Early diagnosis and immediate treatment of Cushing’s syndrome (CS) are critical for a better prognosis but remain a challenge. However, few comprehensive reports have focused on this issue or investigated whether patient-reported manifestations are consistent with physician-assessed symptoms of CS. This study aimed to clarify the differences in patient-reported and physician-assessed manifestations of signs and symptoms of CS that prevent early diagnosis.
Methods
This single-center retrospective study included 52 patients with CS (16 with Cushing’s disease and 36 with adrenal CS). Upon clinical diagnosis, medical records were used to independently review the patient-reported and physician-assessed manifestations of typical (such as purple striae and proximal myopathy) and nonspecific features (such as hirsutism and hypertension). The correlations and differences between the patient-reported and physician-assessed manifestations were then analyzed.
Results
We observed a positive correlation between the total number of manifestations of nonspecific features reported by patients and those assessed by physicians, but not for typical features. Moreover, manifestations reported by the patients were less frequent than those assessed by physicians for typical features, leading to discrepancies between the two groups. In contrast, there were no differences in most nonspecific features between the patient-reported and physician-assessed manifestations. Notably, the concordance between patient-reported and physician-assessed manifestations of typical features was not associated with urinary free cortisol levels.
Conclusion
Regardless of disease severity, patients often do not complain of the typical features of CS that are crucial for formulating a diagnosis.
Introduction
Endogenous Cushing’s syndrome (CS) is caused by chronic and excessive glucocorticoid exposure. This occurs primarily due to adrenocorticotropic hormone (ACTH)-producing pituitary tumors (Cushing’s disease; CD) or cortisol-producing adrenal tumors (adrenal Cushing’s syndrome; ACS) [1]—and has a high mortality rate owing to cardiovascular disease, severe infection, and suicide, even when diagnosed and treated appropriately [1, 2]. Moreover, the prognosis is poor if the disease is not adequately treated or remains undiagnosed [2]. Therefore, early diagnosis and immediate intervention are important, as remission of CS due to surgical and pharmacological treatment can reduce the risk of mortality [3, 4].
CS is a rare disease with a prevalence of 57 per million individuals and an annual incidence of 3.2 per million, and its epidemiology is consistent across various regions worldwide [5, 6]. Most symptoms and signs of CS are common in general metabolic disorders, including obesity, hypertension, osteoporosis, and diabetes mellitus [7]. However, CS should be suspected if these symptoms appear as unusual features for their age [1, 8]. Consequently, the identification of CS is challenging and labor-intensive [1, 9, 10]. In fact, recent research revealed that a definitive diagnosis of CD (the most common form of CS), took an average of 3.8 ± 4.8 years from the onset of symptoms, and patients typically consulted 4.6 ± 3.8 medical professionals before this disease was identified [11]. Typical features of CS include symptoms of moon face, central obesity, or buffalo hump [12], which are similar to other symptoms such as primary obesity and therefore can lead to misdiagnosis. Furthermore, although purple striae or thin skin with an increased propensity for bruising are other typical features of CS [12], these attributes are not commonly acknowledged by the general population [1, 9].
Anzeige
Attempts have been made to diagnose CS early, including the development of scoring systems to estimate the pre-test probability of CS and facial image analysis software to diagnose the specific facial features of CS [13‐15]; however, these have not yet been used widespread or fully and the early diagnosis of CS remains dependent on the experience-based medical skills of the clinical staffs [16].
Additionally, although it is difficult for patients to recognize complex and nonspecific symptoms [17, 18], the significance of patients recognizing their illness has recently been reported for various diseases such as heart failure and malignant carcinoma [19‐21]. It is widely acknowledged that patients’ self-recognition can result in early detection of the disease, reduce its severity and recurrence, and enhance their quality of life [19]. In patients with endocrine diseases, there is increasing focus on issues surrounding self-recognition [22‐24]. For example, a previous study focusing on acromegaly reported a discrepancy between patient-reported and physician-reported manifestations and indicated that resolving this discrepancy could shorten the time to diagnosis [25].
Identifying CS may be challenging for primary care physicians who are yet to specialize. Therefore, endocrinologists with extensive experience in CS have often noticed that patients and these physicians struggle to identify the symptoms of CS; however, few comprehensive reports have focused on this issue or investigated whether patient-reported manifestations are consistent with physician-assessed symptoms of CS.
Therefore, this study aimed to investigate the unreported manifestations of CS among individuals referred to non-specialist healthcare providers, including primary care physicians, and to recognize potential challenges with the current diagnosis of CS with the goal of facilitating early detection.
Anzeige
Materials and methods
Patients, study design, and data collection
This single-center retrospective study was conducted to identify the discrepancies between patient-reported and physician-assessed symptoms and investigate the factors causing these differences.
From September 2004 to December 2022, 199 patients were referred to our department at a tertiary medical institution upon suspicion, evaluation, or follow-up for hypercortisolism. Of these patients, 92 were newly diagnosed with CS (36 with CD, 51 with ACS, and 5 with ectopic ACTH syndrome) based on the diagnostic guidelines [3, 8, 12], with a diagnosis confirmed by pathological evaluation after surgical resection [26]. However, 35 patients were excluded due to a lack of detailed clinical data on the manifestations at diagnosis. Similarly, we excluded individuals diagnosed with ectopic ACTH syndrome because of the lack of comprehensive information on symptoms reported by the patients and primary care physicians due to the rapid progression and severity of this disease. Therefore, 52 patients (16 with CD and 36 with ACS) were enrolled in this study.
Upon clinical diagnosis, the manifestations included in the comprehensive standardized interview at the time of diagnosis and those assessed by the physician through collaborative assessment with multiple board-certified endocrinologists as routine practice were independently reviewed from the medical records. We categorized these manifestations reviewed from the medical records into the following two categories based on the diagnostic guidelines including those of the Japan Endocrine Society: typical features, including moon face, central obesity or buffalo hump, purple striae of ≥1 cm, thin skin and easy bruising, and proximal myopathy; and nonspecific features (shown as atypical in Japan Endocrine Society’s guideline), including hypertension, menstrual abnormalities, acne, hirsutism, peripheral edema, glucose metabolism impairment, osteoporosis, pigmentation (which is not expected in patients with ACS), and mental abnormalities [1, 8, 12]. Central obesity or buffalo hump can also be observed in pseudo CS. However, in this study, features were classified as the same typical feature according to clinical guidelines [12, 27]. We also reviewed the biochemical findings, comorbidities, duration from the initial recognition of CS-related symptoms to diagnosis, and number of medical institutions visited before diagnosis.
The present retrospective study was performed in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Kobe University Hospital (Approval No. 1351). The patients had the option of an opt-out process, and all procedures were part of routine medical care.
Definition of patient-reported and physician-assessed manifestations
In the context of routine clinical care, physicians asked the patients about the presence or absence of manifestations and comorbidities (e.g., hypertension, menstrual abnormalities, glucose metabolism impairment, osteoporosis, and mental abnormalities), which were documented in the medical records. These reports in the medical records were defined as patient-reported manifestations in this study. In contrast, the manifestations and comorbidities of CS were assessed within several weeks after the patient was referred to our department for suspected CS. Additional diagnostic information on comorbidities is provided in the subsequent section. Physician-assessed manifestations were subsequently defined based on these findings.
Comorbidities of Cushing’s syndrome
All comorbidities were diagnosed according to the appropriate guidelines [28‐30]. For example, hypertension was diagnosed if patients were taking oral antihypertensive medication or had more than grade 1 hypertension (≥140/90 mmHg) in a treatment-naïve state [28]. Moreover, glucose metabolism impairment—including diabetes mellitus, impaired glucose tolerance, and impaired fasting glucose—was diagnosed based on the results of blood glucose levels during fasting and after a 75-g oral glucose tolerance test, as well as hemoglobin A1c (HbA1c) levels [29]. Patients taking medications for diabetes mellitus at the time of CS diagnosis were also categorized as having diabetes.
Other comorbidities included mental abnormalities, menstrual abnormalities, and the presence of osteoporosis. Mental abnormalities were defined as the use of anxiolytic medications, sleeping pills, or antidepressants prescribed by experienced psychologists, and menstrual abnormalities were defined as women with irregular menstrual cycles. Furthermore, the presence of osteoporosis was defined as bone mineral density (BMD) of <–2.5 standard deviations (SD) of the T-score of the lumbar vertebrae (L2–L4), femoral neck, or distal radius measured using dual-energy x-ray absorptiometry (DXA; Horizon A DXA System), and/or an experience of a fragility fracture [30]. As per the specifications of the measurement system employed, L1 was not included in the assessment. The Z-score was also employed as a diagnostic reference among young adults. Patients also diagnosed with osteoporosis who were receiving medications for this disease.
Hormone assay
In this study, blood samples were collected after an overnight fast. Subsequently, serum cortisol levels were measured using a chemiluminescent enzyme immunoassay [CLEIA] (TOSOH, Tokyo, Japan, RRID:AB_3099658) or enzyme immunoassay [EIA] (TOSOH, Tokyo, Japan, RRID:AB_3076600). Similarly, plasma ACTH levels were measured using a CLEIA (TOSOH, Tokyo, Japan, RRID:AB_3099657, or Siemens, Tokyo, Japan, RRID:AB_2909441) and EIA (TOSOH, Tokyo, Japan, RRID:AB_2783633). In both methods, the measurements showed good correlation and no conversion was required [31, 32].
