Vitamin D receptor hypermethylation as a biomarker for pediatric adrenocortical tumors
Ana Carolina Bueno (1, Mônica F Stecchini1, Junier Marrero-Gutierrez1, Candy Bellido More1, Leticia Ferro Leal1,1, Débora Cristiane Gomes2, Daniel Ferreira de Lima Neto1, Silvia Regina Brandalise3, Izilda Aparecida Cardinalli3, José Andres Yunes3, Thais Junqueira3, Carlos Alberto Scrideli1, Carlos Augusto Fernandes Molina4, Fernando Silva Ramalho5, Silvio Tucci4, Fernanda Borchers Coeli-Lacchini6, Ayrton Custodio Moreira6, Leandra Ramalho5, Ricardo Zorzetto Nicoliello Vêncio7, Margaret De Castro6 and Sonir Roberto R Antonini DD1
1Department of Pediatrics, Ribeirao Preto Medical School, University of Sao Paulo, Ribeirao Preto, Sao Paulo, Brazil, 2Federal University of Uberlandia, Uberlandia, Minas Gerais, Brazil, 3Boldrini Children’s Center, State University of Campinas, Campinas, Sao Paulo, Brazil, 4Department of Surgery and Anatomy, 5Department of Pathology, 6Department of Internal Medicine, Ribeirao Preto Medical School, University of Sao Paulo, Ribeirao Preto, Sao Paulo, Brazil, and 7Department of Computation and Mathematics, Faculty of Philosophy, Sciences and Letters at Ribeirao Preto, University of São Paulo, Ribeirao Preto, Sao Paulo, Brazil
*(L F Leal is now at Molecular Oncology Research Center, Barretos Cancer Hospital, Barretos, SP, Brazil; Barretos School of Health Sciences, Dr Paulo Prata - FACISB, Sao Paulo, Brazil and is supported by Public Ministry of Labor Campinas (Research, Prevention, and Education of Occupational Cancer))
Correspondence should be addressed to S R Antonini Email antonini@fmrp.usp.br
Abstract
Objective: Pediatric adrenocortical tumors (pACT) display complex genomic backgrounds, lacking robust prognostic markers and targeted therapeutic options. Vitamin D3 receptor (VDR) promoter hypermethylation and underexpression were reported in adrenocortical carcinomas from adult patients. In this study, we aimed to investigate VDR expression levels and methylation status in pACT and their clinical and prognostic significance.
Design: Retrospective cross-sectional study enrolling pediatric patients with ACT from two tertiary referral institutions. Methods: We evaluated clinicopathological features, VDR mRNA (qPCR) and protein (immunohistochemistry) expression, and VDR-wide methylation of ACT samples from 108 pediatric patients. Fourteen pediatric and 32 fetal and postnatal normal adrenals were used as controls.
Results: Unlike in pre- and post-natal normal adrenals, most pACT lacked nuclear VDR expression and had reduced mRNA levels, especially the carcinomas. Unsupervised analysis of VDR methylation data revealed two groups of pACT with distinct disease features and outcomes. Tumors with high VDR methylation presented lower mRNA levels, and the respective patients presented advanced disease and reduced disease-free and overall survival.
Conclusions: VDR has a role in normal adrenocortical development and homeostasis, which is impaired during tumorigenesis. VDR hypermethylation and underexpression may be both predictive and prognostic biomarkers for pACT.
European Journal of Endocrinology (2022) 186, 573-585
Introduction
Pediatric adrenocortical tumors (pACT) are typically hormone-secreting tumors that present a benign clinical
course but, rarely, may behave as aggressive malignancies (1, 2). pACT are uncommon (3), but their incidence is specifically elevated in Southern Brazil, where it associates with the prevalent p53 p.R337H mutation (4, 5). Pediatric
| Clinical Study | A C Bueno and others | VDR impairment in pediatric | 186:5 | 574 |
|---|---|---|---|---|
| ACT |
and adult ACT present specific epidemiological, clinical, histopathological features and survival rates that characterize them as different entities (2, 3).
ACT display complex genomic background and lack robust prognostic and predictive biomarkers. Surgical resection is the only curative treatment for patients with localized disease (1, 6); systemic therapies render limited improvement in the survival of patients with advanced or metastatic disease (7). Recent studies have associated tumor genomic methylation profiles with the prognosis of patients with ACT (8, 9, 10). However, despite the advances in the identification of driver genes and molecular subtypes of tumors with distinct prognostic features (8, 11, 12, 13), the development of new target-specific therapies has not advanced significantly.
The vitamin D3 (VD3) receptor (VDR) is a member of the nuclear receptors superfamily that mediates the expression of numerous genes in a 1,25(OH)2D-dependent fashion. VDR is ubiquitously expressed in the endocrine system, signaling in a cell-specific manner (reviewed in (14, 15)), but its function in adrenal development and homeostasis remains unsettled. Increased VDR gene promoter methylation and reduced expression were reported in ACT from adult patients (16, 17), demonstrating the impairment of the receptor in adrenocortical carcinomas (ACC). VDR transactivation was shown to repress aberrant Wnt/B- catenin pathway signaling - which is a hallmark of ACT and a key feature of unfavorable patients’ outcomes (18, 19, 20) - reducing tumor proliferation in different cancers (21, 22, 23). Additionally, VDR mRNA expression was recently suggested as a predictive biomarker for immunotherapy in primary and metastatic melanomas (24).
In the present study, we aimed to investigate VDR temporal and spatial expression during normal human adrenal cortex development and in pACT. We also investigated tumor VDR methylation profile and its association with VDR mRNA levels and pediatric patients’ outcomes.
Subjects and methods
Study design, patients and biological specimens
This study is a retrospective cross-sectional study comprising 108 pediatric patients who were diagnosed with ACT and followed-up between 1991 and 2019 tertiary referral centers in the Southeast region of Brazil: Ribeirao Preto Medical School - University of Sao Paulo (FMRP-USP) and Boldrini Children’s Cancer Center - State University of Campinas (BCC-SUC). Diagnostic evaluation, treatment,
follow-up and the sequencing of exon 3 of the CTNNB1 gene and exon 10 of the TP53 gene - to investigate the presence of the p53 p.R337H hypomorphic variation - were performed as previously reported (19, 20). Disease stage was classified according to the International Pediatric Adrenocortical Tumor Registry (IPACTR) (25). Tumor histopathological classification was based on the criteria proposed by Wieneke et al. (26, 27, 28, 29). According to this system, tumors with ≤2 criteria are diagnosed as adenomas (ACA), with 3, as of uncertain malignant potential (UMP), and with ≥4, as carcinomas (ACC). Individual features of the study participants are detailed in Supplementary Table 1 (see section on supplementary materials given at the end of this article).
We included 14 pediatric normal adrenal cortices as control specimens, obtained during autopsy (n= 4) and nephrectomy (n= 10) from patients with Wilms tumors without clinical features of Beckwith-Wiedemann syndrome (BWS) and before chemotherapy. We also included a panel of 32 normal adrenals obtained from spontaneously miscarried fetuses (20-37 weeks of gestation) and children aged up to 10 years who underwent routine autopsy examination in the FMRP-USP’s Department of Pathology. All samples were reviewed by experienced pathologists. The number of samples submitted to the different VDR investigation levels was defined upon high-quality sample availability; their features are summarized in Fig. 1.
This study was performed in compliance with the Declaration of Helsinki and was approved by the FMRP- USP University Hospital Ethics Committee (#7534/2010, #43758/2013) and BCC-SUC Ethics Committee (#1.7- 050809). Written informed consent was obtained from the patients’ and controls’ parents or legal guardians.
