Mild Autonomous Cortisol Secretion (MACS) – Related Osteoporosis

Abstract

Mild autonomous cortisol secretion (MACS) has thus far been associated with several comorbidities, among which osteoporosis and fractures appear to be highly prevalent. Recent guidelines for adrenal incidentalomas have updated the definition of MACS, currently formulated on serum cortisol after a 1-mg dexamethasone test above 1.8 µg/dL or 50 nmol/L. Previous studies on bone health in adrenal incidentalomas had adopted different definitions of MACS, producing heterogeneous results in terms of fracture prevalence. This review aims to summarize the clinical impact of MACS in relation to fractures, bone quantity and quality, by providing a thorough update on MACS-related osteoporosis (MACS-ROP). This area has a large room for research, and management of this comorbidity still needs to be elucidated.

Keywords

adrenal tumors - osteoporosis - MACS - hypercortisolism - mild autonomous cortisol secretion - fractures

Introduction

The terminology mild autonomous cortisol secretion (MACS), previously known as subclinical hypercortisolism or subclinical Cushing’s Syndrome, was recently adopted by the new European guidelines for adrenal incidentalomas (AI), released by the European Society of Endocrinology (ESE) in collaboration with the European Network for the Study of Adrenal Tumors (ENSAT). According to these guidelines, MACS is diagnosed in patients with AI based on the presence of serum cortisol after a 1-mg dexamethasone suppression test (1-mg DST) above 50 nmol/L or 1.8 µg/dL, provided there are no signs or symptoms of overt Cushing’s syndrome (mostly catabolic signs) [1] [2].

While signs or symptoms of overt Cushing’s syndrome are absent, MACS can be associated with similar metabolic, cardiovascular, and bone complications as in Cushing’s syndrome. Among them, osteoporosis and fragility fractures have been associated with MACS, although a paucity of data has been published using the newly adopted criteria to define MACS. Therefore, the ESE-ENSAT guidelines still consider the association of MACS and osteoporosis not well established, and large prospective cohort studies are required. Moreover, the role of noninvasive radiological tools in evaluating the impact of MACS on bone microarchitecture and estimating fracture risk in AI with MACS is yet to be determined. Nevertheless, screening for vertebral fractures in patients with MACS is encouraged by the guidelines, although no definite guidance on medical management was reported. Very recent data [3] [4], however, seem to reinforce the concept that a full evaluation of bone metabolism should be performed in patients with MACS.

This narrative review aims to provide updates on the connections between MACS and bone fragility/osteoporosis. The term ‘MACS-related osteoporosis’ (MACS-ROP) will be used to describe this still-debated type of secondary osteoporosis, which will certainly need further research for its evaluation and management.

Pathophysiology

Most data on the detrimental effects of cortisol exposure to bone metabolism derive from literature on glucocorticoid-induced osteoporosis (GIOP), due to exogenous glucocorticoids [5]. While there might be similarities between GIOP and MACS-ROP, some distinct features should also be highlighted. In MACS-ROP, the time of onset of the cortisol excess is virtually unknown because AIs are discovered, by definition, for reasons unrelated to the adrenal mass. Therefore, cortisol exposure to the bone might be of variable duration, with stable or possibly progressive patterns. By contrast, in GIOP, the dose and duration of exogenous glucocorticoids is obvious, and this has allowed derivation of fracture risk estimates with stratification according to the dose of administered prednisone-equivalents [6]. Exposure to low doses of corticosteroids to reach a cumulative dose of 10 g or more is also detrimental to bone, with clinical data showing decreased fracture risk within the first year after cessation of low-dose glucocorticoid administration [5]. The last scenario might be remarkably similar to what may occur in MACS-ROP, where one of the main pathogenetic factors might be the exposure time in conjunction with the degree of cortisol secretion. Although the degree of cortisol secretion can be empirically graded by 1-mg DST cortisol, the excess cortisol secretion during the 24 hours cannot be derived by this single threshold, which means that the amount and time of onset of cortisol excess in MACS are still difficult to estimate and will be so until further research.

Nevertheless, glucocorticoid-induced bone loss can be linked to multiple factors. Beyond dose, duration, and route of administration, which all apply to GIOP, some pathogenetic factors that can also apply to MACS-ROP can be highlighted. First, two isoenzymes affect the biological activity of glucocorticoids, 11-beta-hydroxysteroid dehydrogenase type 1 (11β-HSD1) and type 2 (11β-HSD2), which respectively catalyze the conversion of (inactive) cortisone into (active) cortisol, and vice versa. 11β-HSD2 converts cortisol into cortisone, thereby protecting bone cells and the skeleton. This may influence the skeletal sensitivity to hypercortisolism [7]. 11β-HSD1 activity increases with age, with potentially major clinical relevance in the older population [8].

With exogenous glucocorticoids, a transient increase in bone resorption followed by maintained bone resorption with persistent blunted bone formation is usually the main mechanism behind bone loss and fractures. Osteoblast and osteocyte apoptosis are promoted by glucocorticoid excess, with osteoblast precursors differentiation shifted to adipocytes rather than mature osteoblasts. Decreased intestinal calcium absorption due to disruption of active transport of calcium [9] and vitamin D metabolism [10] and increased Parathyroid hormone within the physiologic range further contribute to bone loss [11]. Glucocorticoids also enhance renal excretion of calcium and decrease levels of growth hormone, gonadotropins, and adrenocorticotropic hormone (ACTH), which in turn cause lower levels of bone active hormones like insulin-like growth factor-1, estrogens, and androgens [12]. Not only the bone but also the muscle seems to be negatively affected by mild glucocorticoid excess. Compared to referent subjects, patients with MACS demonstrated reduced muscle strength, as evaluated by the nondominant hand grip strength and sit-to-stand test [13].

