While the need for personalised treatment approaches grows in recognition, predicting treatment outcomes for adults with Attention-Deficit/Hyperactivity Disorder (ADHD) remains underexplored. Recent interest has turned to the brain's surface and its association with treatment response. Although the precise interplay between cortical gyrification and ADHD treatment outcomes remains to be elucidated, preliminary investigations suggest a promising avenue for diagnostic innovation. Expanding upon the Comparison of Methylphenidate and Psychotherapy in Adult ADHD Study (COMPAS), we investigated the prognostic value of cortical gyrification in predicting treatment response. Specifically, we explored how pre-treatment cortical gyrification might predict response to psychotherapy or clinical management in combination with either methylphenidate or placebo following a 12-week intensive treatment period. Cortical gyrification was assessed using 121 T1-weighted anatomical scans. Linear regression models investigated the predictive value of cortical gyrification, regressing baseline cortical structure against post-treatment severity. All brain structural analyses were conducted using the threshold-free cluster enhancement (TFCE) approach and the Computational Anatomy Toolbox (CAT12) within the Statistical Parametric Mapping Software (Matlab Version R2021a). Results revealed significant positive region-specific associations between cortical gyrification and treatment response across three symptom dimensions, with significant associations localised predominantly in frontal regions of the left hemisphere. Our findings emphasise that increased cortical gyrification in frontal cortical regions signifies enhanced treatment efficacy following a 12-week intervention. Further research in this area is imperative to verify the reliability of biological markers in view of treatment success to potentially reduce unnecessary drug-related side-effects, minimising delay from receiving more effective treatments, and increase treatment adherence.
While historically considered a neurodevelopmental disorder confined to childhood, Attention-Deficit/Hyperactivity Disorder (ADHD) persistently impacts up to 60% of individuals into adulthood [1, 2]. Characterised by its intrinsic heterogeneity, ADHD manifests itself across a spectrum of severity among those affected. Patients present distinct manifestations of symptom domains, aetiologies, developmental trajectories, and response to treatment [3, 4].
Effective management of ADHD in adults often necessitates a multifaceted approach, encompassing both pharmaco- and psychotherapeutic interventions. Although the administration of stimulants (e.g., methylphenidate) and nonstimulants (e.g., atomoxetine) has been demonstrated to reduce core ADHD symptoms in the short term [5,6,7,8], the effectiveness of pharmacotherapy varies considerably. Approximately 30% of patients do not respond to pharmacotherapy and the inter-individual variability in tolerability of associated side effects often results in discontinuation of medication [9, 10]. Where patients respond, the benefit of pharmacologic treatment may decline over time [11, 12], demonstrating the necessity of other strategies to improve the functional outcomes of patients.
Alongside pharmacotherapy, numerous nonpharmacological interventions have been introduced for adults with ADHD. Among others, these approaches include cognitive behavioural therapy (CBT), psychoeducation, emotional regulation therapy, mindfulness therapy, and neurofeedback [13]. Nevertheless, previous systematic reviews and meta-analyses on the efficacy of psychotherapy on ADHD core symptoms [14,15,16], internalising symptoms [17, 18] and emotional regulation [19] yielded inconclusive results with either small or modest effects due to limited sample size or methodology (i.e., open-label studies). Recently, however, a meta-analysis by Liu et al. [20] focusing on previous methodological shortcomings arrived at a more definitive conclusion regarding the effectiveness of CBT in adults with ADHD. First, CBT significantly reduced the severity of both core ADHD symptoms and emotional symptoms, with medium to large effect sizes. Second, ADHD severity total scores, inattention and hyperactivity/impulsivity symptoms were significantly lower after CBT. Third, the reduction in ADHD symptoms correlated with an improvement of quality of life. Together, these findings suggest that CBT is a promising strategy for adults with ADHD in its ability to offer symptom remission and improved life quality.
In the light of precision medicine, there have been only few studies aiming to predict treatment success in ADHD [21]. For this reason, there is currently no clear evidence to guide decision making about who will tolerate and/or respond best to which treatment approach. Presently, it is not possible to predict course or outcome at the level of the individual patient [22]. The inherent complexity and diversity within ADHD pose significant challenges to predicting treatment efficacy. Still, some research groups tried to detect biological predictors of psychostimulant treatment response in adults affected by ADHD [23,24,25,26]. Among the findings that significantly predicted treatment response to psychostimulants were genetic markers [27], task-based- [28] and structural neuroimaging markers [25], neurophysiological- [29] as well as electrophysiological markers [30], constructing a complex and diverse clinical picture [23]. Even though these results may be promising for improving diagnostic accuracy, additional research will have to verify the reliability of the different biological markers in view of the complex pathophysiology of ADHD [23].
