Functional Characterization of Adrenal Lesions Using [123]]IMTO-SPECT/CT
Stefanie Hahner,* Michael C. Kreissl,* Martin Fassnacht, Heribert Haenscheid, Stefanie Bock, Frederik A. Verburg, Pascal Knoedler, Katharina Lang, Christoph Reiners, Andreas K. Buck, Bruno Allolio, and Andreas Schirbel
Endocrinology and Diabetes Unit (S.H., M.F., S.B., K.L., B.A.), Department of Internal Medicine I, Department of Nuclear Medicine (M.C.K., H.H., C.R., A.K.B., A.S.), and Department of Radiology (P.K.), University Hospital of Würzburg, University of Würzburg, D-97080 Würzburg, Germany; and Department of Nuclear Medicine (F.A.V.), University Hospital of Aachen, Aachen University, D-52074 Aachen, Germany
Context: Adrenal tumors are highly prevalent and represent a wide range of different pathological entities. Conventional imaging often provides only limited information on the origin of these lesions. Novel specific imaging methods are, therefore, of great clinical interest.
Objective: We evaluated [123I]iodometomidate ([123I]IMTO) imaging for noninvasive character- ization of adrenal masses.
Design/Setting: This was a prospective monocentric diagnostic study in a tertiary care center.
Patients and Intervention: A total of 51 patients with an adrenal lesion underwent [123I]IMTO imaging after injection of 185 MBq of [123I]IMTO. Sequential planar whole-body scans until 24 hours postinjection and single photon emission computed tomography (SPECT)/computed tomog- raphy imaging 4 to 6 hours postinjection were performed.
Main Outcome Measure: Sensitivity and specificity of [123I]IMTO imaging for the noninvasive characterization of adrenal lesions were measured.
Results: Adrenocortical tissue showed high and specific tracer uptake with a short investigation time and low radiation exposure. Qualitative analysis of SPECT/computed tomography data re- sulted in a sensitivity of 89% and a specificity of 85% for differentiating adrenocortical tumors from lesions of nonadrenocortical origin. Receiver-operating characteristic analysis of semiquantitative data revealed a sensitivity of 83% and a specificity of 86% for identification of adrenocortical lesions at a cutoff value of tumor to liver ratio of 1.3.
Conclusions: [123I]IMTO is a highly specific radiotracer for imaging of adrenocortical tissue with a short investigation time and low radiation exposure. Because of the general availability of SPECT technology, [123I]IMTO scintigraphy has the potential to become a widely used tool to noninva- sively characterize the biology of adrenal lesions. (J Clin Endocrinol Metab 98: 1508-1518, 2013)
T he prevalence of incidentally discovered adrenal masses on computed tomography (CT) examinations has been reported to be 0.4% to 5.0% (1-3). The key question is whether or not such a tumor needs to be re- moved by surgery. This is the case for tumors leading to
relevant hormone excess or if morphological features sug- gest malignancy. Although CT and magnetic resonance imaging (MRI) protocols often allow malignancy to be ruled out because of the high fat content of the adrenal lesion, a significant portion of patients have indeterminate
* S.H. and M.C.K. contributed equally to the study.
Abbreviations: ACC, adrenocortical carcinoma; CT, computed tomography; ED, effective dose; FDG, fludeoxyglucose; IMTO, iodometomidate; MRI, magnetic resonance imaging; PET, positron emission tomography; pi, postinjection; ROC, receiver-operating character- istic; SPECT, single photon emission computed tomography; VOI, volume of interest.
tumors (4-9). Furthermore, in hormonally nonfunction- ing tumors, differentiating adrenocortical lesions from other lesions is a major diagnostic challenge. Because ther- apeutic strategies differ among the various entities (eg, adrenocortical neoplasia vs metastatic disease), noninva- sive characterization of adrenal masses would be of great value.
In recent years, etomidate derivatives that bind with high specificity and avidity to CYP11B enzymes exclu- sively expressed in adrenocortical cells have been devel- oped as tracers for adrenal imaging (10-15). For example, [11C]metomidate has been successfully introduced as a tracer for positron emission tomography (PET), differen- tiating adrenocortical from nonadrenocortical tissue with high specificity. Its use, however, is limited by the short half-life of 11℃ (20 minutes), which restricts the availabil- ity of [11C]metomidate to PET centers with access to an on-site cyclotron. Furthermore, because of the short half- life, only early uptake of the tracer can be monitored, potentially missing the optimum target to background ratio.
We have recently demonstrated that iodometomidate (IMTO) binds to adrenal membranes with high-affinity in vitro and in vivo (16). Therefore, we hypothesized that the use of [123I]IMTO for single photon emission computed tomography (SPECT) and planar scintigraphy may pro- vide a valuable alternative to PET imaging.
Patients and Methods
Patient selection
We aimed to include patients with adrenal lesions of different adrenocortical and nonadrenocortical origin. Inclusion criteria were an adrenal lesion ≥2 cm in diameter detected by CT or MRI, complete hormonal workup of the tumor lesion, negative pregnancy test and established contraception in women of child- bearing age, and written informed consent. Exclusion criteria were pregnancy or breast-feeding and renal failure (serum cre- atinine >2 mg/dL, modification of diet in renal disease value ≤60 mL/min). The article comprises data obtained from patients included in a phase I/II study (n = 33) (Clinical Trial NCT00454103) and from patients investigated according to the study protocol on a compassionate use basis after completion of the clinical trial (n = 18). The phase I/II study included both patients with an adrenal lesion and patients with known meta- static adrenocortical carcinoma (ACC) who had already under- gone resection of the primary tumor (n = 50). This article focuses on the potential of [123IJIMTO to characterize adrenal lesions, excluding imaging of metastasis in advanced ACC. Thus, only the subgroup of 33 study patients with an adrenal lesion still in situ at the time of imaging was analyzed. The results of the re- maining patients with metastatic ACC and data from the 14 patients with ACC included in this analysis were also included in an analysis of the utility of [123I]IMTO in patients with ACC, which will be published elsewhere.
Radiopharmaceutical preparation
[123I]IMTO was prepared as described previously (16). La- beling was performed in a sealed conical vial containing 30 µg of the stannylated precursor in 30 uL of ethanol and [123I]Nal in 0.02 N NaOH. To initiate the reaction, 6 pL of 1 N HCI and 10 uL of chloramine-T (1.5 mg/mL) were added. The reaction was allowed to proceed for 3 minutes at room temperature and was quenched by the addition of 7 AL of 1 N NaOH and 10 uL of Na2S2O5 (4 mg/mL). Purification of [123IJIMTO was performed by HPLC. The [123I]IMTO-containing fraction was evaporated to dryness and dissolved in saline and passed through a sterile filter.