VDR immunohistochemistry
We performed immunohistochemical staining (IHC) in formalin-fixed paraffin-embedded (FFPE) samples from the panel of fetal and postnatal adrenals and 42 pACT from FMRP-USP. Tumor samples were arranged in tissue microarrays. Kidney and liver tissue sections were used as positive and negative controls, respectively (data not shown). We used VDR monoclonal antibody (clone 9A7gamma.E10.E4; aa 89-105 epitope; Cat#MA5-14617, Thermo Fisher Scientific), diluted to 1:100, following signal detection with the REVEAL Biotin-Free Polyvalent HRP kit (REVEAL®, Amsbio, Mainz, Germany). The antigen/antibody complexes were labeled with 3.3’-diaminobenzidine (DAB, Vector Laboratories Inc.) and counterstained with Harris hematoxylin. Staining evaluation and image acquisition were performed using
Patients and samples Primary pACT from 2 reference centers: FMRP-USP and BCC-SUC
Controls Normal adrenal tissues from 2 reference centers: FMRP-USP and BCC-SUC
108 pACT (100%): VDR mRNA Real time qPCR (FF; FMRP-USP and BCC-SUC)
14 Pediatric adrenals: VDR mRNA - Real time qPCR (FF; FMRP-USP and BCC-SUC)
57 pACT (53%): VDR methylation EPIC Array (FF; FMRP-USP and BCC-SUC)
42 pACT (39%): VDR expression IHQ (FFPE; FMRP-USP)
32 Pre and postnatal adrenals: VDR expression IHQ (FFPE; FMRP-USP)
Figure 1 Patients and control samples analyzed in the study according to high-quality sample availability. Samples’ numbers, level of VDR investigation, type of conservation, and reference center of origin. pACT, pediatric adrenocortical tumors; FMRP-USP, Ribeirao Preto Medical School; BCC-SUC, Boldrini’s Children Center, State University of Campinas; FF, fresh frozen; FFPE, formalin-fixed paraffin-embedded; IHQ, immunohistochemistry.
a Zeiss optical microscope (AxioCam MRC) and an image acquisition software (Zeiss AG).
VDR nuclear immunoreactivity in pACT samples was evaluated independently by two experienced pathologists using the H-score (30). VDR nuclear immunoreactivity was considered positive in specimens with H-score ≥ 1. Concordance between pathologists was calculated using Cohen’s kappa coefficient and interpreted according to Landis and Koch guidelines (31).
Quantitative real-time PCR
We isolated total RNA from microdissected fresh-frozen (FF) normal pediatric adrenals (n= 14) and pACT samples (n= 108) using TRIzol® Reagent (Life Technologies) or RNeasy kit (Qiagen), according to the manufacturers’ recommendations. Extracted RNA was further cleaned-up using RNeasy kit columns. RNA quantification was performed by fluorimetry (Qubit Fluorometer, Invitrogen) and integrity, by electrophoresis, using the Tape Station 4200 System (Agilent Technologies), considering an adequate RNA integrity number (RIN) ≥7. mRNA samples (500 ng) were reverse transcribed using the High-Capacity cDNA RT kit and MultiScribe® enzyme (Life Technologies). We used TaqMan® assays (Applied Biosystems) for qPCR: VDR (Hs00172113_m1) and GUSB (4326320E) as endogenous control, as previously described (32). Relative mRNA expression levels were determined using the 2-44Ct method.
VDR methylation profiling
We investigated VDR methylation in a subset of the pACT (n= 57), which was enrolled in our group’s pACT DNA methylation profiling study (manuscript in preparation). DNA was extracted from microdissected FF pACT samples using the QIAamp DNA Mini Kit (Qiagen), according to the manufacturer’s instructions. DNA quantification was performed using the Qubit dsDNA BR Assay (Thermo Fisher Scientific), and integrity was assessed using the Tape Station 4200 System (Agilent). DNA integrity number (DIN) ≥ 6 was considered adequate. Methylation EPIC BeadChip Array (Illumina ) was commissioned to the University of Southern California Keck Genomics Platform (Los Angeles, CA, USA) and performed according to Illumina’s standard protocol. Hybridization intensities raw files were provided by the facility and subjected to a standard quality control pipeline in our group. The minfi R package v.1.30.0 (33) was used for data preprocessing and M-value estimation. M-value was calculated as the log2 ratio of the intensities of methylated probe vs unmethylated probe (log2(methylated/unmethylated)), in which M-values close to 0 indicate a similar intensity between methylated and unmethylated probes, whereas positive M-values mean that more molecules are methylated than unmethylated, and negative M-values mean the opposite (34). Bias correction was performed using the preprocessQuantile function implemented in R statistical language (35). Data filtering
removed probes with a detection P-value higher than 0.01, probes located in sex chromosomes, probes containing single nucleotide polymorphisms in CpG-sites, and cross- reactive probes (36). In the present study, we considered only VDR-associated probes and its immediate vicinity, according to the annotation file provided by Illumina (53 probes; Supplementary Table 2). Additional annotation information was recovered from Ensembl human genome version GRCh38.p13.
Unsupervised hierarchical clustering analysis (UHCA) was performed using Euclidian distance and Ward’s method in R (version 3.6). Differentially methylated probes (DMPs) were identified using R Bioconductor’s ChAMP package v.2.13.5 (37), function champ.DMP, and increasingly ranked according to statistical significance (P-values).
Statistical analysis
Continuous or discrete variables were reported individually, collapsed (mean and/or median, and range) or as a percentage, as informed in figure and table legends. Appropriate statistical tests were used accordingly: Mann- Whitney, Kruskal-Wallis, Fisher’s exact test, and chi- squared test. Disease-free (DFS) and overall (OS) survivals were analyzed using Kaplan-Meier curves and were defined as the time elapsed from diagnosis until the last follow-up, considering metastasis/recurrence or death, respectively, as unfavorable events. Patients who were lost to follow-up were censored considering their last follow-up visit. Log- rank test was used for the comparison of survival rates between groups. Multiple hypothesis test correction methods were used when appropriate: Dunn’s, Dunnett’s, or Bonferroni’s. GraphPad Prism software (v.9.0.0) was used for statistical analyses. The minimum statistical significance level was set at P < 0.05, but other stringent options are also presented when appropriate.
Results
Clinical presentation and mortality- related features
Baseline demographic, clinical, hormonal, histopathological, and molecular features of the 108 pediatric patients with ACT are summarized in Table 1. We studied 78 girls (73%) and 30 boys (female-to-male ratio of 2.6:1), with median age at diagnosis of 1.8 years (range: 0.18- 16.6). Twenty-three (21%) children were older than 4 years of age (18 (17%) between 4 and 12 years and 5 (4%) older than 13 years). The most frequent clinical presentation was
virilization (77%), followed by asymptomatic presentation (11%), Cushing syndrome (6.5%) and virilization associated with Cushing syndrome (5.5%). Among the 12 asymptomatic patients, 3 were actively screened for ACT prior to any hormone excess manifestation due to altered 17-hydroxyprogesterone levels in neonatal screening (pACT003), pACT family history (pACT038), and BWS (pACT060). One asymptomatic patient was further diagnosed with hypopituitarism (pACT097). Most of the tumors were androgen and cortisol-secreting (85%), whereas isolated androgen-secretion was present in six (11%) and isolated cortisol-secretion in two (4%) patients. Disease stage was I in 59 (55%), II in 23 (21%), III in 14 (13%), and IV in 11 (10%) patients. Thirty-six patients (33%) received adjuvant chemotherapy, and 19 (18%) evolved with post-surgery recurrence or metastasis. Median follow-up was 6.8 years, and to date, 17 (16%) patients died. Histopathological classification diagnosed 43 (40%) tumors as ACA, 29 (27%) as of UMP, and 27 (25%) as ACC. There was an association between tumor histopathological classification and patient’s age at diagnosis: the percentage of ACC was 18% in those diagnosed <4 years, 56% in those with 4-12 years, and 60% in those with 13-20 years of age (P=0.01).
