The Effect of Methylphenidate on Cognition in Patients with Comorbid Attention Deficit/Hyperactivity Disorder and Amphetamine Use Disorder: An Exploratory Single-Blinded within-Subject Study

Introduction: Attention deficit/hyperactivity disorder (ADHD) with co-occurring substance use disorder (SUD) is associated with poor treatment outcomes. Two randomized controlled trials, utilizing robust doses of stimulants, demonstrated a significant effect on treatment outcomes in patients with ADHD/SUD. This study aimed to investigate differences in executive functioning and explore the dose-dependent effect of OROS-methylphenidate (MPH) in patients with comorbid ADHD and amphetamine use disorder (ADHD+AMPH) and patients with ADHD only. Methods: Three groups (ADHD+AMPH, ADHD only, and healthy controls) were assessed repeatedly with a neuropsychological test battery. An exploratory within-subject single-blinded design was employed where the ADHD only group received a maximum dose of 72 mg OROS-MPH, the ADHD+AMPH group a maximum dose of 180 mg, whereas the healthy subjects did not receive any study medication. Both ADHD groups received the same dose titration up to 72 mg OROS-MPH. Results: The ADHD+AMPH group demonstrated a significantly poorer motor inhibition and spatial working memory and reported more severe ADHD symptoms compared to the ADHD only group. 180 mg OROS-MPH was associated with a significant improvement in executive functioning in the dual diagnosis group. However, the exploratory study design and recruitment issues do not allow for any conclusion to be drawn regarding the effect of 180 mg OROS-MPH. Conclusion: Patients with ADHD+AMPH present with more severe neurocognitive deficits compared to ADHD only. The effect of 180 mg OROS-MPH on cognition in patients with ADHD+AMPH was inconclusive. Future studies should consider recruitment issues and high drop-out rates in this study population.

Stimulant treatments, such as methylphenidate (MPH) and amphetamine salts, are effective first-line pharmacological treatments for ADHD [21][22][23].Their primary effect is thought to be through increasing dopamine and norepinephrine availability in brain regions that have been found to be structurally and functionally altered in ADHD, such as corticostriatal systems involved in, e.g., executive functioning [24].
However, results of studies on pharmacological treatment of ADHD in patients with concurrent SUD have been discouraging.Randomized trials utilizing standard doses of MPH have showed limited effects on ADHD symptoms and no demonstrable effects on substance use, compared to placebo, in the comorbid populations [25].In patients with stimulant use disorder, one of many possible explanations for this lack of effect might be tolerance due to long-term exposure to high doses of stimulants, suggesting that more robust pharmacological doses might be warranted in this comorbid population.This hypothesis is supported by two randomized placebocontrolled studies utilizing higher doses than typically prescribed to individuals with ADHD only.Konstenius and colleagues [26] demonstrated a significant effect of doses up to 180 mg OROS-MPH on substance use, in individuals with amphetamine use disorder and ADHD.Levin and colleagues [27] investigated the response to extended-release mixed amphetamine salts in individuals with cocaine use disorder and ADHD.The latter study demonstrated a significant dose-dependent reduction on substance use by 60 mg and 80 mg, respectively, compared to placebo.
It remains unclear whether the clinical effect by such robust doses is driven by improving cognitive deficits seen in ADHD or by direct effect on the target symptoms of SUD (such as craving), analogous to the effects of opioid maintenance treatment for opioid use disorder.Although stimulants are known cognitive enhancers, doses that are too high reverse this effect and impair cognitive functioning [28,29].To achieve a better understanding of the pharmacological mechanism and medication safety, it is important to systematically assess the effect of robust doses of MPH on cognitive functioning, specifically targeting core neurocognitive deficits of ADHD and SUD in this dual-diagnosis population.
The most consistent differences in neurocognitive functioning between ADHD and healthy controls are found in measures of response inhibition, attention, spatial working memory (SWM), and some measures of planning, signal detection (arousal), and set shifting [30,31].Similar deficits have also been found in SUD, including disruptive reward processing [7,8].Moreover, studies suggest that the combination of ADHD/SUD is associated with more severe deficits in executive functions, such as working memory, attention, and inhibitory control [9][10][11][12][13].MPH improves neurocognitive deficits in ADHD, and inhibitory control and working memory specifically are hypothesized to mediate the clinical effect of MPH [32].In addition, self-reported emotional dysregulation is highly prevalent in both SUD [33] and ADHD [34] compared to human control (HC), which may also have important clinical implications.Whether individuals with ADHD + SUD present with higher emotional dysregulation compared to individuals with ADHD only has not been investigated.
A previous RCT demonstrated that MPH doses of 180 mg were safe and attenuated drug use and craving in patients with comorbid ADHD and amphetamine use disorder (ADHD+AMPH) [26].The purpose of the current human laboratory study was to explore possible mechanisms for this therapeutic effect by evaluating a dose-response effect of MPH on the core neurocognitive deficits seen in ADHD and SUD, and furthermore, investigate differences in neurocognitive function between individuals with ADHD+AMPH and ADHD only.We investigated the effect of OROS-MPH on executive functioning in individuals with comorbid ADH-D+AMPH after titration to the maximum dose of 180 mg.We hypothesized that participants would improve significantly with regard to SWM and stop signal reaction time (SSRT), compared to baseline.