Anzeige
Urinary free cortisol (UFC) levels were also measured using radioimmunoassays (RIA; TFB, Tokyo, Japan, RRID:AB_2894408) or chemiluminescent immunoassays (CLIA; Siemens, Tokyo, Japan, RRID:AB_2893154). Using the following formula, the UFC levels measured by RIA were then corrected to the value measured by CLIA: Y = 0.832X − 4.23 (Y = UFC levels using CLIA, X = UFC levels using RIA) [33].
Statistical analysis
All statistical analyses were performed using SPSS ver. 28.0 software (IBM Corp., Armonk, NY, USA). All continuous variables were analyzed using the Shapiro–Wilk normality test to confirm a normal distribution, whereas Fisher’s exact test was used to analyze categorical data. Between the two groups, differences in normally or non-normally distributed data were compared using the unpaired Student’s t-test or the Mann–Whitney U test, respectively.
Cohen’s kappa coefficient was used to describe the concordance between the patient-reported and physician-assessed manifestations. As previously reported [19, 20, 34], the concordance based on the value of Cohen’s kappa coefficient was rated as follows: 0.00–0.20 for “Slight,” 0.21–0.40 for “Fair,” 0.41–0.60 for “Moderate,” 0.61–0.80 for “Substantial,” and 0.81–1.00 for “Almost Perfect.” For correlation analysis between two variables of non-normally distributed data, we used Spearman’s rank correlation coefficient. Multivariate logistic regression analyses were then performed to investigate variables associated with the discrepancies between patient-reported and physician-assessed manifestations.
The results are presented as mean ± SD for normally distributed data and median [interquartile range] for non-normally distributed data, and differences were considered statistically significant when the P value was <0.05.
Anzeige
Results
Clinical characteristics of the patients
We included 52 patients diagnosed with CS in this study. Their clinical characteristics are presented in Table 1. Notably, this group consisted of 5 males and 47 females, with a mean age of 49.4 ± 15.8 years, median body mass index (BMI) of 23.0 [21.3–28.0] kg/m2, and median UFC level of 272.1 [126.0–435.0] µg/day. Of the CS patients, 16 had CD and 36 had ACS, which is consistent with epidemiological data on CS observed in Asians (including Japanese individuals); however, this differed from epidemiological data from Western countries [35, 36]. Regarding comorbidities, 43 patients were diagnosed with hypertension—of which 34 were prescribed antihypertensive medications—with a mean systolic blood pressure (BP) of 136.4 ± 21.5 mmHg and diastolic BP of 83.5 ± 15.0 mmHg. In addition, 44 patients were diagnosed with glucose metabolism impairment—of which, 20 were prescribed oral hypoglycemic agents and/or insulin—with a median fasting serum glucose level of 99.5 [87.3–116.5] mg/dL and median HbA1c level of 6.3% [5.7–7.4]. Moreover, 29 patients were diagnosed with osteoporosis, of which 4 were prescribed antiosteoporosis medication, with BMD T-score SDs of -1.54 ± 1.39, -1.76 ± 1.12, and -0.50 [-1.53–0.50] for the lumber spine, femoral neck, and distal radius, respectively. Notably, the UFC levels were higher in patients with CD than in those with ACS (412.6 [243.2–1,100.3] vs. 215.3 [114.0–387.8] µg/day); however, there were no significant differences attributed to sex, age, BMI, or the proportion of patients with respect to comorbidities, including hypertension and glucose metabolism impairment, between patients with CD and ACS.
Table 1
Clinical characteristics of the patients
Total
CD
ACS
CD vs. ACS P value
Number of men/women
5/47
1/15
4/32
1.00
Age (years)
49.4 ± 15.8
54.3 ± 19.2
47.2 ± 13.8
0.14
BMI (kg/m2)
23.0 [21.3–28.0]
24.7 [22.2–30.0]
22.8 [20.8–26.4]
0.17
Midnight F (µg/dL)
20.1 [16.0–23.5]
20.2 [13.9–24.7]
20.1 [16.9–23.0]
0.97
F after LDDST (μg/dL)
21.2 ± 6.9
24.2 ± 10.1
19.7 ± 4.2
0.11
UFC (μg/day)
272.1 [126.0–435.0]
412.6 [243.2–1,100.3]
215.3 [114.0–387.8]
0.02
Basal ACTH (pg/mL)
2.0 [0.0–53.9]
83.2 [57.4–169.9]
0.0 [0.0–2.1]
<0.01
Systolic BP (mmHg)
136.4 ± 21.5
140.5 ± 20.7
134.6 ± 21.8
0.36
Diastolic BP (mmHg)
83.5 ± 15.0
83.1 ± 14.3
83.6 ± 15.5
0.90
Use of antihypertensive drugs, n (%)
34 (65)
13 (81)
21 (58)
0.13
FSG (mg/dL)
99.5 [87.3–116.5]
110.0 [102.0–142.8]
92.5 [83.3–114.3]
0.01
HbA1c (%)
6.3 [5.7–7.4]
6.8 [5.9–8.6]
6.0 [5.7–7.1]
0.08
Use of OHA and/or insulin, n (%)
20 (38)
9 (56)
11 (31)
0.12
LS BMD T-score (SD)
−1.54 ± 1.39
−1.00 ± 1.38
−1.79 ± 1.35
0.07
LS BMD Z-score (SD)
−0.78 ± 1.37
0.13 ± 1.11
−1.20 ± 1.28
<0.01
FN BMD T-score (SD)
−1.76 ± 1.12
−1.73 ± 1.54
−1.78 ± 0.88
0.92
FN BMD Z-score (SD)
−0.79 ± 1.01
−0.39 ± 1.10
−0.99 ± 0.92
0.10
Radius BMD T-score (SD)
−0.50 [−1.53–0.50]
−0.30 [−2.50–0.40]
−0.60 [−1.30–0.60]
0.79
Radius BMD Z-score (SD)
0.60 [−0.60–1.50]
1.50[−0.60–1.80]
0.50[−0.50–1.00]
0.33
Use of antiosteoporosis drugs, n (%)
4 (8)
1 (6)
3 (8)
1.00
Time to diagnosis (months)
44.0 [13.3–125.3]
43.0 [15.0–128.3]
47.5 [12.5–125.3]
0.87
Number of medical institutions before diagnosis
3.0 [2.0–5.0]
3.0 [2.0–5.0]
3.0 [3.0–5.8]
0.23
The results are presented as mean ± SD for normally distributed data and median [interquartile range] for non-normally distributed data
CD Cushing’s disease, ACS adrenal Cushing’s syndrome, BMI body mass index, F cortisol, LDDST low-dose dexamethasone suppression test, UFC urinary free cortisol, ACTH adrenocorticotropic hormone, BP blood pressure, FSG fasting serum glucose, HbA1c hemoglobin A1c, OHA oral hypoglycemic agents, BMD bone mineral density, LS lumber spine, FN femoral neck
The median duration from the patients’ initial recognition of CS-related manifestations to diagnosis was 44.0 [13.3–125.3] months, and it took more than 3 years to diagnose CS in 30 patients (58%). Furthermore, the median number of medical facilities visited by patients before diagnosis was 3.0 [2.0–5.0]; however, there were no significant differences in the duration or number of medical institutions between patients with CD and those with ACS.
Frequency and concordance between patient-reported and physician-assessed CS-related manifestations
Each manifestation reported by a patient or assessed by a physician is shown vertically for individual cases in Fig. 1. Compared with nonspecific features, typical features appeared to not be reported by the patients but were only assessed by the physicians. In addition, compared to nonspecific features, there were fewer cases in which the manifestations reported by the patients were consistent with those assessed by physicians for typical features.
Fig. 1
Consistency between patient-reported and physician-assessed manifestations for each individual case. The consistencies or discrepancies between patient-reported and physician-assessed manifestations are shown. Vertical lines represent manifestations in individual patients. CD Cushing’s disease, ACS adrenal Cushing’s syndrome
Consistent with the impact of these visually distinctive presentations shown in Fig. 1, no correlation was observed in the number of typical features between patient-reported and physician-assessed manifestations (r = –0.20, P = 0.16) (Fig. 2A), whereas a positive correlation was found for nonspecific features (r = 0.62, P < 0.01) (Fig. 2B). Moreover, the total number of patient-reported manifestations of typical features was lower than that of physician-assessed manifestations (1.0 [0.0–2.0] vs. 3.5 [3.0–4.0], P < 0.01), and four of the five typical features were reported less frequently by patients than by physicians, except for proximal myopathy (Table 2A). According to Cohen’s kappa coefficient, the concordance between patient-reported and physician-assessed manifestations was marked as “Fair” to “Slight,” indicating a discrepancy for all typical features. Similarly, the total number of patient-reported manifestations of nonspecific features was also lower than that in physicians (2.5 [2.0–3.0] vs. 4.0 [3.0–5.0], P < 0.01). However, except for glucose metabolism impairment or osteoporosis, there were no differences in the frequencies of nonspecific features between patient-reported and physician-assessed manifestations, and the concordance of the nonspecific features between the patient-reported and physician-assessed manifestations was “Almost perfect” for menstrual abnormality and “Substantial” for mental abnormality and hypertension, whereas that for glucose metabolism impairment and osteoporosis was “Fair.” This suggests that the discrepancy between patient-reported and physician-assessed manifestations was more significant for typical than for nonspecific features. However, no differences in these discrepancies were observed between patients with CD and those with ACS (Table 2B, C).
