Data mining analyses for precision medicine in acromegaly

<b>Context:</b> Predicting which acromegaly patients could benefit from somatostatin receptor ligand (SRL) is crucial to avoid months of ineffective treatment for non-responding cases. Although many biomarkers linked to SRL response have been identified, there is no consensus criterion on how to assign pharmacologic treatment according to biomarker levels. <b>Objective:</b> Our aim is to provide better predictive tools for a more accurate acromegaly patient stratification regarding the ability to respond to SRL. <b>Design and patients:</b> Retrospective multicenter study of 71 acromegaly patients. <b>Methods:</b> We used advanced mathematical modelling and artificial intelligence to predict SRL response combining molecular and clinical information. <b>Results:</b> Different models of patient stratification were obtained regarding SRL response, with a much higher accuracy when the studied cohort is fragmented according to relevant clinical characteristics. Considering all the models, a patient stratification based on the extrasellar growth of the tumor, sex, age and the expression of E-cadherin, <i>GHRL</i>, <i>IN1-GHRL</i>, <i>DRD2</i>, <i>SSTR5</i> and <i>PEBP1</i> is proposed, with accuracies that stand between 71 to 95%. Furthermore, we show an association between extrasellar growth and high BMI for SRL non-responding patients. <b>Conclusion.</b> The use of data mining is necessary for implementation of personalized medicine in acromegaly and requires an interdisciplinary effort between computer science, mathematics, biology and medicine. This new methodology opens a door to more precise personalized medicine for acromegaly patients.