European S3‐Guideline on the systemic treatment of psoriasis vulgaris – Update Apremilast and Secukinumab – <scp>EDF</scp> in cooperation with <scp>EADV</scp> and <scp>IPC</scp>

Abstract

An update of the European S3-Guidelines on the systemic treatment of psoriasis vulgaris – the European Dermatology Forum (EDF) in cooperation with the European Academy of Dermatology and Venereology (EADV) and the International Psoriasis Council (IPC – was published in December 2015.1, 2 In addition to the interventions discussed in the update, the Committee for Medicinal Products for Human Use (CHMP) of the European Medicines Agency (EMA) approved apremilast and secukinumab as new treatment options for psoriasis. The European Commission granted a marketing authorization for both treatments on 15 January 2015.3, 4 In February 2016, EMA also approved ixekizumab,5 which will be discussed in a further update as the expert group felt that at the time of the consensus conference, expert experience with ixekizumab was still too limited to allow conclusive discussion. The methods used to develop this amendment were in accordance with those of the previously published European S3-Guidelines on the systemic treatment of psoriasis vulgaris–Update 2015.1, 2 Details for the fast update can be found in the methods report (https://doi.org/10.1111/jdv.14470). A systematic search to identify relevant randomized controlled trials investigating apremilast and/or secukinumab was conducted in Medline, Medline In-Process, Embase and the Cochrane Library on 23.02.2016. Autoalerts were screened until 27 July 2016. Briefly, studies assessing the efficacy and safety of apremilast or secukinumab in patients with moderate-to-severe psoriasis were included. The literature screening was performed in collaboration (Stefanie Rosumeck, Corinna Dressler, Gayle van der Kraaij, Paula van Lumig, Marlies Wakkee). Originally, 559 hits were identified, 173 duplicates were removed. Through autoalerts, 14 additional hits were identified. After double inspection of 400 relevant title/abstracts, 16 full texts were included. The data were extracted using a standardized form, and outcome data were analysed using Review Manager.6 The risk of bias in included studies was assessed using the Cochrane Risk of Bias Assessment Tool.7 Evidence was summarized according to the system recommended by the GRADE (Grading of Recommendations Assessment, Development and Evaluation) working group.8 Drafts of the Dutch Psoriasis Guidelines9, 10 were taken into consideration and partly adapted. An online consensus conference was held on 02. November 2016 using the formal consensus methodology of the nominal group technique to agree upon recommendations. These recommendations along with their strengths are highlighted in grey boxes throughout the document. Standardized languages were used based on GRADE: Pretreatment During treatment Post-treatment No information is given in the summary of products characteristics (SmPC)11 for the duration of contraception after discontinuation of apremilast. Continuation of the contraception for five half-lives (5 × 9 h) is suggested. Strong consensus The recommended dose is 30 mg BID. An initial titration schedule is required as shown above (Table 2). ‘The most commonly reported adverse reactions in Phase III clinical studies have been gastrointestinal (GI) disorders including diarrhoea (15.7%) and nausea (13.9%). These GI adverse reactions were mostly mild to moderate in severity, with 0.3% of diarrhoea and 0.3% of nausea reported as being severe. These adverse reactions generally occurred within the first 2 weeks of treatment and usually resolved within 4 weeks (Table 3)’.11 Vomiting, dyspepsia, frequent bowel movements, upper abdominal pain gastroesophageal reflux disease, decreased appetite, upper respiratory infection, nasopharyngitis, bronchitis, cough, back pain, fatigue, insomnia, tension headache, migraine, depression ‘Patient weight was measured routinely in clinical studies. The mean observed weight loss in patients treated for up to 52 weeks with apremilast was 1.99 kg. A total of 14.3% of patients receiving apremilast had observed weight loss between 5% and 10% while 5.7% of the patients receiving apremilast had observed weight loss >10%. None of these patients had overt clinical consequences resulting from weight loss. A total of 0.1% of patients treated with apremilast discontinued due to adverse reaction of weight decreased’.11 The weight of underweight patients should be monitored from start of treatment. In case of inexplicable and significant weight loss, discontinuation of treatment should be considered. See below special patient population 5) Neurological and psychiatric disease. Phase 2/3 studies reported more upper respiratory infections with apremilast compared to placebo.12-14 There are no