Abstract
Bilateral macronodular adrenocortical disease (BMAD) is characterized by the development of adrenal macronodules resulting in a pituitary-ACTH independent Cushing’s syndrome. Although there are important similarities observed between the rare microscopic descriptions of this disease, the small series published are not representative of the molecular and genetic heterogenicity recently described in BMAD. We analyzed the pathological features in a series of BMAD and determined if there is correlation between these criteria and the characteristics of the patients. Two pathologists reviewed the slides of 35 patients who underwent surgery for suspicion of BMAD in our center between 1998 and 2021. An unsupervised multiple factor analysis based on microscopic characteristics divided the cases into 4 subtypes according to the architecture of the macronodules (containing or not round fibrous septa) and the proportion of the different cell types: clear, eosinophilic compact, and oncocytic cells. The correlation study with genetic revealed subtype 1 and subtype 2 are associated with the presence of ARMC5 and KDM1A pathogenic variants, respectively. By immunohistochemistry, all cell types expressed CYP11B1 and HSD3B1. HSD3B2 staining was predominantly expressed by clear cells whereas CYP17A1 staining was predominant on compact eosinophilic cells. This partial expression of steroidogenic enzymes may explain the low efficiency of cortisol production in BMAD. In subtype 1, trabeculae of eosinophilic cylindrical cells expressed DAB2 but not CYP11B2. In subtype 2, KDM1A expression was weaker in nodule cells than in normal adrenal cells; alpha inhibin expression was strong in compact cells. This first microscopic description of a series of 35 BMAD reveals the existence of 4 histopathological subtypes, 2 of which are strongly correlated with the presence of known germline genetic alterations. This classification emphasizes that BMAD has heterogeneous pathological characteristics that correlate with some genetic alterations identified in patients.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Primary bilateral macronodular adrenal hyperplasia is a rare cause of Cushing’s syndrome, first described in 1964 [1] with changing terminology over the years. The recent description of clonal molecular alterations has led to consider the neoplastic nature of the disease and the name BMAD for the bilateral macronodular adrenocortical disease was first proposed by Pakbaz and Mete [2] and then by Hodgson et al. and Juhlin et al. [3, 4] and endorsed by the WHO Adrenal Cortex Tumor Classification Consensus Conference 2022 [5].
BMAD is mainly diagnosed in patients in the 5th or 6th decade [6]. It is defined by the development of several bilateral macronodules larger than one centimeter. Some cases can be asymmetrical. This disease is also heterogeneous as demonstrated by the high variability in the clinical signs of Cushing’s syndrome [7], the level of pituitary ACTH-independent hypercortisolism [8, 9], and the number of nodules seen on CT [7]. Steroidogenesis inhibitors (such as metopirone or ketoconazol) have been reported in some cases to control hypercortisolism, but adrenalectomy (bilateral or in selected cases unilateral) is the standard treatment indicated on the basis of the comorbidities and the level of cortisol excess [6, 8, 10].
In contrast with this clinical heterogeneity, to date, pathological descriptions are rather homogeneous in the 55 published case reports of BMAD including a pathological description. Indeed, the enlarged adrenal glands distorted by multiple yellow macronodules larger than 1 cm and up to more than 4 cm, well bounded by fibrous trabeculae appears as the main macroscopic description of BMAD. Stratakis and then Hsiao and collaborators proposed to divide BMAD into two subtypes, type I and II based on the presence of residual adrenal gland [11, 12]. Type I is characterized by nodules separated by residual adrenal gland while type II is characterized by no residual adrenal gland or atrophic cortex around the nodules. The nodules were composed of a majority of large, clear, lipid-rich, microvacuolated cells similar to those of the zona fasciculata. However, the proportion of smaller, eosinophilic compact cells was variable between cases [1, 13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45]. These observations support the hypothesis that a pathological heterogeneity could exist in BMAD.
Most cases are isolated. They can be sporadic or familial [7]. Indeed, rare BMAD cases have initially been described in patients with a genetic syndrome [4, 46] such as multiple endocrine neoplasia type 1 (MEN1 gene) [4, 46,47,48], familial adenomatous polyposis (APC gene) [4, 46, 49], hereditary leiomyomatosis and renal cell cancer (FH gene) [4, 38, 46], and McCune Albright syndrome (GNAS gene) [4].
The recent molecular descriptions demonstrated a genetic heterogeneity in isolated BMAD. To date, these genetic studies divide the isolated BMAD into 3 genetic groups [50]. The first group presents with biallelic inactivation of the ARMC5 (Armadillo containing protein 5) gene [51]. ARMC5 pathological variants are found in about 20 to 25% of isolated BMAD cases [52, 53], 50% of the operated patients (due to more severe forms) [51, 54], and in 80% of cases with a familial presentation [7]. The second group presents with biallelic inactivation of the KDM1A (lysine (K)-specific demethylase 1A/LSD1) gene [50, 55]. The presence of this inactivation leads to aberrant expression of the GIP (gastric inhibitory polypeptide) receptor by the adrenal cortical cells, resulting in food-dependent hypercortisolism [19]. To date, the third group has no identified genetic alteration [50].
So far, no pathological classification of this clinically, molecularly, and genetically heterogeneous disease has been described. We propose the first study based on pathological characteristics and immunoexpression of adrenocortical markers in a large series of 35 BMAD. We then looked for potential correlations between the pathological features identified and the characteristics of the patients, in particular, their genotype.
Material and Methods
Patients
The series concerned 35 index cases with Cushing’s syndrome who underwent surgery for BMAD at the Cochin Hospital Paris, France, between 1998 and 2021. All patients had a constitutional genetic profile study (COMETE network). Leucocyte DNA from BMAD patients has been sequenced for ARMC5, KDM1A, APC, PRKACA, PRKAR1A, and MEN1 genes by next-generation sequencing using the Ion Personal Genome Machine system (Ion Torrent, Thermo Fisher Scientific, USA) or a NextSeq 500 sequencer (Illumina). All clinical, biological, radiological, and pathological information were collected without the knowledge of the genetic data.
