European Journal of Endocrinology (2012) 167 311 326
`
`ISSN 0804-4643
`
`REVIEW
`MANAGEMENT OF ENDOCRINE DISEASE
`
`The burden of Cushing’s disease: clinical and health-related
`quality of life aspects
`R A Feelders, S J Pulgar1, A Kempel2 and A M Pereira3
`Endocrine Section, Department of Internal Medicine, Erasmus Medical Center Rotterdam, Gravendijkwal 230, 3015 CE Rotterdam, The Netherlands,
`1Global Health Outcomes, GlaxoSmithKline Research and Development, Research Triangle Park, North Carolina, USA, 2Pharmametrics GmbH, Institute
`for Health Economics and Epidemiology, Freiburg, Germany and 3Department of Endocrinology and Metabolism, Leiden University Medical Center, Leiden,
`The Netherlands
`
`(Correspondence should be addressed to R A Feelders; Email: r.feelders@erasmusmc.nl)
`
`Abstract
`
`Objective: Cushing’s disease (CD) is a rare endocrine disorder characterized by excess secretion of
`ACTH due to a pituitary adenoma. Current treatment options are limited and may pose additional
`risks. A literature review was conducted to assess the holistic burden of CD.
`Design: Studies published in English were evaluated to address questions regarding the
`epidemiology of CD, time to diagnosis, health related quality of life (HRQoL), treatment outcomes,
`mortality, prevalence of comorbidities at diagnosis, and reversibility of comorbidities following the
`treatment.
`Methods: A two stage literature search was performed in Medline, EMBASE, and Science Citation
`Index, using keywords related to the epidemiology, treatment, and outcomes of CD: i) articles published
`from 2000 to 2012 were identified and ii) an additional hand search (all years) was conducted on the
`basis of bibliography of identified articles.
`Results: At the time of diagnosis, 58 85% of patients have hypertension, 32 41% are obese, 20 47%
`have diabetes mellitus, 50 81% have major depression, 31 50% have osteoporosis, and 38 71% have
`dyslipidemia. Remission rates following transsphenoidal surgery (TSS) are high when performed by
`expert pituitary surgeons (rates of 65 90%), but the potential for relapse remains (rates of 5 36%).
`Although some complications can be partially reversed, time to reversal can take years. The HRQoL of
`patients with CD also remains severely compromised after remission.
`Conclusions: These findings highlight the significant burden associated with CD. As current treatment
`options may not fully reverse the burden of chronic hypercortisolism, there is a need for both improved
`diagnostic tools to reduce the time to diagnosis and effective therapy, particularly a targeted medical therapy.
`
`European Journal of Endocrinology 167 311 326
`
`Introduction
`
`Cushing’s disease (CD) is a rare condition caused by a
`pituitary adenoma that secretes excess ACTH (1), which
`promotes excess cortisol production from the adrenal
`glands. Excess cortisol induces a clinical phenotype that
`harbors all components of the metabolic syndrome,
`such as central obesity, diabetes mellitus, dyslipidemia,
`and hypertension, as well as muscle weakness,
`hirsutism, increased bruisability, psychological dysfunc-
`tion, and osteoporosis (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11).
`Patients with CD experience a significant clinical
`burden due to comorbidities, increased mortality, and
`impaired health-related quality of life (HRQoL) due to
`prolonged exposure to elevated cortisol levels (3, 5, 11,
`12, 13, 14, 15, 16, 17, 18, 19, 20). In particular,
`
`patients with CD often experience severe fatigue and
`weakness, physical changes, emotional
`instability,
`depression, and cognitive impairments, which have a
`profound impact on daily life (13, 21).
`Although there have been several consensus state-
`ments published recently on the definition of remission,
`diagnosis, and the management of CD, the severity and
`diversity of the clinical scenario and associated morbidities
`continue to present a management challenge (1, 22, 23).
`Additionally, there is recent evidence of persistent
`deleterious effects after remission, most notably persistent
`elevated cardiovascular risk (3, 22). The main objective of
`the current literature review is to describe the current
`burden of the disease and to summarize data on specific
`aspects of this burden, which underscores the need for
`improved diagnostic and therapeutic approaches.