reactivations of tuberculosis or opportunistic infections reported.12-15 Screening for latent tuberculosis was not required before enrolment in the randomized clinical trials; however, a history of incompletely treated tuberculosis was an exclusion criterion.12-15 Surgery: Real-life data on perioperative management of apremilast have not yet become available. However, there is no evidence to date that continuous treatment with apremilast will lead to perioperative complications. Patients who need minor surgical treatments including dental treatments and skin surgery may continue apremilast treatment. In the case of major surgery, the decision of apremilast withdrawal should be taken case-by-case considering patient characteristics, the risk of infection, the risk of psoriasis worsening and after counselling with the surgeon. Co-administration of strong cytochrome P450 3A4 (CYP3A4) enzyme inducer, rifampicin, resulted in a reduction of systemic exposure of apremilast, which may result in a loss of efficacy of apremilast.16 Therefore, the use of strong CYP3A4 enzyme inducers (Table 4) with apremilast is not recommended. There was no clinically meaningful drug–drug interaction with ketoconazole, methotrexate and oral contraceptives.16 ‘In case of an overdose, it is recommended that the patient is monitored for any signs or symptoms of adverse effects and appropriate symptomatic treatment is instituted’.11 No dosage adjustment is necessary for elderly patients.11 No overall differences were observed in the safety profile of elderly patients ≥65 years of age and younger adult patients <65 years of age in the clinical studies. There is limited pharmacokinetic data in subjects over 75 years of age in clinical trials. The maximum concentration (Cmax) for apremilast in elderly subjects (65–85 years of age) is about 6% higher than that in young subjects (18–55 years of age).11 Five studies evaluating apremilast12-15, 17-19 were included in the evidence-based assessment. Summary of findings tables is available as Supporting information. Apremilast was found to be more effective than placebo during induction therapy based on PASI 75/90 (low/moderate quality), percentage reduction in PASI (low quality), PGA ‘clear/almost clear’ (low quality) and for absolute DLQI reduction (very low quality) but not based on PGA ‘clear’ (low quality). Significantly more patients experienced at least one AE in the apremilast groups compared to placebo (low quality). No difference was found for patients with at least one SAE (low quality) and for withdrawal due to AE (low quality). After response to apremilast, there was no difference in relapse rate after 16 weeks of treatment (moderate quality). Time till onset of action was faster for apremilast than placebo (very low quality). Apremilast was also found to be more effective than placebo in long-term therapy based on PASI 75/90 (moderate/low quality) and PGA ‘clear/almost clear’ (low quality). No safety data were available. No differences were found between apremilast 30 mg BID and apremilast 20 mg BID for induction therapy based on all included outcomes: PASI 75/90 (moderate quality), percentage PASI reduction (moderate quality), PGA ‘clear/almost clear’ (moderate quality), PGA ‘clear’ (moderate quality), absolute reduction in DLQI (high quality), patients with at least one AE (high quality), patients with at least one SAE (low quality) and withdrawal due to AE (moderate quality). Time until onset of action was marginally faster in apremilast 30 mg BID than apremilast 20 mg BID (low quality). No differences could be found for apremilast 30 mg BID compared to apremilast 20 mg BID for long-term therapy for all included outcomes: PASI 75/90 (moderate quality), PGA ‘clear/almost clear’ (moderate quality), PGA ‘clear’ (moderate quality), absolute reduction in DLQI (high quality), patients with at least one AE (moderate quality) and withdrawal due to AE (high quality). Apremilast 30 mg BID is superior to apremilast 10 mg BID in the induction therapy based on PASI 75 (high quality), percentage PASI reduction (moderate quality) and PGA ‘clear/almost clear’ (high quality). The higher dose was not superior to the lower dose based on PASI 90 (moderate quality), PGA ‘clear’ (low quality) and absolute reduction in DLQI (high quality). Significantly more patients experienced at least one AE (moderate quality) or withdrew due to an AE (high quality) in the higher dose groups but no differences could be found for patients with at least one SAE (low quality). Time until onset of action was faster in apremilast 30 mg BID than apremilast 10 mg BID/20 mg QW (low quality). Apremilast 30 mg BID is superior to apremilast 10 mg BID in long-term treatment based on PASI 75/90 (high quality) and PGA ‘clear/almost clear’ (high quality). The higher dose was not superior to the