While attempts have been made to predict treatment response to pharmacotherapy, research investigating biological predictors of treatment response to psychotherapy in adults with ADHD has yet to begin. This is despite substantial evidence to suggest that such predictors may indeed exist. Psychological interventions have the capacity to modify brain function across various psychopathological conditions [31]. First, research suggests that changes in brain function following psychotherapy correlate with symptom improvement [32]. Second, psychotherapy can lead to normalisation of brain structure and function in patients with conditions such as obsessive compulsive disorder, depression, or schizophrenia [33,34,35]. Third, comparisons between neurobiological changes post-psychological treatment and those following pharmacological treatment reveal that psychotherapy can have effects on brain function comparable to medication, though not universally across all disorders [31]. While direct evidence may be lacking, the existing body of research strongly implies a significant relationship between psychotherapy and neurobiological changes, highlighting the potential for further investigation in this area.
Recently, there has been increasing interest toward exploring the brain surface as predictor of treatment outcomes across a variety of psychiatric conditions [36,37,38]. Especially local cortical gyrification and the overall gyrification index (GI) have received most attention. Nevertheless, research investigating the GI in relation to ADHD in adulthood are notably scarce and yielded inconclusive results. Ambrosino et al. [39] found decreased cortical folding in the left rostral middle frontal cortex and right pars opercularis, and Mous et al. [40] reported increased gyrification in the left medial temporal lobe. Contrastingly, neither Forde et al. [41], Shaw et al. [42] and Gharehgazlou et al. [43], nor our research group [44] report differences in either global or local gyrification between patients with ADHD and healthy controls. Such inconsistencies likely reflect both methodological variability and sample heterogeneity. Differences in imaging protocols, including the choice of smoothing parameters, and whether gyrification is assessed local or globally can substantially influence outcomes. Moreover, variability in sample characteristics such as age distribution, comorbidity profiles, ADHD subtypes, and prior treatment exposure may further contribute to variability across studies. Although the general consensus holds that the GI remains relatively stable after the third trimester of pregnancy [45], more recent work by White et al. [46] propose that the GI may also undergo significant modulation during adolescence and maturity into early adulthood. In this light, cortical gyrification may not only reflect early neurodevelopmental processes, but also serve as a dynamic marker with potential utility in predicting treatment response. In fact, there have been prognostic studies on the relationship between cortical gyrification and psychotic disorders predicting treatment response [36, 37], subsequent recurrence in patients with first-episode schizophrenia [47] or conversion to psychosis [48, 49]. Studies in adults with ADHD, however, are yet to be conducted but may hold promising diagnostic value.
Based on the considerations given above, the present study investigated whether alterations in cortical gyrification hold prognostic value in predicting treatment response in adults with ADHD. Expanding upon our prior work [50] within the framework of the Comparison of Methylphenidate and Psychotherapy in Adult ADHD Study (COMPAS) cohort, we explored cortical gyrification in relation to treatment response following a 12-week intervention period, both across the entire sample and within each treatment group. COMPAS was the first, and so far largest, multi-centre randomised clinical study that evaluated the effects of group psychotherapy (GPT) compared with clinical management (CM) combined with methylphenidate (MPH) or placebo (PLA) in adults with ADHD [51,52,53]. In pursuit of adequately addressing the complexity of neuroanatomical variability, we applied the threshold-free cluster enhancement (TFCE) approach. By addressing key limitations of voxel-based morphometry, TFCE enables the detection of both spatially diffuse and sharply localized structural changes while maintaining strict control over family-wise error [54]. Its sensitivity to nuanced structural patterns has solidified its utility in capturing neuroanatomical variability, particularly in studies of neurodevelopmental disorders [55]. Within this methodological and clinical framework, our objective offers the potential to discern individuals prone to suboptimal treatment responses before the commencement of MPH prescription or the initiation of psychotherapeutic interventions. Such initiatives bear clinical significance to reduce unnecessary drug-related side-effects, minimising delay from receiving more effective treatments, and patient discouragement. Given the limited research available on predicting treatment response in adults with ADHD, encompassing both pharmacological- and psychotherapeutic intervention in relation to cortical gyrification, we will explore gyrification without predefined hypotheses. Nonetheless, based on existing research in schizophrenia [36, 37], we anticipate a positive relationship between treatment response and cortical gyrification.