Efficacy and Safety of Biologics in Erythrodermic Psoriasis
Efficacy and Safety of Biologics in Erythrodermic Psoriasis
EP is recognized as one of the rare psoriatic disease clinical subtypes associated with significant direct mortality. Indeed, two studies published 40 and 20 years ago reported mortality rates of 15% and 9%, respectively, with a 5% and 2% mortality rate attributable to psoriasis, despite the use of conventional systemic treatments such as retinoids and methotrexate. A French national survey focusing on lethal outcome in psoriasis directly related to the skin disease or its therapy observed that among 46 patients who died, 28 (60%) had EP. The overlap between plaque-type psoriasis and EP on the one hand, and the recent labelling of several biological drugs in moderate-to-severe plaque-type psoriasis on the other, opened new perspectives for the treatment of patients with EP. Up to now, only one prospective study addressed the risk/benefit ratio of a biological drug in EP, investigating in this latter case the impact of etanercept treatment in 10 patients, and showing promising results regarding short-term efficacy (Table 5). Indeed, after 12 weeks of treatment with etanercept, 50% of patients reached a PASI 75 response following a 25 mg twice weekly dosage, a rate which exceeds the 30–34% response rate observed with this dosage in clinical trials in moderate-to-severe psoriasis. Regarding infliximab, short series of five to seven patients have been published with excellent short-term results as shown by the 40–100% rates of patients reaching a PASI 90 improvement. A case series of eight patients with EP from Asia treated with ustekinumab has recently been published with 37·5% of patients reaching PASI 90 at week 28 (Table 5). So far, the efficacy of efalizumab or adalimumab in EP has only been reported in isolated cases. Our results, obtained in patients showing failure to conventional systemic treatments in most cases, support the good short-term efficacy of biologics in EP, with a rate of patients reaching a PASI 75 or BSA improvement close to 50% for the different TNF-α inhibitors at the 12-week endpoint. Infliximab was by far the biologic the most frequently used as first line, which is in keeping with its high, rapid efficacy, as stated in most current guidelines supporting infliximab as first-line BT for severe, unstable forms of psoriasis. Indeed, half of our patients who received infliximab reached a significant improvement of their skin condition 4 weeks after treatment onset, providing striking confirmation of its rapidly efficient profile, particularly important to obtain in a life-threatening disease. On the other hand, good efficacy was also observed in the subgroup of patients receiving etanercept and adalimumab, although the small sample size and the retrospective nature of the study preclude any comparative analysis between fusion protein and the different monoclonal antibodies. Nevertheless, efficacy results observed in the etanercept group of the present study provide additional support to conclusions from the study previously published by Italian authors. Regarding efalizumab, a CD11a antagonist now withdrawn from market authorization, no efficacy was observed in two different patients, as first-line or as third-line BT for EP. The limited efficacy observed in the small group of patients receiving ustekinumab should be interpreted cautiously, as this IL-12/23 p40 antagonist was prescribed as a fourth-line BT in all cases, e.g. in most recalcitrant cases.
Only one-third of study patients still received the initial biological drug 1 year after onset of treatment, without major differences between the three drugs targeting TNF-α. The main reason for drug discontinuation was lack of efficacy (62·5%), followed by incidence of side-effects (33%). This drug survival rate in patients with EP appears lower than the one observed in patients with psoriasis vulgaris treated with TNF-α blockers, as stated by a recent analysis of data extracted from the Danish nationwide database DERMBIO. In this latter study, the 4-year drug survival rate in patients naive of biologics was 70% for infliximab and 40% for adalimumab and etanercept, with a much lower rate of 10% in patients who failed one or more TNF-α blocker. The reasons for discontinuation were predominantly the loss or lack of efficacy (75%) and side-effects (12%). The low TNF-α blocker survival rate observed in patients with EP can be explained by a lesser efficacy in EP than in psoriasis vulgaris but also by an increased number of side-effects in these predisposed patients.
Considering safety, the AEs observed in our study were in keeping with previously identified safety risks in patients receiving these biologics, mainly bacterial infection for all currently approved biologics, and immunoallergic reactions related to infliximab due to the chimeric structure of this latter drug. Most SAEs in our patients were of infectious nature with skin origin, as expected in patients with EP who show major alteration of the skin barrier. Of note, these infectious events were observed with all types of biologics (anti-TNF-α, anti-IL-12/23 or anti-CD11a) and were not restricted to monoclonal antibodies among the class of TNF-α inhibitors. Thus, while it remains true that the incidence of bacterial and mycobacterial severe infection has been shown to be higher in patients with rheumatoid arthritis receiving anti-TNF-α monoclonal antibodies in comparison with etanercept, no firm conclusion can be drawn about respective safety evaluation of different biologics regarding infection in the frame of the present study, due to its limited sample size and its retrospective nature. The fatal outcome observed in one patient with several cardiovascular risk factors receiving ustekinumab raised, among other hypotheses, a cardiovascular origin, but we were unable to ascertain that this patient's fatal course met all criteria for sudden death. However, recent evidence for nonstatistically significant imbalance of major adverse cardiovascular events between anti-IL-12/23 p40 antibodies and placebo groups in randomized phases of clinical trials emphasizes the need to address this risk in prospective large-scale cohorts of patients.
Finally, only prospective comparative controlled trials will allow us to address the risk/benefit ratio of each drug in patients with EP, as well as to balance this tolerance profile with the main disease risk. However, the present study provides additional rational basis for the use of TNF-α inhibitors as first-line treatment in EP showing failure, contraindication or intolerance to conventional antipsoriatic systemic therapies.
