The usefulness of the combined high-dose dexamethasone suppression test and desmopressin stimulation test in establishing the source of ACTH secretion in ACTH-dependent Cushing’s syndrome

Abstract

Bilateral inferior petrosal sinus sampling (BIPSS) is the current gold standard test for differentially diagnosing ACTH-dependent Cushing’s syndrome (CS). However, BIPSS is an invasive procedure, and its availability is limited. We retrospectively analysed the 24-hour urinary free cortisol (UFC) level during the high-dose dexamethasone suppression test (HDDST) and plasma ACTH/cortisol levels after the desmopressin stimulation test (DDAVP test) in subjects with confirmed Cushing’s disease (CD) (n = 92) and ectopic ACTH-dependent CS (EAS) (n = 16), and evaluated the positive predictive value (PPV) of the two combined-tests in the aetiological diagnosis of ACTH-dependent CS. The percent changes in UFC levels after the HDDST and in ACTH/cortisol levels after DDAVP administration relative to the corresponding basal levels and the area under the receiver operating characteristic (ROC) curve (AUC) were analysed. UFC suppression below 62.7% suggested a pituitary origin with a sensitivity (SE) of 80% (95% CI: 70–88) and a specificity (SP) of 80% (95% CI: 52–96). A threshold increase in the ACTH level after DDAVP stimulation of 44.6% identified CD with an SE of 91% (95% CI: 83–97) and an SP of 75% (95% CI: 48–93). The combination of both tests yielded an SE of 95.5% and PPV of 98.4% for CD, and significantly improved the efficiency of the differential diagnosis between CD and EAS. These dual non-invasive endocrine tests may substantially reduce the need for BIPSS in the etiological investigation of ACTH-dependent CS.

CUSHING’S SYNDROME (CS) is one of the most challenging diseases for endocrinologists, undoubtedly because the clinical manifestations are often nonspecific, and a single diagnostic test combining an optimal sensitivity (SE) and specificity (SP) for hypercortisolism and the determination of its etiology is unavailable [1]. The greatest challenge in the investigation of CS involves the differentiation between Cushing’s disease (CD) and ectopic ACTH-dependent CS (EAS), which should be sought if a diagnosis of ACTH-dependent CS has been established. This procedure includes the administration of the high-dose dexamethasone suppression test (HDDST), CRH stimulation test, imaging studies, and bilateral inferior petrosal sinus sampling (BIPSS).

Imaging studies are not completely reliable, as pituitary incidentaloma has been uncovered in individuals with EAS [2, 3] and the general population [4]. Meanwhile, not all potential adenomas will be identified in CD patients [2, 5-8]. BIPSS, an invasive examination, is costly and time-consuming, requires a dedicated neuroradiological team, and may also lead to some complications [9]. The HDDST is the most widely used test because it is easy to perform, but its performance is not satisfactory. There is no consensus as to how to interpret the results of the CRH test, and the commercialized drugs required for it are unavailable in many countries, especially in Asian countries, including China.

Desmopressin (DDAVP) is a vasopressin analogue selective for V2R that mainly mediates renal water retention and is less selective for V3R [10]. Some studies have confirmed that DDAVP can induce an ACTH and cortisol response in a considerable proportion of patients with CD [11], in contrast to the minimal responses obtained in subjects with pseudo-Cushing’s syndrome [12-14] and EAS [11]. The DDAVP stimulation test (DDAVP test) has also been documented to predict prolonged remission in the early postoperative period and herald recurrence after transsphenoidal surgery (TSS) [11, 15, 16]. Currently, the HDDST and the CRH test are used more often in clinical practice, and the value of DDAVP test in the differential diagnosis of CD and EAS remains controversial [6]. Early studies [17-19] explored concordant results from the CRH stimulation test and the HDDST, and from the CRH test and DDAVP test, but have rarely focused on whether informative findings can be obtained by combining the DDAVP test and the HDDST.

In this study, we retrospectively analysed the clinical data, pituitary magnetic resonance imaging (MRI), other imaging data, HDDST and DDAVP test results from subjects with ACTH-dependent CS. We used the results of surgery and/or BIPSS as the gold standard to determine the diagnostic efficiency of the HDDST and DDAVP test alone or in combination in differentiating CD and EAS. The expression of V2R and V3R in tumour tissues was examined to elucidate the relationships between the ACTH response to DDAVP and the extent of arginine vasopressin receptor (AVPR) expression in patients with CD.

Patients and Methods

Patients

We performed a retrospective analysis of 108 ACTH-dependent CS patients who presented to West China Hospital from January 1, 2013, to July 31, 2020. A diagnosis of CS was suspected based on clinical features and confirmed by at least two biochemical tests indicating elevated 24-hours urinary free cortisol (UFC) levels, loss of circadian rhythm in plasma cortisol levels at 0800 h (morning plasma cortisol) and 2,400 h (midnight plasma cortisol), and cortisol after 1-mg overnight dexamethasone (1-mg DST) higher than 5 μg/dL. The diagnosis of ACTH-dependent CS was confirmed if the ACTH level was >10 pg/dL.

