https://orcid.org/0000-0002-2957-6148
Abstract
Primary aldosteronism (PA) is the most common cause of endocrine hypertension (HT). HT remission (defined as blood pressure <140/90 mm Hg without antihypertensive drugs) has been reported in approximately 50% of patients with unilateral PA after adrenalectomy. HT duration and severity are predictors of blood pressure response, but the prognostic role of somatic KCNJ5 mutations is unclear.
To determine clinical and molecular features associated with HT remission after adrenalectomy in patients with unilateral PA.
We retrospectively evaluated 100 patients with PA (60 women; median age at diagnosis 48 years with a median follow-up of 26 months). Anatomopathological analysis revealed 90 aldosterone-producing adenomas, 1 carcinoma, and 9 unilateral adrenal hyperplasias. All patients had biochemical cure after unilateral adrenalectomy. KCNJ5 gene was sequenced in 76 cases.
KCNJ5 mutations were identified in 33 of 76 (43.4%) tumors: p.Gly151Arg (n = 17), p.Leu168Arg (n = 15), and p.Glu145Gln (n = 1). HT remission was reported in 37 of 100 (37%) patients. Among patients with HT remission, 73% were women (P = 0.04), 48.6% used more than three antihypertensive medications (P = 0.0001), and 64.9% had HT duration <10 years (P = 0.0015) compared with those without HT remission. Somatic KCNJ5 mutations were associated with female sex (P = 0.004), larger nodules (P = 0.001), and HT remission (P = 0.0001). In multivariate analysis, only a somatic KCNJ5 mutation was an independent predictor of HT remission after adrenalectomy (P = 0.004).
The presence of a KCNJ5 somatic mutation is an independent predictor of HT remission after unilateral adrenalectomy in patients with unilateral PA.
Primary aldosteronism (PA) is the most common cause of endocrine hypertension (HT), with an estimated prevalence of 10% in referred populations and 15% to 20% in patients with resistant HT (1). Because of the adverse cardiovascular effects of excess aldosterone that are independent of high blood pressure (BP) levels, patients with PA have higher cardiovascular morbidity and mortality than age- and sex-matched patients with essential HT and the same degree of BP elevation (2–4). The frequency of PA subtypes and hypokalemia in different cohorts depends on whether PA is routinely screened among hypertensive patients and if adrenal venous sampling (AVS) is available in the specialized center (5). In general, aldosterone-producing adenomas (APAs) account for ∼40% and idiopathic hyperaldosteronism (bilateral hyperplasia) for ∼60% of PA cases. Hypokalemia is more frequent in APAs (50%) than in idiopathic hyperaldosteronism (20%) (5). The most common cause of unilateral PA is APAs. Unilateral primary hyperplasia is considered a rare cause of PA but is probably underestimated.
Surgical treatment of unilateral PA should resolve aldosterone excess in all patients. Normalization of serum potassium levels occurs in all patients (1). Although almost all patients with PA improve HT control after surgery, complete clinical success or HT remission (defined as BP <140/90 mm Hg without antihypertensive drugs) has been reported in about 50% (range, 35% to 80%) of patients with unilateral PA after adrenalectomy (1, 6–8). In 2001, Sawaka et al. (9) evaluated factors associated with resolution of HT after adrenalectomy in 93 patients with PA. Complete clinical success (33% of the cases) was associated with having one or no first-degree relative with HT and preoperative use of two or fewer antihypertensive drugs in multivariate analysis (9). Shorter HT duration, higher preoperative ratio of aldosterone to plasma renin activity (A/PRA) and higher urine aldosterone levels were associated with resolution of HT only in univariate analysis or when the cutoff for BP resolution was 160/95 mm Hg (8–11).
Over the past few years, somatic mutations in KCNJ5, CACNA1D, ATP1A1, and ATP2B3 genes, which are involved in maintaining intracellular ionic homeostasis and cell membrane potential, have been described in APAs (12–15). Recurrent KCNJ5 mutations have been found in around 40% and 79% of APAs in European and Japanese populations, respectively (16, 17). In addition, somatic activating mutations in exon 3 of the CTNNB1 gene were initially identified in three APAs diagnosed in women with PA, two who presented while pregnant and one after menopause (18). Despite these advances in the molecular pathogenesis of APAs, few reports evaluated the prognostic role of genetic findings in the clinical outcome after adrenalectomy (19–21). In 2013, Arnesen et al. (19) showed that KCNJ5 somatic mutations are associated with a better surgical outcome in 28 APAs. In an Australian cohort, HT cure and age <50 years at PA diagnosis were associated with a higher chance of harboring a KCNJ5 somatic mutation (20). However, the biochemical criteria to assess PA cure after surgery was not stated in these previous reports (19–21).