Participants
Three groups were recruited through public advertising and active recruitment from relevant clinics in the Stockholm region: (1) individuals with ADHD+AMPH, (2) ADHD only, and (3) healthy controls.The study was performed at Stockholm Centre for Dependency Disorders between August 2015 and December 2019.This trial was registered before patient enrollment, and the trial registration number was withheld for review.
The main inclusion criteria were as follows: meeting DSM-5 criteria for ADHD and amphetamine use disorder, with minimum 7 and maximum 90 continuous amphetamine-free days before inclusion, and an IQ estimate of >75.The main exclusion criteria were as follows: currently meeting DSM-5 criteria for a diagnosis of any major psychiatric illness, e.g., schizophrenia, bipolar disorder, or major depression.Ongoing treatments with psychoactive substances, fulfilling DSM-5 criteria of any other severe SUD (except nicotine) during the 6-month period prior to study entry.
The main inclusion criteria for the ADHD only group were as follows: meeting DSM-5 criteria for ADHD, no history of any SUDs other than nicotine, when applying a 14-day-long wash-out period from pharmacological ADHD treatment prior to inclusion, and an IQ estimate of >75.The main exclusion criteria were as follows: currently meeting DSM-5 criteria for a diagnosis of any major psychiatric illness, e.g., schizophrenia, bipolar disorder, or major depression, ongoing treatment with psychoactive substances.
For HC, the main inclusion criteria were an IQ estimate of >75 and no history of any SUDs other than nicotine.The main exclusion criteria were as follows: currently meeting DSM-5 criteria for a diagnosis of any major psychiatric illness, e.g., schizophrenia, bipolar disorder, or major depression; ongoing treatments with psychoactive substances; to screen negative for ADHD.