All patients gave their written consent for research purposes including genetic analysis in the national COMETE network. This project was approved as a monocentric retrospective study by the data protection office (bureau de la protection des données, registre d’enregistrement AP-HP) (number 20220221155734) and the CLEP, (comité local d’éthique des publications de l’hôpital Cochin) (number AAA-2022–08,019). It complies with the principles of the declaration of Helsinki.
Clinic and Biology
The following clinical information were collected: age, sex, weight, and height. The following biological results were collected: 24-h urinary free cortisol, plasma cortisol at 8:00 am, at midnight, after dexamethasone suppression, and testosterone levels.
Radiology
Computed tomography (CT) scans were available for review in 29/35 patients (83%). An abdominal radiologist (M.B) reported the following morphological criteria: maximal adrenal gland size, number of nodules, and spontaneous adrenal gland attenuation.
Pathology
Macroscopic photographs and stained slides were reviewed by two pathologists (FV, MS) in consensus, without knowledge of clinical, biological, radiological, and genetic data. The following macroscopic criteria were collected: adrenal weight, number and maximum nodule size, color, separate or coalescent nature of nodules, and presence of residual adrenal parenchyma (type I) or not (type II) [12].
A microscopic study was performed on all available nodules on 3 µm sections stained with a Tissue-Tek Prisma Plus automat (Sakura Finetek Europe BV; Zoeterwoude, The Netherlands). We reviewed a median of 10 H&E slides by the case (minimum: 4, maximum: 20). The microscopic criteria collected were architecture (round fibrous septa or sparse fibrous trabeculae inside macronodules, usually evaluated by reticulin [56, 57] or by Sirius red histochemistry [58], presence or absence of trabeculae, pseudoglandular aspects), type and proportion of cells (clear, compact, and oncocytic cells), inflammatory infiltrate, and adipocytic metaplasia.
Exhaustive immunohistochemical study was performed on dewaxed sections of 3 µm thickness on a Leica BOND-III System (Leica, Berlin, Germany). HIER (heat-induced epitope retrieval) treatment was performed for 20 min in a pH6 buffer solution (EDTA buffer, Bond Epitope Retrieval Solution 1, Leica, Berlin, Germany) for anti-CYP17A1 (clone HPA048533, diluted 1:800, Sigma Aldricht, Saint Louis, USA), anti-DAB2 (clone HPA028888, diluted 1:800, Sigma Aldricht, Saint Louis, USA) anti-CYP11B2 (clone 41-17B, diluted 1:500; Millipore, Frankfurt, Germany), and anti-KDM1A antibodies (clone ab-17721, diluted 1:400; Abcam, Boston USA). HIER treatment was performed for 20 min in pH9 buffer (EDTA buffer, Bond Epitope Retrieval Solution 2, Leica, Berlin, Germany) for anti-alpha inhibin (clone R1 IR058 ready to use, Dako, Santa Clara, USA), anti-HSD3B1 (clone WH0003283M1, diluted 1:8000, Sigma Aldricht, Saint Louis, USA), anti-HSD3B2 (clone SAB1402232, diluted 1:8000, Sigma Aldricht, Saint Louis, USA), and anti-CYP11B1 antibodies (clone 80–7, diluted 1:100, Millipore, Frankfurt, Germany). We performed double immunostaining using anti-HSD3B2 and anti-alpha inhibin antibody and with anti-CYP17A1 and anti-DAB2 antibody.
In the normal adult adrenal gland, HSD3B2 is expressed in the zona fasciculata and reticularis and has a lower expression in the zona glomerulosa [59, 60]. HSD3B1 is expressed in the three zones of the adrenal cortex [59, 60]. CYP17A1 is expressed in the zona fasciculata and the zona reticularis [59]. CYP11B1 is expressed in the zona fasciculata [59, 60]. DAB2 is expressed diffusely in the zona glomerulosa and CYP11B2 is expressed by aldosterone-secreting cells in the zona glomerulosa [60,61,62]. Alpha inhibin is expressed in the zona reticularis and can be expressed in the zona fasciculata [60, 63].
Statistical Analyses
The multiple factor analysis was performed with all microscopic criteria using R software (version 4.0.5, R Core Team (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.) with Factoshiny library. All microscopic criteria, qualitative and quantitative, had the same weight. Clustering was performed without pre-treatment with K-means or consolidation. Metric was Euclidian. Statistical analyses were performed using nonparametric tests appropriate for small numbers. Qualitative data were analyzed using Fisher exact test using R software (version 4.0.5). Quantitative data were analyzed using Kruskal–Wallis and Wilcoxon tests using R software (version 4.0.5).
Results
Patients
Most patients were female (66%), overweight (45%), or obese (20.5%). The median age was in the 6th decade (51.3 years). All patients had pituitary ACTH-independent clinical or subclinical Cushing’s syndrome. CT scans showed enlarged, multinodular adrenal glands with a spontaneous density of less than 20 HU in most cases (88%). Demographic, clinical, biological, and radiological characteristics are presented in Table 1.
Macroscopic Characteristics
By definition, all 35 patients had multinodular adrenal glands that increased in weight and volume. The nodules were mainly coalescent (86%) and yellow (72%). Some nodules contained red or brown areas. Most patients presented with type II with atrophic or no adrenal parenchyma around the nodules (57%). All macroscopic features are presented in Table 2.
Microscopic Characteristics
The nodules were densely cellular, well-bounded, and separated by fibrous trabeculae that contained capillaries. Most of the cases (63%) showed round fibrous septa around several nests of adrenal cells inside nodules. The architecture within the nodules was nest-like.
In most cases (33/35; 94%), the main cell population was composed of polygonal clear cells with microvacuolized cytoplasm and a medium-sized, round centrally located nucleus. Their proportion varied from 60 to more than 90% except for 2 cases with only 20% of clear cells.
In all cases, there were compact eosinophilic cells of smaller size. They were polygonal, regular with the cytoplasm of variable abundance, and a medium-sized, round centrally located nucleus. Their proportion varied from less than 10% to 40%. They were organized in islands, bands, or sheets intermingled with clear cells. In 11 cases, different cylindrical compact eosinophilic cells forming trabeculae were observed especially at the periphery of the nodules, close to the adipose tissue.
