Comments on Effect of Food on Early Drug Exposure from Extended-Release Stimulants: Results from the Concerta®, Adderall XR® Food Evaluation Study

Mario A. González, James E. Polli, Jill A. Morgan


Curr Med Res Opin. 2002;18(7) 


Dear Sir,

Auiler et al. recently reported in Current Medical Research and Opinion a food-effect study involving Concerta® and Adderall XR®.[1] The study was unusual in that it attempted to correlate plasma level concentrations of two different chemical compounds used to treat attention-deficit, hyperactivity disorder (ADHD), and it attempted to use partial areas under the plasma concentration-time curves (AUC) to evaluate the food effect on early drug exposure. Partial AUCs for both Concerta® and Adderall XR®, AUCp4h, AUCp6h, and AUCp8h, were calculated after fasting and fed treatments.[1] Based on pharmacokinetic profiles up to 8 h after dosing, a high-fat breakfast resulted in lower partial areas for Adderall XR® relative to the fasting treatment, whereas food had no appreciable effect on early drug exposures for Concerta®. Auiler et al. noted that Concerta®, an osmotic-controlled OROS® tablet formulation, provides reliable and consistent delivery of methylphenidate and that less day-to-day variation would be expected in methylphenidate exposures during the first 8 h after dosing compared with amphetamine exposure following dosing of Adderall XR® (a two-component, bead formulation). Furthermore, they concluded that the reliability of Concerta® is an important clinical consideration when choosing a once-daily ADHD therapy.

The bioavailability of amphetamine from Adderall XR® taken with or without food was originally demonstrated in a randomized, open-label, crossover design study in which healthy subjects received Adderall XR® 30 mg in the morning after an overnight fast or a high-fat breakfast.[2,3] Although the TMAX was delayed and CMAX was decreased, the 90% confidence intervals for the mean AUCINF and CMAX FED/FASTED ratios were within the range of bioequivalence criterion stipulated by the Food and Drug Administration (FDA). In other words, the total drug absorbed or total exposure as measured by the total area, AUCINF, was not affected by food as judged by the FDA. The maximum drug exposure, CMAX, is also unaffected by food. Finally, contrary to what Auiler et al. suggest, no published data currently exist on how closely the plasma levels of stimulants correlate to objective efficacy measures for these medications in the treatment of ADHD. The bioequivalence of amphetamine from Adderall XR® in the presence or absence of food has been unequivocally demonstrated and these data are included in the current Adderall XR® approved package insert.[3] Unfortunately, Auiler et al. only report plasma levels for the first 8 h, and while this is interesting, the entire plasma profile would have allowed the reader to see the similarity in total amphetamine exposure when Adderall XR® is taken with or without food. Auiler et al. did briefly mention that total drug exposure as measured by AUCINF with or without food was consistent with previous reports for both Adderall XR® and Concerta®, but failed to highlight this fact in the representative charts and graphs.

To date, small changes in plasma levels of stimulant medications have not been shown to alter clinical outcome or impact individual patient response. Therapeutic efficacy is not solely related to plasma levels of drugs, a concept that is particularly true for psychoactive drugs for which drug effects are often measured by psychological and behavioral responses. Thus, the implication by Auiler et al. that small decreases in plasma amphetamine levels could somehow have a deleterious effect on the treatment of children with ADHD when Adderall XR® is administered with food, is not substantiated, especially when the therapeutic window for amphetamine is unknown. The apparent reduction in absorption seen with Adderall XR® after the first 4 h of dosing has not been identified as a problem in efficacy trials nor have investigators attempted to control or regulate dietary intake of study subjects. The efficacy of Adderall XR® as early as 1.5 h after dosing was demonstrated in a randomized, double-blind, placebo-controlled study of children with ADHD.[4] Adderall XR® produced consistent, dose-related improvements in all measures of efficacy in the morning, afternoon, and late-afternoon time periods. In this study's naturalistic school and home setting, the efficacy of Adderall XR® early in the day was demonstrated regardless of breakfast content or timing of meals in relation to the dose.