Heterogeneity in HT remission (ranging from 35% to 80%) might reflect the absence of standardized criteria to classify outcomes of adrenalectomy for unilateral PA (1, 22). Recently, the Primary Aldosteronism Surgical Outcome (PASO) study established criteria for clinical and biochemical success in patients with unilateral PA after adrenalectomy (22). In our study, we utilized PASO criteria to evaluate clinical and molecular features associated with HT remission in a large cohort of patients with unilateral PA with complete biochemical success after unilateral adrenalectomy.
Patients and Methods
The study was approved by the Ethics Committees of the Hospital das Clínicas, University of São Paulo and informed written consent was obtained from all patients. One hundred forty-five consecutive patients from our institution who underwent unilateral adrenalectomy for unilateral PA were retrospectively evaluated from 1993 to 2018 (Fig. 1). Clinical, biochemical, and imaging data were collected from patient records. The algorithm for PA investigation was detailed described by Vilela and Almeida (23) and follows the 2016 Endocrine Society Guideline for PA management (1). A positive screening for PA was defined as an A/PRA ≥30 or aldosterone to renin (A/R) ratio ≥2 with aldosterone levels ≥10 ng/dL. Seventy-one patients (55%) underwent at least one confirmatory test (saline infusion test, captopril challenge test, or furosemide upright test). Confirmatory testing was not performed in patients with PA with spontaneous hypokalemia, suppressed renin levels plus plasma aldosterone concentration >20 ng/dL (1). All patients with PA not eligible for confirmatory testing according the 2016 Endocrine Society Guideline presented an A/PRA ratio >100 and also fulfilled the criteria proposed by Namba et al. (24) to not undergo confirmatory testing. AVS was performed successfully in 33 patients. Abdominal CT was performed in all patients. CT accuracy in predicting surgical cure of PA is ∼60% (25). Because 92.4% of patients with PA who underwent unilateral adrenalectomy in our cohort had unilateral lesions with median size of 1.8 cm and only 7.6% of the cases had bilateral lesions with AVS lateralization, our data did not reflect the truly overall accuracy of CT on PA diagnosis.
Flowchart for study cohort: 100 patients with unilateral PA met the inclusion criteria and were enrolled in this analysis.
Inclusion criteria were at least six months of follow-up and complete biochemical success after adrenalectomy defined according to PASO criteria: correction of hypokalemia and normalization of the A/R ratio; in patients with a raised A/R ratio postsurgery, aldosterone secretion should be suppressed in a confirmatory test (22). If baseline plasma aldosterone was <10 ng/dL, then the confirmatory tests were unnecessary even if the A/R ratio was increased. After evaluation of inclusion criteria, 100 patients were enrolled in this study. Anatomopathological analysis was reviewed by a single experienced pathologist (M.C.N.Z.) and found 90 APAs,1 carcinoma, and 9 unilateral adrenal hyperplasias.
Complete clinical success or HT remission was defined as a BP <140/90 mm Hg without antihypertensive drugs after six months of follow-up. The cutoff of BP ≥140/90 mm Hg was used to define stage 1 HT in both European and Brazilian guidelines for the management of arterial HT (26, 27). Aldosterone and direct renin concentrations were measured using the LIAISON® kit (DiaSorin, Salugia, Italy). A radioimmunoassay was used to determine plasma renin activity until 2011.
Molecular analysis
After surgical resection, representative areas of tumor or hyperplastic tissue were macrodissected by a pathologist. Tumor fragments were immediately frozen in liquid nitrogen and stored at −80°C until total RNA and DNA extraction using the AllPrep DNA/RNA Mini Kit (Qiagen, Courtaboeuf Cedex, France). Before tumor RNA and DNA extraction, tumor fragment was cut in a cryostat and slides stained by hematoxylin and eosin to confirm the representativeness of tumor or hyperplastic tissue. cDNA was generated from 1 μg of total RNA using the commercial kit Superscript III First Strand S (Invitrogen, Carlsbad, CA). Genomic DNA was extracted using standard procedures. PCR products were sequenced in an automated ABI Prism 3700 sequencer (Applied Biosystems, Carlsbad, CA).