Discussion
EP is recognized as one of the rare psoriatic disease clinical subtypes associated with significant direct mortality. Indeed, two studies published 40 and 20 years ago reported mortality rates of 15% and 9%, respectively, with a 5% and 2% mortality rate attributable to psoriasis, despite the use of conventional systemic treatments such as retinoids and methotrexate. A French national survey focusing on lethal outcome in psoriasis directly related to the skin disease or its therapy observed that among 46 patients who died, 28 (60%) had EP. The overlap between plaque-type psoriasis and EP on the one hand, and the recent labelling of several biological drugs in moderate-to-severe plaque-type psoriasis on the other, opened new perspectives for the treatment of patients with EP. Up to now, only one prospective study addressed the risk/benefit ratio of a biological drug in EP, investigating in this latter case the impact of etanercept treatment in 10 patients, and showing promising results regarding short-term efficacy (Table 5). Indeed, after 12 weeks of treatment with etanercept, 50% of patients reached a PASI 75 response following a 25 mg twice weekly dosage, a rate which exceeds the 30–34% response rate observed with this dosage in clinical trials in moderate-to-severe psoriasis. Regarding infliximab, short series of five to seven patients have been published with excellent short-term results as shown by the 40–100% rates of patients reaching a PASI 90 improvement. A case series of eight patients with EP from Asia treated with ustekinumab has recently been published with 37·5% of patients reaching PASI 90 at week 28 (Table 5). So far, the efficacy of efalizumab or adalimumab in EP has only been reported in isolated cases. Our results, obtained in patients showing failure to conventional systemic treatments in most cases, support the good short-term efficacy of biologics in EP, with a rate of patients reaching a PASI 75 or BSA improvement close to 50% for the different TNF-α inhibitors at the 12-week endpoint. Infliximab was by far the biologic the most frequently used as first line, which is in keeping with its high, rapid efficacy, as stated in most current guidelines supporting infliximab as first-line BT for severe, unstable forms of psoriasis. Indeed, half of our patients who received infliximab reached a significant improvement of their skin condition 4 weeks after treatment onset, providing striking confirmation of its rapidly efficient profile, particularly important to obtain in a life-threatening disease. On the other hand, good efficacy was also observed in the subgroup of patients receiving etanercept and adalimumab, although the small sample size and the retrospective nature of the study preclude any comparative analysis between fusion protein and the different monoclonal antibodies. Nevertheless, efficacy results observed in the etanercept group of the present study provide additional support to conclusions from the study previously published by Italian authors. Regarding efalizumab, a CD11a antagonist now withdrawn from market authorization, no efficacy was observed in two different patients, as first-line or as third-line BT for EP. The limited efficacy observed in the small group of patients receiving ustekinumab should be interpreted cautiously, as this IL-12/23 p40 antagonist was prescribed as a fourth-line BT in all cases, e.g. in most recalcitrant cases.
Only one-third of study patients still received the initial biological drug 1 year after onset of treatment, without major differences between the three drugs targeting TNF-α. The main reason for drug discontinuation was lack of efficacy (62·5%), followed by incidence of side-effects (33%). This drug survival rate in patients with EP appears lower than the one observed in patients with psoriasis vulgaris treated with TNF-α blockers, as stated by a recent analysis of data extracted from the Danish nationwide database DERMBIO. In this latter study, the 4-year drug survival rate in patients naive of biologics was 70% for infliximab and 40% for adalimumab and etanercept, with a much lower rate of 10% in patients who failed one or more TNF-α blocker. The reasons for discontinuation were predominantly the loss or lack of efficacy (75%) and side-effects (12%). The low TNF-α blocker survival rate observed in patients with EP can be explained by a lesser efficacy in EP than in psoriasis vulgaris but also by an increased number of side-effects in these predisposed patients.
Considering safety, the AEs observed in our study were in keeping with previously identified safety risks in patients receiving these biologics, mainly bacterial infection for all currently approved biologics, and immunoallergic reactions related to infliximab due to the chimeric structure of this latter drug. Most SAEs in our patients were of infectious nature with skin origin, as expected in patients with EP who show major alteration of the skin barrier. Of note, these infectious events were observed with all types of biologics (anti-TNF-α, anti-IL-12/23 or anti-CD11a) and were not restricted to monoclonal antibodies among the class of TNF-α inhibitors. Thus, while it remains true that the incidence of bacterial and mycobacterial severe infection has been shown to be higher in patients with rheumatoid arthritis receiving anti-TNF-α monoclonal antibodies in comparison with etanercept, no firm conclusion can be drawn about respective safety evaluation of different biologics regarding infection in the frame of the present study, due to its limited sample size and its retrospective nature. The fatal outcome observed in one patient with several cardiovascular risk factors receiving ustekinumab raised, among other hypotheses, a cardiovascular origin, but we were unable to ascertain that this patient's fatal course met all criteria for sudden death. However, recent evidence for nonstatistically significant imbalance of major adverse cardiovascular events between anti-IL-12/23 p40 antibodies and placebo groups in randomized phases of clinical trials emphasizes the need to address this risk in prospective large-scale cohorts of patients.
Finally, only prospective comparative controlled trials will allow us to address the risk/benefit ratio of each drug in patients with EP, as well as to balance this tolerance profile with the main disease risk. However, the present study provides additional rational basis for the use of TNF-α inhibitors as first-line treatment in EP showing failure, contraindication or intolerance to conventional antipsoriatic systemic therapies.