`
`q 2012 European Society of Endocrinology
`
`DOI: 10.1530/EJE 11 1095
`Online version via www.eje online.org
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2009, Page 1
`
`

`

`312
`
`R A Feelders and others
`
`Materials and methods
`
`Available literature were evaluated to address questions
`regarding the epidemiology of CD, time to diagnosis,
`mortality, prevalence of comorbidities at diagnosis,
`reversibility of comorbidities after treatment (in particu-
`lar, after disease remission), outcomes and compli-
`cations of current
`treatment options, and HRQoL
`associated with CD and interventions.
`The literature search was performed in Medline,
`EMBASE, and Science Citation Index, using keywords
`related to the epidemiology, treatment, and outcomes of
`CD. It was conducted in two stages: i) articles published
`between 2000 and 2012 were identified through a
`PubMed search using the following keywords: CD,
`incidence, prevalence, mortality, treatment, remission,
`cure, excess cortisol, outcomes, cost, QoL, morbidities,
`transsphenoidal surgery (TSS), adrenalectomy, radio-
`therapy, steroidogenesis inhibitors, ketoconazole, mito-
`tane, aminoglutethimide, etomidate, metyrapone,
`pasireotide, and cortisol receptor antagonists; and ii) an
`additional hand search was conducted on the basis of
`the bibliographies of identified articles. All studies that
`provided data (regardless of publication year) related to
`these research questions were retained.
`
`Definitions
`
`Different criteria for defining the remission of hyper-
`cortisolism have been proposed, ranging from the
`occurrence of definitive or transient postoperative
`hypocortisolemia to the adequate suppression of cortisol
`after dexamethasone administration. According to a
`recent consensus statement (23), persistent postoperative
`morning serum cortisol levels of !2 mg/dl (w50 nmol/l)
`are associated with remission and a low recurrence rate of
`w10% at 10 years. Persistent serum cortisol levels above
`5 mg/dl (w140 nmol/l) for up to 6 weeks following
`surgery require further evaluation. When serum cortisol
`levels are between 2 and 5 mg/dl, the patient can be
`considered in remission and can be observed without
`additional treatment for CD. A subset of patients can even
`develop complete adrenal insufficiency (serum cortisol
`levels below 2 mg/dl (w50 nmol/l)) up to 12 weeks
`postsurgery (24, 25). Therefore, repeated evaluation in
`the early postoperative period is recommended. However,
`long-term follow-up is necessary for all patients because
`no single cortisol cutoff value excludes those who later
`experience disease recurrence, and up to 25% of patients
`develop a recurrent adenoma within 10 years after
`surgery (26, 27, 28).
`
`Results
`
`Incidence and prevalence of CD
`
`Although epidemiologic data on CD are limited, several
`population-based studies indicate an incidence of
`
`www.eje online.org
`
`EUROPEAN JOURNAL OF ENDOCRINOLOGY (2012) 167
`
`1.2–2.4 per million (14, 19) and the prevalence of
`diagnosed cases to be w39 per million population (14).
`Lindholm et al. (19) used the case definition as either the
`presence of a corticotroph adenoma or remission after
`neurosurgery, which yielded an estimated incidence
`rate of 1.2–1.7 per million per year. Etxabe & Vazquez
`(14) reported an incidence of 2.4 per million in Vizcaya,
`Spain. A large-scale retrospective survey carried out
`in New Zealand by Bolland et al. (29) found the
`approximate prevalence of all
`forms of Cushing’s
`syndrome (CS) (the majority of these cases were of
`pituitary origin) to be 79 per million and the incidence
`to be 1.8 per million per year. Differences in epidemio-
`logic estimates may be attributable to varying case
`definitions (for instance,
`the study by Lindholm
`excluded cases in which the adenoma could not be
`localized or those that could not achieve remission from
`surgery), geographical differences, and temporal effects.
`The prevalence of CD may be underestimated due to
`unrecognized patients with mild symptoms and patients
`with a cyclic form of CD (30).
`
`Time to diagnosis
`
`Data on the time from onset of symptoms to diagnosis
`are also limited. In a prospective study by Flitsch et al.
