Two stimulant studies can report different remission percentages even if their treatments are not meaningfully different. Before comparing the numbers, determine whether “remission” meant the same thing.
The mixed-salts reanalysis
Adler and colleagues reported post hoc analyses from two adult ADHD trials of oral SHP465 mixed amphetamine salts extended release. One used dose optimization over seven weeks, 12.5–75 mg; the other used fixed doses of 25, 50 or 75 mg over six weeks.
The intention-to-treat populations were 268 in the optimization study, comprising 136 active and 132 placebo participants, and 405 in the fixed-dose study, comprising 302 active and 103 placebo participants. These were reanalyses of existing trials, not newly recruited replication cohorts.
Several response definitions were examined
The authors considered a 30% symptom-score reduction plus clinician improvement, a stricter 50% reduction plus clinician improvement, and an absolute ADHD rating score of 18 or less. Remission, however, required a score of 12 or less.
At the final treatment week, the remission criterion was met by 37.2% receiving mixed salts versus 19.7% placebo in the optimization study. Corresponding fixed-dose percentages were 39.7% and 10.0%. These are percentages under a particular threshold and observation rule.
Why a lisdexamfetamine percentage is not directly comparable
A separate 2013 lisdexamfetamine analysis defined symptomatic remission as a score of 18 or less. The same score that counted as remission there was treated as one response definition in the mixed-salts report.
A stricter cutoff generally classifies fewer observations as meeting the target. Thus a lower reported remission rate cannot automatically establish inferior pharmacologic efficacy. Different populations, doses, trial lengths and missing-data handling add further obstacles to comparison.
Final week and endpoint also differ
The mixed-salts report distinguished the final treatment week from endpoint, which could include earlier termination. Remission percentages at endpoint were lower than the final-week figures. Quoting one study’s final-week percentage against another study’s last-observation percentage creates another mismatch.
Time-to-event analyses also answer a different question from maintaining remission at every later visit. Their nominal significance does not establish permanent recovery.
How to use these findings
The post hoc analyses support symptom improvement over placebo under several definitions. They do not create a universal remission standard, validate cross-trial treatment rankings or establish normal work and driving performance.
Any comparison should specify the scale, cutoff, required change, denominator and timing. The secondary lisdexamfetamine source used for the definition comparison is Mattingly et al., BMC Psychiatry 2013;13:39, doi:10.1186/1471-244X-13-39. Both reports are historical research, not current prescribing guidance.
Scientific source
Adler LA, Robertson B, Chen J, Sarkis E. Post hoc Responder and Remission Analyses from Two Studies of SHP465 Mixed Amphetamine Salts Extended-Release Among Adults with Attention-Deficit/Hyperactivity Disorder. Journal of Child and Adolescent Psychopharmacology. 2020;30:427–438. doi:10.1089/cap.2020.0012. Original commentary on the supplied paper.
Leave a Reply