A diagnosis of CD was confirmed when patients met at least one of the following criteria: histopathology; BIPSS results were consistent with a pituitary source of ACTH; Some small pituitary tumours were lost during the procedure of negative pressure aspiration during TSS, but the patients exhibited a biochemical remission state after TSS. These patients were not confirmed by pathology and were diagnosed with CD; Biochemical remission of ACTH and plasma cortisol levels were observed during the 1–4-year follow-up after Gamma-Knife treatment for pituitary lesions.

A diagnosis of EAS was confirmed if postoperative pathology revealed positivity for neuroendocrine tumour (NET) and ACTH immunohistochemical (IHC) staining; or if the BIPSS findings were consistent with ectopic ACTH production.

This study was approved by the Ethics Committee of West China Hospital of Sichuan University (2019: No. 828).

HDDST

Before the HDDST, the UFC levels were measured twice using samples from different periods. The average UFC level was recorded as the baseline. For the HDDST, 2 mg of dexamethasone was administered orally every 6 h for 2 days (Q6 h × 2 d), and the UFC level on the second day was measured. A reduction in the UFC level greater than 50% (positive result) was considered indicative of CD.

DDAVP test

The DDAVP test was performed in the morning after the patients had fasted and rested. In the DDAVP test, plasma ACTH and cortisol levels were measured at –15 and 0 min before and 15, 30, 45, 60, 90, and 120 min after the IV administration of DDAVP (10 μg). After DDAVP administration, a ≥35% increase in ACTH levels and a ≥20% increase in cortisol levels over the corresponding basal levels (positive result) was considered an indication of CD.

BIPSS

The BIPSS procedures were conducted as described by Miller and Doppman [20]. The bilateral femoral veins were cannulated, and catheters were guided into the bilateral petrosal sinuses. Peripheral venous blood was collected from the peripheral veins and the left and right petrosal sinuses and transferred to the laboratory for ACTH and prolactin (PRL) assays. Then, 10 μg of DDAVP were given intravenously, and blood samples were obtained 3, 5, and 10 min later from the petrosal sinuses for another ACTH assay. A bilateral inferior petrosal sinus to peripheral venous blood (IPS/P) ACTH gradient ≥2 (basal) or ≥3 (after DDAVP) was considered consistent with CD, while lower gradient values were assumed to indicate of EAS, as previously suggested [21].

ACTH, cortisol and UFC levels were measured using an electrochemical chemiluminescence immunoassay (ECLI, MODULAR E170 automatic electrochemical luminescence immune-analyzer, Roche Company, Shanghai) in the Department of Laboratory Medicine, West China Hospital.

Imaging

All included subjects underwent sellar region MRI with a 3.0-T MRI scanner (Siemens, Germany; matrix size, 256 × 256 pixels), including T1-weighted (repetition time/echo time (TR/TE): 600/8.1 ms), T2-weighted (TR/TE: 4,000/93 ms) and contrast-enhanced T1-weighted imaging (TR/TE: 232/8.1 ms), with a 1.5-mm slice thickness (ST) in the coronal plane and 6-mm ST in the axial plane. The imaging data were reviewed by a radiologist, neurosurgeon and endocrinologist. Positive MRI findings were based on full agreement among the three clinicians. When the pituitary appeared normal or occupied lesions were not undefined, the MRI findings were considered negative and recorded as 0 mm in the subsequent analysis. Patients with suspected ectopic tumours underwent other imaging examinations, such as chest and/or pelvic computed tomography (CT)/MRI and positron emission tomography (PET)-CT, to search for the primary lesions.

Surgery and pathology

Once a tumour was identified, selective surgical excision was performed via TSS in 76/92 CD patients by neurosurgeons, and surgeries for other primary lesions were performed in 9/16 EAS patients by professional surgeons. A Gamma-Knife was implemented for 7/92 CD patients with absolute contraindications to or difficulties during the operation. The other patients, who were reluctant to undergo surgical treatment, received a final aetiological diagnosis by BIPSS.

The specimens of corticotrophin adenomas, including 10 specimens positive for the DDAVP test and 6 negative specimens, were obtained through TSS. The sections were incubated with rabbit anti-AVPR2 IgG (ab188748, antibody diluted 1:100, Abcam, USA) and rabbit anti-AVPR3 IgG (HPA075404, antibody diluted 1:1,000, Sigma, USA), followed by secondary antibodies (Cell Signaling Technology, USA). Normal pituitary tissue was obtained from tissue adjacent to the pituitary adenoma as a positive control for the AVPR3 exploration, and we only used the corresponding secondary antibody to detect pituitary sections as the negative controls. All slices were photographed under a ZEISS Axio Imager A2 microscope. For the semiquantitative analysis, we randomly assessed 5–10 horizons (400× magnification) using Image-Pro Plus 6.0, and the final mean density was obtained from the ratio of the sum of the integrated optical density (IOD) and area of interest (AOI) in each slide (sample sum of IOD/sample AOI).