`(31) of 48 patients with pituitary adenomas, including
`19 who had ACTH-secreting adenomas causing CD, the
`reported time from onset of symptoms to diagnosis was
`4.3 years. A study by Martinez Ruiz et al. (32), which
`was based on only four pediatric CD patients, reported
`the time between onset of symptoms and diagnosis as
`ranging from 2.5 to 5 years. Etxabe & Vazquez (14)
`estimated that the average time from onset of clinical
`symptoms to diagnosis in 49 CD patients was 45.8G2.7
`months (6–144 months),
`thus 3.8 years. This is
`corroborated by the findings from a Belgian cross-
`sectional study on pituitary adenomas including CD,
`which estimated that patients experienced symptoms
`for an average of 45 months before diagnosis (33).
`However, the reliability and generalizability of these
`data are limited by small sample sizes and the retro-
`spective nature of the studies. Indeed, the New Zealand
`data from Bolland et al. (29) report that on presentation,
`patients experienced symptoms for a median of 2.0
`years (but ranging up to 20 years) before diagnosis. On
`the basis of data from the prospective European Registry
`on Cushing’s syndrome (ERCUSYN) (total number of
`patients 481, of whom 66% of patients had CD),
`median delay in diagnosis was 2 years (34).
`
`Mortality in patients with CD
`
`Mortality in patients with CD has been analyzed in
`several small studies, with overall rates reported as
`standardized mortality ratio (SMR) ranging from 1.7
`to 4.8 (Table 1) (14, 15, 17, 19). In studies in which
`mortality was assessed among those in remission and
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2009, Page 2
`
`

`

`EUROPEAN JOURNAL OF ENDOCRINOLOGY (2012) 167
`
`The burden of Cushing’s disease
`
`313
`
`Table 1 Mortality in patients with CD.
`
`Study design
`
`Population-based study
`18-year follow-up (14)
`
`n
`
`49
`
`SMR (95% CI)
`
`3.8 (2.5 17.9)
`
`Retrospective study
`Follow-up 1981 1996 (17)
`
`74 CS
`48 CD
`
`1.68 (0.8 30.9)
`
`National Patient Registry study
`11-year period (19)
`
`166 CS
`73 CD
`
`3.68 (2.34 5.33)
`1.7 (with proven etiology)
`
`Single-center study.
`Mean duration of follow-up was
`10.1G7.2 years for the whole
`cohort (15)
`
`Single-center study.
`Median duration of follow-up
`was 15 years (36)
`
`248
`
`60
`
`2.39 (1.22 3.9)
`
`4.8 (2.8 8.3)
`
`Systematic review and
`meta-analysis (40)
`Single-center study (38)
`
`7 studies
`
`80
`
`1.84 (1.28 2.65)
`
`3.17 (1.70 5.43)
`
`Nationwide retrospective
`survey (29)
`
`253 CS; 158 (62.4%)
`microadenoma; 30
`(11.9%) macroadenoma
`
`Macroadenoma: 3.5
`(1.3 7.8); microadenoma:
`3.2 (2.0 4.8)
`
`Comments
`
`SMR of vascular disease was 5.0. Older
`age, persistence of
`hypertension, and abnormalities
`of glucose metabolism
`after treatment were independent
`predictors of mortality
`Mortality risk was moderately increased
`compared with that of the general
`population
`The authors in this study reported
`excess mortality in the first year after
`diagnosis. After the first year,
`mortality was not significantly higher
`than that of the background popu-
`lation
`The SMR in patients with CD with
`remission after TSS was 1.80
`(95% CI: 0.71 3.37), whereas the
`SMR in patients with persistent
`disease was 4.38 (95% CI: 1.38 9.07)
`Vascular disease SMR: 13.8 (95% CI:
`7.2 36.5). SMR for persistent disease
`(nZ6): 16 (95% CI:
`6.7 38.4) vs remission (nZ54):
`3.3 (95% CI: 1.7 6.7)
`SMR for persistent disease after surgery
`3.73 (95% CI: 2.31 6.01)
`Cure group SMR 2.47 (95% CI:
`0.80 5.77)
`Recurrent/persistent disease group
`SMR: 4.12 (95% CI: 1.12 10.54)
`In all 253 patients with CS: SMR 4.1
`(95% CI 2.9 5.6)
`
`those with persistent disease separately, patients with
`persistent hypercortisolemia consistently had the high-
`est mortality risk (15, 19, 35, 36). In addition, TSS as a
`first-line treatment has been an important advance as
`high remission rates after initial surgery have been
`accompanied by mortality rates that mirror those
