Predicting Treatment Response in Primary Aldosteronism Using 11C-Metomidate Positron Emission Tomography

Primary aldosteronism (PA) is the most common cause of secondary hypertension and is characterized by the overproduction of aldosterone from the adrenal glands that can lead to hypertension and sometimes hypokalemia. Primary aldosteronism was considered a rare disease during the period after its discovery by Conn and Louis.1 Recently, PA has been emphasized as a public health issue.2 The impact of PA on the health care system can be explained in 3 aspects. First, the prevalence of PA is 10% in hypertensive patients and increases to 30% in specific populations at referral centers.3 Second, patients with PA experience organ damage, significantly increasing the risk of cardiovascular events, diabetes, metabolic syndromes, renal impairment, and worsened quality of life compared with patients with essential hypertension.4–6 Finally, such PA-induced damages are reversible, and PA is curable with the application of appropriate treatment. Adrenalectomy is the treatment of choice for unilateral disease, and mineralocorticoid antagonist (MRA) is the drug of choice for bilateral disease.7 The key point of PA treatment is accurate lateralization.

Currently, adrenal venous sampling (AVS) is the criterion standard for the lateralization of PA as it directly measures the difference in the aldosterone levels between the bilateral adrenal veins.8 The limitations of AVS include the high cost, technical dependence, risk of complication, and lack of protocol consensus.9 More importantly, AVS does not show the value of criterion standard in outcome prediction.10 A randomized controlled trial indicated that PA treatment on the basis of AVS and CT findings was not different in terms of outcomes.11 There was still a 10% risk of biochemical failure in AVS-based management.12 NP-59 adrenal scintigraphy is an alternative functional approach for lateralization. The degree of tracer uptake is proportional to the hyperfunctional status of the adrenal gland. NP-59 adrenal scintigraphy is considered outdated because of its poor sensitivity and inability to detect small lesion.13,14 Although SPECT had high sensitivity and added prognostic value through semiquantification, its clinical application was limited because of a shortage of tracers, high radiation exposure, and complicated protocol.15,16

Metomidate is a potent inhibitor of CYP11B1 and CYP11B2. Metomidate labeled with 11C (11C-metomidate [MTO]) has a high affinity for the adrenal cortex and shows a high contrast (CON) of tracer uptake between the adrenal glands with and without the tumor, including hormone-secreting tumor, adrenocortical carcinoma, and pheochromocytoma.17–20 As MTO is nonselective to 11β-hydroxylase and aldosterone synthase, the optimal pretreatment steroid suppression dose has been studied before MTO PET in PA.21 Thereafter, MTO PET for PA lateralization has been studied to correlate with AVS, histopathology, and postsurgical outcomes, which showed promising results.21–23 In this study, we aimed to use MTO PET as an observation method for patients with PA who were managed according to the guidelines. We also attempted to correlate it with other lateralization modalities and outcome prediction methods.

METHODS Patients

The study protocol was approved by the institutional review board of the National Taiwan University Hospital (201612110MINB), and written consent was obtained from each patient. Patients with PA clinically confirmed via saline loading test or captopril test were prospectively recruited using the following inclusion criteria: (1) willingness to undergo MTO PET, (2) underwent at least 1 lateralization test (only AVS or NP-59 adrenal scintigraphy) in the 3 months before or after MTO PET, and (3) regular follow-up with biochemical and clinical data at least 1 year. The only exclusion criterion was known malignancy with adrenal gland involvement.

MTO PET Protocol

Low-dose dexamethasone (0.5 mg) as pretreatment was administered before the study 4 times daily for 3 days.2111C-metomidate PET was performed using GE Discovery 710 (GE Medical Systems). Non-CON CT of the adrenal glands was performed with 140 kV, 64 mA, and width of 3.75 mm. PET was performed 35 minutes after administering an IV injection of MTO (10 MBq/kg) for 10 minutes. Images were reconstructed in Xeleris 3.1 (GE Medical Systems) with ordered subset expectation maximization of 2 iterations, 30 subsets, and 6-mm Hanning filter; this was stored in a 128 × 128 matrix with pixel size of 4 × 4 mm. Semiquantification was described as SUVmax manipulated by manually setting the region of interest on both adrenal glands. SUVmax on the higher and lower uptake sides of the adrenal gland was denoted as HSUVmax and LSUVmax, respectively. The ratio of HSUVmax to LSUVmax was considered to indicate CON. Images were interpreted by 2 experienced nuclear medicine physicians (C.-C.L., with 16 years of experience; R.-F.Y., with 36 years of experience) who were blinded to the patient's clinical history.