We observed reduced DFS and OS in patients diagnosed after 4 years of age (4-12 and 13-20 years, P < 0.0001), in those presenting with metastatic disease (P < 0.0001), diagnosed with ACC (P < 0.0001), and Cushing syndrome alone or associated with virilization (P < 0.0001) (Fig. 2A, B, C, and D).
Most patients (85%) were carriers of the p53 p.R337H germline mutation; this feature was not associated with disease progression nor with patients’ outcome. Few patients had tumors harboring CTNNB1 mutations (n= 10; 9%); these patients had inferior DFS and OS (P=0.02 and P=0.008, respectively, Fig. 2E) and their median age at diagnosis was 1.95 years (range: 0.8-15.6), being eight diagnosed <4 years, one between 4 and 12 years and only one >13 years of age.
VDR expression during normal human adrenal cortex development
We observed that, during the first and second trimesters of gestation, VDR was expressed in the nucleus and cytoplasm of cells in the definitive and in the transitional adrenal zones. As gestational age progressed toward term, VDR expression progressively diminished in the cytoplasm and became present mainly in cell nuclei and spread throughout all layers of the adrenal cortex in late
Table 1 Demographic, clinical, and molecular features of the pediatric patients with ACT at baseline according to VDR DNA methylation group. Differences between the variables in the methylations clusters were assessed using chi-square or Fisher’s exact tests (discrete) and Mann-Whitney test (continuous). Statistically significant (P < 0.05) values are presented in bold.
| Features | All patients | VDR methylation | P-value* | |
|---|---|---|---|---|
| Low | High | |||
| Demographic | ||||
| Patients, n (%) | 108 | 43 (75) | 14 (25) | – |
| Reference center, n (%) | 0.54 | |||
| FMRP-USP | 52 (48) | 26 (60) | 10 (71) | |
| BCC-SUC | 56 (52) | 17 (40) | 4 (29) | |
| Sex, n (%) | 0.09 | |||
| Girls | 78 (73) | 33 (77) | 7 (50) | |
| Boys | 30 (27) | 10 (23) | 7 (50) | |
| Age at diagnosis (years) | 0.16 | |||
| Median (range) | 1.8 (0.18-16.6) | 2 (0.2-13.2) | 6.7 (0.4-16.6) | |
| Mean | 3.4 | 3 | 6.9 | |
| <4 years of age, n (%) | 85 (79) | 37 (86) | 6 (43) | 0.003 |
| 4-12 years of age, n (%) | 18 (17) | 5 (12) | 5 (36) | |
| 13-20 years of age, n (%) | 5 (4) | 1 (2) | 3 (21) | |
| Clinical | ||||
| Clinical presentation, n (%) | 0.03 | |||
| Asymptomatic | 12 (11) | 4 (9) | 3 (21) | |
| Virilization only | 83 (77) | 36 (84) | 7 (50) | |
| Cushing only | 7 (6.5) | 2 (5) | 4 (29) | |
| Virilization and Cushing | 6 (5.5) | 1 (2) | 0 | |
| Tumor steroid secretion*, n (%) | 0.69 | |||
| Androgen-secreting | 6 (11) | 4 (15) | 2 (20) | |
| Cortisol-secreting | 2 (4) | 1 (4) | 1 (10) | |
| Androgen- and cortisol-secreting | 44 (85) | 21 (81) | 7 (70) | |
| Disease stage, n (%) | 0.003/0.01+ | |||
| I | 59 (55) | 29 (67) | 2 (14) | |
| II | 23 (21) | 9 (21) | 6 (43) | |
| III | 14 (13) | 3 (7) | 2 (14) | |
| IV | 11 (10) | 2 (5) | 4 (29) | |
| Missing data | 1 (1) | 0 | 0 | |
| Histopathological classification, n (%) | 0.06+ | |||
| Adenoma (≤2) | 43 (40) | 21 (49) | 3 (21) | |
| Uncertain malignant potential (3) | 29 (27) | 10 (23) | 4 (29) | |
| Carcinoma (≥4) | 27 (25) | 8 (19) | 7 (50) | |
| Missing data | 9 (8) | 4 (9) | 0 | |
| Treatment, n (%) | 0.0005 | |||
| Surgery only | 70 (65) | 35 (81) | 4 (29) | |
| Surgery and chemotherapy | 36 (33) | 8 (19) | 10 (71) | |
| Missing data | 2 (2) | 0 | 0 | |
| Post-surgery recurrence/metastasis, n (%) | <0.0001* | |||
| No | 86 (79) | 40 (93) | 5 (36) | |
| Yes | 19 (18) | 3 (7) | 9 (64) | |
| Missing data | 3 (3) | 0 | 0 | |
| Outcome, n (%) | <0.0001* | |||
| Alive and free of disease | 74 (69) | 36 (84) | 5 (36) | |
| Alive with active disease | 6 (5) | 2 (5) | 1 (7) | |
| Deceased | 17 (16) | 1 (2) | 8 (57) | |
| Lost to follow-up | 11 (10) | 4 (9) | 0 | |
| Follow-up (years) | 0.3 | |||
| Median (range) | 6.8 (0.02 to 22.3) | 5.1 (0.2 to 22.3) | 2.6 (0.5 to 18.2) | |
| Mean | 8 | 5.6 | 5.2 | |
(Continued)
Downloaded from https://academic.oup.com/ejendo/article/186/5/573/6853694 by National Library of Medicine user on 03 April 2026
| Clinical Study | A C Bueno and others | VDR impairment in pediatric ACT | 186:5 | 578 |
|---|---|---|---|---|
| Features | All patients | VDR methylation | P-value* | |
|---|---|---|---|---|
| Low | High | |||
| Molecular | ||||
| p53 p.R337H germline mutation, n (%) | 0.99* | |||
| Non-carrier | 10 (9) | 4 (9) | 1 (7) | |
| Carrier | 92 (85) | 37 (86) | 12 (86) | |
| Missing data | 6 (6) | 2 (5) | 1 (7) | |
| CTNNB1-activating somatic mutations, n (%) | 0.34* | |||
| WT | 89 (83) | 36 (84) | 10 (72) | |
| Mutant | 10 (9) | 4 (9) | 3 (21) | |
| Missing data | 9 (8) | 3 (7) | 1 (7) | |
| VDR mRNA (2-44Ct) | 0.027 | |||
| Median (range) | 0.15 (0.0008 to 3.509) | 0.19 (0.0008 to -3.51) | 0.09 (0.0008 to 1.47) | |
| Mean | 0.5 | 0.55 | 0.17 | |
| VDR nuclear immunoreactivity*, n (%) | 0.46+ | |||
| Positive (H-score ≥ 1) | 20 (38) | 11 (26) | 6 (43) | |
| Negative (H-score < 1) | 22 (42) | 11 (26) | 3 (21) | |
| Not available | 10 (20) | 21 (48) | 5 (36) | |
| VDR mean methylation ((M-values), n = 57) | <0.0001 | |||
| Median (range) | 0.2 (-0.497 to 1.124) | 0.142 (-0.497 to 0.429) | 0.67 (0.412 to 1.124) | |
| Mean | 0.28 | 0.136 | 0.72 | |
*Data available only for children from FMRP-USP; +Local disease tumor stages (I/II) were compared to non-localized disease tumor stages (III/IV); * Patients with missing data or those who were lost to follow-up were not included in the analysis; * Comparing methylation groups. FMRP-USP, Ribeirao Preto Medical School, University of Sao Paulo; BCC-SUC, Boldrini Children’s Center, State University of Campinas.