At the molecular level, the receptor activator of nuclear factor-kB ligand (RANKL)-RANK-osteoprotegerin (OPG) system is significantly affected. RANKL is secreted by osteoblasts and osteocytes to promote osteoclast recruitment, activation, and bone resorption through interaction with its receptor RANK. OPG derives from osteoblasts only and acts as a decoy receptor for RANKL, preventing it from binding to RANK on osteoclasts. After prolonged exposure to glucocorticoids, similar to what may occur in MACS-ROP, OPG decreases, thereby promoting the activation of osteoclasts. Depletion of osteoblasts, though, leads to decreased RANKL expression, with maintenance of a higher-than-normal RANKL-OPG ratio, which seems to be the main mechanism behind persistent bone loss over time during a state of low bone turnover [14].

In conclusion, GIOP and possibly MACS-ROP, as well, result from reduced bone formation due to osteoblast damage, continued bone resorption, and reduced skeletal sensing of biomechanical forces by osteocyte apoptosis [15]. While in-vitro and in-vivo data on GIOP are abundant, in MACS-ROP, a bone biopsy study, the gold standard method of assessing bone turnover, has never been performed, thereby limiting speculations on the underlying pathophysiology.

[Fig. 1] briefly summarizes the putative mechanisms implicated in the pathophysiology of MACS-ROP.

Fig. 1 Pathophysiology of MACS-Related Osteoporosis (MACS-ROP). MACS-ROP is likely characterized by a state of normal- to low bone turnover, promoted by osteoblast and and osteocyte apoptosis, although bone loss may be maintained over time because of relative osteoclast activation compared to persistently reduced bone formation. This could be explained by RANK-L prevailing over osteoprotegerin due to cortisol excess. Sclerostin is reduced after a prolonged glucocorticoid exposure likely due to reduction in osteocyte number and function, which cause impairment to biomechanical properties of bone. Abbreviations: AI, adrenal incidentaloma; MACS, mild autonomous cortisol secretion; OPG, osteoprotegerin; RANKL, receptor activator of nuclear factor-κβ ligand.

Fracture prevalence in patients with mild autonomous cortisol secretion

Several studies have reported an increased prevalence of osteoporosis and fragility fractures in patients with AI and MACS [7]. Studies reporting fractures in MACS are presented in [Table 1]. The definition of MACS (previously described as subclinical hypercortisolism) was different across the studies because of the heterogeneous consensus around this pathological entity/state over the past 10–15 years. A meta-analysis published in 2016 [7] showed that the prevalence of vertebral fractures in MACS is, on average, 63%, compared with 28% in patients with adrenal incidentalomas without MACS. When a control group was present, MACS showed a higher prevalence of vertebral fractures compared to patients without MACS. As regards patients’ country of origin, most data derive from Italian patients, with fewer data coming from the US, Japan, or Brazil.

Table 1 Mild autonomous cortisol secretion-related osteoporosis (MACS-ROP): prevalence and incidence of fractures among study cohorts investigating MACS.