Occasionally, there were large, polygonal, oncocytic cells with a round, nucleolated, eccentric nucleus. Their cytoplasm was abundant, strongly eosinophilic, and granular. These cells frequently demonstrated a marked anisocytosis and anisokaryosis. Their proportion varied from less than 10% to 80%.
Most cases had an inflammatory lymphocytic infiltrate (63%) and adipocytic metaplasia (63%). Some cases had pseudoglandular aspects (cavitations in the center of several nests) of clear cells (29%). In one patient (3%), there was a myelolipomatous territory. All microscopic features are presented in Table 3.
Morphological Subtypes
We used all microscopic criteria listed in Table 4 to perform an unsupervised multiple-factor analysis. This analysis subdivided the 35 patients of our cohort into 4 distinct subtypes (Fig. 1).
Those subtypes showed specific macroscopic (Fig. 2) and microscopic (Figs. 3–7) characteristics that are presented in Table 5.
In subtype 1 (n = 17), macroscopically, the adrenal parenchyma was distorted by coalescent, yellow macronodules. Areas of residual non-nodular adrenal gland were rare (type II) (Fig. 2a, e). Microscopically, the macronodules contained round fibrous septa composed of collagen fibers (Fig. 3a, e, i) well demonstrated by reticulin and sirius red histochemistry. Clear cells occupied 70 to 90% of the nodules. Their size and shape were regular. Compact cells represented 10 to 30% of the nodules and formed islands or bands at the periphery and in the center of the nodules (Fig. 4a, e, i). Pseudoglandular aspects inside nodules and trabeculae of cylindrical eosinophilic cells at the periphery and toward the center of the nodules were only seen in this subtype (Fig. 5). Oncocytic cells were rare. Most cases did not show a lymphocytic infiltrate. Areas of adipocytic metaplasia were common (Fig. 4a, e).
In subtype 2 (n = 4), macroscopically, the adrenal gland was predominantly occupied by coalescent yellow nodules with some light brown areas. Most often, areas of residual non-nodular adrenal gland persisted (Fig. 2b, f). Microscopically, the macronodules contained round fibrous septa composed of collagen fibers (Fig. 3b, f, j). Clear cells occupied 60 to 70% of the nodules (Fig. 4b, f, j). Their size and shape were irregular (Fig. 6). Compact eosinophilic cells had abundant cytoplasm and represented 30 to 40% of the nodules (Figs. 4b, f, and 6). They were organized in extensive sheets closely intermingled with clear cells. Oncocytic cells were rare. All cases had a lymphocytic infiltrate. Most cases had areas of adipocytic metaplasia. One of the 4 patients had a myelolipomatous territory.
In subtype 3 (n = 9), the adrenal glands were occupied by large, coalescent, yellow nodules. The residual non-nodular adrenal gland was sometimes visible (Fig. 2c, g). Microscopically, the macronodules contained sparse arciform fibrosis (Fig. 3c, g, k). Clear cells occupied more than 90% of the nodules. Their size and shape were regular. Compact cells were uncommon and formed rare small nests scattered in the nodules. Oncocytic cells were rare. Most cases showed a lymphocytic infiltrate and areas of adipocytic metaplasia (Fig. 4c, g, k).
In subtype 4 (n = 5), macroscopically, nodules were separated by areas of the non-nodular adrenal gland. The nodules were yellow and brown (Fig. 2d, h). Microscopically, the macronodules included sparse arciform fibrosis (Fig. 3d, h, l). Oncocytic cells represented 40 to 80% of the nodules (Fig. 7). The proportion of clear cells varied between 20 and 60%. They were regular in size and shape. Compact cells were rare and formed small islands at the periphery or in the center of the nodules. Most cases had a lymphocytic infiltrate and adipocytic metaplasia (Fig. 4d, h, l).
Interestingly, this subtype classification concerned both adrenal glands when patients had a bilateral adrenalectomy. The presence of round fibrous septa inside nodules and the proportion of cells (i.e., clear, compact, and oncocytic) were the most discriminating criteria to distinguish these 4 subtypes (Table 6).
Immunohistochemistry
Immunohistochemical studies were performed on 34 out of 35 patients (Figs. 8, 9, 10, 11). The remaining patient had non-contributive results due to technical problems.
In all cases, HSD3B2 was preferentially expressed by clear cells (Fig. 8b–e, g–j). Fifteen cases (14 subtypes 1 and 1 subtype 3) showed diffuse staining on all clear cells (Fig. 8g). The remaining 19 cases showed heterogeneous staining with 40–80% of clear cells expressing HSD3B2 (Fig. 8h–j). In all cases, CYP17A1 was preferentially expressed in compact eosinophilic cells (Fig. 8b–e, l–o). HSD3B1 and CYP11B1 were expressed by all nodular cells, regardless of their type (Fig. 9).
DAB2 was expressed by peripheral trabecular cylindrical eosinophilic cells, only seen in subtype 1 (Fig. 10e). These cells also express HSD3B2 but did not express CYP17A1 or CYP11B2 (Fig. 10f).
Alpha inhibin was preferentially expressed in islands and bands of polygonal compact cells. It was more abundant in subtype 2 than in the other subtypes (Fig. 8g–j).
KDM1A expression was weaker only in subtype 2 nodules (Fig. 11).
The oncocytic cells took up anti-HSD3B2, CYP17A1, and anti-alpha inhibin antibodies (Fig. 8e, j, o).
In cases where it was visible, the non-nodular adrenal showed an immunohistochemical profile similar to the one observed in a normal adrenal.
In summary, in all subtypes, HSD3B2 was mainly expressed by clear cells and CYP17A1 preferentially stained compact cells. HSD3B1 and CYP11B1 were expressed in all cell types. Specifically, in subtype 1, DAB2 stained trabeculae of cylindrical eosinophilic cells with no co-expression of CYP11B2. Specifically, in subtype 2, alpha inhibin was strongly expressed in compact cells and KDM1A immunoexpression was weaker in the nodules than in the adjacent normal parenchyma.