`observed in the general population (17, 35, 37). In a
`case series from the UK, it was found that the majority of
`deaths occurred before 1985, which was before TSS was
`employed as the routine first-line treatment at the
`center (36). In a recent retrospective study, 80 patients
`undergoing TSS for CD between 1988 and 2009 were
`evaluated, and long-term cure (defined as ongoing
`absence of hypercortisolism at
`last
`follow-up) was
`reported in 72% of patients. However, overall elevated
`mortality persisted in patients (SMR 3.17 (95% CI:
`1.70–5.43)), including those who achieved ‘cure’ (SMR
`2.47 (95% CI: 0.80–5.77)), although even higher
`mortality was seen in those with postoperative
`recurrence/persistent disease (SMR 4.12 (95% CI:
`1.12–10.54) (38). Additionally, a nationwide, retro-
`spective study in New Zealand reported significant
`persistently increased mortality both in macro- and
`microadenomas (SMR 3.5 (1.3–7.8) and 3.2 (2.0–4.8)
`
`respectively), despite long-term biochemical remission
`rates of 93 and 91% of patients, respectively (29).
`In a single-center study designed to assess the impact
`of hypercortisolism vs the occurrence of a pituitary
`adenoma in mediating increased mortality risk, patients
`with CD had higher mortality rates than patients with
`nonfunctioning pituitary macroadenomas (SMR 2.4 vs
`1.4) (15). Patients with persistent hypercortisolism after
`initial TSS had the highest elevated mortality (SMR 4.4)
`(15), which suggests that hypercortisolism and the
`length of exposure to hypercortisolism is associated with
`increased mortality. This finding is similar to the results
`of a Danish population study in which patients with
`persistent hypercortisolism after initial surgery had an
`SMR of 5 (19). Some deleterious effects of excess cortisol
`exposure appear to persist as even patients who
`achieved surgical remission still had elevated mortality
`(SMR 1.8) (15). Patients with possible CD but whose
`disease etiology was unproven had the highest reported
`mortality risk in the literature (SMR 11.5) (19). Among
`these patients, etiology could not be confirmed because
`the adenoma could not be localized, patients were
`unable to achieve remission after surgery, or patients
`died before a full clinical workup could be done (19).
`
`www.eje online.org
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2009, Page 3
`
`

`

`314
`
`R A Feelders and others
`
`EUROPEAN JOURNAL OF ENDOCRINOLOGY (2012) 167
`
`Mortality has been reported to be higher in females
`(SMR 4.5) (14), particularly from vascular causes (SMR 5),
`although it is important to note that this study had
`a 15:1 female to male preponderance and other
`studies have not observed gender differences in
`mortality (14, 19). However, questions remain regard-
`ing the relationship between specific dimensions of
`hypercortisolism in CD, such as the duration of exposure
`to hypercortisolism and the severity of this exposure,
`and mortality.
`die
`to
`likely
`Patients with CD are most
`causes
`from cardiovascular or
`cerebrovascular
`(14, 15, 19, 37). In the study by Dekkers et al. (15),
`the leading causes of death were cardiovascular disease
`(23.4%), cerebrovascular disease (12.8%), malignancy
`(19%), and infectious diseases (17%). Additionally, the
`average age at death in this Dutch study was reported to
`be 62.4 years, which is significantly lower than the life
`expectancy of the general population (80 years) (9, 39).
`Similarly, Hassan-Smith et al. (38) reported that the
`leading causes of death were cardiovascular disease
`(n 8), cancer (n 3), infection (n 1), and CD (n 1),
`with the median age at death being 57 years. A recent
`systematic review and meta-analysis of mortality in
`patients with CS drawing from seven studies estimated
`that the SMR for patients with CD is 1.84 (95% CI:
`1.28–2.65), with patients with persistent disease
`after surgery having an SMR of 3.73 (95% CI:
`2.31–6.01) (40).