Fig. 2
Correlation between the total number of patient-reported and physician-assessed manifestations. Correlations between the total number of patient-reported and physician-assessed manifestations are shown for typical (A) and nonspecific features (B). CD is plotted by ×, and ACS is plotted by ○. The Spearman’s rank correlation coefficients and P value are presented. CI confidence interval, CD Cushing’s disease, ACS adrenal Cushing’s syndrome
Table 2
Frequencies of patient-reported and physician-assessed manifestations and their concordance. A. All patients (n = 52). B. Patients with CD (n = 16). C. Patients with ACS (n = 36)
Patient-reported
Physician-assessed
P value of Fisher’s exact test
Concordance with Cohen’s kappa coefficient
A
Typical features
Moon face, n (%)
20 (39)
48 (92)
<0.01
Slight
Central obesity or buffalo hump, n (%)
13 (25)
44 (85)
<0.01
Slight
Purple striae, n (%)
3 (6)
15 (29)
<0.01
Fair
Thin skin and easy bruising, n (%)
15 (29)
43 (83)
<0.01
Slight
Proximal myopathy, n (%)
21 (40)
27 (52)
0.33
Slight
Nonspecific features
Hypertension, n (%)
39 (75)
43 (83)
0.47
Substantial
Menstrual abnormalities, n (%)
11 (21)
11 (21)
1.00
Almost perfect
Acne, n (%)
7 (14)
13 (25)
0.21
Moderate
Hirsutism, n (%)
3 (6)
10 (19)
0.07
Moderate
Peripheral edema, n (%)
24 (46)
28 (54)
0.56
Fair
Glucose metabolism impairment, n (%)
24 (46)
44 (85)
<0.01
Fair
Osteoporosis, n (%)
7 (14)
29 (56)
<0.01
Slight
Pigmentation, n (%)
0 (0)
5 (10)
0.06
–
Mental abnormalities, n (%)
17 (33)
17 (33)
1.00
Substantial
B
Typical features
Moon face, n (%)
6 (38)
14 (88)
0.01
Slight
Central obesity or buffalo hump, n (%)
6 (38)
15 (94)
<0.01
Slight
Purple striae, n (%)
2 (13)
4 (25)
0.56
Moderate
Thin skin and easy bruising, n (%)
4 (25)
13 (81)
0.06
Slight
Proximal myopathy, n (%)
8 (50)
8 (50)
1.00
Slight
Nonspecific features
Hypertension, n (%)
16 (100)
15 (94)
0.78
Slight
Menstrual abnormalities, n (%)
5 (31)
5 (31)
1.00
Almost perfect
Acne, n (%)
1 (6)
3 (19)
0.56
Moderate
Hirsutism, n (%)
2 (13)
4 (25)
0.56
Moderate
Peripheral edema, n (%)
8 (50)
10 (63)
0.56
Slight
Glucose metabolism impairment, n (%)
10 (63)
15 (94)
0.14
Slight
Osteoporosis, n (%)
4 (25)
9 (56)
0.15
Slight
Pigmentation, n (%)
0 (0)
5 (31)
0.14
–
Mental abnormalities, n (%)
5 (31)
6 (38)
0.78
Moderate
C
Typical features
Moon face, n (%)
14 (39)
34 (94)
<0.01
Slight
Central obesity or buffalo hump, n (%)
7 (19)
29 (81)
<0.01
Slight
Purple striae, n (%)
1 (3)
11 (31)
<0.01
Slight
Thin skin and easy bruising, n (%)
11 (31)
30 (83)
<0.01
Slight
Proximal myopathy, n (%)
13 (36)
19 (53)
0.24
Slight
Nonspecific features
Hypertension, n (%)
23 (64)
28 (78)
0.30
Substantial
Menstrual abnormalities, n (%)
6 (17)
6 (17)
1.00
Almost perfect
Acne, n (%)
6 (17)
10 (28)
0.40
Moderate
Hirsutism, n (%)
1 (3)
6 (17)
0.11
Fair
Peripheral edema, n (%)
16 (44)
18 (50)
0.81
Fair
Glucose metabolism impairment, n (%)
14 (39)
29 (81)
<0.01
Fair
Osteoporosis, n (%)
3 (8)
20 (56)
<0.01
Slight
Pigmentation, n (%)
0 (0)
0 (0)
–
–
Mental abnormalities, n (%)
12 (33)
11 (31)
1.00
Almost perfect
The frequencies of patient-reported and physician-assessed manifestations were compared using Fisher’s exact test. The concordance between patient-reported and physician-assessed manifestations was evaluated with Cohen’s kappa coefficient, and its coefficients were defined as follows: 0.00–0.20 for “Slight,” 0.21–0.40 for “Fair,” 0.41–0.60 for “Moderate,” 0.61–0.80 for “Substantial,” and 0.81–1.00 for “Almost perfect”
CD Cushing’s disease, ACS adrenal Cushing’s syndrome
Anzeige
We performed logistic regression analyses using UFC to investigate whether excess cortisol levels influenced the discrepancy between patient-reported and physician-assessed manifestations. Notably, we observed no association between UFC levels and discrepancies between patient-reported and physician-assessed manifestations in the univariate or multivariate logistic regression analyses adjusted for sex and age (Table 3A). In addition, no association was observed after adjusting for other variables such as BMI and disease duration. Similarly, we found that the serum cortisol levels after the low-dose dexamethasone suppression test (LDDST) were not associated with discrepancies between patient-reported and physician-assessed manifestations (Table 3B). Thus, these disparities were shown to be insignificant when directly related to the severity of CS.
Table 3
Logistic regression analyses of the discrepancies between the patient-reported and physician-assessed manifestations. A. Variables associated with UFC levels. B. Variables associated with serum cortisol levels after the LDDST
Univariate
Multivariate 1 (sex- and age-adjusted)
Multivariate 2 (BMI-adjusted)
Multivariate 3 (disease duration-adjusted)
A
Moon face
1.000 (0.999–1.001)
1.000 (0.999–1.001)
1.000 (0.998–1.002)
1.000 (0.999–1.001)
Proximal myopathy
1.000 (0.999–1.001)
1.000 (0.999–1.001)
1.000 (0.998–1.001)
1.000 (0.998–1.001)
Thin skin and easy bruising
1.000 (0.998–1.001)
1.000 (0.999–1.001)
1.000 (0.999–1.001)
1.000 (0.998–1.001)
Central obesity or buffalo hump
1.001 (1.000–1.003)
1.001 (1.000–1.003)
1.001 (1.000–1.003)
1.001 (1.000–1.003)
Purple striae
1.000 (0.999–1.002)
1.000 (0.998–1.002)
1.001 (0.999–1.003)
1.000 (0.999–1.002)
B
Moon face
0.998 (0.919–1.084)
0.999 (0.919–1.086)
1.000 (0.920–1.088)
0.997 (0.918–1.082)
Proximal myopathy
1.007 (0.925–1.097)
1.007 (0.924–1.097)
1.007 (0.925–1.097)
1.006 (0.924–1.096)
Thin skin and easy bruising
1.022 (0.939–1.112)
1.018 (0.934–1.109)
1.023 (0.940–1.113)
1.019 (0.937–1.109)
Central obesity or buffalo hump
0.979 (0.890–1.078)
0.978 (0.865–1.105)
0.981 (0.875–1.099)
0.978 (0.887–1.078)
Purple striae
0.998 (0.919–1.084)
0.999 (0.919–1.086)
1.000 (0.920–1.088)
0.997 (0.918–1.082)
The results are presented as odds ratios (95% confidence intervals)
UFC urinary free cortisol, BMI Body Mass Index, LDDST low-dose dexamethasone suppression test
Discussion
In the present study, we highlight the challenges associated with the diagnosis of CS—a condition resulting from excessive glucocorticoid exposure—and elucidate the divergence between patient-reported and physician-assessed manifestations. Thus, this study may aid in the early detection of CS by identifying symptoms that patients are unable to recognize based on the disparities between patient-reported and physician-assessed manifestations of CS.
In this study, the number of patient-reported manifestations of both typical and nonspecific features was lower than that of physician-assessed manifestations, suggesting that CS symptoms may have been overlooked by relying solely on patient reports. Additionally, analysis of the concordance between patient-reported and physician-assessed manifestations revealed a tendency for these manifestations to be inconsistent for both typical and nonspecific features, with a tendency to be more significant for typical features. Furthermore, the UFC and serum cortisol levels after the LDDST, which represent the severity of CS, were not associated with the concordance of manifestations between patients and physicians, suggesting that even in cases of severe CS, patients may not recognize their symptoms. These findings imply that typical features, which are essential for diagnosing CS, may be difficult for patients to recognize and poorly identified or conveyed to patients by non-specialist physicians, who are typically the first to interact with individuals with CS. The importance of educating healthcare providers such as primary care physicians, family physicians and gynecologists for early diagnosis of CS should be highlighted.