Statistical analysis

The statistical analysis and graph creation were performed with GraphPad Prism version 6.0 (GraphPad Software, San Diego, CA, USA) and SPSS version 25.0 (IBM SPSS, Chicago, IL, US). Data with a normal distribution are expressed as the mean ± standard deviation (x ± sd) and were statistically compared using Student’s t-test; the data with a nonnormal distribution are presented as the median (interquartile range), and the Mann-Whitney U test was used for the comparison. The receiver operating characteristic (ROC) curve was generated using MedCalc version 19.6.4 (MedCalc Application Software, Belgium). The predictive variables were fitted using a logistic regression model, and a Z test was performed to compare the area under the curve (AUC). We calculated the SE, SP, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), negative likelihood ratio (LR–) and Youden’s Index of the results of the individual and combined diagnostic tests. Spearman’s or Pearson’s correlation coefficients were calculated to describe the correlations between the variables. p < 0.05 was considered statistically significant.

Results

Baseline characteristics

In total, 108 patients were included in this study, including 92 CD patients and 16 EAS patients. The CD diagnosis were confirmed by BIPSS/pathological results in 87/92 patients, 4/92 patients were confirmed by biochemical remission after Gamma-Knife treatment for pituitary lesions, and 1/92 patients with pituitary microadenoma achieved relief of clinical symptoms and biochemical indicators after TSS, but the pathological result was negative. The EAS diagnosis was confirmed by BIPSS/pathological results in 9/16 patients, including 1 bronchial carcinoid and anterior mediastinum atypical carcinoid simultaneously, 1 lung carcinoid, 1 anterior mediastinum atypical carcinoid, 1 thymus NET, 1 extraocular muscles NET, 1 liver NET, 1 pancreas NET, 1 anterior mediastinum NET and 1 paraganglioma, and another 7 patients were diagnosed by BIPSS, including 2 patients with a detected lung-occupying lesion and 5 without an obvious primary lesion.

No differences in age and sex distributions were observed between the two groups, although the male/female ratio among CD patients was 15/77 compared with 6/10 among EAS patients. The UFC, ACTH, morning plasma cortisol, midnight plasma cortisol and cortisol levels after 1-mg DST were significantly higher in the EAS group than in the CD group. EAS patients suffered from more severe hypokalaemia and metabolic alkalosis than CD patients (Table 1).

Table 1 Baseline characteristics of CD and EAS patients

	CD	EAS	p value
Age (years)	37.98 ± 12.23	42.94 ± 10.46	0.130
BMI (kg/m2)	25.1 (23.0, 27.9)	27.023 ± 2.83	0.026
AC (cm)	92.18 ± 10.32	95.83 ± 8.75	0.405
UFC (μg/24 hr)	808.65 (453.40, 1,578.15)	1,993.23 (1,445.39, 6,791.33)	<0.0001
ACTH (pg/mL)	75.29 (47.65, 110.23)	171.76 ± 91.82	<0.0001
Morning plasma cortisol (μg/dL)	32.25 (24.39, 40.92)	47.91 ± 17.42	0.002
Midnight plasma cortisol (μg/dL)	24.29 (18.87, 34.22)	41.49 ± 20.11	0.001
1-mg DST(μg/dL)	23.52 (15.92, 35.33)	38.99 ± 16.26	0.015
SBP (mmHg)	160 (137, 173)	156.38 ± 17.19	0.986
DBP (mmHg)	100 (90.75, 109.25)	100 (89.75, 107.5)	0.825
TG (mg/dL)	136.44 (84.17, 201.12)	164.80 ± 81.51	0.601
CHOL (mg/dL)	207.99 ± 51.42	225.39 ± 65.72	0.255
UA (mg/dL)	5.66 (4.50, 6.96)	5.58 ± 1.76	0.785
AG (mmol/L)	19.30 (16.6, 21.5)	18.00 ± 4.15	0.481
CO2CP (mmol/L)	25.45 ± 4.61	29.45 ± 5.27	0.003
pH	7.45 ± 0.04	7.47 ± 0.05	0.158
BE (mmol/L)	1.60 (–0.4, 4.4)	4.12 ± 3.61	0.078
HCO3– (mmol/L)	25.55 (23.95, 28.03)	28.31 ± 3.13	0.044
K+ (mmol/L)	3.47 ± 0.65	2.97 ± 0.79	0.007
Glu-0 (mg/dL)	98.46 (86.04, 119.88)	103.14 (90.72, 136.44)	0.360
Glu-2h (mg/dL)	201.78 (158.94, 266.94)	240.66 ± 62.64	0.234
HbA1c (%)	6.20 (5.50, 6.88)	6.60 ± 0.96	0.392

Abbreviations: BMI, body mass index; AC, abdominal circumference; UFC, 24-hour urinary free cortisol level (normal range: 20.26–127.55 μg/24 hr); ACTH, adrenocorticotropic hormone (normal range: 5.0–78 pg/mL); morning plasma cortisol: plasma cortisol level at 0800 h (normal range: 5.33–22.04 μg/dL); midnight plasma cortisol: plasma cortisol level at 2,400 h; 1-mg DST, cortisol level after 1-mg overnight dexamethasone suppression; SBP, systolic blood pressure; DBP, diastolic blood pressure; TG, triglyceride (normal range: 25.69–162.15 mg/dL); CHOL, total cholesterol (normal range: 108.25–220.36 mg/dL); UA, uric acid (normal range: 4.03–8.23 mg/dL); AG, anion gap (normal range: 12.0–20.0 mmol/L); CO₂CP, carbon dioxide combining power (normal range: 18.0–28.0 mmol/L); pH, power of hydrogen (normal range: 7.35–7.45); BE, base excess (normal range: –2–+3 mmol/L); HCO3^–, bicarbonate ion (normal range: 22–27 mmol/L); K⁺, serum potassium ion (normal range: 3.5–5.3 mmol/L); Glu-0, fasting blood glucose level (normal range: 70.2–106.2 mg/dL); Glu-2h, 2-hour postprandial sugar levels (normal range: 59.4–140.4 mg/dL); HbA1c, glycosylated haemoglobin (normal range: 4.5–6.1%).

