Exploring Determinants of Fibromyalgia with Artificial Intelligence - from Epigenetics to the Clinic: A Brazilian Intergenerational Cohort (FIBROGEN)

• Grant number: 24/09675-5
• Support Opportunities: Regular Research Grants
• Start date: November 01, 2024
• End date: October 31, 2026
• Field of knowledge: Health Sciences - Medicine
• Principal Investigator: João Ricardo Sato
• Grantee: João Ricardo Sato
• Host Institution: Centro de Matemática, Computação e Cognição (CMCC). Universidade Federal do ABC (UFABC). Ministério da Educação (Brasil). Santo André , SP, Brazil
• Associated researchers: Clevia Rosset ; Felipe Fregni ; Mariana Rodrigues Botton ; Suellen Mary Marinho dos Santos Andrade
• Associated scholarship(s): 24/20442-2 - Prediction of Fibromyalgia Symptoms and Pain Intensity Using Machine Learning Methods and Neurobiological Variables, BP.TT

Abstract

Fibromyalgia is a multifaceted condition, associated with a history of abuse and with a significant impact on quality of life. Treatment aimed solely at relieving symptoms has been ineffective and expensive, leading to excessive use of analgesics, including opioids. Given the high failure rates, the search for composite biomarkers that integrate clinical, biological and psychological aspects is necessary. Therefore, the objective of this study is to investigate the integrated influence of traumatic, social, physical, psychological, genetic and epigenetic adverse events on the development and severity of fibromyalgia, using artificial intelligence (AI) resources. Furthermore, it aims to advance the identification of diagnostic biomarkers to guide personalized therapeutic strategies, to improve the effectiveness of treating this condition from a biopsychosocial perspective. The primary outcomes of this intergenerational cohort include measures of the impact on pain levels, functional capacity, quality of life, identification of clusters of phenotypes associated with severity and the function of the descending pain inhibition modulatory system. The main exposure factor is adverse events in childhood and adolescence (AEI), assessed by the Adverse Childhood Experiences International Questionnaire Scale (ACE-IQ), developed by the World Health Organization (WHO). The ACE-IQ assesses 13 different categories of EAI, including sexual, emotional and physical abuse, emotional and physical neglect, family violence, alcohol/drug use, mental illness or suicide in the home, family involvement in criminal activity, separation or parental divorce, community violence, collective violence and bullying. In this study, the index subjects will be 604 subjects diagnosed with fibromyalgia according to the criteria of the American Society of Rheumatology (ACR-2016) aged between 18 and 75 years. The estimated calculated sample size was 604 individuals with fibromyalgia (422 without a history of childhood abuse and 182 with a history of abuse). The descendant sample will include 2 individuals of both sexes up to the third generation for each index individual. Participants with fibromyalgia will be selected from the HCPA Pain and Neuromodulation Laboratory database, which recruited approximately 6000 fibromyalgia individuals recruited through a population-based survey disseminated in wide-reaching mass media. The assessment of this cohort comprises questionnaires for sociodemographic data, health condition, lifestyle habits and adherence to treatments. The assessment instruments are validated for Brazilian Portuguese. In addition, we will evaluate the resting activity rhythm by actimetry, Cold Pressure Test, Conditioned Pain Modulation Test (CPM-Test), measure of autonomic function by the R-R Interval, measurements of frequency waves and functional connectivity of brain areas by EEG. The blood samples will be stored in the HCPA biobank and used to identify polymorphisms and epigenetic markers of the BDNF, COMT, OPRM1 and PER2 genes, and to measure inflammatory markers, which include C-reactive protein, TNF-±, IL-1 interleukins , IL-2, IL-6, IL-10. In this study, we plan to include the first wave of the cohort, with planning for the first 10 years of follow-up. Analysis will include Cox regression, latent growth class analysis, and machine learning capabilities to identify subject clusters and composite markers indicative of risk. This innovative project seeks to identify markers for early diagnosis and influence public health policies aimed at FM. (AU)