Lateralization and Clinical Management

All patients were free to choose from AVS, NP-59 adrenal scintigraphy, or both as lateralization tests. A radiologist with 15 years of experience performed AVS via sequential sampling without adrenocorticotropic hormone stimulation.24 Successful AVS was defined as a selective index of >2, and unilateral disease was considered when the lateralization index (LI) was >2.0.25

NP-95 adrenal scintigraphy was performed after the administration of oral dexamethasone (8 mg for 8 days) and NP-59 (37 MBq) injected on the fourth day. SPECT of the abdomen was performed 96 hours after NP-59 injection. Patients also received 1 mL of diluted Lugol solution daily to prevent free 131I uptake by the thyroid.15 Semiquantitative parameters, including adrenal-to-liver ratio (ALR; higher uptake side of adrenal gland was referred to as HALR; lower uptake side of adrenal gland was referred to as LALR) and CON (defined as HALR divided by LALR), were calculated. HALR >2.28 or CON >1.10 was considered to indicate unilateral disease as previously described.16 Clinical management was based on the results of AVS or NP-59 adrenal scintigraphy (only if AVS is not feasible) and was in accordance with the guidelines.

Outcome Definition

Postsurgical outcomes were evaluated at 1 year after adrenalectomy based on clinical and biochemical factors as per the Primary Aldosteronism Surgical Outcome (PASO) classification system and categorized into complete success, partial success, and failure.26 We simplified the outcomes into 2 categories: responder (complete success and partial success) and nonresponder (failure), as the patients will benefit for each unit improvement in blood pressure (BP) and biochemical profiles. The biochemical outcomes of medical treatment were based on plasma renin activity (PRA). Increased PRA (≥1 ng/mL per hour) under the action of MRA was considered as responder, whereas persistent suppressed PRA (<1 ng/mL per hour) was considered as nonresponder.27,28 The clinical outcomes under medical control are not well established and may be affected by various factors, such as concurrent primary hypertension and drug compliance; however, we still attempted to apply the PASO classification system to the 2 categories (responder and nonresponder) as this interpretation is relevant and reasonable.

Statistical Analyses

Descriptive statistics were used to describe patient characteristics. Continuous data were expressed as mean ± SD. Independent t test was used for variable comparison, and the Pearson correlation coefficient (r) was used to describe the correlation between variables. Receiver operating characteristic curves were plotted, and areas under the curve were calculated for each variable. The sensitivity and specificity were calculated according to the optimal cutoff value selected by Youden index. P < 0.05 was deemed statistically significant. All statistical analyses were performed using MedCalc Statistical Software version 17.9.2 (MedCalc Software BVBA, Ostend, Belgium).

RESULTS Patient Characteristics

From October 2016 to November 2017, 17 patients were enrolled in this study, and their characteristics are presented in Table 1. There were 10 men (59%) and 7 (41%) women, with an average age of 53 years and an average body mass index (BMI) of 25.2 kg/m2. Of these, 16 patients had a history of hypertension (<2 years, 2 patients; 1–10 years, 8 patients; >10 years, 6 patients). Systolic BP, diastolic BP, PAC, PRA, aldosterone-to-renin ratio, and serum potassium levels were 165 ± 15 mm Hg, 99 ± 15 mm Hg, 53.16 ± 35.50 ng/dL, 0.52 ± 0.31 ng/mL per hour, 189.03 ± 266.33 ng/dL/(ng/mL per hour), and 3.6 ± 0.6 mmol/L, respectively. Imaging results, treatments, and outcomes are shown in Table 2. All patients underwent lateralization testing via AVS (n = 12), NP-59 adrenal scintigraphy (n = 15), or both modalities (n = 10).