lower dose based on PGA ‘clear’ (moderate quality) and absolute reduction in DLQI (high quality). No differences could be found for patients with at least one AE (moderate quality) or withdrawal due to AE (low quality). Apremilast 20 mg BID is superior to apremilast 10 mg BID/20 mg QD in the induction therapy based on PASI 75 (high quality), percentage PASI reduction (moderate quality) and PGA ‘clear/almost clear’ (moderate quality) and absolute reduction in DLQI (moderate quality). The higher dose was not superior based on PASI 90 (moderate quality) or PGA ‘clear’ (low quality). No differences could be found for patients with at least one AE (low quality), patients with at least one SAE (moderate quality) or withdrawal due to AE (very low quality). After response to apremilast, there was no difference in relapse rate within 16 weeks of treatment (moderate quality). Time until onset of action was faster for apremilast 20 mg BID than apremilast 10 mg BID/20 mg QW (low quality). No differences could be found between apremilast 20 mg BID and apremilast 10 mg BID/QD in long-term treatment based on PASI 75/90 (moderate quality), PGA ‘clear/almost clear’ (moderate quality) and PGA ‘clear’ (moderate quality). The higher dose was only superior based on absolute reduction in DLQI (moderate quality). No differences could be found for patients with at least one AE (moderate quality) or withdrawal due to AE (low quality). No differences were found between apremilast and etanercept for induction therapy based on all included outcomes: PASI 75 (low quality), PGA ‘clear/almost clear’ (low quality), patients with at least one AE (low quality), patients with at least one SAE (very low quality). TB screening and monitoring in patients receiving apremilast is not required (EMA/FDA approved) according to label. Screening for latent tuberculosis was not required before enrolment in the randomized clinical trials. As of 20 June 2016, a total of 117 728 patients have been exposed to apremilast. Among these patients, three reports of tuberculosis have been reported. Of the three reports of tuberculosis, two had insufficient information of assessment. The last report was a case of latent tuberculosis. It is not known when the diagnosis of latent tuberculosis was made in relation to the initiation of apremilast, but the patient continued treatment with apremilast. (personal communication Celgene, email Ian Parson, 19 December 2016) ‘The pharmacokinetics of apremilast and its major metabolite M12 are not affected by moderate or severe hepatic impairment. This was studied in patients without psoriasis or psoriatic arthritis. No dose adjustment is necessary for patients with hepatic impairment’.11 There are no available data on HIV-infected patients treated with apremilast. For general recommendations on treatment of psoriasis in patients with HIV see 2015 update of EU Psoriasis guidelines.1, 2 If treatment with apremilast is to be considered in patients with no other alternatives, the decision should be taken in collaboration with the infectious disease specialist. The two serious neurological events that have complicated therapy for psoriasis to date (demyelinating diseases with TNF inhibitors and progressive multifocal leucoencephalopathy with efalizumab, fumaric acid esters) are rare and became evident some years after licensing. This needs to be remembered when considering the safety profile of apremilast and secukinumab in the context of neurological disorders. Headaches (tension), which can be serious enough to precipitate drug withdrawal, are frequent and reported with other PDE4 inhibitors. Depression is mentioned in the summary of product characteristics (SmPC) as a potential side effect, based on findings during the placebo-controlled period of phase III clinical trials with 1.2% (14/1184) of patients treated with apremilast reporting depression compared to 0.5% (2/418) with placebo.11 ‘Postmarketing data up to 20 March 2016 reported 65 cases distributed as follows: five completed suicides, four suicide attempts, 50 cases of suicidal ideation, five cases of depression suicidal and one case of suicidal behaviour. In 32 cases of 65, for which information was available, the patients reported improvement after treatment discontinuation. (From launch to 20 March 2016, there were approximately 105 000 patients exposed to apremilast.)’20 There is no evidence that apremilast could affect cardiovascular risk. Congestive heart failure is not a contraindication to apremilast use. There is no evidence that apremilast could affect insulin resistance. Diabetes is not a contraindication for apremilast use. According to the label, patients with mild-to-moderate renal impairment do not require dose adjustment. ‘The dose of apremilast should be reduced to 30 mg once daily in patients with severe renal