According to a previous report on the diagnostic history of 176 patients with CD, 83% of the patients visited their family physician for manifestations such as weight gain and hypertension, while 46% visited a gynecologist for menstrual abnormalities before the diagnosis of CD [11]. Thus, the typical features of CS were not recognized. The examination may reveal nonspecific features. However, individuals who are non-specialists may not recognize these features as indications of CS. Therefore, patients are often unaware of the potential complications associated with CS. This is consistent with the results of our study, in which patient-reported and physician-assessed manifestations were more consistent for hypertension and menstrual abnormalities than for other manifestations such as typical features, glucose metabolism impairment, and osteoporosis. This makes diagnosis challenging as non-specialist physicians and, more prominently, patients may not recognize the full range of symptoms associated with CS, especially the typical features with high diagnostic value. In addition, older patients diagnosed with CS present with a lower BMI and waist circumference than younger patients [37], and they typically do not exhibit symptoms commonly associated with CS such as skin alterations, depression, hair loss, hirsutism, and reduced libido. These findings may further complicate the diagnosis of CS in elderly patients.
By evaluating only the patient-reported manifestations, it appears that manifestations such as peripheral edema and proximal myopathy were more common. Possibly, these symptoms were not considered features of CS by physicians, in comparison to the degree of symptoms experienced by the patients. However, this may not necessarily imply diminishing the significance of the patient’s signs and symptoms, as these manifestations can be considered as the unidentified complaints and may result in a postponement of the diagnosis of CS. Patients may be experiencing symptoms that physicians do not perceive, indicating the importance of interview and physical examination. Further investigation is needed to elucidate underlying factors.
Considering the rarity of CS, it is crucial to suspect and diagnose the condition based on clinical symptoms and perform the appropriate screening tests without over- or under-screening [7]. Although CS screening in patients with diabetes mellitus and hypertension has been reported to lead to a diagnosis in only 0–0.7% and 0.1–0.5% of these patients, respectively [38‐41], it is ineffective in terms of false positives and cost [9]. Therefore, patients with typical features that are highly specific for CS, such as purple striae, easy bruising, and proximal myopathy [1, 8, 12], as well as those with obesity, diabetes mellitus, or hypertension in combination with these features, should be screened for CS [7, 27]. However, our results suggest that these symptoms are unlikely to be self-recognized. Therefore, the appropriate screening measures must be implemented to establish an early and effective diagnosis of CS.
In these situations, it is crucial for physicians to utilize their knowledge and experience to suspect CS based on symptoms such as typical features [10]. It has been reported that years of clinical experience in endocrine practice can contribute to the estimation of the pre-test probability of CS [16]. In contrast, non-specialists are less likely to encounter patients with CS in their lifetime, which can make it difficult to properly suspect CS [9]. From this perspective, it is of utmost importance that family physicians and general internists are knowledgeable regarding the manifestations that require screening for CS, as early diagnosis of this uncommon and severe condition is crucial [11]. Therefore, it is important for physicians who routinely treat patients presenting with common symptoms such as obesity, diabetes mellitus, and hypertension to meticulously interview and observe for any indicators of CS, even if the patient does not recognize them. Failure to adopt an appropriate tone in these situations may cause the disease to become undetectable.
In rare disorders such as CS, in addition to enhancing public recognition of the disease, the appropriate sharing of information and provision of specialized care in clinical practice remain important issues [42]. Early identification of such rare diseases can be achieved by promoting an understanding of the disease and its symptoms among family, friends, and patients who may be the first to recognize the signs and symptoms in an individual. In fact, in a questionnaire survey of 340 patients with CS across 30 countries, the diagnosis of CS was made in 5.6% of cases by the patients themselves and in 0.9% by their family or friends [43]. In the present study, we found that it took more than 3 years to diagnose CS in 58% of the cases. If CS and its symptoms are popularized among the public, the typical features of CS could be more readily reported to physicians and the time to diagnosis might be shorter. Furthermore, a primary care physician who is well-educated and knowledgeable is crucial in ensuring that the concerns of such individuals are not overlooked.
This study has some limitations. First, this single-center retrospective study included a relatively small sample size with few male patients. Second, CD and ACS have different pathologies; therefore, the frequencies of several CS-related manifestations will differ depending on their subtypes [3, 44]. However, in this study, there was no difference in the discrepancies between patient-reported and physician-assessed manifestations in patients with CD or ACS. Nonetheless, it is crucial that comprehensive research is conducted in larger patient populations with a focus on employing methods that accurately reflect the pathophysiology of CD and ACS. Third, patient reports may be inaccurate in terms of onset and duration because they depend on the patient’s memory. Fourth, the endocrinologists who examined the patients differed, which may have affected the presence or absence of physician-assessed manifestations. Finally, this study investigated the differences between the manifestations reported by patients and those assessed by endocrinologists, although the evaluations conducted by primary care physicians, which are crucial for the early detection of CS, were not available. Future research is needed to investigate the differences in recognizing manifestations between non-specialist physicians and endocrinologists with extensive experience in CS and to examine the changes before and after education for these non-specialists to determine if they can lead to earlier diagnosis of CS.
In conclusion, endocrinologists have been shown to be aware of CS-related symptoms, especially typical features, whereas patients do not recognize these manifestations, even when the disease is severe. Therefore, the key to the early diagnosis and treatment of CS is a more proactive approach of questioning and examining patients suspected of having the disease.
Acknowledgements
We thank all the physicians and medical assistants who were involved in this study. We are grateful to all the laboratory members for their excellent discussions and fruitful suggestions. We also thank Editage (www.editage.jp) for English language editing.
Compliance with ethical standards
Conflict of interest
The authors declare no competing interests.
Ethics approval
This study conformed to the Declaration of Helsinki guidelines and was approved by the Ethics Committee of Kobe University Hospital (Approval No. 1351).
Informed consent
Informed consent was obtained from all the participants using an opt-out approach.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
I think members of the Cushing’s Help boards have been saying this forever! Cushing’s isn’t all that rare. Just rarely diagnosed,
BOSTON — The prevalence of Cushing syndrome (CS) in the United States may be considerably higher than currently appreciated, new data from a single US institution suggest.
In contrast to estimates of 1 to 3 cases per million patient-years from population-based European studies, researchers at the University of Wisconsin, Milwaukee, estimated that the incidence of CS in Wisconsin is a minimum of 7.2 cases per million patient-years. What’s more, contrary to all previous studies, they found that adrenal Cushing syndrome was more common than pituitary adrenocorticotropic hormone (ACTH)– secreting tumors (Cushing disease), and that fewer than half of individuals with adrenal Cushing syndrome had classic physical features of hypercortisolism, such as weight gain, round face, excessive hair growth, and stretch marks.
“Cases are absolutely being missed…. Clinicians should realize that cortisol excess is not rare. It may not be common, but it needs to be considered in patients with any constellation of features that are seen in cortisol excess,” study investigator Ty B. Carroll, MD, Associate Professor of Medicine, Endocrinology and Molecular Medicine, and the Endocrine Fellowship Program Director at Medical College of Wisconsin in Milwaukee, told Medscape Medical News.
There are several contributing factors, he noted, “including the obesity and diabetes epidemics which make some clinical features of cortisol excess more common and less notable. Providers get used to seeing patients with some features of cortisol excess and don’t think to screen. The consequence of this is more difficult-to-control diabetes and hypertension, more advance metabolic bone disease, and likely more advanced cardiovascular disease, all resulting from extended exposure to cortisol excess,” he said.
Are Milder Cases the Ones Being Missed?
Asked to comment, session moderator Sharon L. Wardlaw, MD, professor of medicine at Columbia University College of Physicians and Surgeons, New York City, said “When we talk about Cushing [syndrome], we usually think of pituitary ACTH as more [common], followed by adrenal adenomas, and then ectopic. But they’re seeing more adrenal adenoma…we are probably diagnosing this a little more now.”
She also suggested that the Wisconsin group may have a lower threshold for diagnosing the milder cortisol elevation seen with adrenal Cushing syndrome. “If you screen for Cushing with a dexamethasone suppression test…[i]f you have autonomous secretion by the adrenal, you don’t suppress as much…. When you measure 24-hour urinary cortisol, it may be normal. So you’re in this in-between [state]…. Maybe in Wisconsin they’re diagnosing it more. Or, maybe it’s just being underdiagnosed in other places.”
She also pointed out that “you can’t diagnose it unless you think of it. I’m not so sure that with these mild cases it’s so much that it’s more common, but maybe it’s like thyroid nodules, where we didn’t know about it until everybody started getting all of these CT scans. We’re now seeing all these incidental thyroid nodules…I don’t think we’re missing florid Cushing.”
However, Wardlaw said, it’s probably worthwhile to detect even milder hypercortisolism because it could still have long-term damaging effects, including osteoporosis, muscle weakness, glucose intolerance, and frailty. “You could do something about it and normalize it if you found it. I think that would be the reason to do it.”
Is Wisconsin Representative of Cushing Everywhere?