A
Age at diagnosis
B
Disease stage
C
Histopathological classification
Overall survival (%)
100
100
I (n = 59)
100
< 4 years (n = 85)
Overall survival (%)
Il (n = 23)
Overall survival (%)
ACA (n = 43)
75
75-
III (n = 14)
75
UMP (n = 29)
50
4-12 years (n = 18)
50
Log-rank, p-values:
I vs Il: 0.02
50
ACC (n = 27)
13-20 years (n = 5)
25
Log-rank, p-value:
I vs III: 0.008
< 4 vs 4-12 years: < 0.0001
25-
IV (n = 11)
I vs IV: < 0.0001
25
Log-rank, p-value:
ACA vs UMP: 0.0357
< 4 vs 13-20 years: < 0.0001
Il vs III: 0.74
0
4-12 vs 13-20 years: 0.3728
Il vs IV: 0.001
ACA vs ACC: < 0.0001
0
Ill vs IV: 0.01
0
UMP vs ACC: 0.0253
0
4
8
12
16
20
24
0
4
8
12
16
20
24
0
4
8
12
16
20
24
Time (years)
Time (years)
Time (years)
D
Clinical presentation
E
CTNNB1 exon 3
Overall survival (%)
100
Virilization (n = 83)
Overall survival (%)
100
75-
Asymptomatic (n = 12)
75-
Wild type (n = 89)
Log-rank, p-values:
Virilizaton vs Asymptomatic: 0.25
Virilization vs Virilization + Cushing’s Sd: < 0.0001
50-
Mutant (n = 10)
Virilization vs Cushing’s Sd: < 0.0001
Virilization + Cushing’s Sd (n = 6)
50-
Asymptomatic vs Virilization + Cushing’s Sd: 0.03 Asymptomatic vs Cushing’s Sd: 0.02
25-
Log-rank:
Cushing’s Sd (n = 7)
25-
p = 0.008
Virilization + Cushing’s Sd vs Cushing’s Sd: 0.91
HR: 3.5
0
0
95% CI: 0.6 to 20
0
4
8
12
16
20
24
0
4
8
12
16
20
24
Time (years)
Time (years)
gestation. This pattern of expression persisted in mature postnatal adrenals although expression was lower. Figure 3A exhibits representative cases of each development age.
VDR expression in pACT
H-score concordance coefficient was substantial (kappa=0.79). Among the 42 pACT samples evaluated, 31 were from patients diagnosed before 4 years (17 negative; 55%), 8 from those diagnosed between 4and 12 years (2 negative; 25%), and 3 from those diagnosed after 13 years of age (100% negative). VDR expression was negative in most pACT (52.4%, Fig. 3B and Table 1). In agreement, despite the variability among the tumors, VDR mRNA levels were lower in pACT than in normal adrenal cortices (P=0.002; Fig. 3C), especially in UMP and ACC (P=0.01; Fig. 3D). None of the patients diagnosed after 13 years of age had positive immunoreactivity, and there was a tendency of association between negative VDR immunoreactivity and advanced age at diagnosis (P=0.07). Neither VDR nuclear immunoreactivity nor mRNA expression was associated with other patients’ clinical features or outcomes. The anti-VDR antibody and the gene expression assay did not map to the same genomic location but both targeted VDR transcript and protein variants.
VDR methylation in pediatric ACT
The subset of patients whose tumor DNA methylation profiling was analyzed (n= 57) reflected the baseline features observed in the whole cohort. Most of the patients were girls (n= 40, 57%; female-to-male ratio of 2.3:1), with median age at diagnosis of 2.1 years (range: 0.2-16.6). Fourteen (25%) patients were >4 years of age at diagnosis, among which 10 (18%) were between 4 and 12 years and 4 (7%) were >13 years. Most patients presented with virilization (n= 43, 74%), followed by asymptomatic (n= 11, 13%), Cushing syndrome (11%), and virilization associated with Cushing syndrome (n= 1, 2%). Eleven patients (19%) presented with non-localized/advanced (stages III and IV) and 6 (10%) with metastatic disease (stage IV). Histologically, 24 tumors were diagnosed as ACA (42%), 14 (25%) as of UMP, and 15 (26%) as ACC. Eighteen patients (32%) underwent adjuvant chemotherapy, and 12 (21%) experienced post-surgery recurrence or metastasis. Median follow-up was 4.4 years (range: 0.02-22.3), and to date, nine patients (16%) died from the disease. Forty-nine patients (86%) were carriers of the p53 p.R337H mutation and seven (12%), of CTNNB1 mutations.
UHCA of VDR methylation data identified two clusters of pACT (Fig. 4), which differed regarding methylation
signature and were labeled as high (H-VMC) and low (L-VMC) VDR methylation clusters.
VDR methylation was associated with VDR mRNA expression and with prognostic features (Fig. 5A and Table 1). Compared with L-VMC (n= 43), tumors in the H-VMC (n= 14) presented lower VDR mRNA levels (P=0.027, Fig. 5B). This cluster comprised a higher frequency of patients diagnosed after age 4 years (P=0.002), being enriched for patients between ages 4 and 12 (36%) and older than age 13 (21%) (P=0.003). H-VMC was also enriched with patients who presented with Cushing syndrome (P=0.03), non-localized/advanced disease (P=0.01), and metastatic disease (P=0.027). In addition, H-VMC was enriched with those who, during follow-up, needed adjuvant chemotherapy (P=0.0005), experienced post- surgery recurrence or metastasis (P < 0.0001), died from the disease (P < 0.0001), and had lower DFS and OS (P < 0.0001; Fig. 5C and D). When considering the variables associated with unfavorable outcomes, these patients had worse outcomes. When considering the subgroup of patients diagnosed < 4 years or those between 4-12 and 13-20 years, the patients in H-VMC presented or tended to present reduced DFS and OS (Supplementary Fig. 1). Similar results were observed when considering the presence of non-localized/advanced disease, carcinomas, Cushing syndrome, and CTNNB1 mutations (Supplementary Fig. 2).
Among the VDR methylation probes, 6/53 mapped to VDR intron 2, which comprehends the genomic location interval, but is excluded from the design of the VDR gene expression assay used. Moreover, 4/53 probes mapped to the genomic region encompassing the anti-VDR antibody (Supplementary Table 2). Twenty-two probes (41.5%) were differentially methylated between the clusters (P < 0.01) and, among them, 16 (30%) were more stringently differentially methylated (P < 0.005). We observed DMPs distributed throughout VDR extension (Fig. 4 and Supplementary Table 2): intergenic (n= 2), TSS1500 (n=1), 5’UTR (n=3), 3’UTR (n=3), body (n= 12), and exon boundary (n= 1) regions. The three most DMPs are located in promoter flanks, including one annotated in the gene expression assay region (cg06487630, Rank #2), which is also located in a CTCF-binding region (Supplementary Fig. 3). The two intergenic probes - including the most DMP cg07347128 - are currently annotated within the promoter flank and CTCF binding region of the VDR-202 isoform (Supplementary Fig. 4). Methylation at these sites was associated with patients with unfavorable prognosis, while unmethylation was associated with patients with favorable prognosis.