Genetic Characteristics
Among the 35 patients in the series, 15 (43%) had a constitutional pathogenic ARMC5 variant, 4 (11%) had a constitutional pathogenic KDM1A variant, and 16 (46%) had no specific or recurrent known pathological variant to date. No patient had a PRKAR1A, PRKACA, APC, or MEN1 pathological variant. FH pathogenic variants were not investigated as no patient had clinical suspicion of FH genetic alteration.
Correlations
BMI was significantly higher in subtype 1 patients than in patients from subtype 4. Plasma cortisol at 8:00 am in subtype 2 patients was significantly lower than in the other groups. No significant difference was found in the intensity of 24 h free urinary cortisol. On the CT scan, we found only one significant difference: the spontaneous adrenal density of subtype 4 was higher than in the other groups. The adrenal size of subtype 4 was significantly smaller than in the other subtypes (Table 7). Type II correlated with the presence of a pathogenic ARMC5 variant (p < 0.01). This criterion had a positive predictive value of 62.5% and a negative predictive value of 91.5% to detect a pathogenic ARMC5 variant.
We analyzed the correlation between the morphological subtype and the genetic profile of the patients. Subtype 1 contained most patients with a constitutional pathogenic ARMC5 variant. Subtype 2 contained exclusively patients with a constitutional pathogenic KDM1A variant. Subtypes 3 and 4 mainly concerned patients without a known constitutional variant (Table 7).
The clear cells of the 15 patients with an ARMC5 pathogenic variant uniformly express HSD3B2 in contrast to the 4 patients with a KDM1A pathogenic variant and 15 patients with no known genetic cause who express HSD3B2 heterogeneously on clear cells. Weak KDM1A expression in nodule cells was only seen in the 4 patients with a KDM1A pathogenic variant.
Discussion
We report here the first detailed pathological and exhaustive immunohistochemical description of a cohort of 35 patients with BMAD operated by unilateral or bilateral adrenalectomy in our center. In agreement with the literature, the majority of BMAD cases were female and the patients were mainly in their 6th decade when they underwent surgery [7, 11, 64, 65].
The microscopic characteristics found in our series were used to perform an unsupervised multiple-factor analysis. It separated the 35 patients into four distinct histopathological subtypes according to the presence of round fibrous septa and the proportion of cell types (clear, compact, oncocytic).
These morphological subtypes show a strong correlation with the genetic data of our cohort. Patients with a biallelic inactivation of ARMC5 and KDM1A gene belong to two different microscopically homogeneous groups, subtype 1 and subtype 2, respectively. In contrast, patients without a known constitutional variant are morphologically heterogeneous and are distributed between subtypes 1, 3, and 4. All nodules of a patient have the same morphology. Type II was more frequent in patients with a pathological ARMC5 variant. These data strongly support the hypothesis of a causal link between the driver genes and the morphologic appearance of the adrenal glands in BMAD.
In the literature, the microscopic description of the 55 case reports of BMAD is very similar from one publication to another (Table 8). In most of the case reports, the microscopic description is unfortunately not sufficiently detailed to allow classification of the case according to our model. In our series, fibrous septa are frequent within macronodules but were only mentioned in one case report [18]. The presence of cylindrical eosinophilic cell trabeculae at the periphery of the nodules has only been described in one case [17]. To our knowledge, the presence of oncocytic cells has never been mentioned in the microscopic characteristics, and we found them in 5 out of 35 patients. The presence of myelolipomatous territories was reported in two cases [1, 44] and in some patients of a series of BMAD with pathological KDM1A variant [55].
In BMAD, the two main cell populations (clear and compact) have been studied by several authors, in particular, Sasano et al., using antibodies against steroidogenic enzymes involved in the synthesis of adrenal steroid hormones, in particular: HSD3B2 and CYP17A1 [66]. HSD3B2 is predominantly expressed in clear cells while CYP17A1 is preferentially expressed in compact eosinophilic cells [20, 21, 23, 25, 31, 34, 39]. Our immunohistochemical results, consistent with those previous observations, reinforce that this preferential staining of HSD3B2 in clear cells and CYP17A1 in compact cells as previously described by Sasano et al. [39] may become a diagnostic criterion of BMAD. Indeed, the lack of HSD3B2 and CYP17A1 co-expression may allow to distinguish BMAD from multiple cortisol-producing adenomas (characterized by the co-expression of HSD3B2 and CYP17A1) [59, 67].
BMAD cells produce less cortisol than the normal adrenal gland cells, and hypercortisolism is due to the increase of the number of adrenal cells [68]. Sasano et al. postulated that this low efficiency of cortisol production could be explained by the lack of co-expression of HSD3B2 and CYP17A1. According to this hypothesis, as the different enzymatic steps of steroidogenesis would be performed in different cells, the efficiency of cortisol production would be decreased [39]. Our observations corroborate this hypothesis. We observed that the clear cells of patients with an ARMC5 pathogenic variant uniformly express HSD3B2 in contrast to all other BMAD expressing HSD3B2 heterogeneously. Diffuse HSD3B2 expression could then explain, at least in part, the greater severity of Cushing’s syndromes in ARMC5 mutated patients compared to other BMAD patients [65]. HSD3B1 is expressed by all BMAD nodular cells. It is likely that this isoform of the enzyme, expressed in peripheral tissues, is not sufficient to replace HSD3B2. In conclusion, the partial expression of steroidogenic enzymes observed in BMAD nodule cells may participate to the reduction of cortisol production efficiency.
DAB2, a zona glomerulosa marker distinguishes the two populations of eosinophilic cells observed in subtype 1. The population of cylindrical eosinophilic cells in trabeculae at the periphery expressed DAB2 and did not express CYP11B2 or CYP17A1, whereas the population of compact eosinophilic cells in islands and bands expressed CYP17A1 and not DAB2. None of them appear to be able to produce aldosterone. Consistently, we did not find a correlation between subtype 1 and mineralocorticoid secretion (data not shown). Although hypertension is more frequent in ARMC5 mutated patients than in the other BMAD groups [65], our immunohistochemical and biological data do not support the hypothesis of an aldosterone co-secretion explaining the hypertension in these patients.