Assessment of the biodistribution of [123I]IMTO using SPECT and planar images
For planar imaging, a dual-headed large field-of-view gamma camera (E.CAM Duet; Siemens, Munich, Germany) equipped with a medium-energy parallel-hole collimator was used. Thirty minutes before injection of the radiotracer, radioiodine uptake into the thyroid gland was blocked by oral administration of 1.150 mg of sodium perchlorate. Thyroid blocking was contin- ued for 3 days using 3 doses of 459 mg of sodium perchlorate per day orally. Patients received 185 MBq of [123IJIMTO iv. In 5 patients with adrenocortical adenoma and in 3 patients with ACC, planar scans of the whole body were performed 5 minutes, 45 minutes, 90 minutes, 4 hours, 6 hours, and 21 to 24 hours postinjection (pi) using a standard technique (energy window setting, 159 keV ± 15%; bed speed, 20 cm/min; matrix, 256 × 1024) (phase I/II study). In the remaining patients, planar im- aging included at least 1 planar scan 4 to 6 hours pi. All patients also underwent SPECT/CT imaging between 4 and 6 hours pi, which was performed on a dual-head large field-of-view gamma camera combined with dual detector CT (Symbia T2; Siemens). SPECT parameters were as follows: 15% window at 159 keV, rotation of 180° per detector head with 3° angular step (=2 × 60 frames) at 30 seconds per projection, 128 × 128 matrix, and slice thickness of 4.8 mm. The low-dose CT scans were acquired using 130 kV and 17 mA (reconstructed slice thickness, 5 mm).
SPECT data were corrected for attenuation and reconstructed iteratively using the OSEM algorithm (6 subsets and 6 iterations; 3-dimensional smoothing, 8 mm). For attenuation correction, CT data were reconstructed using a B08s kernel; for visual in- terpretation and image fusion, the data were also reconstructed using a B30S kernel.
Image analysis and interpretation
For the visual analysis, [123IJIMTO data sets, planar imaging together with SPECT/CT data, were interpreted by 2 experienced nuclear medicine physicians (M.C.K. and A.K.B) together with an endocrinologist (S.H.) in consensus. The readers were not aware of the results of the full-dose CT assessment or clinical data. Uptake in each tumor lesion described on CT received a score between 1 and 3. Lesions with high and homogeneous uptake on planar and SPECT/CT imaging were considered to be of definite adrenocortical origin and were scored with 1. Lesions with inhomogeneous uptake were scored with either 2a or 2b (indeterminate lesion). If large areas exhibited obvious tracer uptake, the lesion was considered to probably be of adrenocor- tical origin and was scored with 2a. If only small areas of focally increased uptake were visible or uptake was faint, the lesion was considered less likely to be of adrenocortical origin and was
scored with 2b. Lesions without any tracer uptake were diag- nosed to be not of adrenocortical origin and were scored with 3.
For semiquantitative assessment of the planar scans, dorsal views of the whole-body scans of all time points were aligned. The most suitable view for the visualization was determined for each lesion. A circular region of interest encompassing the lesion or healthy adrenal gland was assigned. A rectangular region of interest (3 X 3 cm) was assigned to the liver. Mean pixel counts were used for the calculation of ratios.
For semiquantitative analysis of the SPECT/CT data, spher- ical volumes of interest (VOIs) (6; 2 cm) were assigned to the center of the lesion using the CT scan and the healthy adrenal gland. Another VOI was placed to the liver adjacent to the right adrenal gland (box 2 × 3 × 3 cm). Mean voxel counts were calculated for the determination of ratios. Because expression of the target enzymes may vary depending on the endogenous cor- ticosteroid levels, the size and the configuration of the adrenal gland may have a significant influence on the detected tracer uptake due to a “partial volume effect,” we used the liver for normalization instead of the contralateral adrenal gland.
A clinical diagnosis was established by an experienced endo- crinologist (M.F.), who was not aware of the [123IJIMTO im- aging results, on the basis of clinical parameters, laboratory as- sessment, conventional imaging, histopathological analysis, and follow-up data.
Dosimetry
[123I]IMTO is rapidly degraded in vivo, and the radioactive metabolites are excreted mainly by renal elimination. The dis- tribution volume of the metabolites is slightly larger than typi- cally observed for iodide and the total body residence time is somewhat longer. An estimate of the effective dose (ED) was deduced for each patient from the measured residence times in the liver and in the total body. Because excretion is similar to that of iodide, a 1.1-hour residence time as listed for 123I in ICRP53 was assumed for the bladder content (17). A variable small frac- tion of the activity is removed by hepatobiliary excretion but has a negligible effect on the ED. The determination of the ED was performed using the computer program OLINDA (Organ Level INternal Dose Assessment Code) (18). A more detailed descrip- tion of the biokinetics and more elaborate dosimetry will be published in a separate article.
Ethical considerations
All patients gave written informed consent before all study- related procedures. The study was approved by the German Fed- eral Institute for Drugs and Medical Devices (Permit 4031230), the local ethics committee of the University of Würzburg (Permit 100/05), and the German Federal Office for Radiation Protec- tion (Permit Z5-22461/2-2006-024). Similar to the study of Hennings et al (14), metomidate imaging was later considered a routine diagnostic tool in our hospital. Therefore, after comple- tion of the clinical trial, more recent patients underwent [123IJIMTO imaging on a compassionate use basis.
Safety assessment
Acute and chronic toxicities after injection were assessed ac- cording to the Common Toxicity Criteria of the National Cancer Institute (version 4.0; http://ctep.cancer.gov/protocolDevelopment/ electronic_applications/ctc.htm).
Statistical analysis
Results are expressed as means ± SD and/or median (range). All data were analyzed using PASW Statistics 20 (IBM SPSS, IBM Corp, White Plains, New York). Differences between groups were tested by Mann-Whitney U test. P values <. 05 were considered statistically significant. Sensitivity of [123IJIMTO SPECT/CT was calculated by determining the proportion of pa- tients in whom results of visual analysis would correctly predict adrenocortical origin of the tumor lesion. Specificity was calcu- lated by determining the proportion of patients in whom visual analysis would correctly predict nonadrenocortical origin of the tumor lesion. A receiver-operating characteristic (ROC) curve was constructed from the pairs of sensitivity and specificity mea- sured at each tumor to liver ratio derived from the semiquanti- tative assessment.