A
B
1 st trimester
1 st trimester
ACA
ACA
2nd trimester
2nd trimester
UMP
UMP
20 jim
3rd trimester
3rd trimester
C
D
**
7
6
p = 0.002
7
*
·
6
mRNA (2-44Ct)
A
4
5
p = 0.01
mRNA (2-44Ct)
4
3
3
2
2
20 pm
1
1
1
Postnatal
Postnatal
0
0
Control (n = 14)
pACT (n = 108)
Control ACA
UMP
ACC
(n = 14) (n = 43) (n = 29) (n = 27)
20 pm
Figure 3 VDR expression in normal adrenal cortex development and in pACT. Representative VDR immunostaining (brown) and H&E (red) images of (A) normal adrenal cortex development in the first and second gestational trimesters, showing VDR expression in cellular nuclei and cytoplasm in both the definitive and the transition zones of fetal adrenals; in late gestation (third trimester) and in all zones of mature postnatal adrenals (7 years of age). In the fetal adrenals, white arrows correspond to capsule, gray arrows to the definitive cortex, and black arrows transitional/fetal cortex. In the postnatal adrenal, the white arrow corresponds to capsule, blue arrow to zona glomerulosa, green arrow to zona fasciculata, and yellow arrow to zona reticularis. (B) Positive (ACA; H-score = 2; pACT003) and negative (UMP; H-score = 0; pACT009) VDR nuclear immunoreactivity in pACT. Microscopic magnification, 40x. Scale bar, 20 um. (C) VDR mRNA levels in normal adrenals (control) and in pACT and (D) according to histopathologic classification. Data are expressed as 2-44Ct and presented as individual sample values and median (black lines). Mann-Whitney’s and Kruskal-Wallis’ P-values are indicated. Dunnett’s multiple comparison test: ** P < 0.01. pACT, pediatric adrenocortical tumors; ACA, adrenocortical adenoma; UMP, uncertain malignant potential; ACC, adrenocortical carcinoma.
Discussion
VDR is broadly expressed in human tissues and regulates a plethora of genomic events involved, among others, with tumor suppression mechanisms (14, 15, 38). Specifically,
VDR has been shown to be underexpressed in different types of cancers and to be a predictive biomarker of disease prognosis (21, 22, 23, 24). Herein, we evaluated how VDR expression behaves during normal human adrenocortical development and observed VDR nuclear expression in
Probes
1
18
19
20
21
2
22
17
3
8
14
5
13
7
12
15
9
16
11
10
6
Ranking probes
adj. p-value < 0.01
Group
Methylation
mRNA
cg02470587
cg22833603 cg07347128
cg26333618
cg16998563
cg13556224
cg16321474
cg06369854
cg03137447
cg11843835
cg25911279
cg02522757
cg23654431
cg13865595
cg10195011
cg27537561
cg14311020
cg02547054
cg06195428
cg23190711
cg25293778
cg05190176 cg05285969
cg06141400
cg02714932
cg19228134
cg01886921
cg24715264
cg10037049
cg13173254
cg13192508
cg21240834
cg18774471
cg06487630
cg21548941
cg19008693
cg09135639
cg00357860
cg24962956
cg08726078
cg01208955
cg07290465
cg25219939 cg01077720
cg10592901
cg16713110
cg12045556
cg17517241
cg03661582
cg14854850
cg13301841
cg22054735
cg21843272
adj. p-value < 0.005
Age
pACT020
PACT058
PACT009
pACT040
PACT090
PACT068
PACT006
PACT061
pACT013
pACT021
pACT010
PACT028
PACT014
PACT017
PACT055
pACT092
pACT054
PACT087
PACT102
LOW
PACT067
PACT098
PACT070
PACT065
pACT045
PACT049
PACT107
PACT095
pACT081
DACT088
PACT053
DACT048
PACT048
PACT059
PACT094
pACT026
PACT037
DACT078
pACT046
PACT003
PACT085
PACT093
pACT096
ACT060
PACT062
PACTO30
PACT025
PACT035
PACT075
PACT036
HIGH
DACT038
PACT097
DACT073
PACT064
PACT001
PACT050
DACT029
pACT101
pACT012
5’
VDR gene
3’
-4
-2
0
2
4
-10
-4
-3
-1
1
0
1
2
4
17
M-values
VDR mRNA (log2)
Age (years)
Figure 4 VDR DNA methylation in pACT. (A) Unsupervised analysis of VDR methylation data showing two groups of pACT: high (red) and Low (blue) VDR methylation. The dendrogram based on pACT methylation data alongside individual samples; median M-values, VDR mRNA levels, and age distributionare shown in vertical on the left. Heatmap demonstrating VDR methylation profiles evidencing the differentially methylated probes (DMPs; top extremity) and their location on VDR span (Supplementary Table 2). Patient identification is exhibited (right extremity).
the definitive zone of adrenal cortices from midgestation to postnatal life. Additionally, studying a large cohort of pACT, we demonstrated that VDR hypermethylation is associated with VDR mRNA underexpression and unfavorable outcomes.
We observed progressively lower levels of VDR expression in the definitive zone of adrenal cortices from midgestation to postnatal life. Interestingly, there was a robust VDR expression in cells in the definitive and transitional zones, as well as in the subcapsular adrenal cortex, which may indicate a role for this receptor during normal adrenal development. The capsule and subcapsular cortex are a niche for progenitor/stem adrenal cells, which are known to differentiate and centripetally originate/ repopulate the definitive adrenal cortex during adrenal development and throughout life (39). Not only progenitor/ stem cells but also cells from the fetal cortex collaborate for the generation of the mature cortex. Different signaling
pathways are involved in this mechanism, including Wnt/ B-catenin and Sonic Hedgehog (Shh)/Gli (40, 41). VD3/VDR signaling interacts with both pathways (22, 42), which are deregulated during adrenocortical tumorigenesis. We have previously demonstrated that SHH and GLI1 are expressed throughout the fetal and the definitive zones of the adrenal cortex in earlier stages life but tended to become more intense in the capsule and adjacencies in later stages and postnatal life (43). Interestingly, our study shows that VDR expression progressively weakened but stood positive in all layers of the adrenal cortex in late gestation and in postnatal adrenals.
A minority of the pACT presented VDR nuclear immunoreactivity, demonstrating that VDR expression is abolished in these tumors. In agreement, VDR mRNA levels were also reduced in pACT, mainly in tumors with UMP and in ACC. Hence, active VDR may play a role in adrenal cortex physiology, which is lost during adrenocortical
A
VDR methylation cluster
p-value
VDR mRNA
0.027
VDR immunoreactivity
0.46
Age at diagnosis
0.003
Clinical presentation
0.03
Metastasis at diagnosis
0.03
Recurrence / metastasis
<0.0001
Histopathological diagnosis
0.06
Outcome
<0.0001
Germline p53 p.R337H mutation
0.99
Somatic CTNNB1 mutations
0.34
Patient ID
pACT012
pACT101
pACT029
pACT050
pACT001
pACT064
pACT073
pACT097
pACT038
pACT036
pACT075
pACT035
pACT025
pACT030
pACT062
pACT060
pACT096
pACT093
pACT085
pACT003
pACT046
pACT078
pACT037
pACT026
pACT094
pACT059
pACT048
pACT053
pACT088
pACT081
pACT095
pACT107
pACT049
pACT045
pACT065
pACT070
pACT098
pACT067
pACT102
pACT087
pACT054
pACT092
pACT055
pACT017
pACT014
pACT028
pACT010
pACT021
pACT013
pACT061
pACT006
pACT068
pACT090
pACT040
pACT009
pACT058
pACT020
VDR methylation
VDR mRNA
VDR immunoreactivity
Age at diagnosis
Clinical presentation
High
Minimum
Positive
< 4 years
☐ Virilization
Low
25% Percentile
Negative
4-12 years
☐ Cushing’s syndrome
Median
Not evaluated
13 years
☐ Cushing’s syndrome and virilization
75% Percentile
Asymptomatic
Maximum
Metastasis at diagnosis Recurrence / metastasis
Histopathological diagnosis
Outcome
Mutational status
ACA
☐ Alive and disease free
Yes
UMP
☐ Alive with active disease
Wild-type
☐ No
Mutant
ACC
☐ Not available
Deceased
☐ Not available
☐ Lost to follow-up
B
C
D
VDR mRNA
Overall survival
-NW.A
Disease free survival
p = 0.027
100
100
0.6
Percent survival
Low (n=43)
Percent survival
Low (n = 43)
75
75
2-44Ct
0.4
50
High (n=14)
50
High (n = 14)
0.2
25
25
Log-rank: p< 0.0001 HR: 10.7; 95% CI: 2.4 to 47.2
Log-rank: p < 0.0001
0.0
0
0
HR: 28.4; 95% CI: 5.9 to 137.8
High (n = 14)
Low (n = 43)
0
4
8
12
16
20
24
0
4
8
12
16
20
24
Time (years)
Time (years)
Methylation
tumorigenesis. Intriguingly, we found a tendency toward VDR absence in pACT from older patients (>13 years), who tend to present more aggressive disease. In our study, all samples were obtained from pharmacological treatment- naïve pediatric patients, suggesting that impaired VDR expression is a feature of ACT, regardless of mitotane treatment. Of note, histology, clinical presentation, outcomes, and molecular features associated to adult ACC are also seen in children, independent of age at diagnosis.
