Machine learning based models for prediction of subtype diagnosis of primary aldosteronism using blood test

Treating Primary Aldosteronism-Induced Hypertension: Novel Approaches and Future Outlooks

Primary aldosteronism (PA) is the most common cause of secondary hypertension and is associated with increased morbidity and mortality when compared with blood pressure-matched cases of primary hypertension. Current limitations in patient care stem from delayed recognition of the condition, limited access to key diagnostic procedures, and lack of a definitive therapy option for nonsurgical candidates. However, several recent advances have the potential to address these barriers to optimal care. From a diagnostic perspective, machine-learning algorithms have shown promise in the prediction of PA subtypes, while the development of noninvasive alternatives to adrenal vein sampling (including molecular positron emission tomography imaging) has made accurate localization of functioning adrenal nodules possible. In parallel, more selective approaches to targeting the causative aldosterone-producing adrenal adenoma/nodule (APA/APN) have emerged with the advent of partial adrenalectomy or precision ablation. Additionally, the development of novel pharmacological agents may help to mitigate off-target effects of aldosterone and improve clinical efficacy and outcomes. Here, we consider how each of these innovations might change our approach to the patient with PA, to allow more tailored investigation and treatment plans, with corresponding improvement in clinical outcomes and resource utilization, for this highly prevalent disorder.

Identifying primary aldosteronism patients who require adrenal venous sampling: a multi-center study

Adrenal venous sampling (AVS) is crucial for subtyping primary aldosteronism (PA) to explore the possibility of curing hypertension. Because AVS availability is limited, efforts have been made to develop strategies to bypass it. However, it has so far proven unsuccessful in applying clinical practice, partly due to heterogeneity and missing values of the cohorts. For this purpose, we retrospectively assessed 210 PA cases from three institutions where segment-selective AVS, which is more accurate and sensitive for detecting PA cases with surgical indications, was available. A machine learning-based classification model featuring a new cross-center domain adaptation capability was developed. The model identified 102 patients with PA who benefited from surgery in the present cohort. A new data imputation technique was used to address cross-center heterogeneity, making a common prediction model applicable across multiple cohorts. Logistic regression demonstrated higher accuracy than Random Forest and Deep Learning [(0.89, 0.86) vs. (0.84, 0.84), (0.82, 0.84) for surgical or medical indications in terms of f-score]. A derived integrated flowchart revealed that 35.2% of PA cases required AVS with 94.1% accuracy. The present model enabled us to reduce the burden of AVS on patients who would benefit the most.

Identifying KCNJ5 Mutation in Aldosterone-Producing Adenoma Patients With Baseline Characteristics Using Machine Learning Technology

Background

Primary aldosteronism is characterized by inappropriate aldosterone production, and unilateral aldosterone-producing adenoma (uPA) is a common type of PA. KCNJ5 mutation is a protective factor in uPA; however, there is no preoperative approach to detect KCNJ5 mutation in patients with uPA.

Objectives

This study aimed to provide a personalized surgical recommendation that enables more confidence in advising patients to pursue surgical treatment.

Methods

We enrolled 328 patients with uPA harboring KCNJ5 mutations (n = 158) or not (n = 170) who had undergone adrenalectomy. Eighty-seven features were collected, including demographics, various blood and urine test results, and clinical comorbidities. We designed 2 versions of the prediction model: one for institutes with complete blood tests (full version), and the other for institutes that may not be equipped with comprehensive testing facilities (condensed version).

Results

The results show that in the full version, the Light Gradient Boosting Machine outperformed other classifiers, achieving area under the curve and accuracy values of 0.905 and 0.864, respectively. The Light Gradient Boosting Machine also showed excellent performance in the condensed version, achieving area under the curve and accuracy values of 0.867 and 0.803, respectively.

Conclusions

We simplified the preoperative diagnosis of KCNJ5 mutations successfully using machine learning. The proposed lightweight tool that requires only baseline characteristics and blood/urine test results can be widely applied and can aid personalized prediction during preoperative counseling for patients with uPA.