Carroll presented the findings at the annual meeting of the Endocrine Society. He began by noting that most of the previous CS incidence studies, with estimates of 1.2-3.2 cases per million per year, come from European data published from 1994 to 2019 and collected as far back as 1955. The method of acquisition of patients and the definitions of confirmed cases varied widely in those studies, which reported CS etiologies of ACTH-secreting neoplasms (pituitary or ectopic) in 75%-85% and adrenal-dependent cortisol excess in 15%-20%.
The current study included data from clinic records between May 1, 2017, and December 31, 2022, of Wisconsin residents newly diagnosed with and treated for CS. The CS diagnosis was established with standard guideline-supported biochemical testing and appropriate imaging. Patients with exogenous and non-neoplastic hypercortisolism and those who did not receive therapy for CS were excluded.
A total of 185 patients (73% female, 27% male) were identified from 27 of the total 72 counties in Wisconsin, representing a population of 4.5 million. On the basis of the total 5.9 million population of Wisconsin, the incidence of CS in the state works out to 7.2 cases per million population per year, Carroll said.
However, data from the Wisconsin Hospital Association show that the University of Wisconsin’s Milwaukee facility treated just about half of patients in the state who are discharged from the hospital with a diagnosis of CS during 2019-2023. “So…that means that an actual or approximate incidence of 14-15 cases per million per year rather than the 7.2 cases that we produce,” he said.
Etiologies were 60% adrenal (111 patients), 36.8% pituitary (68 patients), and 3.2% ectopic (6 patients). Those proportions were similar between genders.
On biochemical testing, values for late-night salivary cortisol, dexamethasone suppression, and urinary free cortisol were highest for the ectopic group (3.189 µg/dL, 42.5 µg/dL, and 1514.2 µg/24 h, respectively) and lowest for the adrenal group (0.236 µg/dL, 6.5 µg/dL, and 64.2 µg/24 h, respectively). All differences between groups were highly statistically significant, at P < .0001, Carroll noted.
Classic physical features of CS were present in 91% of people with pituitary CS and 100% of those ectopic CS but just 44% of individuals with adrenal CS. “We found that adrenal-dependent disease was the most common form of Cushing syndrome. It frequently presented without classic physical features that may be due to the milder biochemical presentation,” he concluded.
Carroll reports consulting and investigator fees from Corcept Therapeutics. Wardlaw has no disclosures.
Miriam E. Tucker is a freelance journalist based in the Washington DC area. She is a regular contributor to Medscape, with other work appearing in The Washington Post, NPR’s Shots blog, and Diatribe. She is on X (formerly Twitter) @MiriamETucker.
Today’s Cushing’s Awareness Challenge post is about kidney cancer (renal cell carcinoma). You might wonder how in the world this is related to Cushing’s. I think it is, either directly or indirectly.
I finally started the Growth Hormone December 7, 2004. Was the hassle and 3 year wait worth it? Stay tuned for tomorrow, April 15, 2016 when all will be revealed.
So, as I said, I started Growth Hormone for my panhypopituitarism on December 7, 2004. I took it for a while but never really felt any better, no more energy, no weight loss. Sigh.
April 14 2006 I went back to the endo and found out that the arginine test that was done in 2004 was done incorrectly. The directions were written unclearly and the test run incorrectly, not just for me but for everyone who had this test done there for a couple years. My endo discovered this when he was writing up a research paper and went to the lab to check on something.
So, I went off GH again for 2 weeks, then was retested. The “good news” was that the arginine test is only 90 minutes now instead of 3 hours.
Wow, what a nightmare my arginine retest started! I went back for that Thursday, April 27, 2006. Although the test was shorter, I got back to my hotel and just slept and slept. I was so glad that I hadn’t decided to go right home after the test.
Friday I felt fine and drove back home, no problem. I picked up my husband for a biopsy he was having and took him to an outpatient surgical center. While I was there waiting for the biopsy to be completed, I started noticing blood in my urine and major abdominal cramps.
There were signs all over that no cellphones were allowed so I sat in the restroom (I had to be in there a lot, anyway!) and I left messages for several of my doctors on what I should do. It was Friday afternoon and most of them were gone 😦 I finally decided to see my PCP after I got my husband home.
When Tom was done with his testing, his doctor took one look at me and asked if I wanted an ambulance. I said no, that I thought I could make it to the emergency room ok – Tom couldn’t drive because of the anaesthetic they had given him. I barely made it to the ER and left the car with Tom to park. Tom’s doctor followed us to the ER and instantly became my new doctor.
They took me in pretty fast since I was in so much pain, and had the blood in my urine. At first, they thought it was a kidney stone. After a CT scan, my new doctor said that, yes, I had a kidney stone but it wasn’t the worst of my problems, that I had kidney cancer. Wow, what a surprise that was! I was admitted to that hospital, had more CT scans, MRIs, bone scans, they looked everywhere.
My new “instant doctor” felt that he wasn’t up to the challenge of my surgery, so he called in someone else. My next new “instant doctor” came to see me in the ER in the middle of the night. He patted my hand, like a loving grandfather might and said “At least you won’t have to do chemotherapy”. And I felt so reassured.
It wasn’t until later, much after my surgery, that I found out that there was no chemo yet that worked for my cancer. I was so thankful for the way he told me. I would have really freaked out if he’d said that nothing they had was strong enough!
My open radical nephrectomy was May 9, 2006 in another hospital from the one where the initial diagnosis was made. My surgeon felt that he needed a specialist from that hospital because he believed preop that my tumor had invaded into the vena cava because of its appearance on the various scans. Luckily, that was not the case.
My entire left kidney and the encapsulated cancer (10 pounds worth!) were removed, along with my left adrenal gland and some lymph nodes. Although the cancer (renal cell carcinoma AKA RCC) was very close to hemorrhaging, the surgeon believed he got it all.
He said I was so lucky. If the surgery had been delayed any longer, the outcome would have been much different. I will be repeating the CT scans every 3 months, just to be sure that there is no cancer hiding anywhere. As it turns out, I can never say I’m cured, just NED (no evidence of disease). This thing can recur at any time, anywhere in my body.
I credit the arginine re-test with somehow aggravating my kidneys and revealing this cancer. Before the test, I had no clue that there was any problem. The arginine test showed that my IGF is still low but due to the kidney cancer I couldn’t take my growth hormone for another 5 years – so the test was useless anyway, except to hasten this newest diagnosis.
So… either Growth Hormone helped my cancer grow or testing for it revealed a cancer I might not have learned about until later.
My five years are up now. When I was 10 years free of this cancer my kidney surgeon *thought* it would be ok to try the growth hormone again. I was a little leery about this, especially where I didn’t notice that much improvement.
Magdalena Stasiak,1 Przemysław Witek,2 Emilia Adamska-Fita,1 Andrzej Lewiński1,3
1Department of Endocrinology and Metabolic Diseases, Polish Mother’s Memorial Hospital—Research Institute, Lodz, Poland; 2Department of Internal Medicine, Endocrinology and Diabetes, Medical University of Warsaw; Mazovian Brodnowski Hospital, Warszawa, Poland; 3Department of Endocrinology and Metabolic Diseases, Medical University of Lodz, Lodz, Poland
Correspondence: Magdalena Stasiak, Department of Endocrinology and Metabolic Diseases, Polish Mother’s Memorial Hospital—Research Institute, 281/289 Rzgowska Street, Lodz, 93-338, Poland, Tel +48502049292, Fax +48422711140, Email mstasiak33@gmail.com
Abstract: Cushing’s disease (CD) is the most common cause of endogenous hypercortisolism. Osilodrostat was demonstrated to be efficient in treating CD, and the mean average dose required for CD control was < 11 mg/day. Potential differences in osilodrostat treatment between cortisol-producing adenoma (CPA) and CD have not been reported. The aim of this study was to present two patients with CPA in whom significant differences in the response to therapy compared to CD were found. We demonstrated a case of inverse response of cortisol levels with adrenal tumor progression during the initial dose escalation (Case 1). Simultaneously, severe exaggeration of hypercortisolism symptoms and life-threatening hypokalemia occurred. A further rapid dose increase resulted in the first noticeable cortisol response at a dose of 20 mg/day, and a full response at a dose of 45 mg/day. We also present a case that was initially resistant to therapy (Case 2). The doses required to achieve the first response and the full response were the same as those for Case 1. Our study demonstrated that osilodrostat therapy in patients with CPA may require a different approach than that in CD, with higher doses, faster dose escalation, and a possible initial inverse response or lack of response.
