Natl Cancer Inst. 1994;86(22):1707-1710. https://doi. org/10.1093/jnci/86.22.1707
8. Kleihues P, Schäuble B, zur Hausen A, Estève J, Ohgaki H. Tumors associated with p53 germline mutations: a synopsis of 91 families. Am J Pathol. 1997;150(1):1-13.
9. Wang Z, Liu G, Sun H, et al. Clinical characteristics and prognosis of adrenocortical tumors in children. Pediatr Surg Int. 2019;35(3): 365-371. https://doi.org/10.1007/s00383-018-4409-z
10. Zambaiti E, Duci M, De Corti F, et al. Clinical prognostic factors in pediatric adrenocortical tumors: a meta-analysis. Pediatr Blood Cancer. 2021;68(3):e28836. https://doi.org/10.1002/pbc.28836
11. Kerkhofs TM, Ettaieb MH, Verhoeven RH, et al. Adrenocortical carcinoma in children: first population-based clinicopathological study with long-term follow-up. Oncol Rep. 2014;32(6):2836-2844. https://doi.org/10.3892/or.2014.3506
12. Riedmeier M, Decarolis B, Haubitz I, et al. Adrenocortical carcin- oma in childhood: a systematic review. Cancers (Basel). 2021;13(21):5266. https://doi.org/10.3390/cancers13215266
13. Mendonca BB, Lucon AM, Menezes CA, et al. Clinical, hormonal and pathological findings in a comparative study of adrenocortical neoplasms in childhood and adulthood. J Urol. 1995;154(6): 2004-2009. https://doi.org/10.1016/S0022-5347(01)66673-4
14. Wieneke JA, Thompson LD, Heffess CS. Adrenal cortical neo- plasms in the pediatric population: a clinicopathologic and immu- nophenotypic analysis of 83 patients. Am J Surg Pathol. 2003;27(7):867-881. https://doi.org/10.1097/00000478-200307000- 00001
15. Michalkiewicz E, Sandrini R, Figueiredo B, et al. Clinical and out- come characteristics of children with adrenocortical tumors: a re- port from the International Pediatric Adrenocortical Tumor Registry. J Clin Oncol. 2004;22(5):838-845. https://doi.org/10. 1200/JCO.2004.08.085
16. Lalli E, Figueiredo BC. Pediatric adrenocortical tumors: what they can tell us on adrenal development and comparison with adult ad- renal tumors. Front Endocrinol (Lausanne). 2015;6:23. https://doi. org/10.3389/fendo.2015.00023
17. Jangir H, Ahuja I, Agarwal S, et al. Pediatric adrenocortical neoplasms: a study comparing three histopathological scoring systems. Endocr Pathol. 2023;34(2):213-223. https://doi.org/10.1007/s12022-023- 09767-z
18. Elhassan YS, Altieri B, Berhane S, et al. S-GRAS score for prognostic classification of adrenocortical carcinoma: an international, multicen- ter ENSAT study. Eur J Endocrinol. 2021;186(1):25-36. https://doi. org/10.1530/EJE-21-0510
19. Lippert J, Appenzeller S, Liang R, et al. Targeted molecular analysis in adrenocortical carcinomas: a strategy toward improved personal- ized prognostication. J Clin Endocrinol Metab. 2018;103(12): 4511-4523. https://doi.org/10.1210/jc.2018-01348
20. Libé R, Borget I, Ronchi CL, et al. Prognostic factors in stage III-IV adrenocortical carcinomas (ACC): an European Network for the Study of Adrenal Tumor (ENSAT) study. Ann Oncol. 2015;26(10): 2119-2125. https://doi.org/10.1093/annonc/mdv329
21. Riedmeier M, Decarolis B, Haubitz I, et al. Assessment of prognos- tic factors in pediatric adrenocortical tumors: a systematic review and evaluation of a modified S-GRAS score. Eur J Endocrinol. 2022;187(6):751-763. https://doi.org/10.1530/EJE-22-0173
22. Fassnacht M, Johanssen S, Quinkler M, et al. Limited prognostic val- ue of the 2004 International Union Against Cancer staging classifica- tion for adrenocortical carcinoma: proposal for a revised TNM classification. Cancer. 2009;115(2):243-250. https://doi.org/10.1002/ cncr.24030
