The Clinical Impact of [⁶⁸Ga]-DOTATATE PET/CT for the Diagnosis and Management of Ectopic Adrenocorticotropic Hormone – Secreting Tumors

Summary

Objectives:

Localization of ectopic ACTH-secreting tumors causing Cushing syndrome (ECS) is essential for clinical management, yet often difficult. [⁶⁸Ga]-DOTATATE PET/CT ([⁶⁸Ga]-DOTA-(Tyr³)-octreotate)] is an FDA-approved high-resolution diagnostic tool for imaging neuroendocrine tumors. Data on the clinical utility of [⁶⁸Ga]-DOTATATE in patients with ECS, however, are scarce. The objectives of this study were to determine the efficacy for ECS localization and the clinical benefit of [⁶⁸Ga]-DOTATATE imaging.

Method:

We conducted a retrospective review of all cases with ECS evaluated with [⁶⁸Ga]-DOTATATE from November 2016 through October 2018 at three referral centers. The clinical benefit of [⁶⁸Ga]-DOTATATE was based on detection of new tumors and resultant changes in management.

Results:

Over the study period, 28 patients with ECS underwent [⁶⁸Ga]-DOTATATE: 17 for identification of the primary tumor and 11 during follow up. [⁶⁸Ga]-DOTATATE identified the suspected primary ECS in 11/17 patients (65%). Of these, 9 patients underwent surgery: 8 with confirmed ECS (5 bronchial, 1 thymic, 1 pancreatic and 1 metastatic neuroendocrine tumor of unknown primary origin) and 1 patient with a false positive scan (adrenal gland). Of the 11 patients with ECS who underwent [⁶⁸Ga]-DOTATATE evaluation during follow-up, the study led to changes in clinical management in 7/11 (64%) patients.

Conclusions:

[⁶⁸Ga]-DOTATATE is sensitive in detecting primary and metastatic ECS, often identifies occult tumors after conventional imaging, and impacts clinical care in the majority of patients.

Introduction

Ectopic ACTH-secreting tumors often lead to fulminant Cushing syndrome (ECS) and are associated with higher mortality compared to other causes of cortisol excess.¹ Prompt identification and treatment of ECS are imperative for optimal clinical outcomes. Surgical cure, however, is achieved in less than half of patients with ECS, partly due to unresectable metastases and also because many such tumors are small and difficult to localize.^2–6

Cross-sectional anatomic imaging is commonly used as first-line radiographic investigation for identification of ECS, but its sensitivity for ECS detection is suboptimal, ranging from 52–66%.^7,8 In a systematic review, various nuclear medicine functional imaging techniques identified 79% of ECS not detected by initial conventional anatomic imaging.⁸ Scintigraphy or single-photon emission computed tomography (SPECT) with [¹¹¹In]-pentetreotide, a somatostatin receptor (SSTR) ligand (SRL), and [¹³¹I]- or [¹²³I]-metaiodobenzylguanidine (MIBG), a catecholamine analog, are commonly used for NET detection, but these studies require prolonged time (imaging at 24 to 48 hours after radiotracer injection) and have relatively limited sensitivity for disease detection.⁹ Positron emission tomography (PET) offers significantly improved imaging quality and shorter protocols ^10,11 using a variety of tracers, including [¹⁸F]-fluorodeoxyglucose (FDG), [¹⁸F]-dihydroxyphenylalanine (DOPA) and [⁶⁸Ga]-DOTATATE (DOTA-(Tyr³)-octreotate).^7,12,13 For SRL imaging, [⁶⁸Ga]-DOTATATE PET/CT has proven to be more sensitive than [¹¹¹In]-pentetreotide SPECT in detecting various NETs.^14–18 The advantages of [⁶⁸Ga]-DOTATATE in patients with ECS, however, have only been described in isolated case reports or small case series.^18,19 Here we retrospectively studied the use of [⁶⁸Ga]-DOTATATE in localizing ECS and the resultant impact on clinical management beyond conventional imaging.