Alpha inhibin had never been studied in BMAD to our knowledge. This marker is more expressed in compact cells of subtype 2 compared to other subtypes but we did not find a significant correlation between this subtype and androgen secretion (data not shown). Furthermore, previous studies showed that patients with a biallelic KDM1A inactivation have an overexpression of LH receptors on the surface of adrenal cells [55]. We can hypothesize that the high expression of alpha inhibin comparatively to the other subtypes leads to the inhibition of LH-FSH secretion in this group and then, activates a compensatory mechanism through overexpression of the LH receptor [69]. As observed before in a recent work from our group, weak expression of KDM1A in the nodules is a strong argument in favor of a KDM1A pathogenic variant [50].
The identification of these morphological and immunohistochemical correlations with genetics could have an impact on the genetic sequencing strategy of BMAD specimens from patients who underwent surgery. Patients with a pathogenic ARMC5 variant frequently presented a type II and a microscopic subtype 1. All of them had a diffuse immunoexpression of HSD3B2 on clear cells unlike other patients. Similarly, all patients with a pathological KDM1A variant had a microscopic subtype 2 phenotype and a weak immunoexpression of KDM1A in the nodules compared to the adjacent adrenal gland. Because of their strong correlation with genetic clusters, the pathological characteristics could be incorporated into a genetic search strategy to target either ARMC5 or KDM1A sequencing in adrenal specimens. We previously showed that in the presence of several bilateral nodules on CT scan and a plasma cortisol greater than 50 nmol/L after 1 mg of dexamethasone, the probability of carrying an ARMC5 pathogenic variant is 20% with a sensitivity of 100% [65]. Pathology could guide the identification of this 20% subgroup of ARMC5 mutated patients.
The main observations of this study are summarized in Fig. 12.
In conclusion, our study proposes the first histopathological classification of BMAD into 4 morphological subtypes based on the architecture of the macronodules and the proportion of cell types: clear, compact, and oncocytic. This histopathological classification needs to be validated on a larger number of cases, but the strong correlation with the germinal genetic characteristics of the patients appears as a solid element of validation. This model highlights the heterogeneity of the pathological characteristics of BMAD as well as their link with the genetic characteristics.
Availability of Data and Materials
Not applicable.
References
Kirschner MA, Powell RD, Lipsett MB (1964) Cushing’s Syndrome: Nodular Cortical Hyperplasia of Adrenal Glands with Clinical and Pathological Features Suggesting Adrenocortical Tumor. J Clin Endocrinol Metab 24:947–955. https://doi.org/10.1210/jcem-24-10-947
Pakbaz S, Mete O (2019) Adrenal cortical neoplasia: from histology to molecular biology. Diagnostic Histopathology 25:178–189. https://doi.org/10.1016/j.mpdhp.2019.02.004
Hodgson A, Pakbaz S, Mete O (2019) A Diagnostic Approach to Adrenocortical Tumors. Surgical Pathology Clinics 12:967–995. https://doi.org/10.1016/j.path.2019.08.005
Juhlin CC, Bertherat J, Giordano TJ, Hammer GD, Sasano H, Mete O (2021) What Did We Learn from the Molecular Biology of Adrenal Cortical Neoplasia? From Histopathology to Translational Genomics. Endocr Pathol 32:102–133
Mete O, Erickson LA, Juhlin CC, de Krijger RR, Sasano H, Volante M, Papotti MG (2022) Overview of the 2022 WHO Classification of Adrenal Cortical Tumors. Endocr Pathol 33:155–196. https://doi.org/10.1007/s12022-022-09710-8
Lacroix A (2009) ACTH-independent macronodular adrenal hyperplasia. Best Pract Res Clin Endocrinol Metab 23:245–259. https://doi.org/10.1016/j.beem.2008.10.011
Bouys L, Chiodini I, Arlt W, Reincke M, Bertherat J (2021) Update on primary bilateral macronodular adrenal hyperplasia (PBMAH). Endocrine 71:595–603. https://doi.org/10.1007/s12020-021-02645-w
Zhang Q, Xiao H, Zhao L, et al (2020) Analysis of clinical and pathological features of primary bilateral macronodular adrenocortical hyperplasia compared with unilateral cortisol-secreting adrenal adenoma. Ann Transl Med 8:1173–1173. https://doi.org/10.21037/atm-20-5963
Lefebvre H, Duparc C, Prévost G, Bertherat J, Louiset E (2015) Cell-To-Cell Communication in Bilateral Macronodular Adrenal Hyperplasia Causing Hypercortisolism. Front Endocrinol. https://doi.org/10.3389/fendo.2015.00034
Lacroix A, Bourdeau I, Lampron A, Mazzuco TL, Tremblay J, Hamet P (2009) Aberrant G-protein coupled receptor expression in relation to adrenocortical overfunction. Clinical Endocrinology. https://doi.org/10.1111/j.1365-2265.2009.03689.x
Stratakis CA (2009) New genes and/or molecular pathways associated with adrenal hyperplasias and related adrenocortical tumors. Mol Cell Endocrinol 300:152–157 https://doi.org/https://doi.org/10.1016/j.mce.2008.11.010
Hsiao H-P, Kirschner LS, Bourdeau I, et al. (2009) Clinical and Genetic Heterogeneity, Overlap with Other Tumor Syndromes, and Atypical Glucocorticoid Hormone Secretion in Adrenocorticotropin-Independent Macronodular Adrenal Hyperplasia Compared with Other Adrenocortical Tumors. J Clin Endocrinol Metab 94:2930–2937. https://doi.org/10.1210/jc.2009-0516
Hidai H, Fujii H, Otsuka K, Abe K, Shimizu N (1975) Cushing’s Syndrome due to Huge Adrenocortical Multinodular Hyperplasia. Endocrinol Jpn 22:555–560. https://doi.org/10.1507/endocrj1954.22.555
Hashimoto K, Kawada Y, Murakami K, et al (1986) Cortisol responsiveness to insulin-induced hypoglycemia in Cushing’s syndrome with huge nodular adrenocortical hyperplasia. Endocrinol Jpn 33:479–487. https://doi.org/10.1507/endocrj1954.33.47915.