Thus, since pediatric patients were enrolled based on age at diagnosis (<20 years), some pACT can behave as ‘adult- like’ tumors. VDR upregulation was previously observed in colorectal and breast tumors during early tumorigenesis as a physiological defense system against epithelial tumor progression, but once tumors were dedifferentiated, VDR expression declined (44, 45). Nevertheless, neither VDR mRNA nor protein expression was associated with disease staging in our study, suggesting that VDR underexpression
is present throughout adrenocortical tumorigenesis and might be a driver for tumor development and progression.
VDR and the histones that regulate the access to its transcriptional machinery are modified by different epigenetic changes, including methylation (15, 38). To our knowledge, there are no data on whether VDR expression is affected by epigenetic modifications in pACT. Our results show VDR high methylation signature in tumors with lower mRNA expression, which may partially explain VDR underexpression in pACT. It is known that gene promoter methylation is involved in the transcriptional silencing of various tumor suppressor genes, which is a common epigenetic event in the early phase of tumorigenesis (46). In agreement, most DMPs distinguishing the pACT clusters were annotated in VDR promoters and/or promoter flanks.
Unlike in adult ACC, besides higher methylation in VDR promoter regions, we observed differential methylation in several regions lengthwise the gene span in pACT. It has been recently shown that adults and pACT have distinct methylation patterns, with prognostic relevance (8, 9, 10). In tumors from adult patients, the CpG island hypermethylation phenotype (CIMP) was associated with unfavorable outcomes (8, 9). In pACT, global tumor hypomethylation was associated with unfavorable outcome (10). Interestingly, unlike global pACT hypomethylation, higher VDR methylation was associated with advanced disease at diagnosis and adverse outcomes in our study. We showed that, among the patients presenting with unfavorable disease features, those whose tumors had high VDR methylation were the ones with poorer DFS and OS. Together, these observations suggest that tumor VDR methylation status itself can be considered a prognostic biomarker for pediatric patients with ACT.
It has been recently demonstrated that higher VDR gene expression, in part associated with VDR promoter hypomethylation, is an independent protective factor against death in adult patients with melanoma (24). Moreover, VDR signaling activation inhibited the Wnt/ B-catenin pathway in melanoma cells (24). The Wnt/B- catenin signaling controls cell proliferation and the upregulation of cell cycle progression is one of the major mechanisms involved in ACT pathogenesis (12, 13, 47, 48).
Preclinical and early phase clinical studies have shown the antiproliferative effects of VD3 analogs and VDR signaling activation (49), highlighting its repression over the B-catenin activity in different types of cancers (21, 22, 23). Preclinical studies using the H295R ACC cell line demonstrated that physiological and supraphysiological doses of calcitriol inhibited adrenocortical tumor proliferation and also reduced B-catenin’s expression (16,
50). An additive effect on subtherapeutic doses of mitotane over H295R cell proliferation was also demonstrated, suggesting a beneficial therapeutic effect of VD3 as a co-adjuvant agent (50). Moreover, in vitro studies showed that calcitriol reduces steroid secretion from H295R cells (16, 51). Thus, the activation of VDR signaling in ACT could also be beneficial by decreasing patient morbidity, especially for children, who may experience severe consequences of prolonged exposure to steroids in adulthood (52).
Due to the retrospective character of our pACT cohort, we were not able to determine 25(OH)VD3 serum levels or VD3 sufficiency status in the patients. Moreover, we did not investigate the presence of VDR mutations and SNPs associated with 25(OH)VD3 levels and cancer predisposition or progression (14). These limitations of our study must be considered and can be addressed in future studies.
In summary, our data suggest that VDR has an important role in normal adrenal differentiation and maintenance, which is impaired in pediatric adrenocortical tumorigenesis. VDR gene hypermethylation may result in its underexpression and is associated with unfavorable outcomes in pediatric patients with ACT, suggesting VDR methylation status as a prognostic biomarker.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/ EJE-21-0879.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by the Coordination for the Improvement of Higher Education Personnel (A C B) and the Sao Paulo State Research Foundation (FAPESP) grants 14/03989-6 (M C and S R A), 15/19663-5 (S R A), 19/00860-6 (A C B), and 20/03835-0 (J M G).
Acknowledgements
The authors thank Prof. Maria J Mastellaro MD, Ph.D., (Boldrini Children’s Center) for the patient’s clinical follow-up, Rogerio Zuliani, Renata Sicchieri, and Wendy Turatti (Laboratory of Molecular Endocrinology; FMRPUSP) for technical support, and Prof. Leandro M Colli, MD, Ph.D., (FMRP-USP) for the revision of the manuscript.