HDDST

The CD patients had a lower baseline cortisol level and a higher percentage reduction in the serum cortisol/UFC levels after dosing than the EAS subjects (Fig. 1A and 1B). After HDDST, >50% cortisol suppression was observed in 75 of 92 (81.5%) patients with CD and in 8 of 16 (50.0%) of those with EAS. The total accuracy was 76.85% (83/108) (Table 2).

Fig. 1

Results of HDDST and DDAVP test.

A: Percentage change in UFC after administration of dexamethasone to CD and EAS patients (p < 0.0001).

B: Percentage change in cortisol after administration of dexamethasone to CD and EAS patients (p = 0.001). Black dots represent CD, blue squares represent EAS.

C: Percentage change in ACTH relative to the baseline level after the DDAVP test for CD and EAS patients (p < 0.0001).

D: Percentage change in cortisol relative to the baseline level after the DDAVP test (p < 0.0001). Black dots represent CD, blue squares represent EAS.

Table 2 Performance of diagnostic tests alone and in combination in the diagnosis of ACTH-dependent Cushing’s syndrome

	N	SE	SP	PPV	NPV	LR+	LR–	OR	Youden’s Index	Accuracy rate (%)
HDDST	108	81.52	50.00	90.36	32.00	1.63	0.37	4.41	0.32	76.85
DDAVP	108	83.70	87.50	97.47	48.28	6.70	0.19	35.93	0.71	84.26
MRI	108	82.61	62.50	92.68	38.46	2.20	0.28	7.92	0.45	79.63
HDDST + DDAVP	74	95.45	87.50	98.44	70.00	7.64	0.05	147.00	0.83	94.59
HDDST + DDAVP + MRI	59	100.00	83.33	98.15	100.00	6.00	0.00	—	0.83	98.31

Abbreviations: N, sample size; SE, sensitivity; SP, specificity; PPV, positive predictive value; NPV, negative predictive value; LR+, positive likelihood ratio; LR–, negative likelihood ratio; OR, odds ratio; Accuracy, diagnostic accuracy relative to pathology or BIPSS results. HDDST + DDAVP refers to patients with concordant results between the two test results; HDDST + DDAVP + MRI refers to patients with concordant results between the three tests. — indicates that the OR was not calculated because LR– = 0.

ROC curves were used to analyse the efficacy of the HDDST in diagnosing CD. The AUC of the percentage of UFC suppression was 0.801 (95% CI: 0.703–0.878), and a cut-off value of 62.7% for the decrease relative to the baseline UFC was identified as achieving the best performance in diagnosing the patients with CD, with a SE of 0.800 (95% CI: 0.692–0.884) and SP of 0.800 (95% CI: 0.519–0.957). Regarding the cortisol level, the corresponding cut-off value was 51.24%, with an SE of 0.663 (95% CI: 0.557–0.758), SP of 0.813 (95% CI: 0.544–0.960), and an AUC of 0.782 (95% CI: 0.682–0.863) (Fig. 2).

Fig. 2

Receiver operating characteristic curves during the HDDST and DDAVP test

Optimal cutoff values and AUCs for delta UFC and cortisol for distinguishing CD and EAS with the HDDST; and optimal cutoff values and AUCs for delta ACTH and cortisol for distinguishing CD and EAS with the DDAVP test.

DDAVP test

The DDAVP injection produced a higher absolute percent increase in ACTH among the CD patients than in the EAS patients. The same tendency was observed for cortisol (p < 0.0001 for both) (Fig. 1C and 1D). An ACTH-positive response (≥35% increase above basal level) was observed for 85/92 (92.4%) CD patients with CD and 5/16 (31.3%) EAS patients. A cortisol-positive response (≥20% increase) after dosing was detected in 78/92 (84.8%) patients with CD and 2/16 (12.5%) with EAS.

Among the patients with CD, the percentage of patients with an ACTH-positive response (≥35% increase) was significantly greater than the percentage of patients with a cortisol response (≥20% increase) (92.39% vs. 84.78%; p value = 0.039). However, the SP of the ACTH and cortisol responses did not significantly differ (68.75% vs. 87.50%, p value = 0.25).

According to the ROC curve analysis, percent increases in ACTH and cortisol levels of 44.63% and 16.22%, respectively, emerged as the best predictors of CD, achieving an SE of 0.911 (95% CI: 0.832–0.961) and 0.870 (95% CI: 0.783–0.931) and an SP of 0.750 (95% CI: 0.476–0.927) and 0.875 (95% CI: 0.617–0.984), respectively. The AUCs of the post-DDAVP percentage increased in ACTH and cortisol were 0.836 (95% CI: 0.743–0.906) and 0.912 (95% CI: 0.833–0.961), respectively (Fig. 2).