`These conclusions drawn from small studies and case
`series with a low number of deaths and variable follow-
`up time should be interpreted with caution. Studies with
`longer follow-up time and more analyzable events are
`needed to assess whether long-term mortality is
`normalized after long-term remission or whether
`persistent cardiovascular risk factors translate into an
`increased mortality risk. The fact that some studies have
`shown persistent elevated mortality after cure,
`in
`particular the study by Hassan-Smith et al. (38) where
`all patients underwent TSS, may suggest that a stricter
`definition of ‘cure’ could be applied. Nonetheless, these
`findings suggest that normalization of cortisol secretion
`improves mortality, but may be insufficient to normalize
`mortality as other factors, like adverse cardiovascular
`risk factors or hypopituitarism, may determine the
`mortality outcome of patients after successful surgery.
`In the large retrospective survey from New Zealand,
`demographic predictors of mortality in patients with a
`micro- or macroadenoma were older age, hypertension,
`and diabetes mellitus. It is of note that patients with an
`adrenal adenoma (n 37) had an elevated SMR but
`very good prognosis following the treatment (29). This
`may suggest that pituitary compromise may account
`for the persistent elevated mortality in patients with
`‘cured’ disease following transsphenoidal intervention.
`The data reported by Bolland et al. (29) also showed a
`relatively consistent elevated SMR over four time
`periods (1960–1980, 1980–1990, 1990–2000, and
`
`www.eje online.org
`
`2000–present) despite a general shift in pituitary-
`directed therapy from radiotherapy to TSS, which is at
`odds with the previously discussed UK data in which the
`majority of deaths occurred before 1985, before TSS
`was employed as the standard first-line therapy (36).
`Nonetheless, early detection and rapid intervention may
`be key in avoiding long-term sequelae of the disease
`(41). As CD is a rare disease and, thus, difficult to
`study, it is hoped that databases such as the ERCUSYN,
`which now includes more than 500 patients, could
`become a key resource to further research and aid
`collaboration (42).
`It should be noted that geographic variability in time
`to diagnosis, the availability of pituitary experts, and the
`awareness levels of general practitioners who may first
`encounter patients with undiagnosed CD can all lead to
`treatment disparities and differences in remission and
`mortality rates between countries and regions.
`
`Comorbidities: prevalence and reversibility
`
`The chronic hypersecretion of cortisol causes central
`obesity, systemic arterial hypertension, impaired glu-
`cose tolerance, dyslipidemia, and hypercoagulability,
`and is associated with other comorbidities, such as
`psychological and cognitive dysfunction, osteoporosis,
`and increased susceptibility to infection (1, 9, 43, 44)
`(Table 2 and Supplementary Table 1, see section on
`supplementary data given at the end of this article).
`When compared with the general population, CD
`patients showed a greater than twofold increase in
`hypertension, diabetes, and osteoporosis, and an
`increase of over five times the rate of major depression
`and impaired glucose tolerance (Fig. 1) (45, 46, 47). A
`recent study, using whole-body magnetic resonance
`imaging (MRI), evaluating body composition and
`cardiovascular parameters after remission of CD found
`that cardiovascular risk persisted despite potential
`dramatic improvements in body composition abnorm-
`alities (48).
`
`Cardiovascular complications Among all systemic
`consequences of hypercortisolism, cardiovascular
`complications are the most dire as these are among
`the leading causes of death in patients with CD (14, 15).
`Cardiovascular risk is increased in CD due to compli-
`cations such as hypertension, atherosclerosis, hyper-
`coagulability, obesity, diabetes, and dyslipidemia, which
`are all consequences of chronic cortisol hypersecretion
`(1, 49).
`In addition, CD is associated with left
`ventricular hypertrophy and diastolic dysfunction (50).
`Years after disease remission, cardiovascular morbid-
`ity persists (3, 51). However, despite numerous studies
`determining the cardiovascular risk in the general
`population, there are few studies that comprehensively
`evaluate the cardiovascular risk factors and assess the
`global risk for patients with CD. A recent paper by
`De Leo et al. (52) suggests that assessing the global
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2009, Page 4
`
`

`

`EUROPEAN JOURNAL OF ENDOCRINOLOGY (2012) 167
`
`The burden of Cushing’s disease
`
`315
`
`Table 2 Comorbidities, prevalence at diagnosis, and reversibility in patients with CD.