Tumor and hyperplastic tissues (n = 76) were screened for somatic mutations in the hot spot regions of KCNJ5, ATP1A1, ATP2B3, and CTNNB1. Exon 3 of the CTNNB1 gene was sequenced in tumor DNA as previously described (28). KCNJ5, ATP1A1, and ATP2B3 cDNA were amplified using intron-spanning primers, as reported previously (16). Germline KCNJ5, ATP1A1, ATP2B3, and CTNNB1 mutations were investigated in all patients with somatic mutations, as previously described (28, 29).
Statistical analysis
Statistical analysis was performed using IBM SPSS Software (version 25.0; SPSS Inc., Chicago, IL). Continuous data are expressed as median (range). Differences in expression levels between two groups were analyzed by means of the two-tailed Mann-Whitney U test. The χ2 test was used to compare dichotomous variables. Multivariate logistic analysis was performed with initial covariates chosen from those with P < 0.2 in the univariate analysis. Predictive factors of HT resolution were evaluated according to a stepwise multivariate logistic regression analysis, which was used to estimate hazard ratios and their 95% CI in multivariate analysis. Final model specification was tested with link test and goodness-of-fit was assessed with Hosmer and Lemeshow test. P < 0.05 was considered significant.
Results
Unilateral PA was diagnosed in 60 women and 40 men with a median follow-up of 26 months (range, 6 to 252). Median age at diagnosis was 48 years (range, 20 to 74). Hypokalemia was identified in 80% of the patients at the diagnosis. Thirty-seven of 100 cases (37%) had HT remission after adrenalectomy. Patients who presented HT remission after surgery more often were women (73% vs. 47.6%; χ2= 4.1, P = 0.042), used fewer than three antihypertensive medications (48.6 vs. 15.9%; χ2= 12.42, P = 0.0001) and had <10 years of HT duration before PA diagnosis (64.9 vs. 39.7%; χ2= 5.91, P = 0.015) when compared with those with residual HT (Table 1). Familial history of HT was not different between patients with residual HT or remission.
Clinical and Biochemical Characteristics of Patients With Unilateral PA According to HT Remission After Adrenalectomy
| Hypertension Remission (n = 37) | No Hypertension Remission (n = 63) | P | |
|---|---|---|---|
| Female sex | 73% | 47.6% | 0.042 |
| Age at diagnosis, ya | 47 (25–68) | 50 (20–74) | 0.32 |
| Familial history of HT | 45.9% | 46% | 0.99 |
| <3 medications for HT treatment | 48.6% | 15.9% | 0.0001 |
| Hypokalemia | 86.4% | 81% | 0.49 |
| BMI <25 kg/m2 | 58.3% | 39.7% | 0.073 |
| HT duration <10 y | 64.9% | 39.7% | 0.015 |
| Aldosterone, ng/dLa | 34.8 (11.7–195) | 30 (12.1–127) | 0.37 |
| A/PRA ratioa | 198 (37.7–1466) | 162 (34–977) | 0.64 |
| A/R ratioa | 17.2 (3.1–121.8) | 17 (2.9–79.4) | 0.88 |
| Nodule size, cma | 2.0 (0.7–5) | 1.6 (0.4–9.1) | 0.25 |
| Follow-up, ma | 24 (6–201) | 36 (6–252) | 0.09 |
| Hypertension Remission (n = 37) | No Hypertension Remission (n = 63) | P | |
|---|---|---|---|
| Female sex | 73% | 47.6% | 0.042 |
| Age at diagnosis, ya | 47 (25–68) | 50 (20–74) | 0.32 |
| Familial history of HT | 45.9% | 46% | 0.99 |
| <3 medications for HT treatment | 48.6% | 15.9% | 0.0001 |
| Hypokalemia | 86.4% | 81% | 0.49 |
| BMI <25 kg/m2 | 58.3% | 39.7% | 0.073 |
| HT duration <10 y | 64.9% | 39.7% | 0.015 |
| Aldosterone, ng/dLa | 34.8 (11.7–195) | 30 (12.1–127) | 0.37 |
| A/PRA ratioa | 198 (37.7–1466) | 162 (34–977) | 0.64 |
| A/R ratioa | 17.2 (3.1–121.8) | 17 (2.9–79.4) | 0.88 |
| Nodule size, cma | 2.0 (0.7–5) | 1.6 (0.4–9.1) | 0.25 |
| Follow-up, ma | 24 (6–201) | 36 (6–252) | 0.09 |
Median (range).