References
1 Fassnacht M, Kroiss M & Allolio B. Update in adrenocortical carcinoma. Journal of Clinical Endocrinology and Metabolism 2013 98 4551-4564. (https://doi.org/10.1210/jc.2013-3020)
2 Antonini SR, Leal LF & Cavalcanti MM. Pediatric adrenocortical tumors: diagnosis, management and advancements in the understanding of the genetic basis and therapeutic implications. Expert Review of Endocrinology and Metabolism 2014 9 445-464. (https:// doi.org/10.1586/17446651.2014.941813)
3 Sandrini R, Ribeiro RC & DeLacerda L. Childhood adrenocortical tumors. Journal of Clinical Endocrinology and Metabolism 1997 82 2027-2031. (https://doi.org/10.1210/jcem.82.7.4057)
4 Ribeiro RC, Sandrini F, Figueiredo B, Zambetti GP, Michalkiewicz E, Lafferty AR, DeLacerda L, Rabin M, Cadwell C, Sampaio G et al. An inherited p53 mutation that contributes in a tissue-specific manner to pediatric adrenal cortical carcinoma. PNAS 2001 98 9330-9335. (https://doi.org/10.1073/pnas.161479898)
5 Latronico AC, Pinto EM, Domenice S, Fragoso MC, Martin RM, Zerbini MC, Lucon AM & Mendonca BB. An inherited mutation outside the highly conserved DNA-binding domain of the p53 tumor suppressor protein in children and adults with sporadic adrenocortical tumors. Journal of Clinical Endocrinology and Metabolism 2001 86 4970-4973. (https://doi.org/10.1210/jcem.86.10.7957)
6 Tucci S, Martins ACP, Suaid HJ, Cologna AJ & Reis RB. The impact of tumor stage on prognosis in children with adrenocortical carcinoma. Journal of Urology 2005 174 2338-2342, discussion 2342. (https://doi. org/10.1097/01.ju.0000180645.89054.fd)
7 Megerle F, Herrmann W, Schloetelburg W, Ronchi CL, Pulzer A, Quinkler M, Beuschlein F, Hahner S, Kroiss M, Fassnacht M et al. Mitotane monotherapy in patients with advanced adrenocortical carcinoma. Journal of Clinical Endocrinology and Metabolism 2018 103 1686-1695. (https://doi.org/10.1210/jc.2017-02591)
8 Zheng S, Cherniack AD, Dewal N, Moffitt RA, Danilova L, Murray BA, Lerario AM, Else T, Knijnenburg TA, Ciriello G et al. Comprehensive pan-genomic characterization of adrenocortical carcinoma. Cancer Cell 2016 29 723-736. (https://doi.org/10.1016/j.ccell.2016.04.002)
9 Barreau O, Assié G, Wilmot-Roussel H, Ragazzon B, Baudry C, Perlemoine K, René-Corail F, Bertagna X, Dousset B, Hamzaoui N et al. Identification of a CpG island methylator phenotype in adrenocortical carcinomas. Journal of Clinical Endocrinology and Metabolism 2013 98 E174-E184. (https://doi.org/10.1210/jc.2012-2993)
10 Clay MR, Pinto EM, Cline C, Tran QT, Lin T, Dyer MA, Shi L, Wu H, Pounds SB, Zambetti GP et al. DNA methylation profiling reveals prognostically significant groups in pediatric adrenocortical tumors: a report from the International Pediatric Adrenocortical Tumor Registry. JCO Precision Oncology 2019 3 1-21. (https://doi.org/10.1200/ PO.19.00163)
11 de Reyniès A, Assié G, Rickman DS, Tissier F, Groussin L, René-Corail F, Dousset B, Bertagna X, Clauser E & Bertherat J. Gene expression profiling reveals a new classification of adrenocortical tumors and identifies molecular predictors of malignancy and survival. Journal of Clinical Oncology 2009 27 1108-1115. (https://doi.org/10.1200/JCO.2008.18.5678)
12 Assié G, Letouzé E, Fassnacht M, Jouinot A, Luscap W, Barreau O, Omeiri H, Rodriguez S, Perlemoine K, René-Corail F et al. Integrated genomic characterization of adrenocortical carcinoma. Nature Genetics 2014 46 607-612. (https://doi.org/10.1038/ng.2953)
13 Pinto EM, Chen X, Easton J, Finkelstein D, Liu Z, Pounds S, Rodriguez- Galindo C, Lund TC, Mardis ER, Wilson RK et al. Genomic landscape of paediatric adrenocortical tumours. Nature Communications 2015 6 6302. (https://doi.org/10.1038/ncomms7302)
14 Bouillon R, Marcocci C, Carmeliet G, Bikle D, White JH, Dawson- Hughes B, Lips P, Munns CF, Lazaretti-Castro M, Giustina A et al. Skeletal and extraskeletal actions of vitamin D: current evidence and outstanding questions. Endocrine Reviews 2019 40 1109-1151. (https:// doi.org/10.1210/er.2018-00126)
15 Bikle DD. Vitamin D: newer concepts of its metabolism and function at the basic and clinical level. Journal of the Endocrine Society 2020 4 bvz038. (https://doi.org/10.1210/jendso/bvz038)
16 Pilon C, Urbanet R, Williams TA, Maekawa T, Vettore S, Sirianni R, Pezzi V, Mulatero P, Fassina A, Sasano H et al. 1a,25-Dihydroxyvitamin
D3 inhibits the human H295R cell proliferation by cell cycle arrest: a model for a protective role of vitamin D receptor against adrenocortical cancer. Journal of Steroid Biochemistry and Molecular Biology 2014 140 26-33. (https://doi.org/10.1016/j.jsbmb.2013.11.008)
17 Pilon C, Rebellato A, Urbanet R, Guzzardo V, Cappellesso R, Sasano H, Fassina A & Fallo F. Methylation status of vitamin D receptor gene promoter in benign and malignant adrenal tumors. International Journal of Endocrinology 2015 2015 375349. (https://doi.org/10.1155/2015/375349)
18 Gaujoux S, Grabar S, Fassnacht M, Ragazzon B, Launay P, Libé R, Chokri I, Audebourg A, Royer B, Sbiera S et al. Beta-catenin activation is associated with specific clinical and pathologic characteristics and a poor outcome in adrenocortical carcinoma. Clinical Cancer Research 2011 17 328-336. (https://doi.org/10.1158/1078-0432.CCR-10-2006)
19 Mermejo LM, Leal LF, Colli LM, Fragoso MC, Latronico AC, Tone LG, Scrideli CA, Tucci S, Martinelli CE, Yunes JA et al. Altered expression of noncanonical Wnt pathway genes in paediatric and adult adrenocortical tumours. Clinical Endocrinology 2014 81 503-510. (https://doi.org/10.1111/cen.12462)
20 Leal LF, Mermejo LM, Ramalho LZ, Martinelli CE Jr, Yunes JA, Seidinger AL, Mastellaro MJ, Cardinalli IA, Brandalise SR, Moreira AC et al. Wnt/ß-catenin pathway deregulation in childhood adrenocortical tumors. Journal of Clinical Endocrinology and Metabolism 2011 96 3106-3114. (https://doi.org/10.1210/jc.2011-0363)
21 Aguilera O, Peña C, García JM, Larriba MJ, Ordóñez-Morán P, Navarro D, Barbáchano A, López de Silanes I, Ballestar E, Fraga MF et al. The Wnt antagonist DICKKOPF-1 gene is induced by 1,25-dihydroxyvitamin D3 associated to the differentiation of human colon cancer cells. Carcinogenesis 2007 28 1877-1884. (https://doi. org/10.1093/carcin/bgm094)
22 Beildeck ME, Islam M, Shah S, Welsh J & Byers SW. Control of TCF-4 expression by VDR and vitamin D in the mouse mammary gland and colorectal cancer cell lines. PLoS ONE 2009 4 e7872. (https://doi. org/10.1371/journal.pone.0007872)
23 Johnson AL, Zinser GM & Waltz SE. Vitamin D3-dependent VDR signaling delays ron-mediated breast tumorigenesis through suppression of ß-catenin activity. Oncotarget 2015 6 16304-16320. (https://doi.org/10.18632/oncotarget.4059)
24 Muralidhar S, Filia A, Nsengimana J, Poźniak J, O’Shea SJ, Diaz JM, Harland M, Randerson-Moor JA, Reichrath J, Laye JP et al. Vitamin D-VDR signaling inhibits Wnt/beta-catenin-mediated melanoma progression and promotes anti-tumor immunity. Cancer Research 2019 79 5986-5998. (https://doi.org/10.1158/0008-5472.CAN-18-3927)
25 Ribeiro RC, Pinto EM, Zambetti GP & Rodriguez-Galindo C. The International Pediatric Adrenocortical Tumor Registry initiative: contributions to clinical, biological, and treatment advances in pediatric adrenocortical tumors. Molecular and Cellular Endocrinology 2012 351 37-43. (https://doi.org/10.1016/j.mce.2011.10.015)