Imaging

All 108 ACTH-dependent Cushing’s syndrome patients underwent pituitary MRI. Among the 92 CD patients, pituitary lesions larger than 6 mm and ≤6 mm were identified in 50 (54.3%) and 26 (28.3%) subjects, respectively, while the MRI was normal in the remaining 16 (17.4%) patients. Among the 16 EAS patients, 6 (37.5%) had pituitary masses, including 2 patients with tumours greater than 6 mm in diameter. The largest pituitary lesion among the EAS patients was approximately 25 mm; the patient had a hepatic ectopic focus, and his pituitary lesion was considered a metastatic lesion.

Among the 16 EAS patients, 4 suspicious ectopic lesions were discovered in the mediastinum, 3 in the lung, 1 in the liver, 1 in the pancreas, 1 in the extraocular muscles, and 1 in the lung and mediastinum simultaneously. The remaining 5 patients had no obvious primary lesions in the imaging examinations.

Combined HDDST and DDAVP test

As shown in Table 3, 64 patients had positive results for both the DDAVP test and HDDST results. Among them, 63 patients were diagnosed with CD by BIPSS and/or postoperative pathology, and the SE and PPV of this diagnosis were 95.5% (63/66) and 98.4% (63/64), respectively. The only EAS patient with concordant positive test results had a paraganglioma. 10 patients had negative results for both the DDAVP test and HDDST results. Among them, 7 patients were diagnosed with EAS by BIPSS and/or postoperative pathology, and the SE and PPV of the EAS diagnosis were 87.5% (7/8) and 70.0% (7/10), respectively. The total accuracy was 94.6% (70/74). (Tables 2 and 3).

Table 3A Concordance between DDAVP test and HDDST among CD patients

HDDST	DDAVP test		Total
	+	–
+	63	12	75
–	14	3	17
Total	77	15	92

Table 3B Concordance between DDAVP test and HDDST among EAS patients

HDDST	DDAVP test		Total
	+	–
+	1	7	8
–	1	7	8
Total	2	14	16

Combined biochemical and imaging tests

54 patients had positive results from the DDAVP test, the HDDST and sellar MRI finding, 53 were confirmed to have CD, while 1 was verified to have EAS. The SE and PPV in diagnosing CD were 100% (53/53) and 98.1% (53/54), respectively. Among the patients with negative results in all tests, 100% (5/5) were diagnosed with EAS, with an SE of 83.3% (5/6) and a PPV of 100% (5/5). The accuracy was 98.3% (58/59), and the SE reached 100%, while the SP was lower than that of the combination of the HDDST and DDAVP test (Table 2).

AVPR immunohistochemical staining

We investigate the immunohistochemical staining density of V2R and V3R in 16 specimens of corticotrophin adenomas, including 10 with positive and 6 with negative DDAVP responses (all available negative DDAVP test tissues), to explore the relationship between the DDAVP test response and the AVPR expressions levels in corticotrophin adenoma cells. There was no significant difference in the mean density of V2R staining between CD patients with or without a response to the DDAVP test (0.0121 ± 0.0056 vs. 0.0160 ± 0.0073, p value = 0.823; (Fig. 3). The density of V3R staining was significantly weaker in 5/6 CD patients who had no response to DDAVP (0.0149 (0.0118, 0.0379) vs. 0.0081 ± 0.0053, p = 0.04), but 1 remaining case showed a V3R staining density similar to that of the CD patients who had a positive response to the DDAVP test (finally integrated analysis result: 0.0149 (0.0118, 0.0379) vs. 0.0112 ± 0.0090, p = 0.118). The correlation analysis showed that neither the IPS/P ratio nor the percent increase in ACTH levels was associated with AVPR2 or AVPR3 expression.

Fig. 3

The expression of V2R and V3R in CD determined by IHC staining.

A: V2R staining in the pituitary adenoma of a patient who did not respond to desmopressin (original magnification: ×400);

B: V2R staining in the pituitary adenoma of a patient who responded to desmopressin (original magnification: ×400);

C: V3R staining in the pituitary adenoma of a patient who did not respond to desmopressin (original magnification: ×400);

D: V3R staining in the pituitary adenoma of a patient who responded to desmopressin (original magnification: ×400);

E: Negative control in pituitary (original magnification: ×200);

F: V3R-positive control in the pituitary (original magnification: ×400).

Discussion

The determination of the origin of abnormally elevated ACTH secretion in ACTH-dependent CS is crucial, as it directly determines the appropriate treatment options and prognosis [22]. Based on the results of our study, EAS patients exhibited a greater degree of ACTH and cortisol hypersecretion and urinary cortisol excretion, weaker cortisol suppression after the overnight administration of 1-mg DST and more substantial hypokalaemia than CD patients, consistent with the findings of other studies [6]. However, because of the considerable overlap between the two conditions, it is impossible to distinguish CD and EAS based on the basal hormone levels, circadian rhythm and clinical manifestations in clinical practice; therefore, more specific tests are needed to establish the correct source of the increased ACTH [1].