First Author Year [Ref]	Design	MACS definition	Number of pts with MACS	Age (years)	Female/Male Ratio	Vertebral fracture definition	Vertebral Fractures Prevalence	Vertebral Fractures Incidence	All Fractures Prevalence	All Fractures Incidence	Vertebral fracture Prevalence in the control population	Country
Chiodini et al. 2004 [17]	R	At least two out of: 1) urinary free cortisol (UFC) levels above 70.0 mcg/24 (193.1 nmol/24 h), 2) 1-mg DST cortisol>3.0 mcg/dL (82.8 nmol/L); and 3) ACTH levels below 10 pg/mL (2.2 pmol/L)	21	42.9 (pre-menopause) 63.9 (post-menopause)	70/0	Radio	42.9% (pre-menopause) 78.6% (post-menopause)	/	/	/	27.3% (0% in premenopausal women; 37.7% in post-menopausal women)	Italy
Tauchmanovà et al. 2007 [18]	R, cross-sectional	1-mg DST cortisol>3 μg/dL	35	46	35/0	Radio	57.0%	/	/	/	1.4%	Italy
Chiodini et al. 2009 [19]	R	At least two out of: 1) urinary free cortisol (UFC) levels above 70.0 mcg/24 (193.1 nmol/24 h), 2) 1-mg DST cortisol>3.0 mcg/dL (82.8 nmol/L); and 3) ACTH levels below 10 pg/mL (2.2 pmol/L)	85	63	53/32	Radio	70.6%	/	/	/	21.8%	Italy
Chiodini et al. 2009 [20]	R, cross-sectional	At least two out of: 1) urinary free cortisol (UFC) levels above 70.0 mcg/24 (193.1 nmol/24 h), 2) 1-mg DST cortisol>3.0 mcg/dL (82.8 nmol/L); and 3) ACTH levels below 10 pg/mL (2.2 pmol/L)	22	66	0/22	Radio	72.7%	/	/	/	20.0%	Italy
Tauchmanova et al. 2009 [21]	RCT on clodronate in MACS	1-mg DST cortisol>3.0 mcg/dL and at least one other abnormality (low ACTH, UFC, daily cortisol average)	46	43	46/0	Radio	63.0%	/	/	/	/	Italy
Morelli et al. 2011 [22]	R, longitudinal	At least two out of: 1) urinary free cortisol (UFC) levels above 70.0 mcg/24 (193.1 nmol/24 h), 2) 1-mg DST cortisol>3.0 mcg/dL (82.8 nmol/L); and 3) ACTH levels below 10 pg/mL (2.2 pmol/L)	27	65	Not reported	Radio	55.6%	81.5% (cumulative percentage at month 24) (+48.1% Vertebral fractures at month 24)	/	/	28.9% (non-functioning AI)	Italy
Eller-Vainicher et al. 2012 [23]	R, cross-sectional	At least two of the following three parameters: (1) 24-h urinary free cortisol (UFC) levels>70 mg/ 24 h; (2) 1-mg DSTcortisol>3 μg/dL;(3) ACTH<10 pg/mL	34	66.3	19/15	Radio	82.4%	/	/	/	45.6% (non-functioning AI)	Italy
Morelli et al. 2013 [24]	P	At least two of the following three parameters: (1) 24-h urinary free cortisol (UFC) levels>70 mg/ 24 h; (2) 1-mg DSTcortisol>3 μg/dL;(3) ACTH<10 pg/mL	51 (41 unilateral; 10 bilateral)	66(unilateral) 62(bilateral)	28/13 (unilateral) 4/6(bilateral)	Radio	46.3%(unilateral) 70%(bilateral)	/	/	/	/	Italy
Lasco et al. 2014 [25]	R	(1) increased urinary free cortisol (UFC) levels>70 g per 24 h (193 nmol/24 h); (2) 1-mg DST cortisol>1.8 μg/dL (50 nmol/L) and (3) cortisol levels after 2-day low dose DST>1.8 μg/dL (50 nmol/l)	3	57	3/0	Radio	100%	/	/	/	/	Italy
Salcuni et al. 2016 [26]	R	1-mg DST cortisol>5.0 mcg/dl (138 nmol/l) or in the presence of greater than or equal to two out of the following alterations: 1 mg-DST cortisol>3.0 mcg/dl (83 nmol/l), ACTH<10 pg/ml (2.2 pmol/l), 24 h urinary free cortisol (UFC) levels>70 mcg/24 h (193 nmol/24 h)	55 (23 conservatively treated, 32 surgically treated) Follow-up: ≈28 months (conservatively treated) ≈40 months (surgically treated)	65 (conservatively treated) 61 (surgically treated)	32/ 23	Radio	54.5%	+12/23 (52.1%) (conservatively treated)+3/32 (9.4%) (surgically treated)	/	/	/	Italy
Kim et al. 2018 [27]	R	1-mg DST cortisol>5.0 μg/dL (138 nmol/L) or 1-mg DST cortisol>2.2 μg/dL (61 nmol/L) plus ACTH<10 pg/mL(2.2 pmol/L) or DHEA-S<80 μg/dL (2.17 μmol/L) in men or<35 μg/dL (0.95 μmol/L) in women	61	59.5 (males) 51.2 (females)	31/30	Radio	0	/	/	/	2.5% (non-functioning AI)	South Korea
Ahn et al. 2019 [28]	R	1-mg DST cortisol>5 μg/dl (138.0 nmol/L) or 1-mg DST cortisol>61.0 nmol/L plus ACTH<2.2 pmol/L or DHEA-S<2.17 μmol/L in men or<0.95 μmol/L in women	109	39 (pre-menopausal women) 59.7 (post-menopausal women) 56.9 (	56/53	Radio	0	/	/	/	1.7%	Korea
Moraes et al. 2020 [29]	Cross-sectional	1-mg DST cortisol between 1.9–5.0 μg/dL (50–138 nmol/L)	30	60	26/4	Radio	81.3%	/	/	/	55.6% (non-significantly different)	Brazil
Ishida et al. 2021 [30]	R	1-mg DST cortisol≥1.8 μg/dl (50 nmol/L) or overnight 8-mg DST cortisol≥1.0 μg/dl	55	62	33/22	Radio	51.0%	/	/	/	70% (non-significantly different)	Japan
Li et al. 2021 [31]	R (Population-based)	1-mg DST cortisol≥1.8 μg/dl (50 nmol/L)	81	63	55/26	Radiology reports	7.4% (only symptomatic vertebral fractures)	/	Any fracture: 44.4%	30% at 10 years (estimate)	3.7%	USA
Izawa et al. 2022 [32]	R	1-mg DST cortisol≥1.8 μg/dl (50 nmol/L)	237	56	182/55	Radiology reports	/	/	14.8% (fragility fractures not otherwise specified)	/	/	Japan
Dogra et al. 2023 [33]	R, cross-sectional	1-mg DST cortisol≥1.8 μg/dl (50 nmol/L)	212	61	155/57	Questionnaire	/	/	Any fragility fracture: 9%	/	/	USA
Zavatta et al. 2023 [3]	R, cross-sectional	1-mg DST cortisol≥1.8 μg/dl (50 nmol/L)	238	66	151/87	Both Radio and Clinical (chart review)	31.7% (4.6% clinical VFs)	/	34.0%	/	24.1% (non-functioning AI)	Italy
Favero et al. 2023 [4]	R	1-mg DST cortisol≥1.8 μg/dl (50 nmol/L)	230 (cross-sectional arm) 66 (Longitudinal arm)	64	137/93	Radio	62.6%	36.4%	/	/	10.0%	Italy

Abbreviations: MACS-IOP, mild autonomous cortisol secretion-induced osteoporosis; AI, adrenal incidentalomas; P, prospective; R, Retrospective; Radio, radiological or morphometric (review of radiological images); VF, vertebral fracture.

Heterogenous MACS definitions across study cohorts [16] have thus far produced heterogeneous fracture prevalences [7]. Several studies mentioning fractures in patients with benign adrenal masses were designed using a 1-mg DST cortisol cut-off of 3 mcg/dL (82.7 nmol/L), while others had lower cut-off values. Importantly, the DST was not even mandatory in a few studies, and different additional criteria were often used to define MACS. The two most recent major studies on MACS and fractures adopted the 1-mg DST threshold currently recommended by the ESE-ENSAT guidelines [1]. Even considering this, there is no evident trend between fracture prevalence and 1-mg DST cortisol. Older studies [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] with a 1-mg DST cortisol cut-off>3 mcg/dL showed a highly variable prevalence of vertebral fractures ranging from 43% to 100%. More recent studies performed with 1-mg DST cortisol cut-off>1.8 mcg/dL (50 nmol/L) [3] [4] [30] [31] [32] [33] demonstrated similar rates regardless of the size of the cohort, with fracture prevalence in larger cohorts ranging from 31.7% to 62.6% ([Table 1]).

