Inclusion and exclusion criteria in phase III trials with systemic agents in psoriasis: the external validity of drug development

Dear Editor, Better understanding of the pathogenesis of psoriasis has allowed the development of new drugs. The efficacy of these drugs was demonstrated in phase III clinical trials which led to drug approval.1 Clinical trials are performed on a specific patient population carefully selected with the help of inclusion criteria (IC) and exclusion criteria (EC).2 These criteria, if too stringent, can reduce dramatically the scope of the trial population and impair the generalizability of the results to the general psoriasis population.3 The current evolution in drug‐trial design in psoriasis suggests an increase in the number of EC over the past 15 years. It has recently been shown that applying the EC from drug trials extracted from publications to a general psoriasis population would exclude a high proportion of patients with psoriasis.4 The objective of our study was to evaluate the evolution in IC and EC in phase III psoriasis trials for systemic agents over the past 15 years. This study represents an analysis of IC and EC in psoriasis trials for biological agents (antitumour necrosis factor‐α, ustekinumab, efalizumab, anti‐interleukin (IL)‐17‐molecules) and small molecules (apremilast, tofacitinib) evaluated in randomized placebo‐controlled multicentre phase III trials performed between 2000 and 2015. Data was extracted from original publications, protocol summaries on www.clinicaltrials.gov and full protocols provided by pharmaceutical companies. As a first step we asked pharmaceutical companies that have developed systemic agents in psoriasis for original protocols or original text in protocols about IC and EC for phase III studies. We focused on phase III placebo‐controlled studies because these are the pivotal studies evaluating efficacy and safety of new molecules. We did not include active comparator studies in order not to contaminate the evaluation with inclusion/exclusion criteria related to the active comparator. For the 12 protocols obtained from pharmaceutical companies on 10 different molecules, we analysed the 12 related original primary publications and the 12 protocol summaries published on clinicaltrial.gov. Criteria were analysed overall and by period: agents developed between 2000 and 2009 (period I) and between 2010 and 2015 (period II). There were no significant changes in the number and in the wording of IC over time. All phase III studies included adult patients aged 18 years or older with moderate to severe plaque‐type psoriasis and eligible for phototherapy and/or systemic treatment. Figure 1 shows the mean number of EC during the two periods. The mean number of EC in protocols during period I was 48 with a maximum of 59, in period II 59 with a maximum of 83. More restrictive wordings of EC were observed during the second period (Table 1). Publications and the www.clinicaltrial.gov database listed only a limited number of EC included in original protocols with a mean number of 14 EC in publications and 14·3 in protocols available on clinicaltrials.gov. Most EC related to comorbidities were listed only in full protocols. For example, the history of viral infection, hypertension, history of basal cell carcinoma or actinic keratosis and EC concerning blood values were listed only in full protocols. Mean number of exclusion criteria over time (two time periods). Clinic.tri.gov, www.clinicaltrials.gov; EC, exclusion criteria Most restrictive exclusion criteria period I vs. period II Age > 70 years nonplaque psoriasis serious infection within 2 months serious infection requiring hospitalization or intravenous antibiotics within the previous 2 months history of cancer (other than nonmelanotic skin cancer and cervix cancer) history of serious cardiac disease within 12 weeks Haemoglobin < 8·5 g dL−1 (for men) and < 8·0 gdL−1 (for women) history of substance abuse within the last 5 years nonplaque psoriasis clinically significant flare of psoriasis during the 12 weeks prior to baseline active infection by mycobacterium tuberculosis within 3 years prior to screening recurrent or chronic infections or other active infection that is clinically significant in the opinion of the investigator within 6 months a serious infection, defined as requiring hospitalization or intravenous anti‐infectives within 6 months prior to the first dose history of disseminated herpes zoster or herpes simplex or dermatomal herpes zoster basal cell carcinoma or actinic keratosis in the past 12 weeks history of cancer (other than nonmelanotic skin cancer and cervix cancer) history of clinically significant cardiac disease history of clinically significant pulmonary, hepatic, renal, haematological, neurological, psychiatric and immunological disease first‐degree relative with a hereditary immunodeficiency haemoglobin < 11·0 g dL−1 ECG abnormalities clinically significant have donated blood of more than 500 mL within the last 4 weeks Age > 70 years nonplaque psoriasis serious infection within 2 months serious infection requiring hospitalization or intravenous antibiotics within the previous 