CT Texture Analysis of Adrenal Pheochromocytomas: A Pilot Study

CT-Based Radiomics Analysis of Different Machine Learning Models for Discriminating the Risk Stratification of Pheochromocytoma and Paraganglioma: A Multicenter Study

RATIONALE AND OBJECTIVES

Using different machine learning models CT-based radiomics to integrate clinical radiological features to discriminating the risk stratification of pheochromocytoma/paragangliomas (PPGLs).

MATERIALS AND METHODS

The present study included 201 patients with PPGLs from three hospitals (training set: n = 125; external validation set: n = 45; external test set: n = 31). Patients were divided into low-risk and high-risk groups using a staging system for adrenal pheochromocytoma and paraganglioma (GAPP). We extracted and selected CT radiomics features, and built radiomics models using support vector machines (SVM), k-nearest neighbors, random forests, and multilayer perceptrons. Using receiver operating characteristic curve analysis to select the optimal radiomics model, a combined model was built using the output of the optimal radiomics model and clinical radiological features, and its accuracy and clinical applicability were evaluated using calibration curves and clinical decision curve analysis (DCA).

RESULTS

Finally, 13 radiomics features were selected to construct machine learning models. In the radiomics model, the SVM model demonstrated higher accuracy and stability, with an AUC value of 0.915 in the training set, 0.846 in external validation set, and 0.857 in external test set. Combining the outputs of SVM models with two clinical radiological features, a combined model constructed has demonstrated optimal risk stratification ability for PPGLs with an AUC of 0.926 for the training set, 0.883 for the external validation set, and 0.899 for the external test set. The calibration curve and DCA show good calibration accuracy and clinical effectiveness for the combined model.

CONCLUSION

Combined model that integrates radiomics and clinical radiological features can discriminate the risk stratification of PPGLs.

CT-based radiomics research for discriminating the risk stratification of pheochromocytoma using different machine learning models: a multi-center study

OBJECTIVES

The purpose of this study was to explore and verify the value of various machine learning models in preoperative risk stratification of pheochromocytoma.

METHODS

A total of 155 patients diagnosed with pheochromocytoma through surgical pathology were included in this research (training cohort: n = 105; test cohort: n = 50); the risk stratification scoring system classified a PASS score of < 4 as low risk and a PASS score of ≥ 4 as high risk. From CT images captured during the non-enhanced, arterial, and portal venous phase, radiomic features were extracted. After reducing dimensions and selecting features, Logistic Regression (LR), Extra Trees, and K-Nearest Neighbor (KNN) were utilized to construct the radiomics models. By adopting ROC curve analysis, the optimal radiomics model was selected. Univariate and multivariate logistic regression analyses of clinical radiological features were used to determine the variables and establish a clinical model. The integration of radiomics and clinical features resulted in the creation of a combined model. ROC curve analysis was used to evaluate the performance of the model, while decision curve analysis (DCA) was employed to assess its clinical value.

RESULTS

3591 radiomics features were extracted from the region of interest in unenhanced and dual-phase (arterial and portal venous phase) CT images. 13 radiomics features were deemed to be valuable. The LR model demonstrated the highest prediction efficiency and robustness among the tested radiomics models, with an AUC of 0.877 in the training cohort and 0.857 in the test cohort. Ultimately, the composite of clinical features was utilized to formulate the clinical model. The combined model demonstrated the best discriminative ability (AUC, training cohort: 0.887; test cohort: 0.874). The DCA of the combined model showed the best clinical efficacy.

CONCLUSION

The combined model integrating radiomics and clinical features had an outstanding performance in differentiating the risk of pheochromocytoma and could offer a non-intrusive and effective approach for making clinical decisions.

Bilateral pheochromocytomas: clinical presentation and morbidity rate related to surgery technique and genetic status

Background

Pheochromocytomas (PHEOs) are rare catecholamine-secreting adrenal tumors. Approximately 60-90% of bilateral PHEOs are hereditary. We retrospectively analyzed the clinical characteristics of patients with bilateral PHEOs and the morbidity rate (malignancy, tumor recurrence and adrenal insufficiency (AI) rate) related to surgery technique and genetic status of the patients.