`
`Morbidity
`
`Hypertension
`
`Prevalence at diagnosis
`
`Reversibility
`
`55 85% (12, 14, 22, 53)
`
`18-Year follow-up: posttreatment 24% (12/49); diag-
`nosis 55% (27/49) (14)
`After 5 years’ cortisol normalization: CD 40% (6/15);
`BMI-matched controls 20% (6/30) (3)
`After 1 year’s cortisol normalization, hypertension was
`reversed in 44% (12)
`Less than 2 years after successful remission, 75% had
`normalized BP (55)
`18-Year follow-up: posttreatment 4% (2/49); diagnosis
`24% (12/49) (14)
`After 5 years’ cortisol normalization: CD 27% (4/15);
`BMI-matched controls 27% (8/30) (3)
`18-Year follow-up: posttreatment 18% (9/49); diag-
`nosis 39% (19/49) (14)
`After 5 years’ cortisol normalization: CD 33% (5/15);
`BMI-matched controls 7% (2/30) (3)
`After 1 year’s cortisol normalization, diabetes was
`reversed in 40% of patients (12)
`After 5 years’ cortisol normalization: CD 33% (5/15);
`sex- and age-matched controls 20% (6/30) (3)
`After 5 years’ cortisol normalization: CD 40% (6/15);
`0 in controls (3)
`After 1 year’s cortisol normalization, 38% were no
`longer obese (12)
`After 5 years’ cortisol normalization, the prevalence
`was 27% (3)
`Adequate prophylaxis with anticoagulants can reverse
`the prothrombotic state and greatly reduce the risk of
`postoperative thromboembolic events (49)
`
`IGT
`
`21 64% (12, 14, 22, 53)
`
`Diabetes mellitus
`
`20 47% (12, 14, 22, 53)
`
`Overweight (BMI 25 30 kg/m2)
`
`21 48% (12, 22, 53)
`
`Obesity (BMIO30 kg/m2)
`
`32 41% (12, 22, 53)
`
`Dyslipidemia
`
`38 71%a (22, 53)
`
`Hypercoagulopathy/hemostatic
`abnormalities
`
`Hemostatic abnormalities (53.6%) (53)
`Vascular morbidity (10%) (49)
`VTE (incidence 2.5 3.1 per 1000 persons
`per year) (57)
`Nephrolithiasis: 50% (116)
`Kidney disease
`Osteoporosis/compression fractures Osteoporosis: 38 50% (6, 71)
`Fracturesb: 15.8% (71)
`Major depression/psychopathologyc MDD/MD/MAD: 54 81% (10, 72, 73)
`Overall psychopathology: 67% (5)
`Atypical depression: 51.5% (5)
`
`Prevalence in patients achieving remission 27% (116)
`Only partially reversible 2 years after normalization of
`cortisol levels (67)
`Prevalence of MD at 3 months: 54%;
`at 6 months: 36%; at 12 months: 24% (5)
`About 70% of patients fully recovered from their
`depression (75)
`Cognitive deficits/loss of brain volume Subjective loss of brain volume: 86% (80) Partially reversible (based on retrospective study in
`38 patients only) (80)
`
`BP, blood pressure; IGT, impaired glucose tolerance; MAD, major affective depression; MD, major depression; MDD, major depressive disorder; VTE, venous
`thromboembolism.
`aLipid profile abnormalities (combined increased cholesterol and triglycerides 25%).
`bPituitary Cushing’s.
`cPsychopathology includes atypical depression, major depressive disorder, and other psychiatric disorders.
`
`cardiovascular risk and managing the risk associated
`with cardiovascular disease such as hypertension,
`obesity, glucose
`intolerance,
`insulin resistance,
`dyslipidemia, endothelial dysfunction, and the hyper-
`coagulable state should be the important goals of
`treatment for CD.
`
`Hypertension Hypertension has been reported in
`55–85% of patients with CD, with most cases being
`mild to moderate (3, 12, 14, 22, 53). Hypertension
`remits in a subset of patients (44–75%, Table 2)
`after successful treatment but persists in w24–56%
`(12, 14, 22), presumably due to microvessel remodeling
`or concomitant underlying essential hypertension (54).