Patients and Methods

We conducted a retrospective study of all cases with ECS evaluated with [⁶⁸Ga]-DOTATATE imaging in three tertiary referral centers (University of Michigan, Mayo Clinic, Rochester, and The University of Texas MD Anderson Cancer Center) since FDA approval (November 2016), through October 2018. Patient demographics, clinical evaluation, imaging modalities, laboratory, histopathological results and treatment data were reviewed. The study was conducted with approval and waiver of informed consent from the Institutional Review Board at the University of Michigan, Mayo Clinic (Rochester), and MD Anderson Cancer Center.

The diagnosis of Cushing syndrome was confirmed by clinical and hormonal evaluation, in accordance with the Endocrine Society practice guidelines (Table 1).⁹ Ectopic ACTH syndrome was defined by hormonal assessment compatible with ACTH-dependent Cushing syndrome, in conjunction with absence of central gradient during bilateral inferior petrosal sinus sampling or known ECS.^3,4 All cross-sectional imaging, [¹¹¹In]-pentetreotide scintigraphy and SPECT, [⁶⁸Ga]-DOTATATE and [¹⁸F]-FDG PET/CT images were reviewed by experienced radiologists and nuclear medicine physicians in each institution.

Table 1.

Characteristics of study participants

	University of Michigan N=17	Mayo Clinic Rochester N=6	MD Anderson Cancer Center N=5	Total N=28
Age (years)	49 [33-49]	50 [39-63]	51 [43-64]	50 [38-64]
Sex (F/M)	15/2	4/2	3/2	22/6
Clinical diagnosis of CS
- Moon face	12	4	5	21 (75%)
- Dorsocervical fat pad	8	3	3	14 (50%)
- Supraclavicular fat pad	8	5	2	15 (54%)
- Purplish striae	9	3	0	12 (43%)
- Facial plethora	11	4	1	16 (57%)
- Central obesity	11	4	4	19 (68%)
- Proximal muscle weakness	16	5	5	26 (93%)
- Hypokalemia	13	4	3	20 (71%)
- Hyperglycemia	12	5	4	21 (75%)
- Hypertension	14	5	4	23 (82%)
Biochemical diagnosis of CS
- Basal cortisol (μg/dL)	61 [41-78]	30 [19-44]	64 [24-103]	56 [28-70]
- Basal ACTH (pg/mL)	160 [101-294]	112 [84-202]	123 [92-167]	127 [89-245]
- Urine free cortisol (μg/24 h)	1229 [315-3234]	289 [167-3342]	441 [404-3151]	730 [225-3456]
ECS diagnosis
- Bronchial NET	5	2	1	8
- Pancreatic NET	2	2	1	5
- MTC	2	0	0	2
- Ileal NET	1	0	0	1
- Thymic NET	1	0	0	1
- Non-small cell lung cancer	1	0	0	1
- NET-unknown primary*	5	2	3	10

^*NET of unknown primary origin included: 7 occult NETs (including 1 false positive lesion) and 3 metastatic NETs with unclear primary site.

Age and biochemical data were expressed as medians [interquartile range].

F, Female; CS, Cushing syndrome; ACTH, adrenocorticotropic hormone; ECS, ectopic ACTH-secreting tumor causing Cushing syndrome; NET, neuroendocrine tumor; MTC, medullary thyroid carcinoma; CS, Cushing syndrome.

The clinical impact of [⁶⁸Ga]-DOTATATE was defined as the detection of primary ECS or new metastatic foci, along with consequent changes in clinical management.