26 Wieneke JA, Thompson LDR & Heffess CS. Adrenal cortical neoplasms in the pediatric population: a clinicopathologic and immunophenotypic analysis of 83 patients. American Journal of Surgical Pathology 2003 27 867-881. (https://doi. org/10.1097/00000478-200307000-00001)
27 Das S, Sengupta M, Islam N, Roy P, Datta C, Mishra PK, Banerjee S, Chaudhuri MK & Chatterjee U. Weineke criteria, Ki-67 index and p53 status to study pediatric adrenocortical tumors: is there a correlation? Journal of Pediatric Surgery 2016 51 1795-1800. (https:// doi.org/10.1016/j.jpedsurg.2016.07.014)
28 Jehangir S, Nanjundaiah P, Sigamani E, Burad D, Manipadam MT, Lea V, Ly T & Holland AJA. Pathological prognostication of paediatric adrenocortical tumours: is a gold standard emerging? Pediatric Blood and Cancer 2018 11 e27567. (https://doi.org/10.1002/pbc.27567)
29 Martins-Filho SN, Almeida MQ, Soares I, Wakamatsu A, Alves VAF, Fragoso MCBV & Zerbini MCN. Clinical impact of pathological features including the Ki-67 labeling index on diagnosis and prognosis of adult and pediatric adrenocortical tumors. Endocrine Pathology 2021 32 288-300. (https://doi.org/10.1007/s12022-020-09654-x)
30 Al-Haddad S, Zhang Z, Leygue E, Snell L, Huang A, Niu Y, Hiller- Hitchcock T, Hole K, Murphy LC & Watson PH. Psoriasin (S100A7) expression and invasive breast cancer. American Journal of Pathology 1999 155 2057-2066. (https://doi.org/10.1016/S0002-9440(10)65524-1)
31 Landis JR & Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977 33 159-174. (https://doi. org/10.2307/2529310)
32 Abduch RH, Bueno AC, Leal LF, Cavalcanti MM, Gomes DC, Brandalise SR, Masterallo MJ, Yunes JA, Martinelli Jr CE, Tone LG et al. Unraveling the expression of the oncogene YAP1, a Wnt/beta-catenin target, in adrenocortical tumors and its association with poor outcome in pediatric patients. Oncotarget 2016 7 84634-84644. (https://doi. org/10.18632/oncotarget.12382)
33 Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD & Irizarry RA. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics 2014 30 1363-1369. (https://doi. org/10.1093/bioinformatics/btu049)
34 Du P, Zhang X, Huang CC, Jafari N, Kibbe WA, Hou L & Lin SM. Comparison of Beta-value and M-value methods for quantifying methylation levels by microarray analysis. BMC Bioinformatics 2010 11 587. (https://doi.org/10.1186/1471-2105-11-587)
35 Touleimat N & Tost J. Complete pipeline for Infinium® human methylation 450K BeadChip data processing using subset quantile normalization for accurate DNA methylation estimation. Epigenomics 2012 4 325-341. (https://doi.org/10.2217/epi.12.21)
36 Pidsley R, Zotenko E, Peters TJ, Lawrence MG, Risbridger GP, Molloy P, Van Djik S, Muhlhausler B, Stirzaker C & Clark SJ. Critical evaluation of the Illumina MethylationEPIC BeadChip microarray for whole- genome DNA methylation profiling. Genome Biology 2016 17 208. (https://doi.org/10.1186/s13059-016-1066-1)
37 Tian Y, Morris TJ, Webster AP, Yang Z, Beck S, Feber A & Teschendorff AE. ChAMP: updated methylation analysis pipeline for Illumina BeadChips. Bioinformatics 2017 33 3982-3984. (https://doi. org/10.1093/bioinformatics/btx513)
38 Long MD, Sucheston-Campbell LE & Campbell MJ. Vitamin D receptor and RXR in the post-genomic era. Journal of Cellular Physiology 2015 230 758-766. (https://doi.org/10.1002/jcp.24847)
39 Wood MA & Hammer GD. Adrenocortical stem and progenitor cells: unifying model of two proposed origins. Molecular and Cellular Endocrinology 2011 336 206-212. (https://doi.org/10.1016/j. mce.2010.11.012)
40 Simon DP & Hammer GD. Adrenocortical stem and progenitor cells: implications for adrenocortical carcinoma. Molecular and Cellular Endocrinology 2012 351 2-11. (https://doi.org/10.1016/j.mce.2011.12.006)
41 King P, Paul A & Laufer E. Shh signaling regulates adrenocortical development and identifies progenitors of steroidogenic lineages. PNAS 2009 106 21185-21190. (https://doi.org/10.1073/ pnas.0909471106)
42 Uhmann A, Niemann H, Lammering B, Henkel C, Hess I, Nitzki F, Fritsch A, Prüfer N, Rosenberger A, Dullin C et al. Antitumoral effects of calcitriol in basal cell carcinomas involve inhibition of
hedgehog signaling and induction of vitamin D receptor signaling and differentiation. Molecular Cancer Therapeutics 2011 10 2179-2188. (https://doi.org/10.1158/1535-7163.MCT-11-0422)
43 Gomes DC, Leal LF, Mermejo LM, Scrideli CA, Martinelli CE Jr, Fragoso MC, Latronico AC, Tone LG, Tucci S, Yunes JA et al. Sonic hedgehog signaling is active in human adrenal cortex development and deregulated in adrenocortical tumors. Journal of Clinical Endocrinology and Metabolism 2014 99 E1209-E1216. (https://doi. org/10.1210/jc.2013-4098)
44 Cross HS, Bareis P, Hofer H, Bischof MG, Bajna E, Kriwanek S, Bonner E & Peterlik M. 25-Hydroxyvitamin D3-1a-hydroxylase and vitamin D receptor gene expression in human colonic mucosa is elevated during early cancerogenesis. Steroids 2001 66 287-292. (https://doi. org/10.1016/s0039-128x(00)00153-7)
45 Lopes N, Sousa B, Martins D, Gomes M, Vieira D, Veronese LA, Milanezi F, Paredes J, Costa JL & Schmitt F. Alterations in vitamin D signaling and metabolic pathways in breast cancer progression: a study of VDR, CYP27B1 and CYP24A1 expression in benign and malignant breast lesions. BMC Cancer 2010 10 483. (https://doi.org/10.1186/1471- 2407-10-483)
46 Laird PW. The power and the promise of DNA methylation markers. Nature Reviews: Cancer 2003 3 253-266. (https://doi.org/10.1038/ nrc1045)
47 Szabó PM, Tamási V, Molnár V, Andrásfalvy M, Tömböl Z, Farkas R, Kövesdi K, Patócs A, Tóth M, Szalai C et al. Meta-analysis of adrenocortical tumour genomics data: novel pathogenic pathways revealed. Oncogene 2010 29 3163-3172. (https://doi.org/10.1038/ onc.2010.80)
48 Tömböl Z, Szabó PM, Molnár V, Wiener Z, Tölgyesi G, Horányi J, Riesz P, Reismann P, Patócs A, Likó I et al. Integrative molecular bioinformatics study of human adrenocortical tumors: microRNA, tissue-specific target prediction, and pathway analysis. Endocrine- Related Cancer 2009 16 895-906. (https://doi.org/10.1677/ERC-09- 0096)
49 Duffy MJ, Murray A, Synnott NC, O’Donovan N & Crown J. Vitamin D analogues: potential use in cancer treatment. Critical Reviews in Oncology/Hematology 2017 112 190-197. (https://doi.org/10.1016/j. critrevonc.2017.02.015)
50 Rubin B, Pilon C, Pezzani R, Rebellato A & Fallo F. The effects of mitotane and 1a,25-dihydroxyvitamin D3 on Wnt/beta-catenin signaling in human adrenocortical carcinoma cells. Journal of Endocrinological Investigation 2020 43 357-367. (https://doi. org/10.1007/s40618-019-01127-1)
51 Lundqvist J, Norlin M & Wikvall K. 1alpha,25-Dihydroxyvitamin D3 affects hormone production and expression of steroidogenic enzymes in human adrenocortical NCI-H295R cells. Biochimica et Biophysica Acta 2010 1801 1056-1062. (https://doi.org/10.1016/j. bbalip.2010.04.009)
52 Else T, Kim AC, Sabolch A, Raymond VM, Kandathil A, Caoili EM, Jolly S, Miller BS, Giordano TJ & Hammer GD. Adrenocortical carcinoma. Endocrine Reviews 2014 35 282-326. (https://doi. org/10.1210/er.2013-1029)
Received 23 August 2021 Revised version received 14 February 2022 Accepted 14 March 2022