Most fractures in MACS are vertebral fractures, while non-vertebral fractures are likely less present or under-reported. In our recent study [3], vertebral fractures were often of mild grade (Genant’s grade 1) and asymptomatic. Mild vertebral fractures were also more common in MACS compared with non-functional adrenal incidentalomas and accounted for the disproportion of vertebral fractures between MACS and non-functional AI groups. Moreover, the evaluation of asymptomatic (detected through radiological imaging only) or symptomatic (the patients seek medical care due to pain or discomfort) vertebral fractures may likely explain the discrepancy in the prevalence among some studies. Indeed, in the study of Li et Al. [31] evaluating symptomatic or clinical vertebral fractures, the prevalence of this condition was 7.4%, much lower than that reported in the Italian studies, which also described morphometric asymptomatic vertebral fractures. Symptomatic vertebral fractures seem infrequent in MACS, with only two studies reporting their prevalences of 4.6% [3] and 7.4% [31].

Non-vertebral fractures have rarely been reported in MACS-ROP studies. Only three studies mention the prevalence of all fragility fractures [3] [31] [33], detected with different methodologies. According to this paucity of data, prevalence of fragility fractures, including vertebral fractures, would lie between 9% and 44.4% in patients with MACS. The reasons why non-vertebral fractures might be less common are unclear, although we could speculate that this might be due to a less severe phenotype compared to overt hypercortisolism; in other words, MACS is likely to cause subclinical vertebral fractures rather than causing symptomatic clinical fractures. The less severe spectrum of hypercortisolism is the most likely explanation, which is also supported by the fact that GRADE 1 – but not GRADE 3 – vertebral fractures are the most prevalent.

Lastly, sex differences are expected to impact bone fragility even in patients with adrenal incidentalomas. Whether an elevated 1-mg DST cortisol might be a risk factor for fractures in both sexes with the same magnitude is still to be demonstrated. In a recent study [3], we showed a net sex dimorphism in terms of fracture risk between women and men, with post-menopausal women with MACS being at significantly higher risk. In our study, the prevalence of fragility fractures in women 65 or older was 48.8% in MACS, while in non-functional AI was much lower (29.5%, P=0.008). When women’s age was less than 65 years, we did not observe significant differences between fracture rates in MACS (12.0%) and non-functional AI (15.8%). In men, fracture prevalence was similar between groups (37.9% in MACS vs 30.3% in non-functional AI, P=0.206). The risk of fractures in men was only dependent on age and not on the presence of MACS, as opposed to post-menopausal women, where age, smoking history, and 1-mg DST cortisol were all independent factors for fracture. The concept of sex dimorphism in MACS was recently supported by a major study investigating age- and sex- disparities in the occurrence of cardiovascular diseases and mortality [34]. Women with MACS, and especially those younger than 65 years, were particularly prone to cardiovascular disease and mortality compared with men, in whom MACS did not seem to be an additional risk factor for those endpoints across all ages. In our study [3], sex dimorphism emerged, especially in women after 65 years, where MACS could well represent an additional risk factor for fracture beyond age and menopause. These findings suggest that women before 65 years should be prioritized for cardiovascular screening, while women older than 65 years should undergo a complete bone metabolism evaluation. These speculations come from our cross-sectional study [3] and must be validated in targeted longitudinal studies. The underlying causative mechanisms remain to be elucidated, although sex hormone levels across different ages, bone microarchitecture, and menopausal status could be important parameters accounting for gender differences in MACS-ROP.

By contrast, the study by Favero et al. [4] found similar prevalences of vertebral fractures in women (62.0%) and in men (63.4%), and MACS predicted the presence of prevalent vertebral fractures independently of gender, with age and low lumbar spine BMD as independent contributors. Earlier studies did not have sufficient numerosity to make any speculation on potential gender differences in MACS-ROP. Based on discordant results in the recently published studies, further research on gender-related fracture risk in MACS should be implemented, hopefully including accurate assessments of baseline estradiol and androgens both in women and in men.

Fracture incidence in patients with mild autonomous cortisol secretion

Compared to studies about prevalence of fractures in MACS, fewer data are available about incident fractures in MACS. Four longitudinal studies were conducted, all of which suggest that probability of fractures in this category of patients is not negligible ([Table 1]).

The first longitudinal assessment of fracture risk was carried out by Morelli et al. who evaluated a cohort of 27 patients with MACS at baseline, 12 months, and 24 months [22]. At the end of follow-up, the MACS group showed a higher prevalence of vertebral fractures (81.5%) compared with baseline (55.6%, P=0.04) regardless of age, gender, BMI, BMD, and time since menopause. The incidence of new vertebral fractures was greater in the MACS group (48%) than in the non-functional group (n=76) (13%; P=0.001). The definition of MACS in this study was based on a combination of different hormonal criteria, including 1-mg DST cortisol>3.0 mcg/dL as an optional criterion, among others.

The second longitudinal study on MACS and fractures was that of Salcuni and colleagues [26], comparing MACS conservatively treated (n=23) to MACS surgically treated (n=32). At the end of follow-up (average of 27.7 months in the conservatively treated group and 39.9 months in the surgically treated group), patients with new vertebral fractures were 12/23 (52.2%) in the conservatively treated group, while only 3/32 (9.4%) in the surgical group. The Authors concluded that surgery in MACS produced a 30% vertebral fracture risk reduction, regardless of age, gender, follow-up duration, 1-mg DST cortisol, lumbar spine BMD, and prevalent vertebral fractures at baseline. The definition of MACS in this study was based on the same criteria as the previous one [22] plus 1-mg DST cortisol>5 µg/dL as an optional additional single criterion.

The third study by Li et al. [31] evaluated the cumulative incidence of clinical fractures at follow-up in MACS compared to non-functional AI. Longitudinal numbers in this study were small, and the estimated incidence of fractures at 10 years was 30% in MACS and 29% in non-functioning AI, with no significant differences. A significant limitation was the number of patients at year 10 of follow-up, respectively 14 in MACS (starting from 42 patients at baseline) and 25 (starting from 78 patients at baseline) in non-functioning AI. The definition of MACS in this study was based on 1-mg DST cortisol>1.8 µg/dL, which is the currently recommended threshold [1].