Recent progress in the diagnosis and treatment of primary aldosteronism

Primary aldosteronism (PA) is caused by excessive secretion of aldosterone from the adrenal glands, with subsequent changes in the renin-angiotensin system. In Japan, chemiluminescent enzyme immunoassay is currently performed for aldosterone assay rather than the earlier method of radioimmunoassay. This change in aldosterone measurement methods has resulted in faster and more accurate measurement of blood aldosterone levels. Since 2019, esaxerenone, a mineralocorticoid receptor antagonist (MRA) with a non-steroidal skeleton, has been available in Japan for the treatment of hypertension. Esaxerenone has been reported to have various effects, such as strong antihypertensive and anti-albuminuric/proteinuric effects. Treatment of PA with MRAs has been reported to improve the patient's quality of life and to suppress the onset of cardiovascular events independent of their effects on blood pressure. Measuring renin levels is recommended for monitoring the extent of mineralocorticoid receptor blockade during MRA treatment. Patients receiving MRAs are prone to developing hyperkalemia, and combining MRAs with sodium/glucose cotransporter 2 inhibitors is expected to prevent severe hyperkalemia and provide additional cardiorenal protection. Mineralocorticoid receptor-associated hypertension is a broad concept of hypertension that includes not only PA, but also hypertension caused by borderline aldosteronism, obesity, diabetes, and sleep apnea syndrome. New findings on primary aldosteronism, which is part of MR-associated hypertension. Aldosterone measurements have been changed to the CLEIA method. Treatment of primary aldosteronism with MRAs has a variety of positive effects. CT-guided radiofrequency ablation and transarterial embolization are alternatives to surgery for aldosterone-producing adenomas. BP blood pressure, CLEIA chemiluminescent enzyme immunoassay, CT computed tomography, K serum potassium, MR mineralocorticoid receptor, MRA mineralocorticoid receptor antagonist, QOL quality of life, SGLT2i sodium/glucose cotransporter 2 inhibitor.

Machine learning-based model for prediction and feature analysis of recurrence in pancreatic neuroendocrine tumors G1/G2

BACKGROUND

Pancreatic neuroendocrine neoplasms (PanNENs) are a heterogeneous group of tumors. Although the prognosis of resected PanNENs is generally considered to be good, a relatively high recurrence rate has been reported. Given the scarcity of large-scale reports about PanNEN recurrence due to their rarity, we aimed to identify the predictors for recurrence in patients with resected PanNENs to improve prognosis.

METHODS

We established a multicenter database of 573 patients with PanNENs, who underwent resection between January 1987 and July 2020 at 22 Japanese centers, mainly in the Kyushu region. We evaluated the clinical characteristics of 371 patients with localized non-functioning pancreatic neuroendocrine tumors (G1/G2). We also constructed a machine learning-based prediction model to analyze the important features to determine recurrence.

RESULTS

Fifty-two patients experienced recurrence (14.0%) during the follow-up period, with the median time of recurrence being 33.7 months. The random survival forest (RSF) model showed better predictive performance than the Cox proportional hazards regression model in terms of the Harrell's C-index (0.841 vs. 0.820). The Ki-67 index, residual tumor, WHO grade, tumor size, and lymph node metastasis were the top five predictors in the RSF model; tumor size above 20 mm was the watershed with increased recurrence probability, whereas the 5-year disease-free survival rate decreased linearly as the Ki-67 index increased.

CONCLUSIONS

Our study revealed the characteristics of resected PanNENs in real-world clinical practice. Machine learning techniques can be powerful analytical tools that provide new insights into the relationship between the Ki-67 index or tumor size and recurrence.

Do multiple types of confirmatory tests improve performance in predicting subtypes of primary aldosteronism?

OBJECTIVE

The clinical practice guideline for primary aldosteronism (PA) places a high value on confirmatory tests to sparing patients with false-positive results in case detection from undergoing adrenal venous sampling (AVS). However, it is unclear whether multiple types of confirmatory tests are more useful than a single type. To evaluate whether the machine-learned combination of two confirmatory tests is more useful in predicting subtypes of PA than each test alone.

DESIGN

A retrospective cross-sectional study in referral centres.

PATIENTS

This study included 615 patients with PA randomly assigned to the training and test data sets. The participants underwent saline infusion test (SIT) and captopril challenge test (CCT) and were subtyped by AVS (unilateral, n = 99; bilateral, n = 516).

MEASUREMENTS

The area under the curve (AUC) and clinical usefulness using decision curve analysis for the subtype prediction in the test data set.

RESULTS

The AUCs for the combination of SIT and CCT, SIT alone and CCT alone were 0.850, 0.813 and 0.786, respectively, with no significant differences between them. The AUC for the baseline clinical characteristics alone was 0.872, whereas the AUCs for these combined with SIT, combined with CCT and combined with both SIT and CCT were 0.868, 0.854 and 0.855, respectively, with no significant improvement in AUC. The additional clinical usefulness of the second confirmatory test was unremarkable on decision curve analysis.

CONCLUSIONS

Our data suggest that patients with positive case detection undergo one confirmatory test to determine the indication for AVS.