Although drug exposure after dosing with Concerta® was not consistently affected by food in the Aulier study, plasma methylphenidate levels varied after a single dose of Concerta® 36 mg, particularly 6-8 h after dosing. From Figure 4 of the Auiler study it appears that, following administration of Concerta® 36 mg with a high-fat meal, nearly 30% of patients had at least a 20% decrease in plasma methylphenidate levels and approximately 25% of patients had at least a 20% increase.[1] From this same figure it can be interpreted that some patients receiving Concerta® had greater than a 100% increase in early drug exposure. In contrast, the effect of food on amphetamine exposure after administration of Adderall XR® 20 mg in the Auiler study was consistent. More than 88% of patients who received Adderall XR® 20 mg with a high-fat meal had at least a 20% decrease in early plasma amphetamine concentrations. Thus, the change in plasma amphetamine levels after dosing of Adderall XR® with a high-fat meal is more consistent and predictable than the change in plasma methylphenidate levels after dosing Concerta® with a high-fat meal.

A review of the literature revealed several published pharmacokinetic studies for Concerta® since it was approved by the FDA. The variability of methylphenidate release from Concerta® tablets was an issue in an earlier published study by Modi et al.[5] In this study, food appeared to have a significant effect on methylphenidate plasma concentrations following administration of an 18 mg Concerta® tablet, while at the same time, no effect was noted on plasma concentrations following a 36 mg dose. The CMAX FED/FASTED ratio was 130% for the 18 mg tablet but only 112% for the 36 mg tablet. When Concerta® was approved, the FDA guidance in effect at the time required a ratio of no more than 120% for CMAX or AUCINF in order to support a no food-effect claim. Physicians should be aware that, in the publication, the Concerta® 18 mg tablets were significantly affected by food while the 36 mg product was not.

In reviewing the published studies, it was observed that the methylphenidate pharmacokinetic parameters used to demonstrate bioequivalence for regulatory purposes (e.g. mean AUCINF and CMAX) differ dramatically between the results from Auiler et al. and those reported in the Concerta® FDA registration data and in earlier publications.[1,5,6] Both fasting and fed AUCINF and CMAX values for methylphenidate following the administration of Concerta® 36 mg published in the Auiler study were substantially higher than the same parameters reported for Concerta® in an earlier publication, as may be seen in Table 1 .[1,5] González et al.[6] also reported fasting methylphenidate pharmacokinetic parameters after dosing with Concerta® 36 mg that are similar to the values reported by Auiler but also much higher than the levels previously reported for Concerta®. These results, which were obtained in similar populations of healthy adult volunteers, are summarized in Figures 1 and 2. It is important to contrast that the mean fed and fasted amphetamine AUCINF and CMAX values reported by Auiler et al. after dosing with Adderall XR® 20 mg do not differ from the corresponding values reported in the NDA for Adderall XR®.[1,2] The similar results seen for Adderall XR® in the Auiler study and in Shire's NDA data[1,2] further support the similarity in the volunteer populations, yet the same reproducibility was not demonstrated for Concerta®.

Methylphenidate AUCINF comparison

Methylphenidate CMAX comparison

The variability in bioequivalence parameters (CMAX and AUCINF) for methylphenidate following administration of Concerta® 36 mg in the recent Auiler study versus earlier published data is not easily explained but may suggest that the Concerta® tablets currently available on the market are somehow different in their release characteristics from those originally approved by the FDA. It is also possible that this difference may be due to lot-to-lot variability. While it is true that neither levels of amphetamine or methylphenidate have been correlated with therapeutic effect, it is not unreasonable to expect that plasma drug levels be reproducible from year to year in the same patient or similar patient populations. Awareness of the high degree of variability in the release of methylphenidate from Concerta® tablets may assist clinicians in their interpretation of individual patient responses and adverse event profiles.