Study Design
This study employed a single-blinded within-subject design.Each day, the participants received five identical capsules, where at least one capsule contained extended-release OROS-MPH (Concerta ® ) and the other ones contained placebo.The ADH-D+AMPH group reached a maximum dose of 180 mg, whereas the ADHD only group reached a maximum dose of 72 mg.It was not considered ethically feasible to provide higher doses to the ADHD only group.Note that both ADHD groups received identical dose titration up to 72 mg.The dose titration schedule is presented in Table 1.
Participants were informed that they would receive a dose ranging from 18 to 72 mg and 18-180 mg, in the ADHD only group and ADHD+AMPH group, respectively.In other words, participants were not told what dose they received on a particular day or whether they received the same dose every day or not.Healthy controls did not receive any study medication.
Each participant in the ADHD+AMPH group visited the clinic daily for supervised intake of study medication.Most participants in the ADHD only group visited the clinic every day, while some at least every third day for 9-10 days.For those who did not visit the clinic daily (in the ADHD only group), daily intake of study medication was confirmed through daily phone calls.Healthy controls were assessed at two time points, no less than 7 days and no longer than 14 days apart.Urine toxicology, alcohol breath analysis, and timeline follow-back (TLFB) were employed daily (for the ADHD-only group, at a minimum of two different time points during the study) to ensure that participants did not drink alcohol or use illicit drugs during the study period.Nicotine use was not allowed during the hour preceding neuropsychological testing.Importantly, due to the effects of psychoactive drugs on cognition, participants were excluded if they consumed alcohol or other psychoactive drugs during the study period.
Neuropsychological assessments were performed on day 1 (baseline/before medication), day 9 (72 mg MPH), and day 18 (180 mg MPH).Participants with SUD were offered a referral for treatment after study completion.Participants received a monetary compensation of SEK 5,500 (ADHD+AMPH), SEK 2,000 (ADHD only group), or SEK 1,250 (healthy controls), equivalent to approximately 660 USD, 240 USD, and 150 USD, respectively.The different compensations reflect how many days and test sessions the different groups were required to participate in.Clinical Instruments Newly referred participants, who were not already diagnosed with ADHD, were assessed in accordance with national guidelines including the Diagnostic Interview for ADHD in Adults (DIVA 2.0) and interviews with family members.In these cases, given the inclusion criterion of an IQ estimate of >75 and for differential diagnostic purposes, a short form of the Wechsler Adult Intelligence Scale (WAIS-IV) was administered.For participants with an earlier diagnosis of ADHD, the study physician assessed patient records to make sure that the diagnostic procedure had been made in accordance with national guidelines, including structured diagnostic instruments and WAIS.Each participant underwent a psychiatric evaluation by the study physician, utilizing the Structured Clinical Interview for DSM-IV/5 (for SUDs) and the Mini International Neuropsychiatric Interview for DSM IV/5 (for other psychiatric disorders).All participants underwent a physical examination, electrocardiogram, and blood sample collection.Substance use was assessed through urine toxicology, breathalyzer, and a timeline follow-back interview.
All participants were assessed using the Adult ADHD Self-Report Scale (ASRS), the brief version of Difficulties in Emotional Regulation Scale (DERS-16), the Drug Use Disorders Identification Test-Extended (DUDIT-E), the Montgomery Åsberg Depression Rating Scale (MADRS), the Lifetime Drinking History interview (LDH; participants with amphetamine use disorder were also assessed regarding lifetime amphetamine use), and the Fagerström Test of Nicotine Dependence.Craving for amphetamine was measured daily by a 1-item visual analog scale, asking patients to assess their present level of craving, ranging from 0 to 100.

Neuropsychological Testing
Cognitive function was assessed through the Cambridge Neuropsychological Test Automated Battery (CANTAB ® ) using a touch-screen tablet PC (MOTION J3500-i7B) and a press pad from Cambridge Cognition Ltd.The three tasks described below constitute the recommended test battery for ADHD by the scientific advisors of the test developers.The primary outcomes were working memory and response inhibition, which can be described as separate dimensions of core executive functions [35] (together with cognitive flexibility) and have been shown to be dysregulated in patients with ADHD as compared to controls [36][37][38].
Spatial Working Memory Task Visuospatial working memory was assessed through a SWM, where participants were presented with several colored squares and instructed to find a hidden token within one of the squares.After locating a hidden token, the participant was instructed not to return to the square where the token was hidden and continue on to finding the next.The test increases in difficulty with more and more squares, reaching a maximum of twelve.The main outcome measure for this test is Between Errors (returning to a square where a token has already been found), which is hypothesized to reflect SWM (a higher number reflects poorer SWM).

Stop Signal Task
Response inhibition was assessed by the stop signal task (SST) during which participants were presented with a visual cue (an arrow pointing right or left) and instructed to press right or left as fast as possible, unless the visual cue was followed by an auditory signal, in which case the participants were instructed to inhibit their response.The time between the visual cue and auditory signal varies depending on how well the participant performs.The main outcome measure is the SSRT, which is the time between the go and the stop stimulus at which the participants are able to successfully inhibit their response on 50% of trials.SSRT is hypothesized to reflect motor inhibition (a higher SSRT reflects poorer motor/response inhibition).
Rapid Visual Processing Sustained attention was assessed through rapid visual processing (RVP), where participants were presented with digits ranging from 2 to 9 in a pseudo-random order.Participants were instructed to detect certain sequences of digits and to respond as soon as possible when a target sequence was detected.The main outcome measure is A prime (A'), in which sensitivity to detect the target sequences is measured regardless of response tendency.A prime is reflected to measure sustained attention.