Clinical and Biochemical Characteristics of Patients With Unilateral PA According to HT Remission After Adrenalectomy
| Hypertension Remission (n = 37) | No Hypertension Remission (n = 63) | P | |
|---|---|---|---|
| Female sex | 73% | 47.6% | 0.042 |
| Age at diagnosis, ya | 47 (25–68) | 50 (20–74) | 0.32 |
| Familial history of HT | 45.9% | 46% | 0.99 |
| <3 medications for HT treatment | 48.6% | 15.9% | 0.0001 |
| Hypokalemia | 86.4% | 81% | 0.49 |
| BMI <25 kg/m2 | 58.3% | 39.7% | 0.073 |
| HT duration <10 y | 64.9% | 39.7% | 0.015 |
| Aldosterone, ng/dLa | 34.8 (11.7–195) | 30 (12.1–127) | 0.37 |
| A/PRA ratioa | 198 (37.7–1466) | 162 (34–977) | 0.64 |
| A/R ratioa | 17.2 (3.1–121.8) | 17 (2.9–79.4) | 0.88 |
| Nodule size, cma | 2.0 (0.7–5) | 1.6 (0.4–9.1) | 0.25 |
| Follow-up, ma | 24 (6–201) | 36 (6–252) | 0.09 |
| Hypertension Remission (n = 37) | No Hypertension Remission (n = 63) | P | |
|---|---|---|---|
| Female sex | 73% | 47.6% | 0.042 |
| Age at diagnosis, ya | 47 (25–68) | 50 (20–74) | 0.32 |
| Familial history of HT | 45.9% | 46% | 0.99 |
| <3 medications for HT treatment | 48.6% | 15.9% | 0.0001 |
| Hypokalemia | 86.4% | 81% | 0.49 |
| BMI <25 kg/m2 | 58.3% | 39.7% | 0.073 |
| HT duration <10 y | 64.9% | 39.7% | 0.015 |
| Aldosterone, ng/dLa | 34.8 (11.7–195) | 30 (12.1–127) | 0.37 |
| A/PRA ratioa | 198 (37.7–1466) | 162 (34–977) | 0.64 |
| A/R ratioa | 17.2 (3.1–121.8) | 17 (2.9–79.4) | 0.88 |
| Nodule size, cma | 2.0 (0.7–5) | 1.6 (0.4–9.1) | 0.25 |
| Follow-up, ma | 24 (6–201) | 36 (6–252) | 0.09 |
Median (range).
Nodule size was not different in patients with and without HT resolution [2.0 cm (0.7 to 5) vs. 1.6 cm (0.4 to 9.1), respectively; P = 0.3]. Regarding biochemical data, frequency of hypokalemia, aldosterone levels, and A/R ratio were not statistically significant between patients with and without complete clinical success. Follow-up duration was similar in both groups [24 m (6 to 201) vs. 36 m (6 to 252), respectively] (Table 1).
KCNJ5 somatic mutations were identified in 33 of 76 (43.4%) cases: p.Gly151Arg (n = 17), p.Leu168Arg (n = 15), and p.Glu145Gln (n = 1) (Fig. 2). All KCNJ5 somatic mutations were identified in APAs. The p.Leu104Arg ATP1A1 somatic mutation was detected in two APAs (2.6%) and the p.Leu425_Val426del ATP2B3 somatic mutation in one APA (1.3%). Additionally, the p.Ser45Pro CTNNB1 somatic mutation was identified in two APAs (2.6%). None of these patients harbored germline mutations in the genes with somatic events.
Electropherograms of KCNJ5 somatic mutations in APAs. KCNJ5 somatic mutations were identified in 33 of 76 (43.4%) cases: p.Gly151Arg (n = 17), p.Leu168Arg (n = 15), and p.Glu145Gln (n = 1).