Results
Study participants
Thirty-three patients participating in the phase I/II study and 18 patients were investigated according to the study protocol on a compassionate use basis after com- pletion of the clinical trial. Altogether 51 patients (25 men and 26 women) with a median age of 55 years (range, 17-88 years) were analyzed. Median tumor diameter was 56 mm (range, 23-217 mm). Histopathological analysis of the tumor lesion was available for 33 patients. In a further 9 patients with known malignancy and metastatic disease, histopathological analysis from other tumor manifesta- tions was available. Clinical diagnosis was ACC (n = 14), metastasis from nonendocrine malignancy (n = 11), pheo- chromocytoma (n = 6), nonfunctioning adrenocortical adenoma (n = 5), cortisol-producing adenoma (n = 4), adrenocortical hyperplasia (n = 4), aldosterone-produc- ing adenoma (n = 3), hematoma (n = 1), myelolipoma (n = 1), and renal cell carcinoma (n = 1). One tumor was rated as most likely being malignant but could not be clas- sified as adrenocortical or nonadrenocortical because hor- monal assessment did not reveal hormonal activity, and histopathological analysis was not available because the patient refused both surgery and biopsy (PID31). This pa- tient was excluded from further analysis. Median tumor size was 61 mm (range, 23-217 mm) (Table 1).
Three of the 14 patients with ACC received mitotane for adrenolytic therapy (PID9, mitotane level 2.9 mg/L; PID13, mitotane level at time of imaging 5.8 mg/L; and PID14, mitotane level 7.4 mg/L). Another patient (PID6) had stopped mitotane 4 weeks before imaging. No mito- tane plasma concentration at the time of imaging was available; the mitotane level at the time of cessation of mitotane treatment was 6.0 mg/L. Three patients had re- ceived cytotoxic chemotherapy with the last cycle >6 weeks before [123IJIMTO imaging (PID6, PID11, and PID13).
| PID | Diagnosis | Age, y | Sex | Hormonal Activity | CT Diameter, mm | Qualitative Assessment of [123I]IMTO Uptake | Semiquantitative Analysis: Tumor to Liver Ratio |
|---|---|---|---|---|---|---|---|
| 1 | ACCª | 55 | F | Androgen | 156 | 1 | 1 |
| 2 | ACCª | 88 | M | NF | 79 | 2b | 0.7 |
| 3 | ACCª | 58 | F | Cortisol, androgen | 130 | 1 | 2.8 |
| 4 | ACCª | 40 | M | Cortisol | 35 | 1 | 5.2 |
| 5 | ACCª | 49 | F | NF | 130 | 2b | 0.6 |
| 6 | ACCª | 64 | F | Cortisol, androgen | 87 | 1 | 15.8 |
| 7 | ACC | 79 | F | Androgen | 129 | 3 | 0.5 |
| 8 | ACCª | 57 | M | NF | 217 | 1 | 0.9 |
| 9 | ACC | 51 | F | NF | 103 | 1 | 2.2 |
| 10 | ACCª | 76 | M | NF | 65 | 3 | 0.8 |
| 11 | ACCª | 23 | M | NF | 49 | 1 | 3.1 |
| 12 | ACCª | 71 | F | Cortisol, androgen | 130 | 2a | 1.4 |
| 13 | ACCª | 41 | F | Cortisol, androgen | 120 | 1 | 1.9 |
| 14 | ACCª | 45 | M | Cortisol, androgen | 141 | 1 | 1.4 |
| 15 | APA | 53 | F | Aldosterone | 26 | 1 | 20.6 |
| 16 | APAª | 46 | M | Aldosterone | 26 | 1 | 6.1 |
| 17 | APAª | 42 | F | Aldosterone | 29 | 1 | 9.5 |
| 18 | CPAª | 44 | M | Cortisol | 38 | 1 | 11 |
| 19 | CPAª | 54 | F | Cortisol | R: 56; L: 30 | R: 1; L: 1 | R: 44.5; L: 9.3 |
| 20 | CPA | 62 | M | Cortisol (subclinical) | R: 31; L: 28 | R: 1; L: 1 | R: 7.7; L: 3.4 |
| 21 | CPAª | 72 | M | Cortisol (subclinical) | 35 | 1 | 4.7 |
| 22 | NFA | 62 | F | NF | 25 | 1 | 3.9 |
| 23 | NFA | 73 | M | NF | 54 | 1 | 21.9 |
| 24 | NFAª | 49 | M | NF | 46 | 1 | 18.2 |
| 25 | NFA | 62 | M | NF | 28 | 1 | 4.4 |
| 26 | NFA | 54 | F | NF | 37 | 1 | 5.1 |
| 27 | Hyperplasia (macronodular)a | 69 | F | Cortisol (subclinical) | R: 61; L: 22 | R: 1; L: 1 | R: 37.5; L: 8.6 |
| 28 | Hyperplasia (macronodular)ª | 42 | F | Cortisol | 23 | R: 1; L: 1 | 9.9 |
| 29 | Hyperplasia (macronodular)ª | 58 | F | Cortisol | R: 52; L: 27 | R: 1; L: 1 | R: 43; L: 6;7 |
| 30 | Hyperplasia (macronodular) | 44 | F | Cortisol | R: 45; L: 33 | R: 1; L: 1 | R: 10.9; L: 7.4 |
| 31 | indeterminate, suspicious for | 67 | M | NF | 58 | 2a | 1.7 |
| malignancy | |||||||
| 32 | Hematomaª | 17 | F | NF | 93 | 3 | 0.1 |
| 33 | Metastasisª | 65 | M | NF | 65 | 2b | 0.8 |
| 34 | Metastasisª | 45 | F | NF | 120 | 3 | 0.3 |
| 35 | Metastasis | 56 | M | NF | 102 | 2b | 0.2 |
| 36 | Metastasis | 51 | F | NF | R: 45; L: 31 | R: 3; L:1 | R: 1.8; L: 3.7 |
| 37 | Metastasisª | 44 | M | NF | 110 | 3 | 0.7 |
| 38 | Metastasis | 69 | M | NF | 50 | 3 | 0.7 |
| 39 | Metastasisª | 67 | F | NF | 102 | 3 | 0.2 |
| 40 | Metastasis | 71 | M | NF | 51 | 3 | 1.1 |
| 41 | Metastasis | 71 | F | NF | 24 | 1 | 1.2 |
| 42 | Metastasisª | 33 | F | NF | R: 72; L: 76 | R: 3; L: 3 | R: 0.3; L: 0.6 |
| 43 | Metastasis | 61 | F | NF | 38 | 3 | 0.8 |
| 44 | Myelolipomaª | 49 | F | NF | 80 | 3 | 0.8 |
| 45 | Pheochromocytomaª | 58 | M | Catecholamines | 81 | 3 | 0.9 |
| 46 | Pheochromocytomaª | 67 | M | Catecholamines | 36 | 1 | 3.3 |
| 47 | Pheochromocytomaª | 48 | M | Catecholamines | 94 | 3 | 0.4 |
| 48 | Malignant pheochromocytoma | 52 | M | Catecholamines | 112 | 2a | 0.6 |
| 49 | Malignant pheochromocytomaª | 51 | M | Catecholamines | 75 | 2b | 1 |
| 50 | Malignant pheochromocytoma | 46 | M | Catecholamines | 55 | 3 | 1.2 |
| 51 | Renal cell carcinomaª | 70 | F | Catecholamines | 75 | 3 | 0.3 |
Abbreviations: APA, aldosterone-producing adenoma; CPA, cortisol-producing adenoma; L, left; NF, nonfunctioning; NFA, nonfunctioning adenoma; R, right.