In clinical practice, BIPSS is recommended for all ACTH-dependent forms of CS with a pituitary adenoma less than 6 mm in size and for subjects with discordant test results between the HDDST/CRH test and MRI findings [23-25]. Because approximately 20–30% of pituitary MRIs performed in CD patients are negative [2, 6-8] and many patients have a pituitary adenoma less than 6 mm in size, an easier non-invasive test for determining the source of the increased ACTH secretion is required to improve the standard of medical treatment for patients with ACTH-dependent CS.

In this study, we used pathology and/or BIPSS results as the gold standard to analyse the value of the HDDST and DDAVP test alone or in combination in the differential diagnosis of CD and EAS. The results of the HDDST showed that when UFC suppression ≥50% was used as the cut-off for a positive response, 75 of the 83 patients (90.4%) with positive HDDST results were diagnosed with CD, and the remaining 8 patients were diagnosed with EAS by BIPSS and/or postoperative pathology. Thus, the coincidence rate between the HDDST and BIPSS and/or surgery-based diagnosis of the final aetiology of CS was 76.9% (83/108). If the results of the HDDST are used as the basis for the diagnosis of CD, the false-positive and false-negative rates are 50.0% and 18.5%, respectively. When we increased this cut-off to 62.7%, the AUC was maximized, with SE and SP values of 0.80. Previous studies [6, 24, 26] found that an HDDST adopting a decrease in plasma cortisol levels ≥50% diagnosed CD with an SE ranging from 79.5% to 88% and a high SP ranging from 71.4% to 90%. Our results and findings from other researchers reveal that the diagnostic accuracy of the HDDST alone is not satisfactory to determine the source of ACTH secretion in ACTH-dependent CS.

As an alternative dynamic test to the CRH stimulation test, the DDAVP test, has recently been applied in clinical practice with debatable diagnostic efficacy. Some studies [6, 26, 27] reported that the DDAVP had a higher SE (80–89%) and a relatively lower SP (40–62%) in distinguishing CD and EAS. In the present study, among the 79 patients with DDAVP-positive results, 77 (97.5%) were diagnosed with CD and 2 were diagnosed with EAS by BIPSS/or surgery. Our results revealed that the DDAVP test had an SE of 83.7% (77/92) and SP of 87.5% (14/16). Although the SP was higher than many published results, it was unable to accurately distinguish the aetiology of ACTH-dependent CS alone.

We further analysed the AVPR expression in the specimens of pituitary ACTH adenoma obtained from TSS. We only obtained 6 tissue samples from patients who were non-responsive to the DDAVP test; therefore, we did not remove the extremum in the analysis of V3R expression. V2R and V3R expression was not significantly different between the CD patients with or without a response to DDAVP stimulation. Some authors have postulated that upregulation of V3R [28, 29], and/or ectopic expression of V2R [30-32] provides insights into the ACTH response to desmopressin. However, other researchers [33] considered that ACTH secretion in response to DDAVP stimulation in CD may be related to external pituitary confounders. Hence, the mechanism of the DDAVP stimulation test needs further investigation.

Because a 100% SP in differentiating between CD and EAS with the HDDST or the DDAVP test alone is impossible to obtain, we further analysed the diagnostic value of the combination of the HDDST and the DDAVP test. Our results indicated that among patients with ACTH-dependent CS, if the DDAVP test and the HDDST results were both positive, the SE and PPV in diagnosing CD were 95.5% (63/66) and 98.4% (63/64), respectively. Additionally, if the two results were both negative, the SE and PPV in diagnosing EAS were 87.5% (7/8) and 70.0% (7/10), respectively. Although the combination of the HDDST and the DDAVP test plays an important role in distinguishing the pseudo-Cushing’s state and Cushing’s syndrome [34], our study is the first to show that the combined use of two non-invasive endocrine tests substantially improves the efficiency of the differential diagnosis between CD and EAS.

Retrospective studies have revealed that the combination of a positive ACTH response in CRH stimulation test and a positive HDDST reached an SE of 92.1–97.6% and PPV of 100% in diagnosing CD [6, 18]. The result of our study is consistent with these studies. We propose that HDDST combined with CRH test or DDAVP test substantially improves the efficiency of non-invasive dynamic endocrine tests in the etiological investigation of ACTH-dependent CS. DDAVP can be used as a substitute for CRH preparations in some countries where CRH preparations are not available.

Another finding in our study was that the combination of these two endocrine tests and imaging studies further improved the specificity of the CD diagnosis. When positive pituitary MRI findings rather than the previously recommended 6-mm boundary value were considered, we found that the accuracy reached 98.3%, and the SE and SP reached up to 100% (53/53) and 83.3% (5/6), respectively, with a 98.2% PPV and 100% NPV for the CD diagnosis (Table 2). Our results are similar to the data reported by Frete C et al. [19].

Although BIPSS is important for distinguishing CD and EAS, this procedure requires professional doctors and advanced equipment and is even associated with a certain probability of complications [9]. Our results suggest that if both the HDDST and the DDAVP test are positive for subjects with ACTH-dependent CS, the PPV is 98.4% in diagnosing CD, especially when pituitary imaging uncovers an adenoma (regardless of whether the size is greater than 6 mm). These dual non-invasive endocrine tests can substantially reduce the need for BIPSS in the etiological investigation of ACTH-dependent CS.