APAs harboring somatic KCNJ5 mutations were diagnosed more often in females (72.7% vs. 39.5%; χ2 = 8.28, P = 0.004) compared with KCNJ5 wild type (WT) APAs (Table 2). The tumor size was larger in KCNJ5 mutant APAs [2.0 cm (0.7 to 4.0) vs. 1.3 cm (0.7 to 9.1); P = 0.001). Interestingly, the postoperative HT cure was observed in 57.6% of patients with APAs harboring a somatic KCNJ5 mutation (χ2 = 14.14, P = 0.0001), whereas only 16.2% of patients with KCNJ5 WT APAs had complete clinical success. Among the 33 patients with PA who underwent AVS before surgery, a somatic KCNJ5 mutation was also associated with HT remission (67% vs. 23%; χ2= 4.55, P = 0.03). The percentage of patients with HT duration <10 years before PA diagnosis, body mass index (BMI) <25 kg/m2 or using fewer than three antihypertensive medications was not statistically different between both groups. Preoperative aldosterone levels, A/R ratio and frequency of hypokalemia were not affected by the presence of KCNJ5 somatic mutation (Table 2).
Clinical and Biochemical Characteristics of Patients With Unilateral PA According to KCNJ5 Mutational Status
| Mutant KCNJ5 (n= 33) | WT KCNJ5 (n= 43) | P | |
|---|---|---|---|
| Female sex | 72.7% | 39.5% | 0.004 |
| Age at diagnosis, ya | 44 (20 – 68) | 52 (33–74) | 0.059 |
| Familial history of HT | 60.6% | 48.8% | 0.31 |
| <3 medications for HT treatment | 27.3% | 18.6% | 0.37 |
| Hypokalemia | 84.4% | 75% | 0.33 |
| BMI <25 kg/m2 | 54.5% | 35.7% | 0.1 |
| HT duration <10 y | 57.6% | 39.5% | 0.11 |
| Aldosterone, ng/dLa | 31.9 (15.2–195) | 27.5 (11.7–84.2) | 0.16 |
| A/PRA ratioa | 209 (66.4–1466) | 139 (34.5–633) | 0.06 |
| A/R ratioa | 17.9 (5.5–121.8) | 12.4 (2.9–52.6) | 0.12 |
| Nodule size, cma | 2.0 (0.7–4.0) | 1.3 (0.7–9.1) | 0.001 |
| HT remission | 57.6% | 16.2% | 0.0001 |
| Follow-up, mo | 25 (6–76) | 24 (6–75) | 0.77 |
| Mutant KCNJ5 (n= 33) | WT KCNJ5 (n= 43) | P | |
|---|---|---|---|
| Female sex | 72.7% | 39.5% | 0.004 |
| Age at diagnosis, ya | 44 (20 – 68) | 52 (33–74) | 0.059 |
| Familial history of HT | 60.6% | 48.8% | 0.31 |
| <3 medications for HT treatment | 27.3% | 18.6% | 0.37 |
| Hypokalemia | 84.4% | 75% | 0.33 |
| BMI <25 kg/m2 | 54.5% | 35.7% | 0.1 |
| HT duration <10 y | 57.6% | 39.5% | 0.11 |
| Aldosterone, ng/dLa | 31.9 (15.2–195) | 27.5 (11.7–84.2) | 0.16 |
| A/PRA ratioa | 209 (66.4–1466) | 139 (34.5–633) | 0.06 |
| A/R ratioa | 17.9 (5.5–121.8) | 12.4 (2.9–52.6) | 0.12 |
| Nodule size, cma | 2.0 (0.7–4.0) | 1.3 (0.7–9.1) | 0.001 |
| HT remission | 57.6% | 16.2% | 0.0001 |
| Follow-up, mo | 25 (6–76) | 24 (6–75) | 0.77 |
Median (range).
