Home-based Transcranial Direct Current Stimulation (tDCS) Compared to Duloxetine: Non-inferiority Clinical Trial (FIBROSTIM)

Fibromyalgia is characterized by widespread pain, fatigue, non-restorative sleep, and psychocognitive alterations, compromising quality of life and leading to absenteeism and early retirement. Up to 70% of patients discontinue treatment with antidepressants and anticonvulsants due to adverse effects or low efficacy, and more than 30% resort to opioid use. Given the treatment challenges and the scarcity of safe alternatives, there is growing interest in interventions such as transcranial direct current stimulation (tDCS), which has shown efficacy in improving symptoms and functionality, with low cost and few side effects. In this context, we designed a randomized, double-blind, double-dummy clinical trial to compare the non-inferiority of 28 home-based anodal tDCS (2 mA) applied over the primary motor cortex (M1) versus duloxetine 60 mg. Both treatments will be combined with physical exercise and pain education. Outcomes will be assessed through multidimensional measures of pain, functionality, global impression of improvement, and the function of the descending pain inhibitory system. Secondary outcomes include quality of life, depressive symptoms, psychophysical pain measures, and treatment adherence. An additional analysis will compare the results of sham tDCS and duloxetine placebo within the non-inferiority model. Predictors of treatment response will also be explored, including symptom severity and oscillatory patterns of cortical electrical activity, rest-activity rhythm, and autonomic function assessed by R-R interval. Furthermore, serum levels of S100-B protein, brain-derived neurotrophic factor (BDNF), and genetic variants related to neuroplasticity in the BDNF Val66Met, Catechol-O-Methyltransferase (COMT) (rs4680) (G\>A), OPRM1, and PER2 genes will be analyzed. Inflammatory markers (TNF-α, IL-1, IL-2, IL-6, IL-10, C-reactive protein) and serum endorphins will also be assessed. A total of 610 women with fibromyalgia (aged 18 to 75 years) will be randomized into three groups (2:2:1): duloxetine + sham tDCS (n=244); active tDCS + placebo (n=244); and sham tDCS + placebo (n=122). Participants will be assessed during treatment and at 3, 6, and 12 months after completing the intervention protocol. An interim analysis will be conducted when \~50% of participants (n ≈ 305) complete the 3-month follow-up by an independent, blinded Data Monitoring Committee (DMC). (i) The trial may be stopped if the conditional probability of demonstrating non-inferiority is \<10%, based on frequentist or Bayesian methods. (i) The trial will be stopped if serious adverse events (SAEs) in the active tDCS group increase by ≥30% compared to duloxetine (p \< 0.01, adjusted). (ii) Early stopping for efficacy will be considered if active tDCS demonstrates clear non-inferiority or superiority over duloxetine on the primary outcome. Superiority requires: (iii) a clinically relevant difference exceeding the non-inferiority margin (≥10% pain reduction); (ii) statistical significance (p \< 0.005, O'Brien-Fleming adjusted); and (iii) a ≥2-point (20%) improvement on the BPI, confirmed in the ITT analysis. This study aims to generate evidence to support the decision-making process of the National Committee for Health Technology Incorporation (CONITEC) regarding the availability of tDCS in the Brazilian Unified Health System (SUS). In addition, identifying predictors of response to tDCS and duloxetine, through the integration of genetic, neurophysiological, inflammatory, and psychosocial markers using machine learning algorithms, will allow for identifying factors that can personalize fibromyalgia treatment. This approach enhances clinical efficacy, reduces costs associated with ineffective interventions, and supports more accurate therapeutic decisions, expanding access to safe, effective, and sustainable care within the public healthcare system