23. Sandrini R, Ribeiro RC, DeLacerda L. Childhood adrenocortical tu- mors. J Clin Endocrinol Metab. 1997;82(7):2027-2031. https://doi. org/10.1210/jcem.82.7.4057
24. Ribeiro RC, Sandrini Neto RS, Schell MJ, Lacerda L, Sambaio GA, Cat I. Adrenocortical carcinoma in children: a study of 40 cases. J Clin Oncol. 1990;8(1):67-74. https://doi.org/10.1200/JCO.1990. 8.1.67
25. Virgone C, Roganovic J, Vorwerk P, et al. Adrenocortical tumours in children and adolescents: the EXPERT/PARTNER diagnostic and therapeutic recommendations. Pediatr Blood Cancer. 2021;68(S4): e29025. https://doi.org/10.1002/pbc.29025
26. Cagle PT, Hough AJ, Pysher TJ, et al. Comparison of adrenal cor- tical tumors in children and adults. Cancer. 1986;57(11):2235- 2237 .< 2235 :: AID-CNCR2820571127>3.0.CO;2-O
27. Dehner LP, Hill DA. Adrenal cortical neoplasms in children: why so many carcinomas and yet so many survivors? Pediatr Dev Pathol. 2009;12(4):284-291. https://doi.org/10.2350/08-06-0489.1
28. Rodriguez-Galindo C, Krailo MD, Pinto EM, et al. Treatment of pediatric adrenocortical carcinoma with surgery, retroperitoneal
lymph node dissection, and chemotherapy: the children’s oncology group ARAR0332 protocol. J Clin Oncol. 2021;39(22):2463-2473. https://doi.org/10.1200/JCO.20.02871
29. Gulack BC, Rialon KL, Englum BR, et al. Factors associated with survival in pediatric adrenocortical carcinoma: an analysis of the National Cancer Data Base (NCDB). J Pediatr Surg. 2016;51(1): 172-177. https://doi.org/10.1016/j.jpedsurg.2015.10.039
30. Kuhlen M, Mier P, Kunstreich M, et al. Locally advanced adreno- cortical carcinoma in children and adolescents-enigmatic and chal- lenging cases. Cancers (Basel). 2023;15(17):4296. https://doi.org/ 10.3390/cancers15174296
31. Beuschlein F, Weigel J, Saeger W, et al. Major prognostic role of Ki67 in localized adrenocortical carcinoma after complete resec- tion. J Clin Endocrinol Metab. 2015;100(3):841-849. https://doi. org/10.1210/jc.2014-3182
32. Fassnacht M, Dekkers OM, Else T, et al. European Society of Endocrinology Clinical Practice guidelines on the management of adre- nocortical carcinoma in adults, in collaboration with the European Network for the Study of Adrenal Tumors. Eur J Endocrinol. 2018;179(4):G1-G46. https://doi.org/10.1530/EJE-18-0608
33. Martins-Filho SN, Almeida MQ, Soares I, et al. Clinical impact of pathological features including the Ki-67 labeling index on diagnosis
and prognosis of adult and pediatric adrenocortical tumors. Endocr Pathol. 2021;32(2):288-300. https://doi.org/10.1007/s12022-020- 09654-x
34. Pinto EM, Rodriguez-Galindo C, Pounds SB, et al. Identification of clinical 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. https://doi.org/10. 1200/JCO.2017.74.2460
35. 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. https://doi.org/10.3390/cancers11111730
36. Mattone MC, Gil S, Costanzo M, et al. Pediatric adrenocortical tumors cohort characteristics and long-term follow-up at a single Argentinian tertiary center. J Pediatr Endocrinol Metab. 2022;35(1):19-27. https:/ doi.org/10.1515/jpem-2021-0392
37. Picard C, Orbach D, Carton M, et al. Revisiting the role of the patho- logical grading in pediatric adrenal cortical tumors: results from a na- tional cohort study with pathological review. Mod Pathol. 2019;32(4): 546-559. https://doi.org/10.1038/s41379-018-0174-8
Downloaded from https://academic.oup.com/ejendo/article/191/1/64/7699771 by National Library of Medicine user on 03 April 2026