TABLE 1 - Patient Characteristics No. Age, y Gender BMI, kg/m2 Hypertension
History Systolic
BP, mm Hg Diastolic
BP, mm Hg PAC, ng/dL PRA, ng/mL per hour ARR Serum
Potassium, mmol/L 1 57 F 19 20 194 109 25.82 0.12 215.17 4 2 48 M 33 19 183 108 89.03 0.26 342.42 2.9 3 56 M 25 10 157 85 100.59 0.53 189.79 4.7 4 52 F 23 1 170 106 148.67 0.13 1143.62 4.2 5 34 M 24 0.1 166 95 53.16 0.95 55.96 3.3 6 64 M 27 11 161 72 56.56 0.44 128.55 4 7 53 M 26 3 151 93 28.91 0.25 115.64 4.4 8 54 F 23 2 175 109 57.65 0.16 360.31 2.7 9 35 M 33 2 165 106 57.04 0.69 82.67 3.6 10 63 M 26 20 170 99 26.4 0.87 30.34 3.5 11 59 M 29 1.5 134 98 23.27 0.13 179.00 3.5 12 46 F 24 0.5 159 96 53.87 0.47 114.62 3.6 13 55 M 22 20 164 93 22.33 0.66 33.83 3.8 14 68 F 24 20 164 99 23.88 0.81 29.48 3.8 15 51 F 22 0 139 72 27.2 0.6 45.33 2.4 16 69 M 21 10 179 140 28.48 1.05 27.12 2.7 17 38 F 27 2 176 100 90.91 0.76 119.62 3.3
TABLE 2 - Findings of CT, MTO PET, and NP-59 Adrenal Scintigraphy, Treatment, and Outcome No. CT Size, cm MTO PET
SUV (L) SUV (R) CON Side NP59
ALR (L) ALR (R) CON Side AVS
LI Side Treatment Clinical
Outcome Biochemical Outcome 1 Left 0.8 14.0 10.6 1.32 L 1.06 1.28 1.21 R 2.28 R Operation (R) 1 1 2 Left 1.6 14.2 6.7 2.12 L 1.18 1.20 1.02 R 188.7 R Operation (R) 1 0 3 Bilateral L 0.7; R 1.5 23.3 24.4 1.05 R NA NA NA NA 1.41 B Medicine 1 0 4 Negative NA 13.1 14.8 1.13 R 1.34 1.42 1.06 R NA NA Medicine 1 1 5 Left 0.7 23.8 20.3 1.17 L 1.26 1.53 1.21 R 4.25 R Operation (R) 1 1 6 Bilateral L 1.2; R 1.5 16.9 23.6 1.40 R NA NA NA NA 8.74 R Medicine 0 1 7 Left 0.9 14.7 15.3 1.04 R 1.06 1.22 1.16 R 1.05 B Medicine 1 1 8 Bilateral L NH; R 0.9 10.6 18.8 1.77 R 1.18 1.79 1.52 R 3.53 R Medicine 1 1 9 Negative NA 9.5 10.8 1.14 R 1.37 1.73 1.26 R 1.63 B Operation (R) 1 1 10 Bilateral L 1.5; R NH 20.6 16.7 1.23 L 1.54 1.39 1.11 L 1.15 B Medicine 0 1 11 Negative NA 11.9 17.6 1.48 R 1.08 1.07 1.01 L 1.58 B Medicine 0 1 12 Left 1.2 32.7 29.7 1.10 L 1.43 1.54 1.08 R 1.64 B Medicine 0 0 13 Negative NA 5.5 9.6 1.75 R 1.54 1.58 1.03 R NA NA Medicine 1 1 14 Left NH 20.1 14.2 1.42 L 1.47 1.15 1.28 L NA NA Medicine 0 0 15 Left hyperplasia 11.2 12.0 1.07 R 1.55 1.69 1.09 R 3.24 R Medicine 1 1 16 Bilateral L NH; R 1.8 26.6 24.0 1.11 L 1.36 2.29 1.68 R NA NA Operation (R) 1 1 17 Left 1.0 22.2 19.1 1.16 L 1.90 2.05 1.08 R NA NA Operation (L) 1 1

F, female; M, male; NA, not available; NH, nodular hyperplasia; NP-59, NP-59 adrenal scintigraphy.


Correlation of MTO PET Between the Clinical Profile and CT

The average HSUVmax, LSUVmax, and CON were 18.9 ± 6.2, 15.2 ± 6.5, and 1.32 ± 0.31, respectively. These 3 PET parameters were compared with 11 clinical profiles, including age, sex, BMI, hypertension history, family history of hypertension, systolic BP, diastolic BP, PAC, PRA, ARR, and serum potassium levels. Only longer hypertension history was associated with higher CON (r = 0.51). Patients were divided into 3 groups based on their adrenal CT findings: bilateral lesions (group 1, n = 5), unilateral lesions (group 2, n = 8), and negative finding (group 3, n = 4) (Table 3). The HSUVmax and LSUVmax of group 1 were significantly higher than those of group 3 (P = 0.0038 and 0.034, respectively). There was no significant difference in terms of CON among the 3 groups; however, a trend of increasing CON was noted in the order of groups 1, 2, and 3 (1.31, 1.32, and 1.38, respectively). Both HSUVmax and LSUVmax were not significantly different between groups 1 and 2 and groups 2 and 3 (Fig. 1).

TABLE 3 - MTO PET Parameters Among Different CT Findings HSUVmax LSUVmax CON Bilateral 22.8 ± 3.1* 18.3 ± 5.5† 1.31 ± 0.29 Unilateral 19.3 ± 6.9 15.9 ± 7.2 1.32 ± 0.37 Negative 13.2 ± 3.7* 10.0 ± 3.4† 1.38 ± 0.30

*P = 0.0038.

†P = 0.0034.


F1FIGURE 1:

11C-metomidate PET parameters among the different CT findings. *Statistically significant.