a Histologically confirmed.
Visual interpretation of [123I]IMTO SPECT/CT and planar images
Both adrenal glands and adrenocortical tumor tissue were first detected within 5 to 360 minutes after injection of [123I]IMTO, with very good delineation of the adrenal glands and the tumor lesions 4 to 6 hours after tracer injection. The best target to background ratios were ob- served at 24 hours postinjection (pi) with specific retention of the tracer exclusively in the adrenocortical tissue (Fig- ures 1-3). According to the scoring system applied, all tumors of the 16 patients with a benign adrenocortical
lesion were graded as 1 (definite adrenocortical origin), 9 tumors of 14 patients with ACC were graded as 1, 1 tumor was graded as 2a (probably adrenocortical), 2 tumors were graded as 2b (probably nonadrenocortical), and 1 tumor was graded as 3 (nonadrenocortical). Of the nona- drenocortical lesions, 3 of 20 lesions were characterized as 1 (1 pheochromocytoma and 2 metastases), 4 tumors were scored as 2a or 2b, and the remaining 13 lesions were correctly graded as 3 (Table 1).
Three lesions with a clinical diagnosis of metastasis (n = 2) or pheochromocytoma (n = 1) had a false-positive
LDR
5 min
45 min
90 min
4 h
6 h
23 h
LDR
·
.
·
·
·
5 min
45 min
90 min
4 h
6 h
diagnosis as adrenocortical lesions according to the qual- itative analysis. Histopathological analysis was only avail- able from the pheochromocytoma, which described a tumor being completely surrounded by normal adreno- cortical tissue. Of the 2 lesions classified as metastases, one exhibited 13 Hounsfield units and demonstrated an in- crease in size at follow-up. The other lesion had 31 Hounsfield units and exhibited high metabolic activity with fludeoxyglucose (FDG)-PET and also showed an in- crease in size at follow-up (Table 2). In addition, 2 ACCs were classified as 2b (probably nonadrenocortical) and 1 malignant pheochromocytoma was classified as 2a (prob- ably adrenocortical origin).
Qualitative analysis resulted in a sensitivity of 89% and a specificity of 85% if the lesions graded with 2 were excluded from analysis and in a sensitivity of 89% and specificity of 80% if the lesions graded with 2a were regarded as adrenocortical and the lesions graded with 2b were re- garded as nonadrenocortical.
After exclusion of the 18 patients without available histopathology of the adrenal lesion, qualitative analysis resulted in a sensitivity of 95% and a specificity of 89% if the lesions graded with 2 were excluded from analysis and in a sensitivity of 88% and specificity of 77% if the lesions graded with 2a were regarded as adrenocortical and the lesions graded with 2b were regarded as nonadrenocortical.
Both renal elimination and hepa- tobiliary excretion with transient tracer accumulation in the gallblad- der and the bowels were noted. Only temporary tracer accumulation in the early phase was observed in the brain, liver, and kidneys.
Semiquantitative assessment of adrenal [123I]IMTO retention based on planar images
22 h Planar whole-body scans with re- peated imaging were available for the 33 patients included in the phase I/II study (13 patients with adreno- cortical adenoma with 16 adrenal le- sions, 15 patients with nonadreno- cortical adenoma with 17 adrenal lesions, and 5 patients with ACC). All 16 adrenocortical adenoma lesions were visible on pla- nar scans as focal tracer retention in projection to the lesion (starting at 5 minutes pi, n = 5; at 45 minutes pi, n = 2; at 90 minutes pi, n = 2; at 4-6 hours pi, n = 4, and at 24 hours pi, n = 1). In 3 of 5 patients with ACC, tumors were detectable on [123IJIMTO planar scans (5 minutes pi, n = 1; and 45 minutes pi, n = 2). In the patients with nonadrenocortical lesions, some tracer accumulation in the region of the adrenal lesion was noted in 11 of 16 lesions (first visible at 90 minutes pi, n = 2; at 4-6 hours pi, n = 8; and at 24 hours pi, n = 1). Tracer accumulation was also visible in 25 of 28 unaffected contralateral ad-
NFA
CPA
APA
ACC
Renal cell carcinoma
Metastasis
Pheochromocytoma
Hematoma
renal glands (starting at 5 minutes pi, n = 1; at 45 minutes pi, n = 3; at 4-6 hours pi, n = 18; and at 24 hours pi, n = 3). Five of the participants had cortisol excess due to a cortisol-producing adenoma (n = 2) or cortisol-producing
100
!
:
.
10
tumor / liver ratio
.
I
.
.
… . I H . .
1
i
.
0.1
ACC
ACA
Metastasis
Pheo
ACC (n = 3). In 3 of these patients tracer uptake into the contralateral adrenal gland was completely diminished.