Chronic persistent hypercortisolism is a life-threatening condition that requires effective treatment. Untreated exposure to excessive cortisol secretion leads to severely increased morbidity and mortality due to cardiovascular diseases, thromboembolic events, sepsis, visceral obesity, impairment of glucose metabolism, and dyslipidaea, as well as musculoskeletal disorders, such as myopathy, osteoporosis, and skeletal fractures. Moreover, neuropsychiatric disorders, such as impairment of cognitive function, depression, or mania, as well as impairment of reproductive function can frequently occur.1,2 Cushing’s disease (CD) – a disorder caused by a pituitary adenoma secreting adrenocorticotropic hormone (ACTH) – is the most common cause of hypercortisolism. Cushing’s syndrome (CS) includes all other causes of cortisol excess, including ectopic ACTH production as well as direct cortisol overproduction by adrenal adenoma (cortisol-producing adenoma [CPA]) or adrenocortical carcinoma (ACC). Approximately 10% of hypercortisolism cases result from CPA. The first line therapy is a surgical resection of the tumor, which is the source of hormone excess. However, in many patients surgery is not fully efficient and other therapies are required to reduce cortisol levels. Additionally, due to severe cardiovascular complications and unstable DM, the surgical approach sometimes entails unacceptable risk and it is frequently postponed until cortisol levels are lowered. Pharmacotherapy with steroidogenesis inhibitors reduces cortisol levels and improves the symptoms of hypercortisolism.1,2 As CD is the most common cause of cortisol excess, most studies have focused on the efficacy and safety of novel steroidogenesis inhibitors, including patients with CD only.3–6 This is exactly the case with osilodrostat – a new potent inhibitor of 11β-hydroxylase.3–6 More data are available for metyrapone efficacy and safety in CSA,7 as the drug has been available much longer than osilodrostat. A study by Detomas et al, which reported results of comparison of efficacy of metyrapone and osilodrostat, included 4 patients with adrenal CS, among whom one CPA patient was treated with osilodrostat.8 Osilodrostat is approved in the United States to treat CD in patients in whom pituitary surgery was not curative or is contraindicated.9 In Poland, osilodrostat therapy is available for patients with all kinds of endogenous hypercortisolism not curative with other approaches, within a national program of emergency access to drug technologies.10 Reports on osilodrostat application in CPA are highly valuable as data on potential differences in the treatment regimens between CD and CPA are scarce.
Here, we present two patients with CPA in whom the response and doses of osilodrostat were different from those reported in patients with CD. The main purpose of this study was to demonstrate that the efficacy of osilodrostat in CPA is high, although initial resistance to treatment or even deterioration of hypercortisolism can occur during the application of lower doses of the drug.
Materials and Methods
Study Design and Patients
We retrospectively analyzed medical files of two consecutive patients with CPA treated with osilodrostat. The analysis included medical history, laboratory and imaging results as well as a detailed reports of adverse events.
Laboratory and Imaging Procedures
Serum cortisol and ACTH levels were measured by electrochemiluminescence immunoassay (ECLIA) using a Cobas e601 analyzer (Roche Diagnostics, Indianapolis, IN, USA). UFC excretion was measured by chemiluminescent microparticle immunoassay (CMIA) using an Abbott Architect ci4100 analyzer (Abbott, Abbott Park, IL, USA). Cross-reactivity with 11-deoxycortisol for this method is very low (2.1% according to the manufacturer’s data). Potassium levels were measured by ion-selective electrode potentiometry using a Beckman Coulter DxC 700 AU Chemistry Analyzer (Beckman Coulter, Brea, CA, USA). Computed tomography (CT) imaging was performed using a Philips Ingenuity Core 128 system (Philips, the Netherlands).
Ethics Procedures
Informed consent was obtained from all subjects involved in the study. Written informed consent was obtained from the patients for publication of this paper. The approval of Institutional Ethics Committee was obtained to publish the case details (approval code KB 33/2023).
Presentation of the Cases
Case 1
A 51-year-old female was referred to our department in November 2021 because of CPA, disqualified from surgery because of severe hypertension with a poor response to antihypertensive therapy and uncontrolled DM despite high doses of insulin. Additionally, the patient presented with hyperlipidemia and severe obesity (BMI=50.7 kg/m2), gastritis, depression, and osteoarthritis. On admission, she complained of a tendency to gain weight, fragile skin that bruised easily, difficulty with wound healing, susceptibility to infections, and insomnia. Physical examination revealed a moon face with plethora, a buffalo hump, central obesity with proximal muscle atrophy, and purple abdominal striae.
The CPA diagnosis was initially made two years earlier, but the patient did not qualify for surgery due to a hypertensive crisis. Soon after this episode, the SARS-CoV-2 pandemic began, and the patient was afraid of visiting any medical center because her son had died of COVID-19. Therefore, she was referred to our center for life-threatening hypercortisolism two years later.
At the time of admission, computed tomography (CT) imaging revealed a right adrenal tumor of 34x24x37mm, with a basal density of 21 HU and a contrast washout rate typical for adenomas (83%). The size and CT characteristics were identical as they were two years earlier. High serum cortisol levels, undetectable ACTH concentrations, and a lack of physiological diurnal rhythm of cortisol secretion were observed (Table 1). Urinary free cortisol (UFC) excretion was 310 µg/24 h, with an upper normal limit (UNL) of 176 µg/24 h. No cortisol suppression was achieved in high-dose dexamethasone suppression test (DST) (Table 1). Other adrenal-related hormonal parameters were within normal ranges, with values as follows: DHEA-S 42.68 µg/dl, aldosterone 3.24 ng/mL, and renin 59.14 µIU/mL.
Table 1 Laboratory Results Before Osilodrostat Therapy – Case 1
Due to multiple severe systemic complications, including uncontrolled hypertension, decompensated DM, and cardiac insufficiency, treatment with osilodrostat was introduced for life-saving pre-surgical management. Osilodrostat was started at a dose of 1 mg twice daily and gradually increased to 6 mg per day with actually an inverse response of serum cortisol level. The late-night cortisol level increased from 16 µg/dl to 25 µg/dl. As the full effect of the osilodrostat dose can occur even after a few weeks, the patient was discharged from hospital and instructed to contact her attending doctor immediately if any health deterioration was noticed. In the case of improvement in the patient’s condition, the next hospitalization was planned 3 weeks later. After three weeks of no contact with the patient, she was readmitted to our department with life-threatening escalation of hypercortisolism, severe hypokalemia, and further deterioration of hypertension, DM, cardiac insufficiency, dyspnea, and significant edemas, including facial edema. Treatments of hypertension, cardiac insufficiency, and DM were intensified, as presented in Table 2. Despite active potassium supplementation, life-threatening hypokalemia of 2.1 mmol/l occurred. Previously observed depression was exaggerated with severe anxiety and fear of death. The dose of osilodrostat was increased to 8 mg/day, and after three days of treatment a further elevation of serum cortisol was found, with an increase in UFC up to 9 × UNL (1546.2 µg/24 h). Due to an entirely unexpected inverse cortisol response, CT imaging was performed and revealed progression of the adenoma size to 39 × 36 × 40 mm, with a slight increase in density up to 27 HU as compared to the previous CT scan performed a month earlier (Figure 1).
Table 2 Changes in the Most Important Parameters During Osilodrostat Therapy – Case 1
Figure 1 Progression of the adrenal adenoma size during the initial doses of osilodrostat: (a) CT scan directly before osilodrostat therapy – solid nodule 34x24x37 mm, basal density 21 HU; (b) CT scan during treatment with 8 mg of osilodrostat daily – solid nodule 39x36x40 mm, basal density of 27 HU.
Considering the extremely high risk associated with such a rapid cortisol increase and related complications, decision of fast osilodrostat dose escalation was made. The dose was increased by 5 mg every other day, up to 45 mg per day, and, finally, a gradual decrease in the cortisol level (Table 2) was achieved, with UFC normalization to 168 µg/24 h. During dose escalation, no deterioration in the adverse effects (AEs) of osilodrostat was observed. Conversely, hypokalemia gradually improved despite a simultaneous reduction in potassium supplementation (Table 2). Facial edema decreased and the level of anxiety improved significantly. The course of hypertension severity as well as a summary of the main parameters controlled during treatment and the medications used are presented in Table 2. As soon as the cortisol level normalized, the patient was referred for surgery and underwent right adrenalectomy without any complications. Histopathology results confirmed a benign adenoma of the right adrenal gland (encapsulated, well-circumscribed tumor consisting of lipid-rich cells with small and uniform nuclei, mostly with eosinophilic intracytoplasmic inclusions). After surgery, hydrocortisone replacement therapy was administered. A few days after surgery, blood pressure and glucose levels gradually decreased, and the patient required reduction of antihypertensive and antidiabetic medications. After 22 months of follow-up, the patient’s general condition is good with no signs of recurrence. Antidepressant treatment is no longer required in this patient. Body mass index was significantly reduced to 40 kg/m2. The antihypertensive medication was completely discontinued, and the glucose level is controlled only with metformin. The patient still requires hydrocortisone substitution at a dose of 30 mg/day.
Case 2
A 39-year-old female was referred to our department in November 2022 with a diagnosis of CPA and unstable hypertension, for which surgery was contraindicated. The patient was unsuccessfully treated with triple antihypertensive therapy (telmisartan 40 mg/day, nebivolol 5 mg/day, and lercanidipine 20 mg/day). The patient reported weight gain, muscle weakness, acne, fragile skin that bruised easily, and secondary amenorrhea. Other comorbidities included gastritis, hypercholesterolemia, and osteoporosis. Physical examination revealed typical signs of Cushing’s syndrome, such as abnormal fat distribution, particularly in the abdomen and supraclavicular fossae, proximal muscle atrophy, moon face, and multiple hematomas. A lack of a serum cortisol diurnal rhythm with high late-night serum cortisol and undetectable ACTH levels was found (Table 3). The short DST revealed no cortisol suppression (Table 3), and the UFC result was 725 µg/24 h, which exceeded the UNL more than four times. The serum levels of renin, aldosterone, and 24-h urine fractionated metanephrines were within the normal ranges. Computed tomography imaging revealed a left adrenal gland tumor measuring 25 × 26 × 22 mm, with a basal density of 32 HU and a washout rate typical for adenoma (76%).