Statistical Analysis
Data were analyzed utilizing R version 4.2.0.Further details on R packages used are presented in online supplementary materials (for all online suppl.material, see https://doi.org/10.1159/000535016).

Primary Analysis
The primary outcome of this study was performance on the SST and SWM on 180 mg MPH compared to baseline in the dual diagnosis group.An estimated effect size of 1.5 was based on a previous study of the effect of MPH on the SST in cocainedependent individuals [39].The required sample size to detect a significant difference between baseline and 180 mg MPH in the ADHD+AMPH group, with the estimated effect size of 1.5, power set at 0.80 and alpha at 0.05, and an expected high drop-out rate [26], was calculated for 24 participants.Primary analysis includes only participants in the dual diagnosis group that completed the study protocol and thus reached the target dose of 180 mg MPH.A one-way repeated measures ANOVA was conducted to determine the effect of 72 mg MPH and 180 mg MPH compared to baseline on the SST and SWM in the dual diagnosis group.
A one-way ANOVA with post hoc pairwise comparison was conducted to determine if there were any group differences at baseline measures for continuous variables.For categorical variables, χ 2 test was utilized.Significant main effects of groups were followed by a post hoc pairwise comparison without adjustment for multiple comparison and adjustment with a Tukey's Honest Significant Difference test (for ANOVA) or Bonferroni correction (for Kruskal-Wallis) for the main and sensitivity analysis, respectively.Missing variables were excluded list wise.
Data were assessed visually by boxplot for outliers, normal distribution by the Shapiro-Wilk test, and sphericity by the Mauchly's test of sphericity.Homogeneity of variances and covariances was assessed by the Levene's test of homogeneity of variances and the Box's M test, respectively.

Sensitivity Analysis
A two-way mixed ANOVA was conducted to determine if the ADHD+AMPH group improved at a different rate compared to the ADHD only group (i.e., if there was an interaction between the ADHD groups and time on each outcome).Since the ADHD only group and HC were only assessed at 2 time points (in contrast to the ADHD+AMPH group who were assessed at three time points), the third assessment (180 mg OROS-MPH) for the ADH-D+AMPH group was excluded from this analysis.Correspondingly, a two-way mixed ANOVA was conducted to determine if the ADHD groups (who received MPH) improved at a different rate compared to HC.
If any of the assumptions of the specific ANOVA analysis were violated, sensitivity analysis was conducted and/or a nonparametric test was used, such as the Kruskal-Wallis H test or the Friedman test.For violations of sphericity, a Greenhouse-Geisser correction was applied.Furthermore, statistical tests were adjusted for available confounders (such as age, sex, stimulant consumption, group, and self-rated ADHD symptoms).See the online supplementary material for further details on sensitivity analysis.

Participants
Nineteen participants with comorbid ADHD+AM-PH+AMPH were included in the study, of which 16 completed the assessments at 72 mg MPH and 11 participants completed all three assessments and thus reached the maximal dose of 180 mg MPH.Recruitment started in August 2015, and due to recruitment problems, time, and budget restraints, we decided to stop inclusion in December 2019 before reaching the target of 24 participants with ADHD+AMPH.Sixteen participants with ADHD-only were included, of which 15 completed the study protocol and thus reached the maximal dose of 72 mg MPH.Twenty-one healthy controls were included and completed two test sessions.See Figure 1 for a flowchart illustrating the flow of study participants.Clinical and sociodemographic characteristics are presented in Table 2.