Makino S, Hashimoto K, Sugiyama M, Hirasawa R, Takao T, Ota Z, Saegusa M, Ohashi T, Omori H (1989) Cushing’s Syndrome Due to Huge Nodular Adrenocortical Hyperplasia with Fluctuation of Urinary 17-OHCS Excretion. Endocrinol Jpn 36:655–663. https://doi.org/10.1507/endocrj1954.36.655
Cugini P, Battisti P, Palma LD, Sepe M, Kawasaki T, Uezono K, Sasaki H (1989) “GIANT” Macronodular Adrenal Hyperplasia Causing Cushing’s Syndrome: Case Report and Review of the Literature on a Clinical Distinction of Adrenocortical Nodular Pathology Associated with Hypercortisolism. Endocrinol Jpn 36:101–116. https://doi.org/10.1507/endocrj1954.36.101
Malchoff CD, Rosa J, Debold CR, et al. (1989) Adrenocorticotropin-Independent Bilateral Macronodular Adrenal Hyperplasia: An Unusual Cause of Cushing’s Syndrome. J Clin Endocrinol Metab 68:855–860. https://doi.org/10.1210/jcem-68-4-855
Aiba M, Hirayama A, Iri H, et al (1991) Adrenocorticotropic Hormone—Independent Bilateral Adrenocortical Macronodular Hyperplasia as a Distinct Subtype of Cushing’s Syndrome Enzyme Histochemical and Ultrastructural Study of Four Cases with a Review of the Literature. Am J Clin Pathol 96:334–340. https://doi.org/10.1093/ajcp/96.3.334
Lacroix A, Bolté E, Tremblay J, et al (1992) Gastric Inhibitory Polypeptide–Dependent Cortisol Hypersecretion — A New Cause of Cushing’s Syndrome. N Engl J Med 327:974–980. https://doi.org/10.1056/NEJM199210013271402
Koizumi S, Beniko M, Ikota A, et al (1994) Adrenocorticotropic Hormone-Independent Bilateral Adrenocortical Macronodular Hyperplasia: A Case Report and Immunohistochemical Studies. Endocr J 41:429–435. https://doi.org/10.1507/endocrj.41.42921
Murakami O, Satoh F, Takahashi K, et al (1995) Three Cases of Clinical or Preclinical Cushing’s Syndrome due to Adrenocorticotropic Hormone-Independent Bilateral Adrenocortical Macronodular Hyperplasia: Pituitary-Adrenocortical Function and Immunohistochemistry. Intern Med 34:1074–1081. https://doi.org/10.2169/internalmedicine.34.1074
Nemoto Y, Aoki A, Katayama Y, et al (1995) Non-Cushingoid Cushing’s Syndrome due to Adrenocorticotropic Hormone-Independent Bilateral Adrenocortical Macronodular Hyperplasia. Intern Med 34:446–450. https://doi.org/10.2169/internalmedicine.34.446
Wada N, Kubo M, Kijima H, Ishizuka T, Saeki T, Koike T, Sasano H (1996) Adrenocorticotropin-independent bilateral macronodular adrenocortical hyperplasia: immunohistochemical studies of steroidogenic enzymes and post-operative course in two men. Eur J Endocrinol 134:583–587. https://doi.org/10.1530/eje.0.1340583
Terzolo M, Boccuzzi A, Alí A, Bollito E, De Risi C, Paccotti P, Angeli A (1997) Cushing’s syndrome due to ACTH-independent bilateral adrenocortical macronodular hyperplasia. J Endocrinol Invest 20:270–275. https://doi.org/10.1007/BF03350299
Tamura H, Sugihara H, Minami S, et al (1997) Cushing’s Syndrome due to Bilateral Adrenocortical Adenomas with Different Pathological Features. Intern Med 36:804–809. https://doi.org/10.2169/internalmedicine.36.804
Hayashi Y, Takeda Y, Kaneko K, Koyama H, Aiba M, Ikeda U, Shimada K (1998) A Case of Cushing’s Syndrome due to ACTH-Independent Bilateral Macronodular Hyperplasia Associated with Excessive Secretion of Mineralocorticoids. Endocr J 45:485–491. https://doi.org/10.1507/endocrj.45.485
Swain JM, Grant CS, Schlinkert RT, Thompson GB, van Heerden JA, Lloyd RV, Young WF (1998) Corticotropin-Independent Macronodular Adrenal Hyperplasia: A Clinicopathologic Correlation. Arch Surg. https://doi.org/10.1001/archsurg.133.5.541
Hsieh M-H, Chang C-C, Lin M-C, Chang T-C (2006) Adrenocorticotropin-Independent Bilateral Adrenal Macronodular Hyperplasia (AIMAH) A Case Report. J Intern Med Taiwan. 17:291-297
Kubo N, Onoda N, Ishikawa T, et al (2006) Simultaneous Bilateral Laparoscopic Adrenalectomy for Adrenocorticotropic Hormone-Independent Macronodular Adrenal Hyerplasia: Report of a Case. Surg Today 36:642–646. https://doi.org/10.1007/s00595-006-3209-6
Sato M, Soma M, Nakayama T, et al (2006) A Case of Adrenocorticotropin-independent Bilateral Adrenal Macronodular Hyperplasia (AIMAH) with Primary Hyperparathyroidism (PHPT). Endocr J 53:111–117. https://doi.org/10.1507/endocrj.53.111