Table 3 Laboratory Results Before Osilodrostat Therapy – Case 2
Osilodrostat therapy was administered for preoperative management. The initial daily dose was 2 mg/day, increased gradually by 2 mg every day with no serum cortisol response (late night cortisol levels 15.8–18.5 µg/dl) and no AEs of the drug (Table 4). After the daily dose of osilodrostat reached 10 mg, it was escalated by 5 mg every other day, initially with no serum cortisol reduction. The dose was increased to 45 mg daily (with the lowest detected late-night serum cortisol of 9.6 µg/dl) (Table 4).
Table 4 Changes in the Most Important Parameters During Osilodrostat Therapy – Case 2
After a week of administration of 45 mg daily, UFC normalization was achieved. Despite rapid dose escalation, no AEs were observed during the entire therapy period. Potassium levels were normal without any supplementation (the lowest detected serum potassium level was 3.9 mmol/l; all other results were over 4.0 mmol/l) (Table 4). After UFC normalization, left adrenalectomy was performed without complications. Histopathological examination revealed benign adrenal adenoma. Antihypertensive therapy was reduced only to 2.5 mg of nebivolol daily. The patient’s general condition improved significantly. Currently, hydrocortisone replacement therapy is administered at a dose of 15 mg/day.
Discussion
Osilodrostat is a novel potent steroidogenesis inhibitor whose efficacy and safety have been thoroughly analyzed in clinical trials of patients with CD, the most common cause of endogenous hypercortisolism. No clinical trial of osilodrostat therapy in CPA has been performed, as this disease constitutes only 10% of all cases of endogenous hypercortisolism. Moreover, osilodrostat is not approved by the FDA for hypercortisolism conditions other than CD.9 Therefore, data on potential differences in the treatment regimen are lacking.
During the course of already reported trials in CD, osilodrostat doses were escalated slowly, every 2–3 weeks,3,5,6 with an excellent response to quite low doses of the drug.3–6 In the LINC 2 extension study the median average dose was 10.6 mg/day,5 while in the LINC 3 extension study and the LINC 4 study it was 7.4 mg/day and 6.9 mg/day, respectively.4,6 In most cases, a significant decrease of hypercortisolism was reported with the low doses of osilodrostat (4 or 10 mg/day). Moreover, some patients received 1 mg/day or even 1 mg every other day, with a good response.6 Even in rare cases of CD in whom initial short-term etomidate therapy was given at the beginning of osilodrostat therapy, due to highly severe life-threatening symptoms of hypercortisolism, the final effective dose of osilodrostat was much lower than that in our patients with CPA (25 mg/day vs 45 mg/day) and no increase of cortisol level was observed.11
It should be underlined that many cases of adrenal insufficiency during osilodrostat therapy in patients with CD have been reported,3–6,12,13 and – therefore – low initial dose with slow gradual dose escalation is recommended in patients with CD.1,6,13
In the cases presented here, CPA led to severe hypercortisolism, the complications of which constituted contraindications for surgery. Therefore, osilodrostat therapy was introduced as a presurgical treatment. In Case 1, the therapy was started at low doses according to the approved product characteristics.14 Due to the severity of hypertension, which was uncontrolled despite of active antihypertensive therapy, as well as to unstable DM, the doses were increased faster than recommended. Surprisingly, we immediately observed a gradual increase in hypercortisolism, in both serum cortisol levels and the UFC, with simultaneous burst of complications related to both hypercortisolism itself and 11β-hydroxylase inhibition. Life-threatening episodes of hypertensive crisis responded poorly to standard therapies. Severe exaggeration of cardiac insufficiency could probably be related to these episodes as well as to deep hypokalemia, which occurred despite potassium supplementation. Hypokalemia is a typical complication of treatment with 11β-hydroxylase inhibitors due to the accumulation of adrenal hormone precursors. However, Patient 1 required much higher doses of potassium supplementation, both parenteral and oral, than ever described during osilodrostat therapy.3–6,13 The dose of 20 mg/day of osilodrostat was the first one which led to noticeable cortisol reduction and a decrease in systolic blood pressure (SBP) to below 170 mmHg. Surprisingly, instead of the expected deterioration of hypokalemia, parenteral potassium administration could be stopped with an osilodrostat dose of 20 mg/day and oral supplementation was gradually reduced simultaneously with osilodrostat dose escalation. The reason why such severe hypokalemia occurred with low doses of osilodrostat and did not deteriorate further seems complex. One possible reason is the administration of high doses of potassium-saving antihypertensive drugs such as spironolactone and the angiotensin II receptor antagonist telmisartan. Additionally, one can consider other possible mechanisms, such as downregulation of the receptors of deoxycorticosterone (DOC) or other adrenal hormone precursors. However, this hypothesis requires further research and confirmation. Such an improvement of the potassium level during osilodrostat dose escalation was previously demonstrated in a patient with CD.11 Interestingly, in our Patient 2, no potassium supplementation was required during the whole time of osilodrostat therapy, although the doses were increased intensively up to the finally effective dose, which was the same (45 mg/day) as for Patient 1. In Patient 2, no actual response to doses lower than 20 mg/day was observed. UFC normalization was achieved after a week of administration of 45 mg/day, five weeks from the beginning of therapy. Although UFC normalization is not always required in pre-surgical treatment, clinical symptoms significantly improved in our patients only after the UFC upper normal level was achieved.
The present paper is one of only a few reports focused on osilodrostat therapy in CPA, and the only one presenting a different therapy course as compared to patients with CD. No case of CPA resistance to low doses of osilodrostat has been described. It should be underlined that in our report “low doses” of osilodrostat were higher than the average mean doses of osilodrostat used in clinical trials in patients with CD.3–6 Therefore, they should not generally be considered low but only much lower than those which were effective in our patients. Malik and Ben-Shlomo presented a case of CPA treated with osilodrostat, with an immediate decrease in cortisol level at 4 mg/day and adrenal insufficiency symptoms after dose escalation to 8 mg/day.15 Similar to our two cases, their patient was a middle-aged female with normal results of all other adrenal parameters, such as renin, angiotensin, or metanephrine levels. However, a CT scan was not performed (or presented), while magnetic resonance imaging revealed an indeterminate adrenal gland mass without a typical contrast phase/out-of-phase dropout for adenoma.15 Therefore, different morphology of cortisol-secreting adrenal tumor can potentially be considered a reason of the different response to treatment. Tanaka et al performed a multicenter study on the efficacy and safety of osilodrostat in Japanese patients with non-CD Cushing’s syndrome.16 Five patients with CPA were included in the study, and none of them required osilodrostat doses higher than 10 mg/day to achieve UFC normalization. However, most of the patients presented by Tanaka et al were previously treated with metyrapone,16 whereas both of our patients were treatment-naive. Previous metyrapone therapy may be considered as a potential reason of better response to osilodrostat. This hypothesis was confirmed in the quoted study by Tanaka et al, who demonstrated that at week 12 the median percent changes in the mUFC values were higher in patients previously treated with metyrapone (–98.97%) than in treatment-naive cases (–86.65%).16 Detomas et al performed a comparison of efficacy and safety of osilodrostat and metyrapone, with one CPA patients included in a group treated with osilodrostat, however no data on a dose required for a disease control are available separately for this particular patient.8 To the best of our knowledge, no more CPA cases have been described and therefore no further comparison is available.
Higher doses of osilodrostat were administered to a group of seven patients with hypercortisolism due to adrenocortical carcinoma (ACC) presented by Tabarin et al.17 A full control of hypercortisolism was achieved in one patient for each dose of 4, 8, 10, and 20 mg/day, and in three patients treated with 40 mg/day.17 These patients, however received other therapies including mitotane and chemotherapy, which can significantly modify the response to osilodrostat.
Several authors have reported the phenomenon of a partial or total loss of response to osilodrostat.5,16,17 In such cases, a response to treatment was initially achieved and then lost during treatment with the same dose. A further increase in osilodrostat dose usually resulted in the response resumption.5,16,17 Such a situation could not be suspected in either of our cases.
The presented cases provide a novel insight into modalities of treatment with osilodrostat in patients with CPA and demonstrate for the first time that an inverse cortisol response is possible in CPA cases, especially those with a higher CT density of adrenal adenoma. Such a situation should not be considered a contraindication to dose escalation. Conversely, the dose should be increased more intensively so as to achieve the initial efficacy threshold, which was 20 mg/day in both of our patients. The fully efficient dose that allowed UFC normalization was more than twice as high (45 mg/day in both cases). A similar approach should be applied in patients who do not respond to lower doses, such as Patient 2. The safety of osilodrostat therapy is strictly individual and not dose dependent in patients with CPA. Adverse events, including hypokalemia, severe hypertension, and edema, can be of life-threatening severity or may not occur regardless of the dose. Moreover, AEs of high severity may decrease with osilodrostat dose escalation. Our study demonstrated that osilodrostat is efficient and can be used in patients with CPA as a pre-surgical therapy if surgery is contraindicated due to hypercortisolism complications.