Group Differences at Baseline
Often, data were not normally distributed, and/or the assumption of homogeneity of variance was not met, or there were some outliers.However, when performing sensitivity analysis as described above, the results were always identical (or very similar) to the main ANOVA analysis, and therefore, we report only the ANOVA results to facilitate the interpretation of the results.The full analysis including nonparametric test results is presented in the online supplementary material.The ADHD+AMPH group performed significantly worse on the SST and SWM compared to the ADHD only group and HC at baseline, meaning they presented with poorer motor inhibition and SWM.There was no significant difference between the groups on RVP, i.e., there were no differences in sustained attention at baseline.The ADH-D+AMPH group presented with significantly higher ASRS scores (self-rated ADHD symptoms) compared to the ADHD only group and HC.Moreover, both ADHD groups had a significantly higher degree of self-rated emotional dysregulation (as measured by DERS-16), whereas there was no significant difference on DERS-16 between the ADH-D+AMPH group and the ADHD only group.See Table 3 for further details on baseline differences between groups.

Sensitivity Analysis of Baseline Assessments
Each one-way ANOVA of above baseline results was adjusted for sex and age as covariates and yielded identical results.Adjustment for multiple comparisons (Tukey adjustment and/or Bonferroni) yielded identical (or very similar) results, which did not change the overall conclusion.Neither sex and number of years with amphetamine use nor number of days with abstinence from amphetamine at baseline was associated with baseline scores of SSRT, RVP, and between error (BE).Age was only significantly associated with BE (higher age was associated with poorer performance).The baseline ASRS score was only significantly associated with SSRT at baseline.Hence, the one-way ANOVA for SSRT at baseline was adjusted for ASRS at baseline, which yielded identical results as primary analysis.Further details on statistical analysis are presented in online supplementary materials.

Effect of 72 mg and 180 mg MPH in the ADHD+AMPH Group
Differences between assessments (baseline, second, and third) were investigated in the ADHD+AMPH group.In some cases, one or several of the assumptions of the one-way repeated measures ANOVA were violated.Since sensitivity analysis yielded identical results (or very similar results), the main analysis is presented.Further details on analysis, including sensitivity analysis, are presented in the supplementary materials.For craving scores, ANOVA could not be used, and instead results from the Friedman test with pairwise comparisons are presented.
As shown in Table 4, there was a significant improvement in motor inhibition, SWM, and sustained attention between the first (baseline/no medication) and the third assessment (at 180 mg OROS-MPH).Furthermore, there was a significant improvement in motor inhibition between the second assessment (72 mg OROS-MPH) and the third assessment (180 mg OROS-MPH) on SSRT.Additionally, the ADHD+AMPH group reported significantly lower self-rated ADHD symptoms and craving scores on the third assessment compared to baseline.

Group Interactions
Group*-time interaction was investigated between the ADHD/AMPH group and the ADHD only group on the first assessment (no study medication) and the second assessment (72 mg OROS-MPH).Results from this analysis are summarized in Table 5.

Stop Signal Reaction Time
There was no significant group *time interaction between the ADHD/AMPH group and the ADHD only group between the first and second assessments (0 vs. 72 mg OROS MPH) (F(1, 29 = 0.043, p = 0.84) on SSRT.There was a significant effect of time (p = 0.003) and a significant main effect of groups (p = 0.001), meaning that the ADHD/AMPH group performed significantly worse on the SST compared to the ADHD only group, and that both groups (collapsed) improved significantly between assessments.There was no difference between groups with regard to the rate of improvement.

Spatial Working Memory
There was no significant group*time interaction between the ADHD+AMPH group and the ADHD only group between the first and second assessment (0 vs. 72 mg OROS MPH) on BE (F(1, 29 = 0.56, p = 0.46).There was no significant effect of time (p = 0.13), i.e., both groups (collapsed) did not significantly improve between assessments.There was a significant main effect of groups (p < 0.001), i.e., the ADHD/AMPH group performed significantly worse on the SWM task compared to the ADHD only group.

Rapid Visual Processing
There was no significant group*time interaction between the ADHD/AMPH group and the ADHD only group between the first and second assessment (0 vs. 72 mg OROS MPH) on A prime (F(1, 29 = 0.68, p = 0.42).There was a significant effect of time (p = 0.008), i.e., both groups (collapsed) significantly improve between assessments.There was no significant main effect of groups (p = 0.89), i.e., there was no significant difference on performance on the RVP task between the groups.There was no difference between groups with regard to the rate of improvement.

