L Puig1. 1. Department of Dermatology, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain; Universitat Autònoma de Barcelona, Barcelona, Spain.
Abstract
WHAT IS KNOWN AND OBJECTIVE: To assess the equivalent efficacy of two or more treatments, clinical relevance must be adequately established, but there appears to be some confusion between clinical relevance and differences in absolute risk reduction or effect size, which quantifies the size of the difference between two groups. Delta values used for sample size calculation in clinical trials should take account of clinical relevance. We aim to illustrate this confusion as regards biologics for treatment of moderate-to-severe psoriasis. COMMENT: In psoriasis, treatment success has been defined as achieving lesion response to 'clear or almost clear', ≥ 90% improvement with respect to baseline Psoriasis Area and Severity Index (PASI), or achieving a PASI score lower than 3 on treatment. A PASI 75 response (≥ 75% improvement with respect to baseline) is often taken as a meaningful cut-off for clinical relevance. A recently published meta-analysis using a random-effects model showed that ustekinumab use was associated with statistically significantly higher odds for achieving PASI 75 response compared with adalimumab use (OR, 1·84; 95% credible interval, 1·01-3·54) and etanercept use (2·07; 1·42-3·06), but lower odds for achieving PASI 75 compared with infliximab use (0·36; 0·14-0·82). The inference is that this magnitude of response in PASI will translate into a perceptible clinical improvement by patients. This need not be the case. In particular, this measure does not take account of the adverse effect profiles of the different agents. Only by taking account of these different and opposing effects, such as through use of validated quality-of-life indices, can one make robust inferences on clinical equivalence. WHAT IS NEW AND CONCLUSION: From a clinical perspective, biologics for the treatment of plaque psoriasis should not be considered equivalents solely on the basis of PASI responses. Choice between these agents requires accounting for their relative safety and efficacy profiles, as well as patient-reported outcome measures. At the individual patient level, other factors such as individual contraindications must be taken into account.
WHAT IS KNOWN AND OBJECTIVE: To assess the equivalent efficacy of two or more treatments, clinical relevance must be adequately established, but there appears to be some confusion between clinical relevance and differences in absolute risk reduction or effect size, which quantifies the size of the difference between two groups. Delta values used for sample size calculation in clinical trials should take account of clinical relevance. We aim to illustrate this confusion as regards biologics for treatment of moderate-to-severe psoriasis. COMMENT: In psoriasis, treatment success has been defined as achieving lesion response to 'clear or almost clear', ≥ 90% improvement with respect to baseline Psoriasis Area and Severity Index (PASI), or achieving a PASI score lower than 3 on treatment. A PASI 75 response (≥ 75% improvement with respect to baseline) is often taken as a meaningful cut-off for clinical relevance. A recently published meta-analysis using a random-effects model showed that ustekinumab use was associated with statistically significantly higher odds for achieving PASI 75 response compared with adalimumab use (OR, 1·84; 95% credible interval, 1·01-3·54) and etanercept use (2·07; 1·42-3·06), but lower odds for achieving PASI 75 compared with infliximab use (0·36; 0·14-0·82). The inference is that this magnitude of response in PASI will translate into a perceptible clinical improvement by patients. This need not be the case. In particular, this measure does not take account of the adverse effect profiles of the different agents. Only by taking account of these different and opposing effects, such as through use of validated quality-of-life indices, can one make robust inferences on clinical equivalence. WHAT IS NEW AND CONCLUSION: From a clinical perspective, biologics for the treatment of plaque psoriasis should not be considered equivalents solely on the basis of PASI responses. Choice between these agents requires accounting for their relative safety and efficacy profiles, as well as patient-reported outcome measures. At the individual patient level, other factors such as individual contraindications must be taken into account.