`The duration of disease has been shown to be longer in
`hypertensive patients (4.8G3.7 years) than in normo-
`tensive (0.7G0.2 years) patients (22) and has been
`
`identified as a significant risk factor in several studies
`(54, 55). Older age and longer duration of hypertension
`before treatment also negatively influenced the normal-
`ization of high blood pressure after resolution of
`hypercortisolism.
`
`Hypercoagulopathy Patients with CD show various
`abnormalities of hemostatic parameters and those with
`active CD show an increased thrombotic tendency (56).
`Increased cortisol
`levels stimulate the synthesis of
`several clotting factors, such as fibrinogen by the liver
`and von Willebrand factor by endothelial cells. Gluco-
`corticoids also upregulate the synthesis of plasminogen
`activator inhibitor type 1, the main inhibitor of the
`fibrinolytic system (49). This hypercoagulability state is
`a crucial factor predisposing CD patients to thromboem-
`bolic events, mostly after surgery and during inferior
`
`www.eje online.org
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2009, Page 5
`
`

`

`316
`
`R A Feelders and others
`
`EUROPEAN JOURNAL OF ENDOCRINOLOGY (2012) 167
`
`Overweight and obesity Obese and overweight
`patients with CD were also exposed to hypercortisolism
`for a longer duration (5.6G3.5 and 5.7G3.9 years
`respectively) than normal-weight patients (2.1G3.5
`years) (22). The prevalence of overweight (defined as a
`BMI of 25–30 kg/m2) ranged from 21 to 48% (12, 22,
`53), but can be reduced after cortisol normalization (3).
`A similar pattern is seen in obese patients. However,
`reversibility of body composition is limited. In a case–
`control study conducted in Spain, patients whose
`endogenous hypercortisolism had been controlled for
`an average of 11 years exhibited higher total fat mass
`and central obesity than age-adjusted controls. In
`addition, these patients also expressed an unfavorable
`adipocytokine profile (increased inflammatory markers
`and reduced adiponectin), which is implicated in
`increased coronary heart disease risk and has been
`hypothesized to be a link between adiposity and
`increased cardiovascular risk (2).
`
`Insulin resistance, impaired glucose tolerance, and
`diabetes mellitus Insulin resistance is another import-
`ant cardiovascular risk factor and glucocorticoids seem
`to play a key role in its development (3, 12, 63). Insulin
`resistance is a well-known complication of CD and also
`seems to persist after biochemical remission (3, 12).
`Studies demonstrate that even a slight cortisol excess, as
`in adrenal
`incidentaloma, could be associated with
`insulin resistance in the metabolic syndrome (64). The
`high values of homeostatic model assessment (HOMA)
`(51) and low values of the insulin sensitivity index (ISI)
`(64) in patients with CD confirm the decrease in insulin
`sensitivity in this condition without
`significant
`differences between obese, overweight and normal-
`weight patients, suggesting that the alteration in insulin
`sensitivity is due not to obesity but to the state of
`hypercortisolism per se (22). Barahona et al. (51)
`observed that although patients achieving remission
`had lower levels of insulin compared with those with
`active disease, insulin levels in both groups were elevated
`when compared with population controls (P!0.05).
`Insulin resistance as measured by HOMA also exhibited
`this trend. However, patients with complications (dia-
`betes and/or hypertension and/or dyslipidemia) pre-
`sented a higher BMI than patients without the respective
`complications. Epidemiological studies have shown a
`variable prevalence of abnormalities of glucose metab-
`olism: 20–47% of patients suffer from overt diabetes
`mellitus, whereas impaired glucose tolerance is present
`in 21–64% of patients (12, 14, 22, 53).
`As patients with CD are likely to have impaired
`glucose tolerance, assessment by an oral glucose
`tolerance test (OGTT) may have utility in monitoring
`changes in glucose homeostasis. OGTT is the only
`means of
`identifying people with impaired glucose
`tolerance. The WHO recommends that OGTT should
`be used in individuals with fasting plasma glucose of
`
`CD
`General population
`
`80
`
`70
`
`60
`
`50
`
`40
`
`30
`
`20
`
`10
`
`0
`
`Prevalence (%)
`
`Overweight
`
`Obesity
`
`Hypertension
`
`IGT
`
`Diabetes
`
`Major
`depression
`
`Osteoporosis
`
`Figure 1 Prevalence of morbidities in patients with CD vs the
`general population in the USA, 2006. Prevalence rates in the
`general population are derived from Centers for Disease Control
`and Prevention (USA) statistics and National Health and Nutrition
`Examination Survey III findings.