Adult ADHD Self-Report Scale
There was no significant group*time interaction between the ADHD+AMPH group and the ADHD only group between the first and second assessment (0 and 72 mg OROS MPH) on ASRS (F(1, 28 = 0.036, p = 0.85).There was a significant effect of time (p < 0.001), i.e., both groups (collapsed) significantly improve between assessments.There was a significant effect of groups (p = 0.003), i.e., the ADHD/AMPH group had a significantly higher ASRS score compared to the ADHD only group.There was no difference between groups in regard to the rate of improvement.The table only includes those participants that reached 72 mg MPH and thus were included in analysis.SSRT, stop signal reaction time; BE, between errors; A prime; ASRS, adult ADHD self-report scale; DERS-16, difficulties in emotion regulation scale-16.Data are expressed as the mean and standard deviation.There were no significant group*time interaction in any measurement: meaning that the rate of any improvement between sessions were similar in both groups.*Significant main effect of time: meaning that both ADHD groups (collapsed) improved significantly between first and second assessments.∧ Significant main effect of the group: meaning that there was a significant difference in overall score between groups (when both test sessions were collapsed).

Difficulty in Emotional Regulation Scale-16
There was no significant group*time interaction between the ADHD/AMPH group and the ADHD only group between the first and second assessment (0 and 72 mg OROS MPH) on DERS-16 (F(1, 26 = 0.041, p = 0.84).There was a significant effect of time (p = 0.016), i.e., both groups (collapsed) significantly improve between assessments.There was no significant effect of groups (p = 0.92), i.e., there was no significant difference of mean DERS-16 score between groups.There was no difference between groups in regard to the rate of improvement.

Sensitivity Analysis on Differences between First and Second Assessment
There were no significant associations between either stimulant consumption, age, sex, or the ASRS baseline score and improvement between the first and second assessments on SSRT, BE, RVP, DERS-16, nor ASRS.In contrast to primary analysis, there was a significant association between the group and improvement on the SSRT, where the ADHD/AMPH group had a significantly lower rate of improvement compared to the ADHD group only (p = 0.049).There were no other significant associations between groups (i.e., ADHD+AMPH, ADHD only, and HC) on improvement on BE, RVP, DERS-16, nor ASRS.
Furthermore, when including all three groups in the two-way mixed ANOVA, we found no group*time interaction between first and second assessments on any of the cognitive tasks.Additionally, with both ADHD groups collapsed, there was no group*time interaction between ADHD groups and HC between the first and second assessments.Meaning all three groups improved between assessments on all cognitive tasks in a similar manner.

Adverse Events
One participant, with a history of chronic stomachache, was hospitalized for one night due to a combination of stomachache and anxiety, while being treated with 144 mg MPH, and thus by definition experienced a serious adverse event.However, the participant went through an extensive examination, including esophagogastroduodenoscopy, and no underlying cause of the patients' pain was found, and the condition was resolved the day after.No other serious, or unexpected, adverse events occurred during the study.Two participants chose to drop out due to anxiety, both at 108 mg MPH.The most common adverse events were anxiety, headache, and gastrointestinal issues, e.g., discomfort and dry mouth.