Clinical and Biochemical Characteristics of Patients With Unilateral PA According to KCNJ5 Mutational Status
| Mutant KCNJ5 (n= 33) | WT KCNJ5 (n= 43) | P | |
|---|---|---|---|
| Female sex | 72.7% | 39.5% | 0.004 |
| Age at diagnosis, ya | 44 (20 – 68) | 52 (33–74) | 0.059 |
| Familial history of HT | 60.6% | 48.8% | 0.31 |
| <3 medications for HT treatment | 27.3% | 18.6% | 0.37 |
| Hypokalemia | 84.4% | 75% | 0.33 |
| BMI <25 kg/m2 | 54.5% | 35.7% | 0.1 |
| HT duration <10 y | 57.6% | 39.5% | 0.11 |
| Aldosterone, ng/dLa | 31.9 (15.2–195) | 27.5 (11.7–84.2) | 0.16 |
| A/PRA ratioa | 209 (66.4–1466) | 139 (34.5–633) | 0.06 |
| A/R ratioa | 17.9 (5.5–121.8) | 12.4 (2.9–52.6) | 0.12 |
| Nodule size, cma | 2.0 (0.7–4.0) | 1.3 (0.7–9.1) | 0.001 |
| HT remission | 57.6% | 16.2% | 0.0001 |
| Follow-up, mo | 25 (6–76) | 24 (6–75) | 0.77 |
| Mutant KCNJ5 (n= 33) | WT KCNJ5 (n= 43) | P | |
|---|---|---|---|
| Female sex | 72.7% | 39.5% | 0.004 |
| Age at diagnosis, ya | 44 (20 – 68) | 52 (33–74) | 0.059 |
| Familial history of HT | 60.6% | 48.8% | 0.31 |
| <3 medications for HT treatment | 27.3% | 18.6% | 0.37 |
| Hypokalemia | 84.4% | 75% | 0.33 |
| BMI <25 kg/m2 | 54.5% | 35.7% | 0.1 |
| HT duration <10 y | 57.6% | 39.5% | 0.11 |
| Aldosterone, ng/dLa | 31.9 (15.2–195) | 27.5 (11.7–84.2) | 0.16 |
| A/PRA ratioa | 209 (66.4–1466) | 139 (34.5–633) | 0.06 |
| A/R ratioa | 17.9 (5.5–121.8) | 12.4 (2.9–52.6) | 0.12 |
| Nodule size, cma | 2.0 (0.7–4.0) | 1.3 (0.7–9.1) | 0.001 |
| HT remission | 57.6% | 16.2% | 0.0001 |
| Follow-up, mo | 25 (6–76) | 24 (6–75) | 0.77 |
Median (range).
Cortisol secretion was investigated in 15 patients with PA. Ten of 15 cases had a normal cortisol after an overnight 1 mg low-dose dexamethasone suppression test (LDDST). In 5 patients with PA, cortisol after LDDST ranged from 1.9 to 8.3 μg/dL (only 1 case >5 μg/dL). ACTH, urinary, and midnight salivary cortisol were normal in all cases except in the patient with aldosterone-producing carcinoma, who presented cortisol after LDDST of 8.3 μg/dL and elevated midnight salivary cortisol. This patient required hydrocortisone replacement for six months after adrenalectomy. Among these 15 cases, cortisol after LDDST did not correlate with HT remission and KCNJ5 mutational status.
According to a stepwise multivariate logistic regression analysis, only the presence of a somatic KCNJ5 mutation was an independent predictor of HT remission after adrenalectomy (relative risk 6.48, 95% CI 1.83 to 22.93; P = 0.004) (Table 3). Female sex, age at PA diagnosis, HT duration, BMI <25 kg/m2, nodule size, and number of antihypertensive medications failed to reach statistical significance in the multiple regression model.
Multivariate Analysis of Predictive Factors of HT Remission in Patients With Unilateral PA After Adrenalectomy
| Relative Risk | 95% CI | P | |
|---|---|---|---|
| Female sex | 1.55 | 0.39–6.2 | 0.53 |
| BMI <25 kg/m2 | 1.53 | 0.43–5.5 | 0.51 |
| Age at diagnosis <50 ya | 0.74 | 0.16–3.45 | 0.7 |
| HT duration <10 y | 1.03 | 0.26–4.04 | 0.96 |
| <3 medications for HT | 3.01 | 0.75–12.13 | 0.12 |
| Nodule size >1.6 cma | 1.19 | 0.34–4.18 | 0.79 |
| KCNJ5 mutation | 6.48 | 1.83–22.93 | 0.004 |
| Relative Risk | 95% CI | P | |
|---|---|---|---|
| Female sex | 1.55 | 0.39–6.2 | 0.53 |
| BMI <25 kg/m2 | 1.53 | 0.43–5.5 | 0.51 |
| Age at diagnosis <50 ya | 0.74 | 0.16–3.45 | 0.7 |
| HT duration <10 y | 1.03 | 0.26–4.04 | 0.96 |
| <3 medications for HT | 3.01 | 0.75–12.13 | 0.12 |
| Nodule size >1.6 cma | 1.19 | 0.34–4.18 | 0.79 |
| KCNJ5 mutation | 6.48 | 1.83–22.93 | 0.004 |
Median values of age and nodule size.