impairment (creatinine clearance of <30 mL per minute)’11 (initial dose titration using only the morning dose). In the pivotal clinical trials, there was no evidence for treatment emergent adverse events related to renal function.5, 6 Apremilast is contraindicated during pregnancy. Pregnancy should be excluded before treatment can be initiated. There are limited data about the use of apremilast in pregnant women. Apremilast should not be used during breast-feeding. ‘No fertility data is available in humans. In animal studies in mice, no adverse effects on fertility were observed in males at exposure levels threefold clinical exposure and in females at exposure levels onefold clinical exposure.’11 Adapted from Ann Rheum Dis 2016;75:499-510, https://doi.org/10.1136/annrheumdis-2015-208337 European League Against Rheumatism (EULAR) recommendations for the management of psoriatic arthritis with pharmacological therapies: 2015 update21 There is neither published information nor mentioning in the SMPC about the use of apremilast in the context of vaccination. However, live vaccinations were permitted in patients enrolled in the randomized controlled clinical trials. Pretreatment During treatment Post-treatment Strong consensus ‘The recommended dose is 300 mg of secukinumab by subcutaneous injection with initial dosing at Weeks 0, 1, 2 and 3, followed by monthly maintenance dosing starting at Week 4. Each 300 mg dose is given as two subcutaneous injections of 150 mg’.22 In the placebo-controlled period of clinical studies in plaque psoriasis, infections were reported in 28.7% of patients treated with secukinumab and 18.9% of patients with placebo. Most cases of infection were mild or moderate upper respiratory tract infections which did not require treatment discontinuation. Mucosal or cutaneous candidiasis was more frequent with secukinumab. Cases responded to standard treatment and did not require treatment discontinuation.22 Most cases of neutropenia were mild transient and reversible. Grade 3 neutropenia was observed in 0.5% of patients with no dose dependency or temporal relationship to infection in most cases. Secukinumab specific antidrug antibodies were detected by a Meso Scale Discovery bridging assay (sensitivity: 4 ng/mL). Among 2842 patients who participated in six phase II clinical studies, a total of 11 patients (0.4%) developed antidrug antibodies of whom three developed neutralizing antidrug antibodies.23 The effect of secukinumab on Crohn's disease was studied in a randomized placebo-controlled proof-of-concept trial.24 Secukinumab 2 × 10 mg/kg was administered i.v. on day 1 and day 22. The study was prematurely discontinued due to lack of effect. Four of 39 patients reported exacerbations of Crohn's disease. In the phase III psoriasis clinical trial program, three cases of Crohn's disease were reported as serious adverse events of which two were exacerbations of pre-existing disease.25 In patients with psoriasis and Crohn's disease, caution should be exercised and may be considered before using secukinumab. Surgery: Real-life data on perioperative management of secukinumab have not yet become available. However, there is no evidence to date that continuous treatment with secukinumab will lead to perioperative complications. Patients who need minor surgical treatments including dental treatments and skin surgery may continue secukinumab treatment. In the case of major surgery, the decision of secukinumab withdrawal should be taken case-by-case considering patient characteristics, the risk of infection, the risk of psoriasis worsening and after counselling with the surgeon. of secukinumab with other for or have not been no effect on The effect of secukinumab may levels and with the of those with a as monitoring of should be considered while starting secukinumab. No cases of have been reported. of up to 30 mg/kg have been administered in clinical studies. In case of overdose, the patient should be monitored and appropriate symptomatic treatment be on population pharmacokinetic with a limited of elderly patients for age ≥65 years and for age clearance in elderly patients and patients <65 years of age was studies evaluating secukinumab were included in the evidence-based assessment. The summary of findings tables are as Supporting information. Secukinumab was found to be more effective than placebo in the induction therapy based on PASI 75/90 (high quality) and PGA ‘clear’ and ‘clear/almost clear’ (high quality). Significantly more patients with at least one AE were reported in the secukinumab groups than in the placebo groups (high quality), but not clinically and no difference was found for patients with at least one SAE (moderate quality) and for withdrawal due to AE (moderate quality). of action was more for 150 mg and 300 mg secukinumab compared to placebo (low