Discussion
This study investigated differences in neurocognitive functioning, and self-reported ADHD symptoms between HC, patients with ADHD and comorbid severe amphetamine use disorder, and patients with ADHD only.Our sample in the dual diagnosis group consisted of a clinically representative sample, with a long history of substance use, where a majority administered amphetamine intravenously and had temporary housing and no income, a population that is often excluded from clinical trials.We found that the dual diagnosis group performed significantly poorer on tasks of inhibitory control and SWM and had a significantly higher ASRS score compared to the ADHD only group, while no statistically significant difference in sustained attention and selfreported emotional dysregulation was observed.This is in line with previous research on patients with comorbid ADHD and cocaine use disorder [12], whereas this is the first study, to the best of our knowledge, to demonstrate that patients with comorbid ADHD+AMPH present with more severe cognitive deficits and ADHD symptoms, compared to patients with ADHD only.This may explain the lack of successful treatment outcomes observed in previous research.
Furthermore, we aimed to explore if the previously observed dose-dependent effect of stimulant treatment on substance use and self-reported ADHD symptoms [26,27] was associated with an improvement in neurocognitive functioning in patients with ADHD and comorbid amphetamine use disorder, specifically, 180 mg OROS-MPH (a higher dose than recommended by current guidelines [40]).Arguably, the therapeutic dosedependent effect of such robust doses of stimulants might be explained by a tolerance effect, following long-term use of illicit stimulants and the subsequent hypoactivation of e.g., the dopaminergic system.Another contributing factor might be the significant impairment in neurocognitive functioning observed in patients with comorbid ADHD/SUD [9][10][11][12][13] that may require higher stimulant doses to produce the desired cognitive-enhancing effect.The current study, specifically, provides further evidence to support the hypothesis that patients with comorbid ADHD+AMPH have more severe neurocognitive deficits than patients with ADHD only and thus may require different treatment strategies.Finally, treatment effects of stimulants in this population may also be related to a reduction in craving.
However, we were not able to recruit the pre-calculated required sample size of 24 participants needed to detect the expected within-subject effect on stop signal task and SWM.This is possibly explained by the strict inclusion criteria and the highly demanding study protocol, where the participants were required to be abstinent from all psychoactive substances for a total of at least 26 days, with daily visits during the study period and extensive, repeated neurocognitive testing.Nevertheless, despite the small sample that completed the study protocol, withinsubject analysis revealed that the comorbid ADH-D+AMPH group had a significant improvement in tasks of inhibitory control, SWM, and sustained attention on 180 mg MPH, as compared to baseline.Importantly, selfreported ADHD symptoms and craving scores were also significantly reduced at 180 mg MPH compared to baseline, while there were no significant differences in self-reported emotional dysregulation.
A pragmatic exploratory design was employed of which the major limitations are the lack of a placebo group and the small sample size.However, given the potent effect of psychostimulants, a placebo design might not be feasible with higher doses of MPH, but more importantly, it was not considered feasible to recruit the number of patients needed to include a placebo group.The latter concern was confirmed by the recruitment issues in the current study.The withinsubject design, and lack of a placebo group, limits the conclusion that can be drawn from the observed improvement between assessments in the dual diagnosis group.In addition, all three groups improved, or did not improve, in a similar manner between the first and second assessments (including HC who did not receive MPH).However, it should be noted that the calculated sample size in this study was in regard to detect a significant difference on task performance between baseline and 180 mg MPH in the ADHD+AMPH group, i.e., the current study may not have the required sample size to detect differences in improvement rate between groups.Other factors, such as repeated visits to a clinic, abstinence from alcohol and other substances, or a practice effect of repeated testing may have impacted test results.However, sensitivity analysis revealed a significant effect of groups on improvement on SSRT (i.e., the ADHD+AMPH group may have improved to a lesser extent compared to the other groups on 72 mg MPH).Moreover, on a group level, there were some important differences in baseline characteristics between the groups, specifically in regard to educational level, sociodemographic data, and nicotine use.Additionally, each participant in both ADHD groups received the same dose titration (up to 72 mg OROS-MPH), regardless of weight, sex, age, or estimated tolerance (on the basis of previous stimulant consumption).However, each statistical test was adjusted for sex and age in sensitivity analysis (weight was not collected), and furthermore, in our sample, we found no evidence to suggest that cognitive task performance was confounded by estimated tolerance to stimulants.Finally, task performance may have been confounded by the differences in reimbursements between groups.However, this would not explain why the ADHD/AMPH group (who received the highest reimbursement) performed worse.Additionally, reimbursements reflected the differences in what was required to participate, i.e., the group that had a higher number of test days in the study received higher reimbursement.
It should be noted that although the ADHD+AMPH group improved significantly on all cognitive tasks and self-reported ADHD symptoms at 180 mg compared to baseline, there was only a significant difference between 180 mg and 72 mg for the SST.Thus, self-reported ADHD symptoms improved already at 72 mg compared to baseline but did not seem to improve further at a higher dose.This is important since task-related behavior often does not correlate well with self-reported constructs [41] and, in general, ADHD symptoms questionnaires are more relevant in a clinical setting [42].However, although both task-related behaviors and self-rated behaviors might capture different aspects, both modalities have proven important in describing ADHD and other psychiatric populations [43].Furthermore, previous research in this specific population (ADHD+AMPH) suggests that there is a dose-dependent effect of stimulants on clinical outcomes [26,27].
Despite these inferential issues, a significant improvement in cognition was observed at 180 mg MPH compared to baseline.Although, it should be noted that two participants chose to drop out due to anxiety, both at a dose of 108 mg OROS-MPH.The observed cognitive improvement is important since the effect of psychostimulants on cognitions is dose dependent, where lower doses enhance cognitive functioning and higher doses tend to reverse this effect [28,29].Doses that are too high are thus not expected to improve cognitive functioning but may instead reverse the cognitive enhancement and furthermore cause a subjective "high" and/or an increase in adverse effects such as anxiety [29].Overall, although inconclusive, these results lend some support to the hypothesis that the previously observed clinical effect of 180 mg MPH on relapse and ADHD symptoms may be driven by enhancement of executive functioning.