Results

Over the study period, 28 patients (age 18–77 years, median 50, 78.6 % females) with ectopic ACTH syndrome underwent [⁶⁸Ga]-DOTATATE scan, including 17 patients to localize the primary ECS (Table 1). Of these, the primary tumor was occult after previous imaging in 15 patients (cross-sectional imaging in all patients, [¹⁸F]-FDG PET/CT scan in 4 patients, and [¹¹¹In]-pentetreotide in 3 other patients), and in 2 patients, [⁶⁸Ga]-DOTATATE scan was the first localizing imaging study used (Figure 1). [⁶⁸Ga]-DOTATATE scan identified an ECS in 11/17 (65%) patients, of which 7 were solitary and 4 metastatic. The diagnosis was confirmed by pathology in 8/11 patients: 5 bronchial NET, 1 thymic NET, 1 pancreatic NET and 1 metastatic NET of unknown primary origin. Surgical cure was achieved in 7 of these patients (Figures 2–3), while 1 remaining patient had widely metastatic disease. In this latter patient who was not cured, [⁶⁸Ga]-DOTATATE identified bronchial and bone lesions but failed to identify liver metastases, which were apparent on MRI and confirmed by biopsy, illustrating false-negative metastases (Figure 4). In another patient with occult ECS, [⁶⁸Ga]-DOTATATE demonstrated a nodular focus of uptake in the left adrenal gland (Figure 3). Clinical cure, however, was not achieved after left adrenalectomy, and pathology showed adrenocortical hyperplasia. This case illustrates a false-positive scan, which caused a delay in contralateral adrenalectomy and cure of the hypercortisolemia. Of the remaining 2 patients, one was lost to follow up, and the second is awaiting surgery. Of the 7 patients in whom the ECS source remained occult after [⁶⁸Ga]-DOTATATE PET/CT scan, 3 patients underwent bilateral adrenalectomy, 2 were treated with mifepristone, and 2 were treated with ketoconazole for management of hypercortisolemia.

Figure 1.

Patients with ECS evaluated at University of Michigan, Mayo Clinic Rochester and The University of Texas MD Anderson Cancer Center who underwent [⁶⁸Ga]-DOTATATE PET/CT between November 2016 to Oct 2018.

The dimensions of ECS detected by [⁶⁸Ga]-DOTATATE PET/CT imaging ranged from 0.8 cm to 2.8 cm for primary tumor localization or 0.4 cm to 1.8 cm for newly found metastases. Pathology revealed poorly-differentiated NET in 2 patients and well-differentiated NET in 6.

ECS, ectopic ACTH-secreting tumor causing Cushing syndrome; NET, neuroendocrine tumor.

Figure 2.

[⁶⁸Ga]-DOTATATE PET/CT demonstrated focal uptake in the upper lobe of the left lung (A, B, C), two foci of uptake in the left hilar region (A), and a left thoracic inlet lymph node (LN) (D).

Figure 3.

[⁶⁸Ga]-DOTATATE PET/CT demonstrated uptake in the thymus (A, C) and mediastinal lymph node (A, B, C). Pathology revealed high-grade thymic NET with mediastinal lymph node metastasis.

[⁶⁸Ga]-DOTATATE PET/CT revealed focal uptake in the left adrenal gland (D, E). CT abdomen showed left adrenal thickening (F). Left adrenalectomy was performed, but pathology demonstrated adrenal cortex hyperplasia but no malignant tumor, and hypercortisolism did not resolve.

NET, neuroendocrine tumor.

Figure 4.

[⁶⁸Ga]-DOTATATE PET/CT demonstrated uptake in the right hilar area (A, B), 8^th thoracic vertebra (C) and right iliac bone (D) but failed to identify three liver metastases, apparent on MRI, and confirmed by biopsy. Right liver lobe metastases (E, F) and left liver lobe metastasis (G) in a patient with NET of unclear primary origin.

NET, neuroendocrine tumor.