In general, a continuous increase in local tracer accu- mulation was noted in adrenocortical adenomas, whereas in some patients with ACC relatively high target to back- ground ratios were observed at 5 minutes with lower val- ues at 45 and 90 minutes pi and a continuous increase in ratios at later time points. The highest tumor to back- ground ratio was most frequently seen after 24 hours. Median maximum tumor to background ratios at any time point for tumors/adrenal glands that were visible in planar images were 2.6 (1.5-7.9) in adrenocortical adenomas, 2.5 (1.9-3.2) in ACCs, 1.9 (1.3-3.2) in nonadrenocortical lesions, and 1.9 (1.3-3.8) in unaffected contralateral ad- renal glands.
Semiquantitative assessment of SPECT imaging
Tumor to liver ratios were significantly higher in the group with adrenocortical adenoma/hyperplasia than in the group with nonadrenocortical lesions. For benign le- sions of adrenocortical origin, the median lesion to liver
| PID | Clinical Diagnosis | HU in Unenhanced CT | Hormonal Findings | SUV in FDG-PET (if Available) | Clinical Follow-up Since PD, mo | Imaging Follow-up Since PD, mo | Results of Follow-up Imaging | Histological Diagnoses From Other Lesions |
|---|---|---|---|---|---|---|---|---|
| 7 | ACC | 23 | SC | NA | 1 | 0 | Deceased | Liver metastases of ACC |
| 9 | ACC | 35 | NF | 7.5 | 9 | 5 | Stable | Bone metastasis of ACC |
| 15 | APA | 9 | NF | NA | 43 | 1 | Stable | NA |
| 20 | CPA | R: 1; L: 12 | NF | NA | 32 | 32 | Stable | NA |
| 22 | NFA | 5 | NF | NA | 57 | 23 | Stable | NA |
| 23 | NFA | 15 | NF | 2.1 | 37 | 37 | Stable | NA |
| 25 | NFA | 6 | SC | NA | 12 | 12 | Stable | NA |
| 26 | NFA | 17 | NF | NA | 39 | 8 | Stable | NA |
| 30 | Hyperplasia | R: 5; L: 25 | SC | NA | 8 | 6 | Stable | NA |
| (macronodular) | ||||||||
| 31 | Indeterminate, probably | 28 | NF | 10.4 | 24 | 0 | Deceased | NA |
| malignant | ||||||||
| 35 | Metastasis | 38 | NF | 14 | 1 | 0 | Deceased | NA |
| 36 | Metastasis | R: 38; L: 13 | NF | NA | 9 | 7 | Progressive | Bone, brain, and pancreatic |
| metastases of lung cancer | ||||||||
| 38 | Metastasis | 33 | NF | NA | 0 | 0 | Deceased | Bone metastasis of undifferentiated carcinoma (suspective of renal cell carcinoma) |
| 40 | Metastasis | 30 | NF | 3.5 | 9 | 2 | Regressive | Lymph node metastasis of a |
| poorly differentiated | ||||||||
| adenocarcinoma of | ||||||||
| unknown origin | ||||||||
| 41 | Metastasis | 31 | NF | 5.8 | 56 | 50 | Progressive | Lung metastasis |
| of colon cancer | ||||||||
| 43 | Metastasis | 30 | NF | 15.5 | 29 | 33 | Progressive | Adenocarcinoma |
| of the lung | ||||||||
| 48 | Malignant | 41 | NF | 8.3 | 50 | 50 | Stable | Bone metastasis |
| pheochromocytoma | of pheochromocytoma | |||||||
| 50 | Malignant | 36 | SC | 8.2 | 8 | 8 | Stable | Bone metastasis |
| pheochromocytoma | of pheochromocytoma |
Abbreviations: APA, aldosterone-producing adenoma; CPA, cortisol-producing adenoma; HU, Hounsfield units; L, left; NA, not applicable; NF, nonfunctioning; NFA, nonfunctioning adenoma; PD, primary diagnosis; R, right; SC, subclinical; SUV, standard uptake value.
ratio in the attenuation corrected data set at 4 to 6 hours pi was 8.6 (range, 1.5-44.5), which was significantly higher than that for ACCs (P < . 01) and nonadrenocor- tical lesions (P < . 01). ACC lesions displayed a highly variable uptake with a median tumor to liver ratio of 1.4 (0.5-15.8) but still significantly differed from nonadreno- cortical lesions (P = . 018). Adrenal lesions of nonadre- nocortical origin had a median ratio of 0.75 (range 0.1- 3.7) (Table 1). The normal contralateral adrenal gland was undetectable in 1 patient with ACC and severe cortisol excess. In the remaining patients, the normal contralateral adrenal gland could be visualized with a median normal adrenal to liver ratio of 3.2 (0.8-11.8). No differences between the left and the right adrenal gland were observed. No significant association of tracer uptake with hormonal activity was observed; however, there was a trend to higher tracer uptake in hormone-secreting tumors. The median target to liver ratio in hormone-secreting adreno- cortical tumors (n = 27) was 6.1 (1-45) vs 3.6 (1-22) in nonfunctioning adrenocortical lesions (n = 10) (P = . 29). ROC analysis revealed a sensitivity of 83.3% and a spec- ificity of 86.4% for identification of an adrenocortical lesion based on a cutoff value for the tumor to liver ratio of 1.3 (Figure 4). Setting specificity to 100%, a sensitivity of 61.1% was reached at a cutoff value of the tumor to liver ratio of 3.8.
Safety
Adverse events were observed in 2 patients. One patient developed hot flushes with short-term dyspnea and tachy- cardia after administration of [123I]IMTO. All symptoms
1.0
0.8
sensitivity
0.6
0.4
0.2
0,0
0.0
0.2
0.4
0.6
0.8
1.0
1-specificity
disappeared within 5 minutes without intervention (CT- CAE [Common Terminology Criteria for Adverse Events] grade 1). The symptoms were potentially associated with the addition of a solubilizer (Tween 80). Another patient with metastatic ACC participating in the phase I/II trial, who is not included in this analysis because resection of the primary tumor had been performed before [123I]IMTO imaging also experienced a short flush. Thereafter, Tween 80 was replaced by phosphate-buffered solution (pH 7.4) containing ethanol (20%). No further flushes were ob- served. Another patient had from back pain during imag- ing (CTCAE grade 1), which was potentially related to the need to stay in an unchanged position during the scanning procedure.
During long-term follow-up (>2 years), no side effects attributable to the study were observed. The mean ED per unit activity administered was calculated to be 0.015 mSv/ MBq (range, 0.011-0.020 mSv/MBq) corresponding to 2.7 mSv for a typical investigation with 185 MBq [123IJIMTO.