The fourth and most recent longitudinal cohort of MACS evaluated for incident vertebral fractures was described in the study by Favero et al. [4]. A group of 66 patients with MACS was compared with 60 patients without MACS and followed for an average of 25.6 months and 24 months, respectively (P=0.083). Incident vertebral fractures occurred in 36.4% of patients with MACS, while only 10% of patients without MACS experienced vertebral fractures. Symptomatic vertebral fractures trended toward significance (9.1% vs 1.7% in MACS vs. non-MACS, P=0.069). Incident vertebral fractures were independently predicted by MACS but not by other anticipated risk factors such as age, baseline prevalent vertebral fractures, type 2 diabetes mellitus, female sex, lumbar spine BMD, and duration of follow-up. MACS were about 2.8 times more likely than non-MACS to develop vertebral fractures. Interestingly, patients with low ACTH levels were more likely to report fractures (31.7%) than patients with normal ACTH levels (15.9%, P<0.036), although this relationship was lost after adjustment for the above-mentioned confounders. The definition of MACS in this study was based on 1-mg DST cortisol>1.8 µg/dL. Of note, prevalent and incident non-vertebral fractures were not reported in this study.

The last of these four studies might have important clinical implications. Since almost a third of patients may experience a vertebral fracture during follow-up, a careful fracture risk evaluation should be carried out as soon as the patient is diagnosed with MACS, considering that vertebral fractures in this setting may occur regardless of their BMD T-score. However, it appears difficult to correctly estimate patients’ fracture risk as opposed to what usually occurs in patients without MACS, due to paucity of clinical predictors.

Overall, because of the limited number of patients at follow-up, fracture incidence in patients with MACS should be further investigated, hopefully with larger longitudinal cohorts, to unveil, if any, predictors of fragility fractures through sufficiently powered multivariate analyses.

Bone density and quality in mild autonomous cortisol secretion

The detrimental effects of glucocorticoids on bone are often exerted regardless of bone mineral density, thereby making bone density less useful to stratify patients’ fracture risk [7]. Trabecular bone is affected more than cortical bone by glucocorticoids, and the association of trabecular BMD reduction (i. e., predominantly lumbar spine BMD) should be considered consistently demonstrated in patients with MACS [7]. By contrast, robust longitudinal data on BMD variations over time are missing.

Bone microarchitecture can be indirectly studied using the Trabecular Bone Score (TBS), which is a relatively innovative tool applied to lumbar DXA to predict fracture risk based on the assessment of trabecular connectivity derived from the gray-scale texture of lumbar spine DXA images. TBS (TBS Insight, Medimaps, Meriganc, France) has been shown to improve fracture risk evaluation, predicting fractures independently of DXA during treatment with exogenous glucocorticoids and has been included in the FRAX algorithm to refine fracture risk [35].

Eller-Vainicher et al. [23] first investigated TBS in MACS. Thirty-four patients with MACS were compared with 68 patients without MACS and 70 controls, finding lower TBS Z-score values in MACS compared with other groups. Low TBS was independently associated with prevalent vertebral fractures in MACS, regardless of age, BMI, gender, or lumbar spine BMD. TBS correlated with the number and severity of vertebral fractures. Importantly, TBS was inversely correlated with 1-mg DST cortisol (R=-0.29, P=0.003). In the same study, in a subgroup of 40 patients who were followed for 24 months, TBS was reported to independently predict new vertebral fractures after adjustment for lumbar spine BMD (not significant), BMI (not significant) and age (not significant) with an Odds-ratio of 11.2 (1.7–71.4) for every TBS Z-score unit decrease.

The second study on TBS was conducted by Vinolas et al. [36], who compared patients with MACS (n=29) to patients with non-functional AI (n=18). Both groups were similar, in terms of age, BMI, and female sex. TBS was lower in MACS (1.30±0.09) as compared to non-functional AI (1.37±0.12), with 52% of patients MACS vs. 33% of those with non-functional AI having a degraded or partially degraded TBS microarchitecture (P=0.05). No longitudinal data on TBS are yet available in patients with MACS, although it seems that after remission of overt Cushing’s syndrome TBS, improves rapidly as opposed to BMD (+10% vs.+3% within an average of 15 months, P<0.02) [7].

The last study on TBS [27] available to date was conducted on 61 patients with MACS (30 men, 31 women) and 355 subjects with non-functional AI used as comparators. The study showed that 1-mg DST cortisol was inversely correlated with TBS in men (β=-0.133, P=0.045) and women (β=-0.140, P=0.048). Women with MACS had 2.2% lower TBS (P=0.040) than women with non-functional AI. A degraded TBS (<1.230) was associated with 1-mg DST cortisol (odds ratio [OR] 2.18; 95% confidence interval [CI], 1.04–4.53).

All these findings support the use of tools to evaluate bone quality in clinical practice; however, larger cohorts should confirm the above findings and the strength of association between TBS and fractures in MACS using the newly adopted ESE-ENSAT threshold of 1-mg DST cortisol>1.8 µg/dL. In other words, further studies should clarify whether the associations found between TBS and cortisol levels are both statistically and clinically significant.

In patients with MACS, androgens are typically low. Some intriguing data show that the cortisol/DHEAs ratio negatively correlates with TBS and lumbar spine BMD. This ratio seems clinically relevant in women but not in men, regardless of 1-mg DST cortisol [27]. Interestingly, it seems that a greater cortisol/DHEAs ratio might negatively impact BMD in post-menopausal and not in pre-menopausal women [28]. Further data are certainly needed to clarify if there might be an independent effect of low androgens, especially in women with MACS, across different ages.