[⁶⁸Ga]-DOTATATE was performed to assess disease burden or recurrence in 11 patients with known ECS (Figure 1): 4 pancreatic NETs; 2 unknown primary neuroendocrine carcinomas; 2 medullary thyroid cancers (MTC); 1 ileal NET; 1 bronchial NET, and 1 non-small cell lung cancer (NSCLC). [⁶⁸Ga]-DOTATATE identified 9 new metastatic foci and 3 recurrent tumors: 5 solitary bone lesions, 3 pancreatic lesions, and 1 intra-abdominal lymph node. Of the 11 patients with known ECS, [⁶⁸Ga]-DOTATATE evaluation during follow-up period led to changes in clinical management in 7 cases: in 3 patients with widely metastatic cancer previously treated with chemotherapy (2 MTC and 1 pancreatic NET), [⁶⁸Ga]-DOTATATE uptake in numerous metastatic foci (Figure 5) led to treatment with [¹⁷⁷Lu]- or [⁹⁰Y]-labeled peptide-receptor radionuclide therapy (PRRT); in 2 patients with prior surgical removal of primary ECS (one bronchial and one pancreatic NET), [⁶⁸Ga]-DOTATATE identified single recurrent left upper lung and pancreatic bed lesions, respectively, which were subsequently removed; in 1 patient with NET of unknown primary origin, [⁶⁸Ga]-DOTATATE demonstrated new focal bone lesions, which led to treatment with everolimus; and in the last patient with pancreatic NET, 2 metastatic foci were found in the pancreatic tail, which prompted treatment with a somatostatin analogue. In the remaining 4 cases (1 pancreatic NET, 1 ileal NET, 1 NSCLC and 1 NET of unclear primary origin; all metastatic), [⁶⁸Ga]-DOTATATE findings were concordant with conventional imaging.

Figure 5.

[⁶⁸Ga]-DOTATATE PET/CT demonstrated new bone metastatic lesions in the right humerus (B) 3^rd thoracic vertebra (C) 4^th lumbar vertebra (D) and right femur (E) in a patient with medullary thyroid cancer with liver and lymph node metastasis (A).

The dimensions of newly detected lesions with [⁶⁸Ga]-DOTATATE PET/CT imaging, either primary or metastatic, ranged from 0.4 cm to 2.8 cm. Of the 8 patients with primary ECS confirmed by pathology, 6 were classified as well-differentiated NET (5 bronchial and 1 pancreatic) and 2 as poorly-differentiated NET (1 thymic and 1 of unknown primary origin). The majority of primary tumors identified by [⁶⁸Ga]-DOTATATE PET/CT were not recognized on cross-sectional imaging, especially tumors <1 cm. In 2 patients, [⁶⁸Ga]-DOTATATE PET/CT detected tumors in the right middle lung fields that were misidentified as vascular in origin. In retrospect, however, an anatomic correlate to all DOTATATE-avid tumors except bone metastases could be identified on cross-sectional imaging (Figures 2–5).

Discussion

In this retrospective series of 28 patients with ectopic ACTH production leading to Cushing syndrome, [⁶⁸Ga]-DOTATATE localized the primary tumor and/or metastases in the majority of study patients. [⁶⁸Ga]-DOTATATE is a recently approved advanced molecular imaging probe with high affinity for SSTR-type 2 (0.2 ±0.04 nM, as compared to 22 ± 3.6 nM for [¹¹¹In]-pentetreotide)²⁰, which when coupled with the advanced spatial resolution of PET/CT imaging systems, can result in successful detection and accurate localization of small tumors characterized by SSTR expression. A review of the literature found that [⁶⁸Ga]-DOTATATE is highly sensitive (81.8%) in localization of ECS; however, only 23 of 231 patients assimilated from these studies received [⁶⁸Ga]-DOTATATE imaging.⁸ Papadakis previously published a case of a ectopic Cushing and CRH co-secreting tumor, identified by [⁶⁸Ga]-DOTATATE PET/CT.²¹ ECS are often highly functional and associated with a fulminant clinical course and high mortality. Thus, ECS localization is critical for managing this aggressive form of Cushing syndrome. Despite extensive conventional cross-sectional and functional imaging, up to 19% of ECS remain occult.^2–4 In our study, [⁶⁸Ga]-DOTATATE identified the primary ECS in 11/17 (65%) of previously occult tumors. The somewhat lower overall accuracy of [⁶⁸Ga]-DOTATATE-PET/CT in our series might reflect a selection bias for difficult cases at our referral centers, with a high prevalence of occult ECS.