Hayakawa E, Yoshimoto T, Hiraishi K, Kato M, Izumiyama H, Sasano H, Hirata Y (2011) A Rare Case of ACTH-independent Macronodular Adrenal Hyperplasia Associated with Aldosterone-producing Adenoma. Intern Med 50:227–232. https://doi.org/10.2169/internalmedicine.50.4351
Kobayashi T, Miwa T, Kan K, et al (2012) Usefulness and Limitations of Unilateral Adrenalectomy for ACTH-independent Macronodular Adrenal Hyperplasia in a Patient with Poor Glycemic Control. Intern Med 51:1709–1713. https://doi.org/10.2169/internalmedicine.51.7041
Rhee H, Jeon YK, Kim SS, Kang YH, Son SM, Kim YK, Kim IJ (2014) Adrenocorticotropic hormone-independent macronodular adrenal hyperplasia with abnormal cortisol secretion mediated by catecholamines. Korean J Intern Med 29:667. https://doi.org/10.3904/kjim.2014.29.5.667
Tokumoto M, Onoda N, Tauchi Y, et al (2017) A case of Adrenocoricotrophic hormone -independent bilateral adrenocortical macronodular hyperplasia concomitant with primary aldosteronism. BMC Surg 17:97. https://doi.org/10.1186/s12893-017-0293-z
Jin P, Janjua MU, Zhang Q, Dong C, Yang Y, Mo Z (2018) Extensive ARMC5 genetic variance in primary bilateral macronodular adrenal hyperplasia that started with exophthalmos: a case report. J Med Case Reports 12:13. https://doi.org/10.1186/s13256-017-1529-3
Higashitani T, Karashima S, Aono D, et al (2020) A case of renovascular hypertension with incidental primary bilateral macronodular adrenocortical hyperplasia. Endocrinol Diabetes Metab Case Rep. 6:19-0163. https://doi.org/10.1530/EDM-19-0163
He W-T, Wang X, Song W, et al (2021) A novel nonsense mutation in ARMC5 causes primary bilateral macronodular adrenocortical hyperplasia. BMC Med Genomics 14:126. https://doi.org/10.1186/s12920-021-00896-0
Matyakhina L, Freedman RJ, Bourdeau I, et al (2005) Hereditary Leiomyomatosis Associated with Bilateral, Massive, Macronodular Adrenocortical Disease and Atypical Cushing Syndrome: A Clinical and Molecular Genetic Investigation. J Clin Endocrinol Metab 90:3773–3779. https://doi.org/10.1210/jc.2004-2377
Sasano H, Suzuki T, Nagura H (1994) ACTH-independent macronodular adrenocortical hyperplasia: immunohistochemical and in situ hybridization studies of steroidogenic enzymes. Mod Pathol 7:215–219
Ishihara T, Uchihira F, Tatsumi M, Mori T, Igarashi T, Takayama H, Ishikawa T (1977) A Case with Cushing Syndrome Due to Huge Bilateral Adrenal Nodular Hyperplasia. Folia Endocrinol 53:1082–1093. https://doi.org/10.1507/endocrine1927.53.9_1082
Krivitzky A, Blondeau P, Camilleri JP, Delzant G, Roche-Sicot J (1980) [Cushing’s syndrome caused by a bilateral adrenal adenoma (author’s transl)]. Ann Med Interne (Paris) 131:361–364
Kawamura M, Shiraha S, Sudo T, et al (1983) A Case Of A Total Bilateral Adrenalectomy For Cushing’s Syndrome Due To Bilateral Nodular Adrenocortical Hyperplasia. The journal of the Japanese Practical Surgeon Society 44:183–190. https://doi.org/10.3919/ringe1963.44.183
Zeiger MA, Nieman LK, Cutler GB, Chrousos GP, Doppman JL, Travis WD, Norton JA (1991) Primary bilateral adrenocortical causes of Cushing’s syndrome. Surgery 110:1106–1115
Lieberman SA, Eccleshall TR, Feldman D (1994) ACTH-independent massive bilateral adrenal disease (AIMBAD): A subtype of Cushing’s syndrome with major diagnostic and therapeutic implications. European Journal of Endocrinology 131:67–73. https://doi.org/10.1530/eje.0.1310067
Yamada Y, Sakaguchi K, Inoue T, et al (1997) Preclinical Cushing’s Syndrome due to Adrenocorticotropin-Independent Bilateral Adrenocortical Macronodular Hyperplasia with Concurrent Excess of Gluco- and Mineralocorticoids. Intern Med 36:628–632. https://doi.org/10.2169/internalmedicine.36.628
Chevalier B, Vantyghem M-C, Espiard S (2021) Bilateral Adrenal Hyperplasia: Pathogenesis and Treatment. Biomedicines 9:1397. https://doi.org/10.3390/biomedicines9101397
Skogseid B, Larsson C, Lindgren PG, Kvanta E, Rastad J, Theodorsson E, Wide L, Wilander E, Oberg K (1992) Clinical and genetic features of adrenocortical lesions in multiple endocrine neoplasia type 1. J Clin Endocrinol Metab 75:76–81. https://doi.org/10.1210/jcem.75.1.1352309
Langer P, Cupisti K, Bartsch DK, Nies C, Goretzki PE, Rothmund M, Röher HD (2002) Adrenal Involvement in Multiple Endocrine Neoplasia Type 1. Wor World J Surg 26:891–89610.