The limitations of this study are similar to those of previous studies that retrospectively categorize a disease. The number of CD patients was much higher than the number of EAS patients since EAS is relatively rare, which may reduce the power of the statistical analysis. Our single-center study used systematic test procedures and included reformative pituitary MRI images. In addition, although our study excluded only cases with distinct clinical diagnoses who lacked BIPSS or pathological verification, selection bias may exist. Additionally, BIPSS may be associated with a non-negligible false-positive and false-negative rate, although it is regarded as the gold standard for the diagnosis of CD. This study should be repeated in multiple centres with a larger sample size to determine the optimal ROC cut-off and verify the findings from other institutions.

In conclusion, this study proposed a non-invasive investigational strategy to determine the aetiological of ACTH-dependent Cushing’s syndrome. The combination of the HDDST and the DDAVP test may significantly improve the efficiency of the differential diagnosis between CD and EAS and significantly reduce the need for BIPSS. These dual non-invasive endocrine tests are easy to perform and far less expensive than BIPSS, providing clinicians easier and more reliable methods for distinguish CD and EAS.

Acknowledgements

This work was supported by grants from the Science & Technology Department of Sichuan Province (19ZDYF1311).

Disclosure

The authors have nothing to disclose. None of authors have any potential conflicts of interest associated with this research.