Discussion
This is the first evaluation of the diagnostic utility of [123I]IMTO for noninvasive characterization of adrenal masses demonstrating high sensitivity and specificity for detecting adrenocortical tumors. Of note, all benign ad- renocortical lesions showed focally increased tracer up- take irrespective of their secretory state, indicating high expression of CYP11B enzymes in adenomas. In contrast, uptake in adrenocortical cancer was more variable with some tumors (21%) showing little or no uptake. Lack of uptake of metomidate tracers in ACC has been described previously for [11C]metomidate and has been associated with extensive tumor necrosis (14, 19). Areas that were in- terpreted as necrotic tissue were seen in a large proportion of the ACCs. The VOIs used for semiquantitative analysis of the SPECT/CT data were standardized and assigned to the center of the lesion. Thus, they may have included necrotic areas, reducing the mean uptake values measured in the ACCs. However, in general, good correlation between visual inter- pretation and semiquantitative analysis was observed. In the 14 patients with ACC, 2 of the tumors characterized as ad- renocortical by visual interpretation would not have been correctly identified by using a cutoff value of 1.3. Conversely, 1 ACChad a tumor to liver ratio of 1.4 and was characterized as probably adrenocortical. Furthermore, reduced or miss- ing CYP11B expression has been reported in some cases of ACC, providing a molecular explanation for a lack of uptake of metomidate tracers (20).
Mitotane has been reported to interfere with [11C]me- tomidate uptake and may also affect [123IJIMTO binding
(19). In our series, the 3 patients with ACC who had re- ceived mitotane had tumor to liver ratios in the lower range for adrenocortical tumors according to semiquan- titative assessment, but the tumors were all classified as adrenocortical lesions by visual interpretation.
Of greater concern is the observation that in 3 neo- plasms clinically diagnosed as nonadrenocortical tumors, positive uptake was noted. One patient had a pheochro- mocytoma. Histopathological analysis revealed that the tumor was surrounded by functional nonhyperplastic nor- mal adrenocortical tissue showing strong staining for CYP11B1/2, leading to significant uptake in the peritu- moral tissue and consecutive misclassification. For 2 [123I]IMTO-positive neoplasias, no histopathological analysis was available, and a metastasis was diagnosed based on the clinical context and evidence of enlargement during follow-up, respectively. However, in 1 of the le- sions, the Hounsfield unit score on unenhanced CT was only 11, indicating that an adrenal adenoma could have been the cause of the adrenal mass rather than a metas- tasis. Furthermore, enlargement of an adrenal lesion has limited specificity for the diagnosis of a metastasis. How- ever, in these patients also, uptake of the surrounding nor- mal tissue may have contributed to the misclassification. This view is supported by the observation that the diam- eters of the false-positive lesions were comparably small (36, 31, and 24 mm, respectively). PET has a higher spatial resolution, and this may explain why in a large series of adrenal tumors [11 C]metomidate exhibited a higher spec- ificity (14), although a head-to-head comparison has not yet been performed. The potentially higher specificity of [11C]metomidate is compensated for by a number of ad- vantages of [123IJIMTO SPECT. [11C]Metomidate imag- ing requires an on-site cyclotron, thereby greatly limiting its availability and the investigation time. In contrast, SPECT imaging is widely available and [123I]iodine has a much longer half-life, greatly facilitating image acquisi- tion. In fact, the best tumor to background ratio was dem- onstrated 24 hours after administration of [123IJIMTO, and the most useful time point for imaging was 4 to 6 hours after injection, a time point at which [11C]metomidate is no longer detectable because of its short half-life. In prin- ciple, use of the PET tracers [18F]fluoroetomidate and [124I]metomidate could combine the advantages of high spatial resolution and a longer half-life. However, no pa- tient data for [124I]metomidate are available yet, and re- sults of studies investigating [18F]fluoroetomidate are lim- ited to a small series of normal subjects (n = 10) and 13 patients with an adrenocortical mass, precluding robust conclusions (21, 22).
In this article, we used both a qualitative and a semi- quantitative approach for defining [123I]IMTO-positive
and -negative lesions. Both assessments revealed similar results, indicating that currently for routine clinical use, the qualitative analysis by an experienced physician is suf- ficient. Whether advances in semiquantitative analysis will alter this approach remains to be seen. Semiquanti- tative evaluation of planar images revealed a relatively high overlap between adrenocortical and nonadrenocor- tical lesions, which is most likely due to tracer uptake in surrounding normal adrenal tissue. Therefore, for better discrimination between tumor tissue and adjacent normal adrenocortical tissue, SPECT/CT imaging is superior. In ACC, bimodal tracer accumulation was seen with high uptake within the first minutes and a decrease thereafter followed by a continuous increase in tracer uptake at later time points. This result may be explained by a perfusion effect.
The radiation exposure with [123IJIMTO (2.7 mSV) compares favorably with the very high exposure with ra- diocholesterol imaging ([1311]6ß-iodomethyl-norcholes- terol [NP-59], 30 mSv; [75Se]6ß-selenomethyl-19-norcho- lesterol [Scintadren], 17 mSv) (23). Furthermore, the duration of the imaging procedure is much shorter and more convenient for the patient. However, although it has been claimed that radiocholesterol may allow differenti- ation between benign and malignant adrenocortical le- sions, this is not possible with metomidate tracers. On the other hand, lack of uptake of radiocholesterol tracers is not restricted to ACC but is also seen in both benign and malignant nonadrenocortical lesions. Furthermore, occa- sional uptake of [1311]6ß-iodomethyl-norcholesterol or [75Se]60-selenomethyl-19-norcholesterol in ACC has been reported (24-28).
For differentiation of benign and malignant adrenal lesions, the use of [1 8F]FDG-PET has been advocated (29- 32). However, some malignant adrenal lesions such as metastases of renal cancer or neuroendocrine tumors may fail to exhibit high [18F]FDG uptake, whereas some ad- renal adenomas may show increased uptake (33-35). The combination of [18F]FDG-PET with metomidate imaging may hold potential for identifying benign adrenocortical adenomas in patients with indeterminate lesions as indi- cated by conventional imaging. This may greatly reduce the need for adrenal surgery, because most hormonally inactive incidentalomas removed by surgeons are adreno- cortical adenomas (36).