quality). Secukinumab was also found to be more effective than placebo in long-term therapy based on PASI 75/90 (high quality) and PGA ‘clear/almost clear’ (high quality). No data on adverse events were available. Secukinumab 300 mg is superior to secukinumab 150 mg during the induction treatment with to efficacy based on PASI 75 (high but not clinically PASI 90 (high quality), PGA ‘clear’ and ‘clear/almost clear’ (moderate quality). No significant difference was found for patients with at least one AE (high quality), patients with at least one SAE (moderate quality) and withdrawal due to AE (moderate quality). Time till onset of action was for 300 mg compared to 150 mg secukinumab (low quality). Secukinumab 300 mg is superior to secukinumab 150 mg in long-term therapy based on PASI 75 (high but not clinically PASI 90 (moderate quality), PGA ‘clear’ (moderate quality) and PGA ‘clear/almost clear’ (high quality). Secukinumab 150 mg was found to be more effective than secukinumab 150 mg in the induction therapy based on PASI 75/90 (moderate quality) and PGA ‘clear/almost clear’ (moderate quality). No difference was found patients with at least one AE (high quality), patients with at least one SAE (moderate quality) and withdrawal due to AE (low quality). Secukinumab was superior to etanercept in the induction phase based on PASI 75/90 (high quality), PGA ‘clear/almost clear’ (high quality) and PGA ‘clear’ (high quality). No difference was found patients with at least one AE (high quality), patients with at least one SAE (high quality) and withdrawal due to AE (high quality). of action was more for 300 mg and 150 mg secukinumab when compared to etanercept (low quality). Secukinumab was superior to etanercept in long-term therapy based on PASI 75/90 (moderate quality), PGA ‘clear/almost clear’ (low quality) and PGA ‘clear’ (moderate quality). No difference was found in to withdrawal due to AE (low quality). Secukinumab was superior to in the induction phase based on PASI 75 (high but not clinically PASI 90 (moderate quality), PGA ‘clear/almost clear’ (moderate quality) and PGA ‘clear’ (high quality). No difference was found patients with at least one AE (high quality), patients with at least one SAE (moderate quality) and withdrawal due to AE (low quality). Time till onset of action was for secukinumab 300 mg compared to (low quality). risk of risk of risk of used in risk of ‘No pharmacokinetic data are available in patients with hepatic impairment. are and hepatic impairment is not to clearance of There are no available data on HIV-infected patients treated with secukinumab. For general recommendations on treatment of psoriasis in patients with HIV see 2015 update of EU Psoriasis If treatment with secukinumab is to be considered in patients with no other alternatives, the decision should be taken in collaboration with the infectious disease with monitoring for The two serious neurological events that have complicated therapy for psoriasis to date (demyelinating diseases with TNF inhibitors and progressive multifocal leucoencephalopathy with efalizumab, fumaric acid esters) are and became evident some years after licensing. This needs to be remembered when considering the safety profile of apremilast and secukinumab in the context of neurological disorders. There is no that secukinumab is with specific adverse events in patients with neurological Secukinumab a reduction in in a proof-of-concept study in not a in heart failure but may be in disease. The studies did not an risk of major cardiovascular events in patients treated with secukinumab compared to placebo. A significant of patients enrolled in the clinical trials had There is no evidence that secukinumab any effect on In the pivotal clinical trials, there was no evidence for treatment emergent adverse events related to renal of potential should use an effective of contraception during treatment and for at least 20 weeks after treatment. There are no data from the use of secukinumab in pregnant women. studies do not or effects with to or As a it is to the use of secukinumab in is not known secukinumab is in are in and it is not known secukinumab is after of the potential for adverse reactions in from a decision on to during treatment and up to 20 weeks after treatment or to therapy with secukinumab be made into the of to the and the of secukinumab therapy to the ‘The effect of secukinumab on fertility not been studies do not or effects with to Adapted from Ann Rheum Dis https://doi.org/10.1136/annrheumdis-2015-208337 European League Against Rheumatism (EULAR) recommendations for the management of psoriatic arthritis with pharmacological therapies: 2015 Patients treated with secukinumab may or There is published evidence that after or in subjects during treatment with is According to the live should not be given secukinumab