Results

Fourteen patients (12.5%, nine women, five men) had synchronous or metachronous bilateral PHEOs (out of 112 PHEO patients who underwent surgery between 1976 and 2021). The median age at diagnosis was 32 years (9-76) (three were children). Nine patients (64.2%) presented synchronous bilateral tumors, five (35.7%) contralateral metachronous tumors, 2-12 years after the first surgical intervention; three (21.4%) were metastatic. Median follow-up: 5 years (1-41), IQR 19 months. A total of 78.5% had a germline mutation (eight RET gene with MEN2A syndrome, three VHL syndrome, three not tested). Post-surgery recurrence was noted in 16.6% of patients (one with MEN2A syndrome and metastatic PHEOs, one with VHL syndrome), with similar rates after total adrenalectomy or cortical-sparing adrenal surgery. AI was avoided in 40% after cortical-sparing surgery.

Conclusion

Bilateral PHEOs are usually associated with genetic syndromes. The surgical technique for patients with hereditary bilateral PHEOs should be chosen based on a personalized approach, as they are at higher risk for developing new adrenal tumors requiring additional surgeries.

Application of a Radiomics Machine Learning Model for Differentiating Aldosterone-Producing Adenoma from Non-Functioning Adrenal Adenoma

To evaluate the secretory function of adrenal incidentaloma, this study explored the usefulness of a contrast-enhanced computed tomography (CECT)-based radiomics model for distinguishing aldosterone-producing adenoma (APA) from non-functioning adrenal adenoma (NAA). Overall, 68 APA and 60 NAA patients were randomly assigned (8:2 ratio) to either a training or a test cohort. In the training cohort, univariate and least absolute shrinkage and selection operator regression analyses were conducted to select the significant features. A logistic regression machine learning (ML) model was then constructed based on the radiomics score and clinical features. Model effectiveness was evaluated according to the receiver operating characteristic, accuracy, sensitivity, specificity, F1 score, calibration plots, and decision curve analysis. In the test cohort, the area under the curve (AUC) of the Radscore model was 0.869 [95% confidence interval (CI), 0.734-1.000], and the accuracy, sensitivity, specificity, and F1 score were 0.731, 1.000, 0.583, and 0.900, respectively. The Clinic-Radscore model had an AUC of 0.994 [95% CI, 0.978-1.000], and the accuracy, sensitivity, specificity, and F1 score values were 0.962, 0.929, 1.000, and 0.931, respectively. In conclusion, the CECT-based radiomics and clinical radiomics ML model exhibited good diagnostic efficacy in differentiating APAs from NAAs; this non-invasive, cost-effective, and efficient method is important for the management of adrenal incidentaloma.

Image-Guided Precision Medicine in the Diagnosis and Treatment of Pheochromocytomas and Paragangliomas

In this comprehensive review, we aimed to discuss the current state-of-the-art medical imaging for pheochromocytomas and paragangliomas (PPGLs) diagnosis and treatment. Despite major medical improvements, PPGLs, as with other neuroendocrine tumors (NETs), leave clinicians facing several challenges; their inherent particularities and their diagnosis and treatment pose several challenges for clinicians due to their inherent complexity, and they require management by multidisciplinary teams. The conventional concepts of medical imaging are currently undergoing a paradigm shift, thanks to developments in radiomic and metabolic imaging. However, despite active research, clinical relevance of these new parameters remains unclear, and further multicentric studies are needed in order to validate and increase widespread use and integration in clinical routine. Use of AI in PPGLs may detect changes in tumor phenotype that precede classical medical imaging biomarkers, such as shape, texture, and size. Since PPGLs are rare, slow-growing, and heterogeneous, multicentric collaboration will be necessary to have enough data in order to develop new PPGL biomarkers. In this nonsystematic review, our aim is to present an exhaustive pedagogical tool based on real-world cases, dedicated to physicians dealing with PPGLs, augmented by perspectives of artificial intelligence and big data.