High-resolution peripheral quantitative computed tomography (HR-pQCT) could be highly informative to outline which bone compartment, either cortical or trabecular or both, might be predominantly affected in MACS-ROP. This technique can quantify volumetric BMD (vBMD) and bone microarchitecture with several parameters at two skeletal sites, distal radius and distal tibia, unless these were previously fractured. The only study reporting data on HR-pQCT was that by Moraes et al. [29], who had 45 patients with non-functional AI and 30 patients with MACS perform HR-pQCT, as well as DXA and morphometric spine X-rays. MACS was defined according to the current guidelines [1] (1-mg DST cortisol>1.8 µg/dL). Both groups showed similar ages (60 years and 59 years, P=0.97) and female/male ratios (71.1% and 86.7%, P=0.16), as well as other common risk factors for fractures (smoking and BMI). At HR-pQCT analysis, several parameters associated with trabecular bone were significantly lower in MACS than in non-functional AI. Moreover, none of the cortical bone parameters differed between MACS and non-functional AI, thereby indicating that this bone compartment might be relatively spared in MACS. Of note, the radius parameters were more affected than the tibia parameters, leading the us to hypothesize that the radial trabecular bone could be more prone to deterioration than the tibial one, because the latter bears more mechanical load than the radius, and the effect of mild cortisol excess might therefore go unnoticed in the lower bones of the skeleton.

TBS and HR-pQCT could be more specific radiological tools to detect the effect of MACS on the bone and, therefore, provide a full picture of MACS-ROP. By contrast, the effect of mild cortisol exposure is not easily captured by areal BMD by DXA. Likely, the impact of subtle cortisol excess might only be captured with radiological techniques investigating trabecular bone and bone microarchitecture, and these might be used to predict vertebral fracture incidence after robust validation.

Bone turnover markers (BTM) in mild autonomous cortisol secretion

In primary osteoporosis, BTM can assist the clinician in decision-making on which anti-osteoporotic drug could be chosen, in monitoring compliance to medications and their early effectiveness because consistent data have shown an association between variations of BTMs and fracture risk reduction [37]. The use of BTMs in GIOP is not well-established because they are significantly affected by glucocorticoids [37]. N-terminal propeptide of type 1 collagen (P1NP) and bone-specific alkaline phosphatase (BSAP) are serum markers of osteoblast activity that are usually decreased in GIOP and increase after withdrawal of glucocorticoid therapy. Osteocalcin (OC) is typically low or suppressed, while β-C-terminal telopeptide (CTX) may vary [14].

Although data on BTMs are very scarce in MACS populations, significant alterations of BTMs have not thus far emerged between patients with MACS and non-functional AI. Three studies on BTMs reported slightly discordant results. In a US [38] (n=92) and European study [3] (n=86), CTX did not differ between MACS and non-functional AI. BSAP [3] and OC [38] were also similar between MACS and non-functional AI. Reduced serum sclerostin levels were observed in patients with MACS vs. those with non-functional AI [38], which may be explained by decreased osteocyte function or numbers likely due to chronic glucocorticoid exposure. By contrast, in a Japanese study, Ishida et al. [30] reported that BSAP and OC were higher in MACS (n=55) than in non-functional AI (n=12). However, in the latter study, the control group consisted of only a few patients, limiting the interpretation of the results.

Data on the changes in BTMs after adrenalectomy are even more scarce. Athimulam et al. [38] observed a significant increase in osteocalcin and CTX levels in 8 patients with MACS after adrenalectomy, while sclerostin and P1NP did not change.

Overall, BTMs might be useful in assessing the effect of glucocorticoids on the bone, although more data is needed to define their clinical relevance in the management of MACS-ROP.

Treatment of mild autonomous cortisol secretion-related osteoporosis

The best approach to MACS with bone complications such as osteoporosis or fragility fractures is still to be elucidated. While the latest ESE-ENSAT guidelines suggested performing the screening for asymptomatic vertebral fractures in patients with MACS, no specific recommendations regarding the treatment of MACS-ROP have been provided due to a lack of robust data. Until further targeted studies, the indications for the management of MACS-ROP can be borrowed from the current recommendation for osteoporosis and GIOP.

Medical management

It is reasonable, as in GIOP [15], to optimize calcium intake through diet and/or supplements, meeting the national daily allowances, and maintain serum 25-hydroxyvitamin D at>30 ng/mL. Especially younger populations could benefit from this conservative approach in MACS, without the need for pharmacological treatment of osteoporosis [7].

At present, only one study has reported data on the effectiveness of anti-osteoporotic drugs in MACS-ROP. In a randomized controlled trial comparing 100 mg weekly clodronate+500 mg of calcium ad 800 UI of vitamin D3 daily to calcium and vitamin D3 supplements only, the bisphosphonate was able to increase lumbar spine BMD, to lower BTMs by up to a third, and to prevent the occurrence of vertebral fractures. The study was small, with 23 patients allocated to each study arm, and only a 12-month follow-up.

Theoretically, until further evidence is reported, patients could be treated according to international guidelines for osteoporosis [39] [40], bearing in mind that the safety and effectiveness of virtually all anti-osteoporotic medications have not been tested in MACS. Some suggest that GIOP guidelines [41] should also be considered when choosing pharmacological options for the treatment of MACS-ROP, although 1-mg DST cortisol values cannot be translated into prednisone-equivalent to estimate fracture risk, making these GIOP recommendations difficult to adapt to MACS-ROP.

Surgical management

The previously mentioned study by Salcuni and colleagues [26], though not randomized, suggests that MACS patients treated with adrenalectomy could reduce vertebral fracture risk by about 30%, although robust evidence should be further provided before recommending adrenal surgery to prevent fracture risk in MACS. An unresolved issue is whether prevalent bone fractures could be considered as an indication to treat mild cortisol excess. Further data are also needed to assess whether osteoporosis could be the only criterion for adrenal surgery referral, even for younger patients with no other comorbidities (e. g., diabetes or hypertension).