Multivariate Analysis of Predictive Factors of HT Remission in Patients With Unilateral PA After Adrenalectomy
| Relative Risk | 95% CI | P | |
|---|---|---|---|
| Female sex | 1.55 | 0.39–6.2 | 0.53 |
| BMI <25 kg/m2 | 1.53 | 0.43–5.5 | 0.51 |
| Age at diagnosis <50 ya | 0.74 | 0.16–3.45 | 0.7 |
| HT duration <10 y | 1.03 | 0.26–4.04 | 0.96 |
| <3 medications for HT | 3.01 | 0.75–12.13 | 0.12 |
| Nodule size >1.6 cma | 1.19 | 0.34–4.18 | 0.79 |
| KCNJ5 mutation | 6.48 | 1.83–22.93 | 0.004 |
| Relative Risk | 95% CI | P | |
|---|---|---|---|
| Female sex | 1.55 | 0.39–6.2 | 0.53 |
| BMI <25 kg/m2 | 1.53 | 0.43–5.5 | 0.51 |
| Age at diagnosis <50 ya | 0.74 | 0.16–3.45 | 0.7 |
| HT duration <10 y | 1.03 | 0.26–4.04 | 0.96 |
| <3 medications for HT | 3.01 | 0.75–12.13 | 0.12 |
| Nodule size >1.6 cma | 1.19 | 0.34–4.18 | 0.79 |
| KCNJ5 mutation | 6.48 | 1.83–22.93 | 0.004 |
Median values of age and nodule size.
Discussion
In this study, we investigated predictive factors of HT remission after adrenalectomy for unilateral PA. Combining clinical and genetic variables in 100 consecutive patients with PA, KCNJ5 somatic mutation was associated with almost sixfold chance of complete clinical success after adrenalectomy for unilateral PA. This finding emphasizes the clinical utility in investigating KCNJ5 somatic mutations in unilateral PA, tailoring personalized approach regarding BP response, and the need of adjunctive therapy after adrenalectomy according to KCNJ5 mutational status.
KCNJ5 gene is the most frequently mutated gene in APAs. The frequency of KCNJ5 somatic mutations in our Brazilian cohort (42%) was similar to that described in European cohorts (40%) (16, 30). KCNJ5 mutational status was first correlated with PA outcome in a small cohort of 28 cases and only in univariate analysis (19). Until recently, the potential predictive role of KCNJ5 somatic mutations in predicting HT remission remained to be determined. KCNJ5 somatic mutations were associated with HT cure in an Australian cohort of PA (20). In contrast, Rossi et al. (31) did not demonstrate association between KCNJ5 mutational status and HT cure, but the frequency of KCNJ5 somatic mutation was lower (24%) than in previous reports from Europe. A meta-analysis of KCNJ5 somatic mutations in APAs did not find any difference in preoperative BP and hypokalemia, but the effect of KCNJ5 mutational status in BP control after adrenalectomy was not evaluated (30). Unlike KCNJ5 mutations, CTNNB1 somatic mutations were correlated with residual HT after adrenalectomy (21). However, because of the rarity of CTNNB1 mutations in PA, only 8 patients had APAs with CTNNB1 mutations (21).
In addition to the differences in the prevalence of somatic KCNJ5 mutations in APAs between Europeans (40%) and Asians (79%) (16, 17, 30), genetic characteristics of APAs were also distinct in African Americans (32). Nanba et al. (32) showed that somatic CACNA1D mutations were the most prevalent genetic alteration (42%) in APAs from African Americans. In the current study, the impact of KCNJ5 somatic mutations in HT remission was evaluated in a Brazilian cohort of patients with PA, which has clearly distinct genetic background from other PA cohorts previously reported.
Recently, the PASO study, an international panel of 31 experts from 28 centers, developed consensus criteria for outcomes and follow-up of adrenalectomy for unilateral PA (22). The definition of criteria to assess success of adrenalectomy is essential to allow the comparison of outcome data between studies. In our study, we evaluated a robust cohort of patients with unilateral PA who presented complete biochemical success after adrenalectomy based on criteria proposed by PASO study. In contrast, a precise definition of biochemical success to determine PA cure was not stated in previous reports that evaluated the impact of genetic findings in clinical outcome after surgery for unilateral PA (19–21).