Another important aspect of [123IJIMTO imaging is related to the use of [131IJIMTO for radionuclide treat- ment of adrenocortical cancer. [131IJIMTO has been dem- onstrated to have therapeutic potential in metastatic ACC (37). However, a prerequisite for successful treatment is high uptake of IMTO in all lesions and a limited tumor burden. [123IJIMTO imaging is, therefore, a mandatory
prerequisite to select patients who can benefit from such a therapy, although [124IJIMTO-PET may also potentially be suitable for this purpose.
Our article describes the first data on a clinical evalu- ation of [123I]IMTO within a phase I/II trial. The setting in which [123I]IMTO imaging has its unique added value cannot be finally defined by a phase I/II trial. However, our findings suggest that there are 4 main clinical conditions in which [123IJIMTO imaging has significant potential to guide clinical decision making: (1) patients with indeter- minate adrenal masses on conventional imaging (mostly adrenocortical adenomas with low fat content; for exam- ple, [123I]IMTO uptake can help to exclude metastatic disease from a nonadrenocortical malignancy); (2) pa- tients with bilateral adrenal lesions and steroid excess that may result from unilateral hypersecretion; (3) patients with unclear lesions suspicious for metastases or local re- currence in ACC; and (4) patients with ACC being eval- uated for radionuclide treatment with [131I]IMTO.
With use of chemical shift MRI imaging or CT imaging with determination of Hounsfield units in nonenhanced and contrast medium-enhanced images (washout CT), a large proportion of adrenal masses may already be suffi- ciently characterized. However, a significant number of adrenal incidentalomas (approximately 30% of tumors >3 cm) are not reliably characterized by these standard imaging procedures, and for these tumors a diameter of 4 cm is considered to be an indication for surgery, although based on histopathological analysis most of these tumors are benign adenomas not requiring surgical removal. Con- ventional imaging gives morphological information but no information on the functional status of a lesion. In contrast, [123IJIMTO gives information on the functional status of a tissue and helps to identify a lesion as adreno- cortical. Compared with [18F]FDG-PET, which detects high metabolic activity and increased glucose utilization as indicators for malignant tumors, [123IJIMTO is not suit- able for discrimination between benign and malignant le- sions because it binds to adrenocortical enzymes, which are expressed in both benign and malignant adrenocorti- cal tumors. In contrast, [123I]IMTO gives information on whether a lesion is of adrenocortical origin or not. This imaging method may be used as a complementary imaging tool. The combined use of [18F]FDG-PET and [123]]IMTO may be an ideal imaging approach to noninvasively char- acterize indeterminate lesions and to reduce the number of operations that have been performed for “indeterminate” lesions.
A limitation of our study is that the distribution of the different diagnoses may not perfectly reflect the general situation, which may have many more benign nonfunc- tioning adrenocortical adenomas and fewer adrenocorti-
cal carcinomas. The number of different tumor entities included in the study was predetermined in the study pro- tocol. However, some selection bias cannot be completely excluded. Our cohort comprises a relatively large number of patients with ACCs (28% of investigated tumors). ACCs often show inhomogeneous or no uptake despite being of adrenocortical origin, resulting in a relatively high percentage of false-negative results. Of the investigated tumors, 40% were of nonadrenocortical origin, including 3 false-positive lesions. In the current literature, approx- imately 80% of adrenal incidentalomas are nonfunction- ing benign adrenocortical adenomas. However, our aim was to represent different potential tumor entities within the study. Additional studies are needed to further define the clinical setting in which [123I]IMTO is most useful and to verify our current findings. It should, however, be noted that most diagnostic studies on imaging of adrenal lesions have too few ACCs, limiting the conclusions despite being closer to the general distribution of adrenal lesions.
In summary, we have demonstrated high sensitivity and specificity of [123I]IMTO for the characterization of ad- renal masses ≥2 cm in diameter. Compared with other metomidate tracers, [123I]IMTO-SPECT has the potential to become a widely available imaging method.
Acknowledgments
We thank the whole team of the Department of Nuclear Medi- cine and the Endocrine Unit of the Department of Internal Med- icine I of the University Hospital of Würzburg for their excellent cooperation.
Address all correspondence and requests for reprints to: Bruno Allolio, MD, Endocrinology & Diabetes Unit, Depart- ment of Medicine I, University Hospital Würzburg, Oberdürr- bacher Strasse 6, D-97080 Würzburg, Germany. E-mail: allolio_b@medizin.uni-wuerzburg.de.
This work was supported by the Wilhelm Sander Foundation (Grant 2003.175.2) and the IZKF Würzburg (Grant F-124 to S.H. and A.S.). S.H. is an awardee of the Else-Kroener-Fresenius Stiftung (Grant 2010_EKES.29).
Disclosure Summary: The authors have nothing to disclose.
References
1. Barzon L, Sonino N, Fallo F, Palu G, Boscaro M. Prevalence and natural history of adrenal incidentalomas. Eur J Endocrinol. 2003; 149:273-285.
2. Mansmann G, Lau J Balk E, Rothberg M, Miyachi Y, Bornstein SR. The clinically inapparent adrenal mass: update in diagnosis and management. Endocr Rev. 2004;25:309-340.
3. Grumbach MM, Biller BM, Braunstein GD, et al. Management of the clinically inapparent adrenal mass (“incidentaloma”). Ann In- tern Med. 2003;138:424-429.
4. Ozcan Kara P, Kara T, Kara Gedik G, et al. The role of fluorode-
oxyglucose-positron emission tomography/computed tomography in differentiating between benign and malignant adrenal lesions. Nucl Med Commun. 2011;32:106-112.
5. Caoili EM, Korobkin M, Francis IR, et al. Delayed enhanced CT of lipid-poor adrenal adenomas. AJR Am J Roentgenol. 2000;175: 1411-1415.
6. Caoili EM, Korobkin M, Francis IR, et al. Adrenal masses: charac- terization with combined unenhanced and delayed enhanced CT. Radiology. 2002;222:629-633.
7. Kebapci M, Kaya T, Gurbuz E, Adapinar B, Kebapci N, Demirustu C. Differentiation of adrenal adenomas (lipid rich and lipid poor) from nonadenomas by use of washout characteristics on delayed enhanced CT. Abdom Imaging. 2003;28:709-715.
8. Korobkin M, Brodeur FJ, Francis IR, Quint LE, Dunnick NR, Londy F. CT time-attenuation washout curves of adrenal adenomas and nonadenomas. AJR Am J Roentgenol. 1998;170:747-752.
9. Park BK, Kim CK, Kim B, Lee JH. Comparison of delayed enhanced CT and chemical shift MR for evaluating hyperattenuating inciden- tal adrenal masses. Radiology. 2007;243:760-765.
