Sumatriptan Transdermal System in Patients With Nausea
Sumatriptan Transdermal System in Patients With Nausea
The current study was part of the pivotal trial (phase 3) assessing the efficacy of sumatriptan TDS in the acute treatment of migraine in adults. Details of this study, as well as its main findings, have been published.
In brief, this was a double-blind, randomized, parallel-group, single-attack, placebo-controlled study conducted at 38 sites in the United States, which were composed of a mix of outpatient clinic and university-based sites. The study was conducted from January to July of 2009. A total of 469 patients were treated. Patients were generally healthy men and women aged 18–66 years who are diagnosed with migraine headache (with or without aura) before the age of 50 years. Patients typically experienced ≥1 moderate-to-severe headache/month and had a ≥1-year history of migraines. Patients who planned to start, stop, or change treatment or dose with prophylactic antimigraine or mood stabilizer treatment ≤3 months before randomization and through the final visit were excluded. Being in a registration phase 3 program, participants fulfilling inclusion criteria were directly recruited by the sites, as well as by Investigation Review Board-approved advertising. The primary end point was the proportion of patients who were headache pain-free 2 hours after patch activation. Other end points included the proportions of patients who reported headache pain relief, and freedom from nausea, photophobia, and phonophobia; rescue medication use; and tolerability. A statistically significant higher proportion of patients treated with the sumatriptan TDS, relative to placebo, experienced freedom from headache pain, nausea, photophobia, and phonophobia 2 hours after patch activation. Treatment-emergent adverse events were reported by 50% and 44% of patients treated with the sumatriptan iontophoretic transdermal system and placebo, respectively. Most events were transient mild-to-moderate application-site reactions.
Participants had migraine with or without aura according to the Second Edition of the International Classification of Headache Disorders and were otherwise healthy. Inclusion criteria required at least one moderate-to-severe headache per month. Preventive medications were allowed if on a stable dose. Patients with less than one or more than six migraines per month, as well as those with headaches on more than 15 days per month were excluded. Significant medical diseases were exclusionary as well.
In order to participate, participants signed Institutional Review Board-approved consent forms. The trial was registered on ClinicalTrials.gov (NCT00724815) and was part of the approval package submitted to the FDA.
Patients were randomized 1:1 to receive the sumatriptan TDS delivering 6.5 mg of sumatriptan over a 4-hour period, or a matching placebo TDS. Patients treated one migraine attack.
Procedures and assessments were those recommended by the Guidelines for Controlled Trials for Drugs in Migraine. In brief, during a migraine attack, patients rated their baseline headache pain on a 4-point scale (0 = none, 1 = mild, 2 = moderate, 3 = severe) and applied the study patch once the pain reached moderate severity. Patients were not permitted to take analgesic, nonsteroidal anti-inflammatory drugs (NSAIDs), or antiemetic medications during the 8 hours before or the 2 hours following TDS activation. In addition to pain severity, participants recorded the presence or absence of aura, nausea, phonophobia, and photophobia, as well as other distinguishing migraine features over several time points within 24 hours. End points included pain-free, pain relief, and associated symptoms relief. For the original study, primary endpoint was pain-free at 2 hours. All adverse events were recorded from screening through 30 days after TDS removal.
Herein, we report on post-hoc analyses conducted on a registration trial. Details of the random allocation, randomization blocks, and sample size determination have been described elsewhere.
For this study, we first identified the subset of patients who treated their migraine attacks while nausea-free (had qualifying headache characteristics, moderate intensity, photophobia and phonophobia but not nausea), and provided clinical data information (including on nausea) 2 hours after treatment end. We also contrasted patients who had nausea or not at baseline regarding their response to the sumatriptan TDS. Of 530 patients screened for participation, 446 were included in the intent-to-treat (ITT) population for the primary analyses, defined as all patients who applied and activated a study TDS, and who had ≥1 postbaseline assessment for headache pain (Fig. 1).
(Enlarge Image)
Figure 1.
Patient disposition by treatment group. AE = adverse event.
For the comparison of effectiveness in participants receiving sumatriptan TDS as a function of nausea at baseline, in-between-group treatment differences for the primary and key secondary efficacy end points were assessed using logistic regression models; a last observation carried forward method was used.
For the assessment of TEN, a modified ITT population – patients who had no nausea at baseline, applied and activated a study patch, and had at least one postactivation assessment for pain – was used in this post-hoc analysis. We used a generalized linear model to provide a framework for relating response and predictor variables by extending traditional linear model theory to nonlinear data, because the outcomes observed are dichotomous in nature and not necessarily normally distributed. Because the disease states are within-subject correlated, a repeated measures application was used in the regression analyses using generalized estimating equations (GEE). By specifying one of a variety of possible working correlation matrix structures to account for the within-subject correlations, the GEE method estimates model parameters by iteratively solving a system of equations based on quasi-likelihood distributional assumptions. Effects were entered for randomized group and time, allowing evaluation of a group, time, and group and time interaction effect prior to interpreting pairwise comparisons at each time point. Interaction effects were first tested to identify whether there was a suspected acceleration or deceleration of the group effect with time. Because herein, there was no such effect, the scale parameter was held fixed. The Wald statistic was used to test the true value of the parameters based on the sample estimate. Adjustments for multiplicity were performed using the Bonferroni correction. The Wald statistics and P values are reported for each of the types of standard error estimates, empirical and model-based. Tests of the long-term differences between groups were conducted using a Bonferroni adjusted alpha level of 0.006 to account for multiplicity of time point measurements (0.05/8). All analyses were carried out using SAS version 9.1.3 (SAS Institute, Cary, NC, USA).