In the current study, KCNJ5 somatic mutations were more frequent in women and larger tumors, similar to previously published data (16, 20, 30). Recently, Kitamoto et al. (33) showed that KCNJ5 somatic mutation, shorter duration of HT, and fewer antihypertensive medications predicted HT remission after adrenalectomy, with the KCNJ5 somatic mutation being the most relevant predictor. Similarly, our findings demonstrated that HT duration <10 years, fewer than three antihypertensive medications, and KCNJ5 somatic mutation correlated with HT remission after adrenalectomy. Of greatest importance, KCNJ5 somatic mutation was the only independent predictor of complete clinical success after surgical treatment of unilateral PA in our study. Based on these data, we can speculate that the benefit of surgical intervention in patients with KCNJ5 somatic mutations goes beyond an earlier PA diagnosis.
In a cohort of 474 APAs from the European Network for the Study of Adrenal Tumors, somatic mutations in KCNJ5, CACNA1D, ATP1A , and ATP2B3 genes were identified in 38%, 9.3%, 5.3%, and 1.7% of APAs, respectively (16). In 2018, Nanba et al. (34) investigated the prevalence of somatic mutations in APAs guided by the aldosterone synthase (CYP11B2) staining in white Americans. The most frequently mutated gene was KCNJ5 (43%), followed by CACNA1D (21%), ATP1A1 (17%), ATP2B3 (4%), and CTNNB1 (3%). Although this approach guided by microdissection of CYP11B2-positive areas increased the frequency of CACNA1D and ATP1A1 somatic mutations, it did not change the percentage of KCNJ5 mutant APAs. In our study, a target approach of CYP11B2-positive areas was not performed to guide DNA extraction. However, because we were focused on KCNJ5 mutations and clinical outcome, the absence of a CYP11B2-guided approach did not impact our results.
The impact of KCNJ5 somatic mutations on clinical outcome after adrenalectomy brings new insight about the postoperative follow-up of patients with PA. Based on our findings, patients with unilateral hyperplasia and APAs with WT KCNJ5 should be monitored closely during the postoperative period, because they are more likely to require ongoing antihypertensive medications. Genetic screening for KCNJ5 somatic mutations can be performed rapidly after surgery in clinical practice, because almost the totality of KCNJ5 somatic mutations described so far can be identified by a single PCR reaction and automated Sanger sequencing. Indeed, two KCNJ5 somatic mutations (p.Gly151Arg and p.Leu168Arg) account for 98% of KCNJ5 genetic mutations in APAs (16).
Recently, a steroid metabolome analysis revealed prevalent glucocorticoid excess in PA (35). Postoperative evidence of adrenal insufficiency was found in 29% of tested patients (35). In addition, cortisol excess appears to have an additional impact on cardiac remodeling and metabolic parameters in patients with PA (36, 37). In our study, we investigated cortisol secretion in only 15 patients with PA. Cortisol levels after LDDST did not correlate with HT remission, but the number of patients investigated for cortisol secretion was small. We are currently investigating cortisol secretion in all patients with PA. In a recent Chinese cohort of APAs, cortisol cosecretion (defined by a cortisol >1.8 μg/dL after LDDST) was identified in 22 of 555 patients with PA (38). Aldosterone and cortisol cosecreting adenomas were larger and associated with more cardiovascular and metabolic complications. However, HT remission without any antihypertensive medication was unexpectedly evidenced in all cases with cortisol secretion (38).
Although the retrospective design has the potential for selection bias, a strong aspect of our study was the precise definition of complete biochemical success using PASO criteria, allowing a better assessment of KCNJ5 mutation status in clinical outcome. In conclusion, we demonstrated here that KCNJ5 somatic mutation was an independent predictor of HT remission after unilateral adrenalectomy in patients with unilateral PA.
Acknowledgments
Financial Support: This work was supported by São Paulo Research Foundation (FAPESP) grant 2015/17049-8 and National Council for Scientific and Technological Development (CNPq) grant 403256/2016-0 (both to M.Q.A).
Additional Information
Disclosure Summary: The authors have nothing to disclose.
Data Availability:
The datasets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.
Abbreviations:
- APA
aldosterone-producing adenoma
- A/PRA
aldosterone to plasma renin activity
- A/R
aldosterone to renin
- AVS
adrenal venous sampling
- BMI
body mass index
- BP
blood pressure
- HT
hypertension
- LDDST
low-dose dexamethasone suppression test
- PA
primary aldosteronism
- PASO
Primary Aldosteronism Surgical Outcome
