Pivotal Studies and Protocol Selected Studies
Description of Studies
Two identical sponsor-submitted pivotal trials, Study 301 and Study 302, met the inclusion criteria for the systematic review. The two trials were multi-centre, double-blind, randomized, parallel-group studies comparing HP/TAZ to vehicle when applied once daily to adult patients with moderate-to-severe plaque psoriasis. To be eligible for inclusion, patients needed to have an area of plaque psoriasis that was appropriate for topical treatment and covered a BSA of between 3% and 12% (inclusive). Patients were also required to have a clinical diagnosis of psoriasis at baseline with an Investigator’s Global Assessment (IGA) score of 3 or 4. Patients were excluded if they had received prior treatment for plaque psoriasis and failed to respond, or if they received treatment for psoriasis within a specified amount of time. Additional details of the inclusion and exclusion criteria are available in Table 5.
Overall, 418 patients were included in the two studies. A total of 203 and 215 patients in Study 301 and Study 302, respectively, were randomized in a 2:1 ratio to receive HP/TAZ or vehicle lotion. Patients applied the study drug to the treatment area determined by the investigator once daily for eight weeks; a follow-up visit was scheduled four weeks after treatment cessation. Concomitant medications, including non-medicated cleansers, moisturizers, and sunscreens that were approved by the sponsor, were permitted for use on the treatment areas. In both studies, health-related quality of life (HRQoL) was assessed using the DLQI, skin clearance was assessed using the IGA and BSA affected by psoriasis, and the signs of psoriasis (erythema, plaque elevation, and scaling). Skin clearance and HRQoL are considered outcomes that are important to patients. Treatment adherence and productivity were not included as outcomes.
At baseline, patients were a mean age of 48.8 (standard deviation [SD] = 13.3) and 51.8 (SD = 14.3) years in Study 301 and Study 302, respectively. The majority of patients were male (61.0% to 69.1%) and White (80.1% to 92.6%) in the two studies. Most patients randomized to Study 301 and Study 302 had moderate disease (82.8% and 87.4%, respectively), indicated by an IGA score of three. The rest of the patients in both studies were classified as having severe disease by an IGA score of four. The percent BSA affected by psoriasis was a mean 6.2% (SD = 2.9%) and 5.6% (SD = 2.6%) for patients in Study 301 and Study 302, respectively. Overall, the disease characteristics of patients in both of the pivotal trials were similar between the HP/TAZ and vehicle treatment groups, with a few differences to note. In Study 301, patients randomized to vehicle had ▬▬▬▬▬ than patients receiving HP/TAZ ▬▬▬▬▬. In Study 302, patients randomized to HP/TAZ, the mean (SD) size of the target lesion for treatment was ▬▬▬▬▬, and the erythema signs of psoriasis for the target lesion were ▬▬▬▬▬
Efficacy Results
HRQoL was assessed using the DLQI, which was included as an exploratory outcome in both trials. At the week 8 (end of treatment) study visit, the DLQI score for patients in Study 301 had ▬▬▬▬▬ in the HP/TAZ and vehicle treatment groups, respectively. In Study 302, the mean (SD) change from baseline in the DLQI score was ▬▬▬▬▬ patients in the HP/TAZ and vehicle treatment groups, respectively. In the absence of formal statistical testing, no conclusions can be made about whether DLQI was actually reduced in either group or whether there were any differences in DLQI between the two treatment groups.
Skin clearance was assessed using the IGA score and percentage of BSA affected by plaque psoriasis in both of the pivotal trials. The primary end point was treatment success at week 8, defined by at least a two-grade improvement from baseline in the IGA score in addition to an IGA score of clear or almost clear. In Study 301, 35.8% and 7.0% of patients treated with HP/TAZ and vehicle, respectively, had treatment success at week 8. In Study 302, 45.3% and 12.5% of patients treated with HP/TAZ and vehicle, respectively, had treatment success at week 8. In both trials, the difference between HP/TAZ and vehicle in the proportion of patients achieving treatment success (P < 0.001) was in favour of HP/TAZ. Treatment success based on the IGA was assessed at week 12, 6, 4, and 2 as secondary end points, which were analyzed following a gated sequential testing procedure in that order. The difference between HP/TAZ and vehicle in the proportion of patients with treatment success was in favour of HP/TAZ (P < 0.05) at week 12, 6, and 4 in both studies. A descriptive subgroup analysis of treatment success based on the IGA at week 8, by baseline disease severity, was reported in both trials. None of the patients in the vehicle treatment groups with severe disease based on the IGA at baseline achieved treatment success. The proportion of patients with treatment success at week 8 was numerically greater for the HP/TAZ groups than vehicle for patients with both moderate and severe disease at baseline. The latter is consistent with the primary analysis; however, no firm conclusions can be drawn about any of the subgroups in the absence of formal pre-specified testing. The subgroup analyses are also limited by their sample size, as less than 20% of the overall population in each of the two trials were included in the subgroup analysis of patients with severe disease at baseline.
Productivity and treatment adherence were not evaluated as efficacy outcomes in any of the clinical trials included in the review, but dosing compliance was reported. Based on the intention-to-treat (ITT) population and compliance defined by using between 80% and 120% of the expected applications of the study drug while enrolled in the study, ▬▬▬▬▬ of patients were compliant with the study medication.