A hyperaldosteronism subtypes predictive model using ensemble learning

This study aimed to develop a machine-learning algorithm to diagnose aldosterone-producing adenoma (APA) for predicting APA probabilities. A retrospective cross-sectional analysis of the Japan Rare/Intractable Adrenal Diseases Study dataset was performed using the nationwide PA registry in Japan comprised of 41 centers. Patients treated between January 2006 and December 2019 were included. Forty-six features at screening and 13 features at confirmatory test were used for model development to calculate APA probability. Seven machine-learning programs were combined to develop the ensemble-learning model (ELM), which was externally validated. The strongest predictive factors for APA were serum potassium (s-K) at first visit, s-K after medication, plasma aldosterone concentration, aldosterone-to-renin ratio, and potassium supplementation dose. The average performance of the screening model had an AUC of 0.899; the confirmatory test model had an AUC of 0.913. In the external validation, the AUC was 0.964 in the screening model using an APA probability of 0.17. The clinical findings at screening predicted the diagnosis of APA with high accuracy. This novel algorithm can support the PA practice in primary care settings and prevent potentially curable APA patients from falling outside the PA diagnostic flowchart.

Detection of factors affecting kidney function using machine learning methods

Due to the increasing prevalence of chronic kidney disease and its high mortality rate, study of risk factors affecting the progression of the disease is of great importance. Here in this work, we aim to develop a framework for using machine learning methods to identify factors affecting kidney function. To this end classification methods are trained to predict the serum creatinine level based on numerical values of other blood test parameters in one of the three classes representing different ranges of the variable values. Models are trained using the data from blood test results of healthy and patient subjects including 46 different blood test parameters. The best developed models are random forest and LightGBM. Interpretation of the resulting model reveals a direct relationship between vitamin D and blood creatinine level. The detected analogy between these two parameters is reliable, regarding the relatively high predictive accuracy of the random forest model reaching the AUC of 0.90 and the accuracy of 0.74. Moreover, in this paper we develop a Bayesian network to infer the direct relationships between blood test parameters which have consistent results with the classification models. The proposed framework uses an inclusive set of advanced imputation methods to deal with the main challenge of working with electronic health data, missing values. Hence it can be applied to similar clinical studies to investigate and discover the relationships between the factors under study.

Primary aldosteronism - a multidimensional syndrome

Primary aldosteronism is a common cause of hypertension and is a risk factor for cardiovascular and renal morbidity and mortality, via mechanisms mediated by both hypertension and direct insults to target organs. Despite its high prevalence and associated complications, primary aldosteronism remains largely under-recognized, with less than 2% of people in at-risk populations ever tested. Fundamental progress made over the past decade has transformed our understanding of the pathogenesis of primary aldosteronism and of its clinical phenotypes. The dichotomous paradigm of primary aldosteronism diagnosis and subtyping is being redefined into a multidimensional spectrum of disease, which spans subclinical stages to florid primary aldosteronism, and from single-focal or multifocal to diffuse aldosterone-producing areas, which can affect one or both adrenal glands. This Review discusses how redefining the primary aldosteronism syndrome as a multidimensional spectrum will affect the approach to the diagnosis and subtyping of primary aldosteronism.

Clinical Translationality of KCNJ5 Mutation in Aldosterone Producing Adenoma

Hypertension due to primary aldosteronism poses a risk of severe cardiovascular complications compared to essential hypertension. The discovery of the KCNJ5 somatic mutation in aldosteroene producing adenoma (APA) in 2011 and the development of specific CYP11B2 antibodies in 2012 have greatly advanced our understanding of the pathophysiology of primary aldosteronism. In particular, the presence of CYP11B2-positive aldosterone-producing micronodules (APMs) in the adrenal glands of normotensive individuals and the presence of renin-independent aldosterone excess in normotensive subjects demonstrated the continuum of the pathogenesis of PA. Furthermore, among the aldosterone driver mutations which incur excessive aldosterone secretion, KCNJ5 was a major somatic mutation in APA, while CACNA1D is a leading somatic mutation in APMs and idiopathic hyperaldosteronism (IHA), suggesting a distinctive pathogenesis between APA and IHA. Although the functional detail of APMs has not been still uncovered, its impact on the pathogenesis of PA is gradually being revealed. In this review, we summarize the integrated findings regarding APA, APM or diffuse hyperplasia defined by novel CYP11B2, and aldosterone driver mutations. Following this, we discuss the clinical implications of KCNJ5 mutations to support better cardiovascular outcomes of primary aldosteronism.