Our study presented two cases of CPA treated with osilodrostat, and a small size of our group is the main limitation of this report. Future research is required to confirm our observations.
Conclusion
In some patients with CPA, the doses of osilodrostat required for disease control can be much higher than those previously reported. Acceleration of the dose increase can be fast, and the risk of overdosing, adrenal insufficiency, and later necessity of dose reduction seem to be much lower than it could be expected. Low initial doses (<20 mg/day in our study) can be entirely ineffective or can even cause exacerbation of hypercortisolism, whereas high doses (45 mg/day in the present study) are efficient in pre-surgery UFC normalization. AEs associated with osilodrostat can be rapid, with severe hypokalemia despite active potassium supplementation, or may not occur even if high doses of osilodrostat are applied. Therefore, close monitoring for potential AEs is necessary.
Acknowledgments
The abstract included some parts of this paper was presented at the European Congress of Endocrinology ECE2023 as a rapid communication. The abstract was published in the Endocrine Abstracts Vol. 90 [https://www.endocrine-abstracts.org/ea/0090/].
Funding
The publication of this report was financially supported by the statutory funds of the Polish Mother’s Memorial Hospital – Research Institute, Lodz, Poland.
Disclosure
Professor Przemysław Witek reports personal fees from Investigator in the clinical trials paid by Novartis and Recordati Rare Diseases, outside the submitted work; lectures fees from Recordati Rare Diseases, Strongbridge, IPSEN. The authors report no other conflicts of interest in this work.
References
1. Fleseriu M, Auchus R, Bancos I, et al. Consensus on diagnosis and management of Cushing’s disease: a guideline update. Lancet Diabetes Endocrinol. 2021;9(12):847–875. doi:10.1016/S2213-8587(21)00235-7
2. Pivonello R, Isidori AM, De Martino MC, et al. Complications of Cushing’s syndrome: state of the art. Lancet Diabetes Endocrinol. 2016;4(7):611–629. doi:10.1016/S2213-8587(16)00086-3
3. Pivonello R, Fleseriu M, Newell-Price J, et al. Efficacy and safety of osilodrostat in patients with Cushing’s disease (LINC 3): a multicentre Phase III study with a double-blind, randomised withdrawal phase. Lancet Diabetes Endocrinol. 2020;8(9):48–761. doi:10.1016/S2213-8587(20)30240-0
4. Fleseriu M, Newell-Price J, Pivonello R, et al. Long-term outcomes of osilodrostat in Cushing’s disease: LINC 3 study extension. Eur J Endocrinol. 2022;187(4):531–541. doi:10.1530/EJE-22-0317
5. Fleseriu M, Biller BMK, Bertherat J, et al. Long-term efficacy and safety of osilodrostat in Cushing’s disease: final results from a Phase II study with an optional extension phase (LINC 2). Pituitary. 2022;25(6):959–970. doi:10.1007/s11102-022-01280-6
6. Gadelha M, Bex M, Feelders RA, et al. Randomized trial of osilodrostat for the treatment of Cushing disease. J Clin Endocrinol Metab. 2022;107(7):e2882–e2895. doi:10.1210/clinem/dgac178
7. Daniel E, Aylwin S, Mustafa O, et al. Effectiveness of metyrapone in treating cushing’s syndrome: a retrospective multicenter study in 195 patients. J Clin Endocrinol Metab. 2015;100(11):4146–4154. doi:10.1210/jc.2015-2616
8. Detomas M, Altieri B, Deutschbein T, et al. Metyrapone versus osilodrostat in the short-term therapy of endogenous cushing’s syndrome: results from a single center cohort study. Front Endocrinol. 2022;13:903545. doi:10.3389/fendo.2022.903545
11. Dzialach L, Sobolewska J, Respondek W, et al. Cushing’s syndrome: a combined treatment with etomidate and osilodrostat in severe life-threatening hypercortisolemia. Hormones. 2022;21(4):735–742. doi:10.1007/s42000-022-00397-4
12. Ekladios C, Khoury J, Mehr S, et al. Osilodrostat-induced adrenal insufficiency in a patient with Cushing’s disease. Clin Case Rep. 2022;10(11):e6607. doi:10.1002/ccr3.6607
13. Fleseriu M, Biller BMK. Treatment of Cushing’s syndrome with osilodrostat: practical applications of recent studies with case examples. Pituitary. 2022;25(6):795–809. doi:10.1007/s11102-022-01268-2
15. Malik RB, Ben-Shlomo A. Adrenal cushing’s syndrome treated with preoperative osilodrostat and adrenalectomy. AACE Clin Case Rep. 2022;8(6):267–270. doi:10.1016/j.aace.2022.10.001
16. Tanaka T, Satoh F, Ujihara M, et al. A multicenter, Phase 2 study to evaluate the efficacy and safety of osilodrostat, a new 11β-hydroxylase inhibitor, in Japanese patients with endogenous Cushing’s syndrome other than Cushing’s disease. Endocr J. 2020;67(8):841–852. doi:10.1507/endocrj.EJ19-0617
17. Tabarin A, Haissaguerre M, Lassole H, et al. Efficacy and tolerance of osilodrostat in patients with Cushing’s syndrome due to adrenocortical carcinomas. Eur J Endocrinol. 2022;186(2):K1–K4. doi:10.1530/EJE-21-1008
Corcept Therapeutics Incorporated (NASDAQ: CORT), a commercial-stage company engaged in the discovery and development of medications to treat severe endocrinologic, oncologic, metabolic and neurologic disorders by modulating the effects of the hormone cortisol, today announced completion of enrollment in GRADIENT, a Phase 3 trial of its proprietary selective cortisol modulator relacorilant in patients with Cushing’s syndrome (hypercortisolism) caused by an adrenal adenoma or adrenal hyperplasia.
“Hypercortisolism with adrenal etiology affects many patients and is associated with serious cardiometabolic comorbidities, including hypertension and hyperglycemia, and increased risk of premature death,” said Bill Guyer, PharmD, Corcept’s Chief Development Officer. “GRADIENT is the first prospective placebo-controlled study to be conducted exclusively in these patients with Cushing’s syndrome. We expect data from GRADIENT in the fourth quarter of this year.”
GRADIENT is a randomized, double-blind, placebo-controlled trial conducted at sites in the United States, Europe and Israel. One-hundred thirty-seven patients were randomized 1:1 to receive relacorilant or placebo for 22 weeks. Primary endpoints are improvement in glucose metabolism and hypertension.
About Cushing’s Syndrome (Hypercortisolism)
Cushing’s syndrome is caused by excessive activity of the hormone cortisol. Endogenous Cushing’s syndrome is an orphan disease that most often affects adults aged 20-50. Symptoms vary, but most patients experience one or more of the following manifestations: high blood sugar, diabetes, high blood pressure, upper-body obesity, rounded face, increased fat around the neck, thinning arms and legs, severe fatigue and weak muscles. Irritability, anxiety, cognitive disturbances and depression are also common. Cushing’s syndrome can affect every organ system and can be lethal if not treated effectively.
About Relacorilant
Relacorilant is a selective cortisol modulator that binds to the glucocorticoid receptor (GR), but does not bind to the body’s other hormone receptors. Corcept is studying relacorilant in a variety of serious disorders, including ovarian, adrenal and prostate cancer and Cushing’s syndrome. Relacorilant is proprietary to Corcept and is protected by composition of matter, method of use and other patents. Relacorilant has orphan drug designation in the United States and the European Union for the treatment of Cushing’s syndrome.
About Corcept Therapeutics
For over 25 years, Corcept’s focus on cortisol modulation and its potential to treat patients across a wide variety of serious disorders has led to the discovery of more than 1,000 proprietary selective cortisol modulators. Corcept’s advanced clinical trials are being conducted in patients with hypercortisolism, solid tumors, amyotrophic lateral sclerosis (ALS) and liver disease (NASH). In February 2012, the company introduced Korlym, the first medication approved by the U.S. Food and Drug Administration for the treatment of patients with Cushing’s syndrome. Corcept is headquartered in Menlo Park, California. For more information, visit Corcept.com.
Forward-Looking Statements
Statements in this press release, other than statements of historical fact, are forward-looking statements based on our current plans and expectations that are subject to risks and uncertainties that might cause our actual results to differ materially from those such statements express or imply. These risks and uncertainties include, but are not limited to, our ability to operate our business; risks related to the study and development of Korlym as well as relacorilant, miricorilant, dazucorilant and our other product candidates, including their clinical attributes, regulatory approvals, mandates, oversight and other requirements; and the scope and protective power of our intellectual property. These and other risks are set forth in our SEC filings, which are available at our website and the SEC’s website.
In this press release, forward-looking statements include those concerning the development of relacorilant as a treatment for Cushing’s syndrome, and design, timing and expectations regarding our GRADIENT trial. We disclaim any intention or duty to update forward-looking statements made in this press release.
© 2024 The Author(s). This work is published and licensed by Dove Medical Press Limited. The full terms of this license are available at 
CushieBlog 