Harms Results
The percentage of patients that reported at least one adverse event (AE) up to the week 8 study visit in Study 301 and Study 302 was 36.8% and 35.0% for HP/TAZ groups and 19.4% and 23.3% for vehicle groups, respectively. At week 8, serious AEs were only reported by three (2.3%) patients in the HP/TAZ group in Study 301. No serious adverse events (SAEs) were reported in Study 302. The percentage of patients who stopped treatment due to AEs at week 8 was 7.5% and 5.1% in the HP/TAZ groups of Study 301 and Study 302, respectively, and 6.8% in the vehicle group of Study 302. There were no withdrawals due to adverse events (WDAEs) from the study drug in the vehicle group of Study 301. No deaths were reported in the two pivotal trials.
Of the notable harms included in the CADTH review protocol, pruritus was the most frequently occurring. It was reported among 3% of patients in the HP/TAZ treatment group of both Study 301 and Study 302, and 5.5% of patients in the vehicle treatment group of Study 302; pruritus was not reported in the vehicle group of Study 301. Skin atrophy and folliculitis was also reported and only occurred in the HP/TAZ treatment groups (skin atrophy, 3.0% and 0.7% and folliculitis, 1.5% and 2.2% in Study 301 and Study 302, respectively). In addition, events of burning sensation, skin irritation, and hypersensitivity were reported in both treatment groups in a small percentage of patients (≤ 1.5%). Severe dryness and hypothalamic pituitary adrenal (HPA) axis suppression were also included as notable harms in the CADTH review protocol, but neither occurred in either study.
Summary of Key Results From Pivotal and Protocol Selected Studies (ITT Set Unless Otherwise Indicated).
Critical Appraisal
The trials included in this review were both double-blind randomized controlled trials (RCTs). Both trials used an acceptable method of randomization and concealment of treatment allocation, and there was no reason to suggest randomization or blinding was compromised. Overall, there were no major differences between the baseline disease characteristics that would have significantly impacted the efficacy results. In terms of prior medication use, the vehicle treatment groups ▬▬▬▬▬. The reason for ▬▬▬▬▬ in the vehicle treatment groups is unknown.
The primary and secondary outcomes in both trials were treatment success, defined as at least a two-grade improvement from baseline in the IGA score and an IGA score of clear or almost clear (0 or 1). No evidence for the validity of the five-point IGA scale, which was used in the two trials, was identified for this review; however, the six-point Physician’s Global Assessment (PGA), which is the same scale as the IGA, but the assessor is a physician rather than investigator, was available. There was evidence of reliability in terms of test-retest data and internal consistency, but the outcome is subjective and has poor inter-rater reliability.12 The limitations of this scale introduce uncertainty to the interpretation of the primary and secondary outcomes, as well as the magnitude of observed treatment response. Also, HRQoL was evaluated using the DLQI, which is a validated disease-specific instrument. Between-group comparisons or formal statistical testing of the DLQI was not performed, therefore the conclusions that can be drawn based on this outcome are limited. An appropriate method was used to handle missing data (imputed using Markov chain Monte Carlo [MCMC]) and secondary outcomes of treatment success by IGA were controlled for multiplicity using a gated sequential testing procedure. Statistical testing was conducted for the improvement in the signs of psoriasis, but multiplicity was not controlled for this outcome. Statistical testing and imputation of missing data were not conducted for the BSA, preventing any conclusions that can be drawn from this data.
The two trials included several limitations regarding the generalizability of the results to the target population identified by the indication and Canadian clinical practice. The information available regarding disease severity of the patients who were included in the two trials may be more representative of Canadian patients with mild-to-moderate disease rather than moderate to severe. The mean percent BSA of patients at baseline in the two trials was between 5.5% and 6.5% for randomized patients, however, patients with a BSA as low as 3% were eligible for inclusion in the study. Definitions of disease severity are described in the Canadian Guidelines for the Management of Plaque Psoriasis, which suggest a BSA of 5% is used as an upper limit for mild disease and a BSA of 10% is used as a lower limit for moderate-to-severe psoriasis.4 However, the definitions of disease severity for psoriasis vary across international guidelines. For example, guidelines published by the American Academy of Dermatology and the National Psoriasis Foundation (US) defined disease severity by BSA involvement with less than 3% BSA considered mild, 3% to 10% BSA considered moderate, and greater than 10% BSA considered severe disease.13,14 The clinical expert consulted by CADTH acknowledged that there is lack of consensus regarding the definition of disease severity, but was of the view that the patient populations in Study 301 and Study 302 were more mild than what would be expected in a population of Canadian patients with moderate-to-severe psoriasis. Further, the evidence reviewed is not aligned with the anticipated use of HP/TAZ in clinical practice. According to the clinical expert consulted by CADTH for this review, it is unlikely that HP/TAZ would be used as monotherapy in patients with moderate-to-severe disease. The clinical expert advised that a more likely use of HP/TAZ in this population would be as an adjunct therapy to systemic treatment; however, evidence of this was not identified. Lastly, there is no direct evidence of HP/TAZ compared to other topical treatments for psoriasis, despite the number of available treatments for this disease area. As such, the comparative effectiveness of HP/TAZ remains uncertain.