10. Bergstrom M, Bonasera TA, Lu L, et al. In vitro and in vivo primate evaluation of carbon-11-etomidate and carbon-11-metomidate as potential tracers for PET imaging of the adrenal cortex and its tu- mors. J Nucl Med. 1998;39:982-989.
11. Mitterhauser M, Wadsak W, Wabnegger L, et al. In vivo and in vitro evaluation of [18F]FETO with respect to the adrenocortical and GABAergic system in rats. Eur J Nucl Med Mol Imaging. 2003;30: 1398-1401.
12. Minn H, Salonen A, Friberg J, et al. Imaging of adrenal incidenta- lomas with PET using 11C-metomidate and 18F-FDG. J Nucl Med. 2004;45:972-979.
13. Zettinig G, Mitterhauser M, Wadsak W, et al. Positron emission tomography imaging of adrenal masses: 18F-fluorodeoxyglucose and the 11B-hydroxylase tracer 11C-metomidate. Eur J Nucl Med Mol Imaging. 2004;31:1224-1230.
14. Hennings J, Lindhe O, Bergstrom M, Langstrom B, Sundin A, Hell- man P. [11C]metomidate positron emission tomography of adreno- cortical tumors in correlation with histopathological findings. J Clin Endocrinol Metab. 2006;91:1410-1414.
15. Hahner S, Sundin A. Metomidate-based imaging of adrenal masses. Horm Cancer. 2011;2:348-353.
16. Hahner S, Stuermer A, Kreissl M, et al. A [123I]Iodometomidate for molecular imaging of adrenocortical cytochrome P450 family 11B enzymes. J Clin Endocrinol Metab. 2008;93:2358-2365.
17. International Commission on Radiological Protection. Radiation dose to patients from radiopharmaceuticals: IRCP publication 53. Ann. ICRP. 1988;18:1-4.
18. Stabin MG, Sparks RB, Crowe E. OLINDA/EXM: the second-gen- eration personal computer software for internal dose assessment in nuclear medicine. J Nucl Med. 2005;46:1023-1027.
19. Khan TS, Sundin A, Juhlin C, Langstrom B, Bergstrom M, Eriksson B. 11C-metomidate PET imaging of adrenocortical cancer. Eur J Nucl Med Mol Imaging. 2003;30:403-410.
20. Ragazzon B, Assie G, Bertherat J. Transcriptome analysis of adre- nocortical cancers: from molecular classification to the identifica- tion of new treatments. Endocr Relat Cancer. 2011;18:R15.
21. Wadsak W, Mitterhauser M, Rendl G, et al. [18F]FETO for adre- nocortical PET imaging: a pilot study in healthy volunteers. Eur J Nucl Med Mol Imaging. 2006;33:669-672.
22. Rendl G, Wadsak W, Mitterhauser M, et al. Usefulness of the 11ß- hydroxylase inhibitor 18F FETO in positron emission tomography imaging of adrenal masses. In: Nuklearmedizin Abstract Band of the International Symposium of Nuclear Medicine 2006; Bad Gastein, Austria; 2006. No 17.
23. Avram AM, Hahner S. Functional Imaging of Adrenocortical Carcinoma. New York, NY: Springer Science+Business Media; 2011:85.
24. Reschini E, Catania A, Silva A, Cantalamessa L. Uptake of 75Se-
selenomethylcholesterol by a nonfunctioning adrenocortical ade- noma. J Nucl Med Allied Sci. 1984;28:221-224.
25. Barzon L, Zucchetta P, Boscaro M, Marzola MC, Bui F, Fallo F. Scintigraphic patterns of adrenocortical carcinoma: morpho-func- tional correlates. Eur J Endocrinol. 2001;145:743748.
26. Chatal JF, Charbonnel B, Le Mevel BP, Guihard D. Uptake of 131I- 19-iodocholesterol by an adrenal cortical carcinoma and its metas- tases. J Clin Endocrinol Metab. 1976;43:248-251.
27. Shenker IR, Shaw H. Diagnosis and early management of anorexia nervosa and bulimia. Semin Adolesc Med. 1986;2:21-25.
28. Avram AM, Fig LM, Gross MD. Adrenal gland scintigraphy. Semin Nucl Med. 2006;36:212-227.
29. Boland GW, Dwamena BA, Jagtiani Sangwaiya M, et al. Charac- terization of adrenal masses by using FDG PET: a systematic review and meta-analysis of diagnostic test performance. Radiology. 2011; 259:117-126.
30. Kumar R, Xiu Y, Yu JQ, et al. A 18F-FDG PET in evaluation of adrenal lesions in patients with lung cancer. J Nucl Med. 2004;45: 2058-2062.
31. Jana S, Zhang T, Milstein DM, Isasi CR, Blaufox MD. FDG-PET
and CT characterization of adrenal lesions in cancer patients. Eur J Nucl Med Mol Imaging. 2006;33:29-35.
32. Metser U, Miller E, Lerman H, Lievshitz G, Avital S, Even-Sapir E. 18F-FDG PET/CT in the evaluation of adrenal masses. J Nucl Med. 2006;47:32-37.
33. Blake MA, Slattery JM, Kalra MK, et al. Adrenal lesions: charac- terization with fused PET/CT image in patients with proved or sus- pected malignancy-initial experience. Radiology. 2006;238:970- 977.
34. Han SJ, Kim TS, Jeon SW, et al. Analysis of adrenal masses by 18F-FDG positron emission tomography scanning. Int J Clin Pract. 2007;61:802-809.
35. Rao SK, Caride VJ, Ponn R, Giakovis E, Lee SH. F-18 fluorode- oxyglucose positron emission tomography-positive benign adrenal cortical adenoma: imaging features and pathologic correlation. Clin Nucl Med. 2004;29:300-302.
36. O’Neill CJ, Spence A, Logan B, et al. Adrenal incidentalomas: risk of adrenocortical carcinoma and clinical outcomes. J Surg Oncol. 2010;102:450-453.
37. Hahner S, Kreissl MC, Fassnacht M, et al. A [131 I]iodometomidate for targeted radionuclide therapy of advanced adrenocortical car- cinoma. J Clin Endocrinol Metab. 2012;97:914-932.
THE ENDOCRINE SOCIETY®
Renew your Society membership by Dec. 31 to maintain access to your Society member benefits. www.endo-society.org/renew