Post-operative management

Previous evidence suggests that a state of relatively high bone turnover could be present after remission of overt Cushing’s Syndrome or MACS [38] [42], and management of this condition is debated [7]. In this circumstance, bone apposition could be blunted by antiresorptives (bisphosphonates or denosumab); therefore, the current consensus is to treat patients with calcium and vitamin D3 supplements to favor BMD recovery after the resolution of hypercortisolism [7] [42]. Whether this phenomenon may be present in all surgically treated patients with MACS should be further studied, along with possible risk factors such as age, sex, time from MACS diagnosis, or others. In this regard, a prospective collection of biochemical mineral indices and DXA parameters in all patients just before adrenalectomy and after 6 and 12 months could be pursued to monitor bone density recovery and unravel its underlying mechanisms.

Conclusions

Evidence on MACS-ROP is increasing, with recent data on fracture prevalence confirming those from earlier studies. Vertebral fracture prevalence may vary depending on the threshold of 1-mg DST cortisol, on the different study methods used to define vertebral fractures (e. g., asymptomatic vs. symptomatic), or possibly on the heterogeneous populations from different countries. Non-vertebral fractures appear to be a relatively spared complication, possibly because of relatively young cohorts or because trabecular bone rather than cortical bone seems more affected in the pathophysiology of MACS-ROP. Bone microarchitecture is mildly but significantly deteriorated in the few studies available thus far. Bone mineral density is non-specific in MACS-ROP and should be evaluated in conjunction with morphometric X-rays of the spine, as currently recommended, and possibly with innovative radiological tools such as TBS or HR-pQCT in tertiary centers. Management of bone health in MACS, until further evidence is available, should be individualized after considering the bone quantity and quality together.

Areas of future research

• The fracture risk assessment in MACS is currently not standardized, and radiological tools for assessing bone quality needs validation in larger prospective cohorts before being implemented in clinical practice.

• Primary prevention of fractures in MACS and gender differences in fracture incidence should be investigated.

• The management of MACS-ROP should be further investigated, as well as whether medical or surgical treatments are more effective in improving bone density and bone quality. Sufficiently powered studies could also be designed with fragility fractures as primary outcomes in medically treated vs. surgically treated MACS.

• The best pharmacological treatment or treatment sequence to speed up bone density and quality recovery after remission of MACS needs further investigation.

Conflict of Interest

The authors declare that they have no conflict of interest.