References

• 1   Nieman  LK,  Biller  BMK,  Findling  JW,  Newell-Price  J,  Savage  MO, et al. (2008) The diagnosis of Cushing’s Syndrome: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab 93: 1526–1540.
• 2   Yogi-Morren  D,  Habra  MA,  Faiman  C,  Bena  J,  Hatipoglu  B, et al. (2015) Pituitary Mri findings in patients with pituitary and ectopic Acth-dependent Cushing syndrome: does a 6-mm pituitary tumor size cut-off value exclude ectopic Acth syndrome? Endocr Pract 21: 1098–1103.
• 3   Invitti  C,  Pecori Giraldi  F,  de Martin MCavagnini  F (1999) Diagnosis and management of Cushing’s syndrome: results of an Italian multicentre study. Study Group of the Italian Society of Endocrinology on the pathophysiology of the hypothalamic-pituitary-adrenal axis. J Clin Endocrinol Metab 84: 440–448.
• 4   Hall  WA,  Luciano  MG,  Doppman  JL,  Patronas  NJ,  Oldfield  EH (1994) Pituitary magnetic resonance imaging in normal human volunteers: occult adenomas in the general population. Ann Intern Med 120: 817–820.
• 5   Sun  Y,  Sun  Q,  Fan  C,  Shen  J,  Zhao  W, et al. (2012) Diagnosis and therapy for Cushing’s disease with negative dynamic MRI finding: a single-centre experience. Clin Endocrinol (Oxf) 76: 868–876.
• 6   Barbot  M,  Trementino  L,  Zilio  M,  Ceccato  F,  Albiger  N, et al. (2016) Second-line tests in the differential diagnosis of ACTH-dependent Cushing’s syndrome. Pituitary 19: 488–495.
• 7   Kakade  HR,  Kasaliwal  R,  Jagtap  VS,  Bukan  A,  Budyal  SR, et al. (2013) Ectopic ACTH-secreting syndrome: a single-center experience. Endocr Pract 19: 1007–1014.
• 8   Ciric  I,  Zhao  JC,  Du  H,  Findling  JW,  Molitch  ME, et al. (2012) Transsphenoidal surgery for Cushing disease: experience with 136 patients. Neurosurgery 70: 70–81.
• 9   Zampetti  B,  Grossrubatscher  E,  Dalino Ciaramella  P,  Boccardi ELoli  P (2016) Bilateral inferior petrosal sinus sampling. Endocr Connect 5: R12–R25.
• 10   Saito  M,  Tahara  A,  Sugimoto  T (1997) 1-desamino-8-D-arginine vasopressin (DDAVP) as an agonist on V1b vasopressin receptor. Biochem Pharmacol 53: 1711–1717.
• 11   Vassiliadi  DA,  Tsagarakis  S (2018) Diagnosis of endocrine disease: the role of the desmopressin test in the diagnosis and follow-up of Cushing’s syndrome. Eur J Endocrinol 178: R201–R214.
• 12   Moro  M,  Putignano  P,  Losa  M,  Invitti  C,  Maraschini  C, et al. (2000) The desmopressin test in the differential diagnosis between Cushing’s disease and pseudo-Cushing states. J Clin Endocrinol Metab 85: 3569–3574.
• 13   Rollin  GA,  Costenaro  F,  Gerchman  F,  Rodrigues TCCzepielewski  MA (2015) Evaluation of the DDAVP test in the diagnosis of Cushing’s Disease. Clin Endocrinol (Oxf) 82: 793–800.
• 14   Tirabassi  G,  Faloia  E,  Papa  R,  Furlani  G,  Boscaro  M, et al. (2010) Use of the desmopressin test in the differential diagnosis of pseudo-Cushing state from Cushing’s disease. J Clin Endocrinol Metab 95: 1115–1122.
• 15   Romanholi  DJ,  Machado  MC,  Pereira  CC,  Danilovic  DS,  Pereira  MA, et al. (2008) Role for postoperative cortisol response to desmopressin in predicting the risk for recurrent Cushing’s disease. Clin Endocrinol (Oxf) 69: 117–122.
• 16   Dall’Asta  C,  Barbetta  L,  Bonavina  L,  Beck-Peccoz  P,  Ambrosi  B (2004) Recurrence of Cushing’s disease preceded by the reappearance of ACTH and cortisol responses to desmopressin test. Pituitary 7: 183–188.
• 17   Sharma  ST,  Nieman LKFeelders  RA (2015) Cushing’s syndrome: epidemiology and developments in disease management. Clin Epidemiol 7: 281–293.
• 18   Ritzel  K,  Beuschlein  F,  Berr  C,  Osswald  A,  Reisch  N, et al. (2015) ACTH after 15 min distinguishes between Cushing’s disease and ectopic Cushing’s syndrome: a proposal for a short and simple CRH test. Eur J Endocrinol 173: 197–204.
• 19   Frete  C,  Corcuff  J,  Kuhn  E,  Salenave  S,  Gaye  D, et al. (2020) Non-invasive diagnostic strategy in ACTH-dependent Cushing’s syndrome. J Clin Endocrinol Metab 105: 3273–3284.
• 20   Miller  DL,  Doppman  JL (1991) Petrosal sinus sampling: technique and rationale. Radiology 178: 37–47.
• 21   Findling  JW,  Kehoe  ME,  Shaker  JL,  Raff  H (1991) Routine inferior petrosal sinus sampling in the differential diagnosis of adrenocorticotropin (ACTH)-dependent Cushing’s syndrome: early recognition of the occult ectopic ACTH syndrome. J Clin Endocrinol Metab 73: 408–413.
• 22   Nieman  LK,  Biller  BM,  Findling  JW,  Murad  MH,  Newell-Price  J, et al. (2015) Treatment of Cushing’s syndrome: an Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab 100: 2807–2831.
• 23   Boscaro  M, Arnaldi GJTJocemetabolism (2009) Approach to the patient with possible Cushing’s syndrome. J Clin Endocrinol Metab 94: 3121–3131.
• 24   Vilar  L,  Freitas  MC,  Naves  LA,  Canadas  V,  Albuquerque  JL, et al. (2008) The role of non-invasive dynamic tests in the diagnosis of Cushing’s syndrome. J Endocrinol Invest 31: 1008–1013.
• 25   Hayes  AR,  Grossman  AB (2018) The ectopic adrenocorticotropic hormone syndrome: rarely easy, always challenging. Endocrinol Metab Clin North Am 47: 409–425.
• 26   Suda  T,  Kageyama  K,  Nigawara  T,  Sakihara  S (2009) Evaluation of diagnostic tests for ACTH-dependent Cushing’s syndrome. Endocr J 56: 469–476.
• 27   Terzolo  M,  Reimondo  G,  Alì  A,  Borretta  G,  Cesario  F, et al. (2001) The limited value of the desmopressin test in the diagnostic approach to Cushing’s syndrome. Clin Endocrinol (Oxf) 54: 609–616.
• 28   Luque  RM,  Ibáñez-Costa  A,  López-Sánchez  LM,  Jiménez-Reina  L,  Venegas-Moreno  E, et al. (2013) A cellular and molecular basis for the selective desmopressin-induced ACTH release in Cushing disease patients: key role of AVPR1b receptor and potential therapeutic implications. J Clin Endocrinol Metab 98: 4160–4169.
• 29   Craighead  M,  Milne  R,  Campbell-Wan  L,  Watson  L,  Presland  J, et al. (2008) Characterization of a novel and selective V1B receptor antagonist. Prog Brain Res 170: 527–535.
• 30   Wang  FF,  Tang  KT,  Yen  YS,  Ho  DM,  Yang  AH, et al. (2012) Plasma corticotrophin response to desmopressin in patients with Cushing’s disease correlates with the expression of vasopressin receptor 2, but not with that of vasopressin receptor 1 or 3, in their pituitary tumours. Clin Endocrinol (Oxf) 76: 253–263.
• 31   Tsagarakis  S,  Tsigos  C,  Vasiliou  V,  Tsiotra  P,  Kaskarelis  J, et al. (2002) The desmopressin and combined CRH-desmopressin tests in the differential diagnosis of ACTH-dependent Cushing’s syndrome: constraints imposed by the expression of V2 vasopressin receptors in tumors with ectopic ACTH secretion. J Clin Endocrinol Metab 87: 1646–1653.
• 32   Yang  J,  Yang  Y,  Wang  Y,  Zhang  S,  Cheng  H, et al. (2020) Role of vasopressin receptor 2 and 3 in ACTH-secreting tumors and their potential therapeutic implications. Exp Clin Endocrinol Diabetes 128: 263–269.
• 33   Pecori Giraldi  F,  Marini  E,  Torchiana  E,  Mortini  P,  Dubini  A, et al. (2003) Corticotrophin-releasing activity of desmopressin in Cushing’s disease: lack of correlation between in vivo and in vitro responsiveness. J Endocrinol 177: 373–379.
• 34   Araya  AV,  Romero  C,  Lemp  M (2017) Combined dexamethasone and desmopressin test in the differential diagnosis of ACTH-dependent Cushing’s syndrome and pseudo-cushing’s states. Pituitary 20: 602–603.