Machine learning approach to predict subtypes of primary aldosteronism is helpful to estimate indication of adrenal vein sampling

INTRODUCTION

Primary aldosteronism (PA) is a common disease. Especially in unilateral PA (UPA), the risk of cardiovascular disease is high and proper localization is important. Adrenal vein sampling (AVS) is commonly used to localize PA, but its availability is limited. Therefore, it is important to predict the unilateral or bilateral PA and to choose the appropriate cases for AVS or watchful observation.

AIM

The purpose of this study is to develop a model using machine learning to predict bilateral or unilateral PA to extract cases for AVS or watchful observation.

METHODS

We retrospectively analyzed 154 patients diagnosed with PA and who underwent AVS at our hospital between January 2010 and June 2021. Based on machine learning, we determined predictors of PA subtypes diagnosis from the results of blood and loading tests.

RESULTS

The accuracy of the machine learning was 88% and the top predictors of the UPA were plasma aldosterone concentration after the saline infusion test, aldosterone to renin ratio after the captopril challenge test, serum potassium and aldosterone-to-renin ratio. By using these factors, the accuracy, sensitivity, specificity and the area under the curve (AUC) were 91%, 70%, 99% and 0.91, respectively. Furthermore, we examined the surgical outcomes of UPA and found that the group diagnosed as unilateral by the predictors showed improvement in clinical findings, while the group diagnosed as bilateral by the predictors showed no improvement.

CONCLUSION

Our predictive model based on machine learning can support to choose the performance of adrenal vein sampling or watchful observation.

Machine learning-based models for predicting clinical outcomes after surgery in unilateral primary aldosteronism

Unilateral subtype of primary aldosteronism (PA) is a common surgically curable form of endocrine hypertension. However, more than half of the patients with PA who undergo unilateral adrenalectomy suffer from persistent hypertension, which may discourage those with PA from undergoing adrenalectomy even when appropriate. The aim of this retrospective cross-sectional study was to develop machine learning-based models for predicting postoperative hypertensive remission using preoperative predictors that are readily available in routine clinical practice. A total of 107 patients with PA who achieved complete biochemical success after adrenalectomy were included and randomly assigned to the training and test datasets. Predictive models of complete clinical success were developed using supervised machine learning algorithms. Of 107 patients, 40 achieved complete clinical success after adrenalectomy in both datasets. Six clinical features associated with complete clinical success (duration of hypertension, defined daily dose (DDD) of antihypertensive medication, plasma aldosterone concentration (PAC), sex, body mass index (BMI), and age) were selected based on predictive performance in the machine learning-based model. The predictive accuracy and area under the curve (AUC) for the developed model in the test dataset were 77.3% and 0.884 (95% confidence interval: 0.737–1.000), respectively. In an independent external cohort, the performance of the predictive model was found to be comparable with an accuracy of 80.4% and AUC of 0.867 (95% confidence interval: 0.763–0.971). The duration of hypertension, DDD of antihypertensive medication, PAC, and BMI were non-linearly related to the prediction of complete clinical success. The developed predictive model may be useful in assessing the benefit of unilateral adrenalectomy and in selecting surgical treatment and antihypertensive medication for patients with PA in clinical practice.

High Prevalence of Autonomous Aldosterone Production in Hypertension: How to Identify and Treat It

PURPOSE OF REVIEW

Primary aldosteronism (PA) affects millions of individuals worldwide. When unrecognized, PA leads to cardiovascular and renal complications via mechanisms independent from those mediated by hypertension. In this review, we emphasize the importance of PA screening in at-risk populations, and we provide options for customized PA therapy, with consideration for a variety of clinical care settings.

RECENT FINDINGS

Compelling evidence puts PA at the forefront of secondary hypertension etiologies. Cardiovascular and renal damage likely begins in early stages of renin-independent aldosterone excess. PA must be considered not only in patients with resistant hypertension or hypokalemia, but also when hypertension is associated with obstructive sleep apnea or atrial fibrillation, or in those with early-onset hypertension. Screening with plasma aldosterone and renin is widely accessible, and targeted PA therapy can successfully circumvent the excess cardiorenal risk relative to equivalent primary hypertension. Identifying and treating PA in early stages provide opportunities for personalized hypertension therapy in a large number of patients. Additionally, early targeted therapy of PA is essential for pivoting the care of such patients from reactive to preventive of cardiovascular and renal morbidity and mortality.

[Expert opinion. Spironolactone: a new twist on an old story]

The article presents recent data on possibilities of a broader use of mineralocorticoid receptor antagonists for existing indications and of expanding indications for the use of this pharmaceutical group in the context of the novel coronavirus infection COVID-19. The authors discussed prospects for expanded detection of aldosteronism using a new diagnostic approach, including an additional evaluation of blood pressure response to spironolactone.