Correlation of MTO PET Among Functional Studies

Twelve patients underwent AVS and MTO PET. Concordant results between AVS and MTO PET were defined as follows: (1) LI of AVS >2, with the same side of higher aldosterone-to-cortisol side by AVS and higher side of MTO uptake or (2) LI of AVS <2 with CON <1.15 (decided by authors' consensus, in which 15% higher function on the contralateral side is sufficient to consider it as unilateral disease). Four patients presented with concordant results (33%) between AVS and MTO PET (patients 3, 6, 7, and 8). The HSUVmax and LSUVmax of these patients were both lower than the patients with discordant results without statistical significance (P = 0.16 and P = 0.22, respectively). There was no difference in CON of patients with concordant results and those with discordant results.

Of the 15 patients who underwent NP-59 adrenal scintigraphy and MTO PET, 9 (60%) showed concordant results, defined as having the same side of higher uptake of NP-59 and MTO. No correlation was observed between the uptake of NP-59 and MTO, including SUVmax of the both adrenal glands and CON (CON was defined as the uptake value of higher side divided by the lower side; r = 0.24, r = 0.40, and r = 0.30, respectively). Both HSUVmax and LSUVmax were higher in the discordant group than those in the concordant group (P = 0.10 and P = 0.21, respectively) without statistical significance. There was no difference in CON in the 2 groups.

Outcome Prediction of MTO PET

Six patients underwent unilateral adrenalectomy, and all surgical specimens were immunohistochemically analyzed. Three of them were based on AVS (patients 1, 2, and 5), and one of them was based on NP-59 adrenal scintigraphy (patient 16). Patient 9 showed bilateral disease on AVS but poor response to medical treatment; therefore, adrenalectomy was performed based on NP-59 adrenal scintigraphy. Patient 17 refused AVS and was found to have bilateral disease on NP-59 adrenal scintigraphy; however, she had severe refractory hypertension under medical control. Adrenalectomy was performed based on CT findings because of the patient's young age.29 The decision of adrenalectomy side was against MTO PET results in 4 patients (patients 1, 2, 9, and 16). However, all patients showed biochemical and clinical improvement, except patient 2. He underwent right adrenalectomy based on AVS findings with partially successful clinical outcome and failed biochemical outcome. 11C-metomidate PET showed prominent left adrenal uptake (SUVmax = 14.2, CON = 2.12; Figure 2). 11C-metomidate PET results were discordant with AVS in 3 patients (75%, patients 1, 2, and 5) and with NP-59 adrenal scintigraphy in 5 patients (83%, patients 1, 2, 5, 16, and 17).

F2FIGURE 2:

Patient 2 underwent right adrenalectomy based on AVS (LI, 189; right) with absent biochemical outcome. 11C-metomidate PET revealed intense uptake at the left adrenal gland (left adrenal gland: A–C, red arrow, SUVmax = 14.2; right adrenal gland: D–F, yellow arrow, SUVmax = 6.7; CON = 2.12).

Eleven patients were treated with MRA and other antihypertensive agents based on their clinical response. Biochemical responders (n = 8) had significantly lower HSUVmax and LSUVmax than nonresponders (HSUVmax, 16.5 ± 4.6 vs 25.7 ± 6.4 [P = 0.0244]; LSUVmax, 12.6 ± 3.7 vs 22.4 ± 7.8 [P = 0.0163]). There was no significant difference in the CON of the biochemical responders and nonresponders. There were 5 clinical nonresponders, 3 of whom were administered 1 antihypertensive agent in addition to the maximal dose of spironolactone. The remaining 2 nonresponders were unable to tolerate higher doses of spironolactone; therefore, their medication was adjusted by administering 2 antihypertensive agents of other class along with the tolerable dose of spironolactone. Clinical responders (n = 6) had borderline lower HSUVmax than nonresponders (15.8 ± 5.2 vs 22.9 ± 5.9 [P = 0.0628]). There was no significant difference in the LSUVmax and CON of clinical responders and nonresponders. Comparison of semiquantification PET parameters between responders and nonresponders to medical treatment is shown in Table 4. Using an HSUVmax cutoff of 18.8, the sensitivity and specificity of distinguishing between biochemical responders and nonresponders were 75% and 100%, respectively. Using an LSUVmax cutoff of 16.9, the sensitivity and specificity of distinguishing between responders and nonresponders were 100% and 66%, respectively. Using an HSUVmax cutoff of 15.3, the sensitivity and specificity of distinguishing between clinical responders and nonresponders were 67% and 100%, respectively.

TABLE 4 - Comparison of Semiquantification PET Parameters Between Responders and Nonresponders to MR Antagonist

留言 (0)

沒有登入
gif