Methods
Overview
The current study was part of the pivotal trial (phase 3) assessing the efficacy of sumatriptan TDS in the acute treatment of migraine in adults. Details of this study, as well as its main findings, have been published.
In brief, this was a double-blind, randomized, parallel-group, single-attack, placebo-controlled study conducted at 38 sites in the United States, which were composed of a mix of outpatient clinic and university-based sites. The study was conducted from January to July of 2009. A total of 469 patients were treated. Patients were generally healthy men and women aged 18–66 years who are diagnosed with migraine headache (with or without aura) before the age of 50 years. Patients typically experienced ≥1 moderate-to-severe headache/month and had a ≥1-year history of migraines. Patients who planned to start, stop, or change treatment or dose with prophylactic antimigraine or mood stabilizer treatment ≤3 months before randomization and through the final visit were excluded. Being in a registration phase 3 program, participants fulfilling inclusion criteria were directly recruited by the sites, as well as by Investigation Review Board-approved advertising. The primary end point was the proportion of patients who were headache pain-free 2 hours after patch activation. Other end points included the proportions of patients who reported headache pain relief, and freedom from nausea, photophobia, and phonophobia; rescue medication use; and tolerability. A statistically significant higher proportion of patients treated with the sumatriptan TDS, relative to placebo, experienced freedom from headache pain, nausea, photophobia, and phonophobia 2 hours after patch activation. Treatment-emergent adverse events were reported by 50% and 44% of patients treated with the sumatriptan iontophoretic transdermal system and placebo, respectively. Most events were transient mild-to-moderate application-site reactions.
Participants had migraine with or without aura according to the Second Edition of the International Classification of Headache Disorders and were otherwise healthy. Inclusion criteria required at least one moderate-to-severe headache per month. Preventive medications were allowed if on a stable dose. Patients with less than one or more than six migraines per month, as well as those with headaches on more than 15 days per month were excluded. Significant medical diseases were exclusionary as well.
In order to participate, participants signed Institutional Review Board-approved consent forms. The trial was registered on ClinicalTrials.gov (NCT00724815) and was part of the approval package submitted to the FDA.
Design and End Points
Patients were randomized 1:1 to receive the sumatriptan TDS delivering 6.5 mg of sumatriptan over a 4-hour period, or a matching placebo TDS. Patients treated one migraine attack.
Procedures and assessments were those recommended by the Guidelines for Controlled Trials for Drugs in Migraine. In brief, during a migraine attack, patients rated their baseline headache pain on a 4-point scale (0 = none, 1 = mild, 2 = moderate, 3 = severe) and applied the study patch once the pain reached moderate severity. Patients were not permitted to take analgesic, nonsteroidal anti-inflammatory drugs (NSAIDs), or antiemetic medications during the 8 hours before or the 2 hours following TDS activation. In addition to pain severity, participants recorded the presence or absence of aura, nausea, phonophobia, and photophobia, as well as other distinguishing migraine features over several time points within 24 hours. End points included pain-free, pain relief, and associated symptoms relief. For the original study, primary endpoint was pain-free at 2 hours. All adverse events were recorded from screening through 30 days after TDS removal.
Analyses
Herein, we report on post-hoc analyses conducted on a registration trial. Details of the random allocation, randomization blocks, and sample size determination have been described elsewhere.
For this study, we first identified the subset of patients who treated their migraine attacks while nausea-free (had qualifying headache characteristics, moderate intensity, photophobia and phonophobia but not nausea), and provided clinical data information (including on nausea) 2 hours after treatment end. We also contrasted patients who had nausea or not at baseline regarding their response to the sumatriptan TDS. Of 530 patients screened for participation, 446 were included in the intent-to-treat (ITT) population for the primary analyses, defined as all patients who applied and activated a study TDS, and who had ≥1 postbaseline assessment for headache pain (Fig. 1).
(Enlarge Image)
Figure 1.
Patient disposition by treatment group. AE = adverse event.
For the comparison of effectiveness in participants receiving sumatriptan TDS as a function of nausea at baseline, in-between-group treatment differences for the primary and key secondary efficacy end points were assessed using logistic regression models; a last observation carried forward method was used.
For the assessment of TEN, a modified ITT population – patients who had no nausea at baseline, applied and activated a study patch, and had at least one postactivation assessment for pain – was used in this post-hoc analysis. We used a generalized linear model to provide a framework for relating response and predictor variables by extending traditional linear model theory to nonlinear data, because the outcomes observed are dichotomous in nature and not necessarily normally distributed. Because the disease states are within-subject correlated, a repeated measures application was used in the regression analyses using generalized estimating equations (GEE). By specifying one of a variety of possible working correlation matrix structures to account for the within-subject correlations, the GEE method estimates model parameters by iteratively solving a system of equations based on quasi-likelihood distributional assumptions. Effects were entered for randomized group and time, allowing evaluation of a group, time, and group and time interaction effect prior to interpreting pairwise comparisons at each time point. Interaction effects were first tested to identify whether there was a suspected acceleration or deceleration of the group effect with time. Because herein, there was no such effect, the scale parameter was held fixed. The Wald statistic was used to test the true value of the parameters based on the sample estimate. Adjustments for multiplicity were performed using the Bonferroni correction. The Wald statistics and P values are reported for each of the types of standard error estimates, empirical and model-based. Tests of the long-term differences between groups were conducted using a Bonferroni adjusted alpha level of 0.006 to account for multiplicity of time point measurements (0.05/8). All analyses were carried out using SAS version 9.1.3 (SAS Institute, Cary, NC, USA).