Marchesa P, Fazio VW, Church JM, McGannon E (1997) Adrenal masses in patients with familial adenomatous polyposis. Dis Colon Rectum 40:1023–1028. https://doi.org/10.1007/BF0205092348
Vaczlavik A, Bouys L, Violon F, et al (2022) KDM1A inactivation causes hereditary food-dependent Cushing syndrome. Genetics in Medicine 24:374–383. https://doi.org/10.1016/j.gim.2021.09.018
Assié G, Rizk-Rabin M, Barreau O, Guignat L, René-Corail F, Poussier A, Borson-Chazot F, Bertagna X, Ragazzon B (2013) ARMC5 Mutations in Macronodular Adrenal Hyperplasia with Cushing’s Syndrome. N Engl J Med 28;369(22):2105. https://doi.org/10.1056/NEJMoa1304603
Espiard S, Drougat L, Libé R, et al (2015) ARMC5 Mutations in a Large Cohort of Primary Macronodular Adrenal Hyperplasia: Clinical and Functional Consequences. J Clin Endocrinol Metab 100:E926–E935. https://doi.org/10.1210/jc.2014-4204
Bonnet-Serrano F, Bertherat J (2018) Genetics of tumors of the adrenal cortex. Endocrine-Related Cancer 25:R131–R152. https://doi.org/10.1530/ERC-17-0361
Drougat L, Espiard S, Bertherat J (2015) Genetics of primary bilateral macronodular adrenal hyperplasia: a model for early diagnosis of Cushing’s syndrome? European Journal of Endocrinology 173:M121–M131. https://doi.org/10.1530/EJE-15-0532
Chasseloup F, Bourdeau I, Tabarin A, et al (2021) Loss of KDM1A in GIP-dependent primary bilateral macronodular adrenal hyperplasia with Cushing’s syndrome: a multicentre, retrospective, cohort study. Lancet Diabetes Endocrinol 9:813–824. https://doi.org/10.1016/S2213-8587(21)00236-9
Volante M, Bollito E, Sperone P, et al (2009) Clinicopathological study of a series of 92 adrenocortical carcinomas: from a proposal of simplified diagnostic algorithm to prognostic stratification. Histopathology 55:535–543. https://doi.org/10.1111/j.1365-2559.2009.03423.x
Duregon E, Fassina A, Volante M, et al (2013) The Reticulin Algorithm for Adrenocortical Tumor Diagnosis: A Multicentric Validation Study on 245 Unpublished Cases. American Journal of Surgical Pathology 37:1433–1440. https://doi.org/10.1097/PAS.0b013e31828d387b
Huang Y, de Boer WB, Adams LA, MacQuillan G, Rossi E, Rigby P, Raftopoulos SC, Bulsara M, Jeffrey GP (2013) Image analysis of liver collagen using sirius red is more accurate and correlates better with serum fibrosis markers than trichrome. Liver Int 33:1249–1256. https://doi.org/10.1111/liv.12184
Kubota-Nakayama F, Nakamura Y, Konosu-Fukaya S, et al (2016) Expression of steroidogenic enzymes and their transcription factors in cortisol-producing adrenocortical adenomas: immunohistochemical analysis and quantitative real-time polymerase chain reaction studies. Hum Pathol 54:165–173. https://doi.org/10.1016/j.humpath.2016.03.016
Mete O, Asa SL, Giordano TJ, Papotti M, Sasano H, Volante M (2018) Immunohistochemical Biomarkers of Adrenal Cortical Neoplasms. Endocr Pathol 29:137–149. https://doi.org/10.1007/s12022-018-9525-8
Boulkroun S, Samson-Couterie B, Dzib J-FG, et al (2010) Adrenal Cortex Remodeling and Functional Zona Glomerulosa Hyperplasia in Primary Aldosteronism. Hypertension 56:885–892. https://doi.org/10.1161/HYPERTENSIONAHA.110.158543
Gomez-Sanchez CE, Gomez-Sanchez EP, Nishimoto K (2020) Immunohistochemistry of the Human Adrenal CYP11B2 in Normal Individuals and in Patients with Primary Aldosteronism. Horm Metab Res 52:421–426. https://doi.org/10.1055/a-1139-2079
Arola J, Liu J, Heikkilä P, Voutilainen R, Kahri A (1998) Expression of inhibin α in the human adrenal gland and adrenocortical tumors. Endocrine Research 24:865–867. https://doi.org/10.3109/07435809809032699
Christopoulos S, Bourdeau I, Lacroix A (2005) Clinical and Subclinical ACTH-Independent Macronodular Adrenal Hyperplasia and Aberrant Hormone Receptors. Horm Res Paediatr 64:119–131. https://doi.org/10.1159/000088818
Bouys L, Vaczlavik A, Jouinot A, et al (2022) Identification of predictive criteria for pathogenic variants of primary bilateral macronodular adrenal hyperplasia (PBMAH) gene ARMC5 in 352 unselected patients. Eur J Endocrinol 187:123–134. https://doi.org/10.1530/EJE-21-1032
Nishimoto K, Nakagawa K, Li D, et al (2010) Adrenocortical Zonation in Humans under Normal and Pathological Conditions. J Clin Endocrinol Metab 95:2296–2305. https://doi.org/10.1210/jc.2009-2010
Rege J, Hoxie J, Liu C-J, et al (2022) Targeted Mutational Analysis of Cortisol-Producing Adenomas. J Clin Endocrinol Metab 107:e594–e603. https://doi.org/10.1210/clinem/dgab682
Assie G, Louiset E, Sturm N, et al (2010) Systematic Analysis of G Protein-Coupled Receptor Gene Expression in Adrenocorticotropin-Independent Macronodular Adrenocortical Hyperplasia Identifies Novel Targets for Pharmacological Control of Adrenal Cushing’s Syndrome. J Clin Endocrinol Metab 95:E253–E262. https://doi.org/10.1210/jc.2009-2281
Menon KMJ, Menon B (2014) Regulation of luteinizing hormone receptor expression by an RNA binding protein: role of ERK signaling. Indian J Med Res 140 Suppl:S112-119.
Acknowledgements
The authors would like to thank all the members of the department of pathology, Cochin Hospital, Paris, France, for their precious technical support.
Funding
This work was supported by the Fondation pour la Recherche Médicale (EQU201903007854) and by the Agence Nationale de la Recherche (18-CE14-0008–01). FV and LB received a fellowship from ARC (association de recherche contre le cancer) foundation and from the CARPEM (Cancer research for personalized medicine). FV received a fellowship from FIRENDO (Filière maladies rares endocriniennes).
Author information
Authors and Affiliations
Contributions
F.V. and M.S. wrote the main manuscript text. F.V. L.B. M.B. and M.S. collected the data. All authors reviewed the manuscript.
Corresponding authors
Ethics declarations
Ethical Approval
All patients gave their written consent for research purposes including genetic analysis in the national COMETE network. This project was approved as a monocentric retrospective study by the data protection office (bureau de la protection des données, registre d’enregistrement AP-HP) (number 20220221155734) and the CLEP, (comité local d’éthique des publications de l’hôpital Cochin) (number AAA-2022–08019). It complies with the principles of the declaration of Helsinki.
Competing Interests
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Violon, F., Bouys, L., Berthon, A. et al. Impact of Morphology in the Genotype and Phenotype Correlation of Bilateral Macronodular Adrenocortical Disease (BMAD): A Series of Clinicopathologically Well-Characterized 35 Cases. Endocr Pathol 34, 179–199 (2023). https://doi.org/10.1007/s12022-023-09751-7
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12022-023-09751-7