• References

• 1 Fassnacht M, Tsagarakis S, Terzolo M. et al. European Society of Endocrinology clinical practice guidelines on the management of adrenal incidentalomas, in collaboration with the European Network for the Study of Adrenal Tumors. Eur J Endocrinol 2023; 189: G1-G42
• 2 Fassnacht M, Arlt W, Bancos I. et al. European Society of Endocrinology Clinical Practice Guideline in collaboration with the European Network for the Study of Adrenal Tumors. Eur J Endocrinol 2016; 175: G1-G34
• 3 Zavatta G, Vicennati V, Altieri P. et al. Mild autonomous cortisol secretion in adrenal incidentalomas and risk of fragility fractures: A large cross-sectional study. Eur J Endocrinol 2023; 188: 343-352
• 4 Favero V, Eller-Vainicher C, Morelli V. et al. Increased risk of vertebral fractures in patients with mild autonomous cortisol secretion. J Clin Endocrinol Metab 2023; 109: e623-e632
• 5 Chotiyarnwong P, McCloskey EV. Pathogenesis of glucocorticoid-induced osteoporosis and options for treatment. Nat Rev Endocrinol 2020; 16: 437-447
• 6 Kanis JA, Johansson H, Oden A. et al. Guidance for the adjustment of FRAX according to the dose of glucocorticoids. Osteoporos Int 2011; 22: 809-816
• 7 Chiodini I, Vainicher CE, Morelli V. et al. Mechanisms in endocrinology: Endogenous subclinical hypercortisolism and bone: A clinical review. Eur J Endocrinol 2016; 175: R265-R282
• 8 Cooper MS, Rabbitt EH, Goddard PE. et al. Osteoblastic 11beta-hydroxysteroid dehydrogenase type 1 activity increases with age and glucocorticoid exposure. J Bone Miner Res 2002; 17: 979-986
• 9 Kimberg DV, Baerg RD, Gershon E. et al. Effect of cortisone treatment on the active transport of calcium by the small intestine. J Clin Invest 1971; 50: 1309-1321
• 10 Favus MJ, Kimberg DV, Millar GN. et al. Effects of cortisone administration on the metabolism and localization of 25-hydroxycholecalciferol in the rat. J Clin Invest 1973; 52: 1328-1335
• 11 Suzuki Y, Ichikawa Y, Saito E. et al. Importance of increased urinary calcium excretion in the development of secondary hyperparathyroidism of patients under glucocorticoid therapy. Metabolism 1983; 32: 151-156
• 12 Mazziotti G, Formenti AM, Adler RA. et al. Glucocorticoid-induced osteoporosis: pathophysiological role of GH/IGF-I and PTH/VITAMIN D axes, treatment options and guidelines. Endocrine 2016; 54: 603-611
• 13 Li D, Zhang CD, Saini J. et al Determinants of muscle function and health-related quality of life in patients with endogenous hypercortisolism: A cross-sectional study. Eur J Endocrinol 2023; 10 188: 603-612
• 14 Humphrey EL, Williams JH, Davie MW. et al. Effects of dissociated glucocorticoids on OPG and RANKL in osteoblastic cells. Bone 2006; 38: 652-661
• 15 Zavatta G, Clarke BL. Glucocorticoid- and transplantation-induced osteoporosis. Endocrinol Metab Clin North Am 2021; 50: 251-273
• 16 Zavatta G, Di Dalmazi G. Recent advances on subclinical hypercortisolism. Endocrinol Metab Clin North Am 2018; 47: 375-383
• 17 Chiodini I, Guglielmi G, Battista C. et al. Spinal volumetric bone mineral density and vertebral fractures in female patients with adrenal incidentalomas: The effects of subclinical hypercortisolism and gonadal status. J Clin Endocrinol Metab 2004; 89: 2237-2241
• 18 Tauchmanovà L, Pivonello R, De Martino MC. et al. Effects of sex steroids on bone in women with subclinical or overt endogenous hypercortisolism. Eur J Endocrinol 2007; 157: 359-366
• 19 Chiodini I, Morelli V, Masserini B. et al. Bone mineral density, prevalence of vertebral fractures, and bone quality in patients with adrenal incidentalomas with and without subclinical hypercortisolism: An Italian multicenter study. J Clin Endocrinol Metab 2009; 94: 3207-3214
• 20 Chiodini I, Viti R, Coletti F. et al. Eugonadal male patients with adrenal incidentalomas and subclinical hypercortisolism have increased rate of vertebral fractures. Clin Endocrinol (Oxf) 2009; 70: 208-213
• 21 Tauchmanova L, Guerra E, Pivonello R. et al. Weekly clodronate treatment prevents bone loss and vertebral fractures in women with subclinical Cushing's syndrome. J Endocrinol Invest 2009; 32: 390-394
• 22 Morelli V, Eller-Vainicher C, Salcuni AS. et al. Risk of new vertebral fractures in patients with adrenal incidentaloma with and without subclinical hypercortisolism: A multicenter longitudinal study. J Bone Miner Res 2011; 26: 1816-1821
• 23 Eller-Vainicher C, Morelli V, Ulivieri FM. et al. Bone quality, as measured by trabecular bone score in patients with adrenal incidentalomas with and without subclinical hypercortisolism. J Bone Miner Res 2012; 27: 2223-2230
• 24 Morelli V, Palmieri S, Salcuni AS. et al. Bilateral and unilateral adrenal incidentalomas: Biochemical and clinical characteristics. Eur J Endocrinol 2013; 168: 235-241
• 25 Lasco A, Catalano A, Pilato A. et al. Subclinical hypercortisol-assessment of bone fragility: Experience of single osteoporosis center in Sicily. Eur Rev Med Pharmacol Si 2014; 18: 352-358
• 26 Salcuni AS, Morelli V, Eller Vainicher C. et al. Adrenalectomy reduces the risk of vertebral fractures in patients with monolateral adrenal incidentalomas and subclinical hypercortisolism. Eur J Endocrinol 2016; 174: 261-269
• 27 Kim B-J, Kwak MK, Ahn SH. et al. The association of cortisol and adrenal androgen with trabecular bone score in patients with adrenal incidentaloma with and without autonomous cortisol secretion. Osteoporos Int 2018; 29: 2299-2307
• 28 Ahn SH, Kim JH, Cho YY. et al. The effects of cortisol and adrenal androgen on bone mass in Asians with and without subclinical hypercortisolism. Osteoporos Int 2019; 30: 1059-1069
• 29 Moraes AB, De Paula MP, De Paula Paranhos-Neto F. et al. Bone evaluation by high-resolution peripheral quantitative computed tomography in patients with adrenal incidentaloma. J Clin Endocrinol Metab 2020; 105: e2726-e2737
• 30 Ishida A, Igarashi K, Ruike Y. et al. Association of urinary free cortisol with bone formation in patients with mild autonomous cortisol secretion. Clin Endocrinol (Oxf.) 2021; 94: 544-550
• 31 Li D, Kaur RJ, Zhang CD. et al. Risk of bone fractures after the diagnosis of adrenal adenomas: A population-based cohort study. Eur J Endocrinol 2021; 184: 597-606
• 32 Izawa S, Matsumoto K, Matsuzawa K. et al. Sex difference in the association of osteoporosis and osteopenia prevalence in patients with adrenal adenoma and different degrees of cortisol excess. Borretta G, ed. Int J Endocrinol. 2022. 2022. 1-9
• 33 Dogra P, Šambula L, Saini J. et al. High prevalence of frailty in patients with adrenal adenomas and adrenocortical hormone excess: A cross-sectional multi-centre study with prospective enrolment. Eur J Endocrinol 2023; 189: 318-326
• 34 Deutschbein T, Reimondo G, Di Dalmazi G. et al. Age-dependent and sex-dependent disparity in mortality in patients with adrenal incidentalomas and autonomous cortisol secretion: An international, retrospective, cohort study. Lancet Diabetes Endocrinol 2022; 10: 499-508
• 35 Martineau P, Leslie WD, Johansson H. et al. In which patients does lumbar spine trabecular bone score (TBS) have the largest effect?. Bone 2018; 113: 161-168
• 36 Vinolas H, Grouthier V, Mehsen-Cetre N. et al. Assessment of vertebral microarchitecture in overt and mild Cushing’s syndrome using trabecular bone score. Clin Endocrinol (Oxf.) 2018; 89: 148-154
• 37 Schini M, Vilaca T, Gossiel F. et al. Bone turnover markers: Basic biology to clinical applications. Endocr Rev 2023; 44: 417-473
• 38 Athimulam S, Delivanis D, Thomas M. et al. The impact of mild autonomous cortisol secretion on bone turnover markers. J Clin Endocrinol Metab 2020; 105: 1469-1477
• 39 Shoback D, Rosen CJ, Black DM. et al. Pharmacological management of osteoporosis in postmenopausal women: An Endocrine Society Guideline pdate. J Clin Endocrinol Metab 2020; 105: dgaa048
• 40 Camacho PM, Petak SM, Binkley N. et al. American Association of Clinical Endocrinologists/American College of Endocrinology Clinical Practice Guidelines for the Diagnosis and Treatment of Postmenopausal Osteoporosis- 2020 update executive summary. Endocr Pract 2020; 26: 564-570
• 41 Humphrey MB, Russell L, Danila MI. et al. 2022 American College of Rheumatology guideline for the prevention and treatment of glucocorticoid-induced osteoporosis. Arthritis Rheumatol 2023; 75: 2088-2102
• 42 Braun LT, Fazel J, Zopp S. et al. The effect of biochemical remission on bone metabolism in Cushing's syndrome: A 2-year follow-up study. J Bone Miner Res 2020; 35: 1711-1717

Publication History

Received: 13 January 2024
Received: 02 April 2024

Accepted: 17 May 2024

Accepted Manuscript online:
17 May 2024

Article published online:
11 December 2024