`
`petrosal sinus sampling. Available studies suggest a high
`risk of venous thrombosis. van Zaane et al. (57) have
`summarized the literature on the effects of endogenous
`hypercortisolism on coagulation and fibrinolysis.
`Venous thromboembolism (VTE) was reported as the
`cause of death in 0–1.9% of CS patients (57). The
`incidence of VTE in CD has been reported as 2.5–3.1 per
`1000 persons per year (57), whereas the incidence in
`the general population is 1.0–2.0 per 1000 persons per
`year (58). A recent retrospective cohort study among
`473 Dutch patients with CS showed an incidence of 14.6
`pretreatment VTE cases per 1000 persons per year. In
`addition, it was found that the postoperative risk of VTE
`(3.4%) in patients with CD is greater than that of
`patients with nonfunctional pituitary adenomas (0%).
`Most postoperative VTE occurred between 1 week and 2
`months after surgery (59). It is not entirely clear yet how
`long and to what extent patients remain at risk of VTE
`after remission. A recent study showed that short-term
`biochemical remission (i.e. 3 months) induced by
`medical therapy is not accompanied by normalization
`of procoagulant and fibrinolytic parameters (60).
`Manetti et al. (61) observed an improvement in these
`parameters 1 year after successful surgery, although
`hemostasis did not completely normalize. Persistence of
`abdominal obesity may in part maintain activation of
`the coagulation cascade. Future studies should further
`examine the time to reversal of the hypercoagulable state
`after successful treatment of CD. Because exogenous
`glucocorticoids can influence components of pro- and
`anticoagulative pathways, the risk of VTE should be
`evaluated in patients who are chronically treated with
`supraphysiological glucocorticoid dosages (62).
`Before introduction of anticoagulant prophylaxis,
`10% of CD patients died due to thromboembolism and
`10% had vascular morbidity. Since the introduction of
`prophylaxis, however, morbidity and mortality due to
`thromboembolic events were reduced to 6 and 0.4%
`respectively (49). However, prophylaxis is not routinely
`applied in all centers. There is a clear need for guidelines
`on the dose and duration of
`thromboprophylaxis
`in patients with CD, both in the pre- and the post-
`operative phases.
`
`www.eje online.org
`
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`Corcept Ex. 2009, Page 6
`
`

`

`EUROPEAN JOURNAL OF ENDOCRINOLOGY (2012) 167
`
`The burden of Cushing’s disease
`
`317
`
`(110–125 mg/dl)
`6.1–6.9 mmol/l
`glucose tolerance status (65).
`
`to determine the
`
`Dyslipidemia The prevalence of dyslipidemia in
`patients with CD ranges from 38 to 71% in the
`literature (22, 53). A study by Mancini et al. (22) has
`shown dyslipidemia to occur less frequently than the
`other metabolic complications and that it was not
`correlated to the degree of hypercortisolism or duration
`of
`the disease. The majority of patients examined
`presented with three or more risk factors and a ‘high’
`or ‘very high’ global cardiovascular risk, calculated
`according to WHO/ISI guidelines (22). However, the
`causative role of cortisol excess for dyslipidemia is not
`extensively described in the literature and the findings
`are controversial.
`In some study populations,
`the
`prevalence of hypertriglyceridemia was lower than in
`BMI-matched controls (12). It is recommended that
`dyslipidemia should be aggressively treated in patients
`with CS to manage increased cardiovascular morbidity
`and mortality risk (66). Although surgical remission is
`often associated with the normalization of lipid profiles,
`patients requiring medical
`treatment
`for persistent
`hypercortisolism present specific challenges according
`to the therapeutic agents used. For example, ketocona-
`zole, a potent inhibitor of cytochrome P450 3A4
`(CYP3A4), may significantly increase plasma concen-
`trations of certain statins (such as simvastatin and
`atorvastatin) that undergo metabolism by the same
`pathway, thus increasin