Conclusions
Patients with comorbid ADHD and amphetamine use disorder present with more severe cognitive deficits and self-reported ADHD symptoms compared to patients with ADHD only, specifically poorer motor inhibition and SWM.Overall, our results on the effects of 180 mg MPH on executive functioning in patients with ADHD and comorbid amphetamine use disorder are inconclusive.The target dose of 180 mg MPH was generally well tolerated in the present sample and was associated with a significant improvement in neurocognitive functioning and on self-reported drug craving and ADHD symptoms.However, due to the exploratory study design and the small sample size, it is not possible to draw any conclusion in regard to causality.
Future studies should take into account the high dropout rate observed in previous research and in this study and aim to employ an even more pragmatic design, shorten the dose titration period, employ less strict inclusion/exclusion criteria to facilitate recruitment, and include a placebo group.For instance, recruitment would have perhaps been easier if treatment started immediately (i.e., to not require participants to be abstinent at baseline) and the exclusion criterion of poly-substance use had been removed.A lower drop-out rate would have been expected if the dose titration period was half as long.Furthermore, repeated testing at baseline is recommended to diminish any practice effect.

Table 1 .
Dose titration and schedule of study medication/placebo

Table 2 .
Clinical and sociodemographic characteristics Data are expressed as the mean and standard deviation unless indicated otherwise.*Significant difference to HC. **Significant difference compared to the other groups.

Table 3 .
Group differences from neuropsychological testing and clinical rating scales at baseline SSRT, stop signal reaction time; BE, between errors, A prime; ASRS, Adult ADHD Rating-Scale; DERS-16, Difficulties in Emotion Regulation Scale -16; MADRS, Montgomery Åsberg Depression Scale.Data expressed as mean and standard deviation.*Significant difference compared to HC in post hoc pairwise comparison.∧ Significant difference compared to ADHD only in post hoc pairwise comparison.

Table 4 .
Results from neuropsychological testing and clinical rating scales in the ADHD+AMPH group at baseline, second assessment (72 mg OROS-MPH), and third assessment (180 mg OROS-MPH) BE, between errors, A prime; MADRS, Montgomery Åsberg Depression Scale; DERS-16, Difficulties in Emotion Regulation Scale-16; DUDIT-E, Craving and Drug Use Disorders Identification Test Extended (n = 11 (for MADRS n = 10, DERS-16 n = 9, and DUDIT-E n = 8).Data are expressed as the mean and standard deviation.*Significant difference compared to baseline.∧ Significant difference between 72 mg and 180 mg.++ Significant difference in median score; p-value refers to the Friedman test.

Table 5 .
Results from neuropsychological testing and clinical rating scales in the ADHD+AMPH group and the ADHD only group at baseline and second assessment (72 mg OROS-MPH)