Most of these ECS were located in the thorax, and some were smaller than 1 cm, which is near the limit of detection for other imaging modalities. Furthermore, [⁶⁸Ga]-DOTATATE detected not only well-differentiated NET, but also a poorly differentiated ECS, which are composed of highly proliferative cells with low SSTR expression and high metabolic activity, as evidenced by [¹⁸F]-FDG PET/CT. ^11,22 Hypercortisolemia can lower SSTR expression in EAS²³ and result in false-negative scintigraphy, which can be reversed with glucocorticoid antagonist therapy.²⁴ In our series of 10 patients with previously occult tumors that were located with [⁶⁸Ga]-DOTATATE scans, 3 and 1 were treated with mifepristone and ketoconazole, respectively. Similarly, of the 7 patients with tumors that were not located with [⁶⁸Ga]-DOTATATE scans, 3 each were treated with mifepristone or ketoconazole prior to scanning. Thus, 6 of 7 false-negative scans failed to identify the source of ACTH despite concomitant treatment of hypercortisolemia. [⁶⁸Ga]-DOTATATE imaging was performed only once in all patients, which precludes any conclusions about longitudinal changes or comparisons with and without treatment of hypercortisolemia.

Of previously identified ECS, [⁶⁸Ga]-DOTATATE impacted clinical management in 7/11 (64%) of patients. [⁶⁸Ga]-DOTATATE allowed cure of recurrent disease in 2 patients and identified 3 other patients as candidates for PRRT with [¹⁷⁷Lu]- or [⁹⁰Y]-labeled somatostatin analogs. ^10,11,25

Physiologic uptake of [⁶⁸Ga]-DOTATATE in several locations such as liver, spleen, adrenal glands, thyroid, pancreas, and kidneys can confound image interpretation.^26,27 A previous systematic review described one case of lung NET with a false-positive focus of uptake in the adrenal gland.²⁸ Similarly, in our study, an ACTH-secreting pheochromocytoma was suspected in one patient with occult tumor who had atypical peripheral [⁶⁸Ga]-DOTATATE uptake in the left adrenal gland and normal plasma metanephrines (Figure 3). Cushing syndrome was not resolved after left adrenalectomy, and histopathology showed adrenal cortical hyperplasia. Conversely, in another patient with metastatic NET of unknown primary origin, [⁶⁸Ga]-DOTATATE-avid osseous and pulmonary metastatic lesions were found, but biopsy-proven liver metastases could not be detected (Figure 4), perhaps due to the high background activity of liver and/or poor SSTR-expression in certain clonal cell populations. One review of [⁶⁸Ga]-DOTATATE imaging for detecting ECS reported 17.4% false-negative studies.⁸

Our study demonstrates the high sensitivity of [⁶⁸Ga]-DOTATATE in the localization of ECS, for both occult primary tumors and metastatic lesions. Importantly, the use of [⁶⁸Ga]-DOTATATE impacted clinical management in 64% patients with ECS overall. High cost and limited availability of PET/CT imaging, however, might preclude the widespread use of [⁶⁸Ga]-DOTATATE for ECS imaging, experience with these scans, which were FDA-approved for clinical use in the USA to localize relatively rare tumors in June of 2016, is somewhat limited compared to other imaging studies. Limitations of our study include the small number of patients and the retrospective study design. In addition, the choice of imaging modalities, the management of hypercortisolemia, and the long-term plan of management were not pre-specified and were subject to many subjective and circumstantial influences. Nonetheless, combining the experience of three large referral centers, our study gathers the largest number of ECS imaged with [⁶⁸Ga]-DOTATATE to date and provides a benchmark for the utility of this diagnostic modality for this rare but highly morbid condition. Further prospective investigations are needed to characterize the benefits and pitfalls of [⁶⁸Ga]-DOTATATE in diagnosis and follow-up of patients with ECS.