2 months history of cancer (other than nonmelanotic skin cancer and cervix cancer) history of serious cardiac disease within 12 weeks Haemoglobin < 8·5 g dL−1 (for men) and < 8·0 gdL−1 (for women) history of substance abuse within the last 5 years nonplaque psoriasis clinically significant flare of psoriasis during the 12 weeks prior to baseline active infection by mycobacterium tuberculosis within 3 years prior to screening recurrent or chronic infections or other active infection that is clinically significant in the opinion of the investigator within 6 months a serious infection, defined as requiring hospitalization or intravenous anti‐infectives within 6 months prior to the first dose history of disseminated herpes zoster or herpes simplex or dermatomal herpes zoster basal cell carcinoma or actinic keratosis in the past 12 weeks history of cancer (other than nonmelanotic skin cancer and cervix cancer) history of clinically significant cardiac disease history of clinically significant pulmonary, hepatic, renal, haematological, neurological, psychiatric and immunological disease first‐degree relative with a hereditary immunodeficiency haemoglobin < 11·0 g dL−1 ECG abnormalities clinically significant have donated blood of more than 500 mL within the last 4 weeks Most restrictive exclusion criteria period I vs. period II Age > 70 years nonplaque psoriasis serious infection within 2 months serious infection requiring hospitalization or intravenous antibiotics within the previous 2 months history of cancer (other than nonmelanotic skin cancer and cervix cancer) history of serious cardiac disease within 12 weeks Haemoglobin < 8·5 g dL−1 (for men) and < 8·0 gdL−1 (for women) history of substance abuse within the last 5 years nonplaque psoriasis clinically significant flare of psoriasis during the 12 weeks prior to baseline active infection by mycobacterium tuberculosis within 3 years prior to screening recurrent or chronic infections or other active infection that is clinically significant in the opinion of the investigator within 6 months a serious infection, defined as requiring hospitalization or intravenous anti‐infectives within 6 months prior to the first dose history of disseminated herpes zoster or herpes simplex or dermatomal herpes zoster basal cell carcinoma or actinic keratosis in the past 12 weeks history of cancer (other than nonmelanotic skin cancer and cervix cancer) history of clinically significant cardiac disease history of clinically significant pulmonary, hepatic, renal, haematological, neurological, psychiatric and immunological disease first‐degree relative with a hereditary immunodeficiency haemoglobin < 11·0 g dL−1 ECG abnormalities clinically significant have donated blood of more than 500 mL within the last 4 weeks Age > 70 years nonplaque psoriasis serious infection within 2 months serious infection requiring hospitalization or intravenous antibiotics within the previous 2 months history of cancer (other than nonmelanotic skin cancer and cervix cancer) history of serious cardiac disease within 12 weeks Haemoglobin < 8·5 g dL−1 (for men) and < 8·0 gdL−1 (for women) history of substance abuse within the last 5 years nonplaque psoriasis clinically significant flare of psoriasis during the 12 weeks prior to baseline active infection by mycobacterium tuberculosis within 3 years prior to screening recurrent or chronic infections or other active infection that is clinically significant in the opinion of the investigator within 6 months a serious infection, defined as requiring hospitalization or intravenous anti‐infectives within 6 months prior to the first dose history of disseminated herpes zoster or herpes simplex or dermatomal herpes zoster basal cell carcinoma or actinic keratosis in the past 12 weeks history of cancer (other than nonmelanotic skin cancer and cervix cancer) history of clinically significant cardiac disease history of clinically significant pulmonary, hepatic, renal, haematological, neurological, psychiatric and immunological disease first‐degree relative with a hereditary immunodeficiency haemoglobin < 11·0 g dL−1 ECG abnormalities clinically significant have donated blood of more than 500 mL within the last 4 weeks EC concerning patient demographics and psoriasis characteristics did not change over time with one exception: one study in the second period excluded patients with a significant flare of psoriasis during the 12 weeks prior to baseline. All protocols excluded patients with positive serology for HIV, hepatitis B and C, positive history of opportunistic infection, and active or latent tuberculosis without prior prophylaxis. There was no trend over time in the increase in the number of EC related to infection (between three and eight per protocol). However, the wording of criteria made them more restrictive over time, excluding a higher number of patients at potential risk. During the first period, patients were excluded if they presented an active or history of an ongoing, chronic, recurrent or serious infection requiring hospitalization or the use of intravenous anti‐infectives within 8 weeks prior to the first dose. This time window was enlarged to 6 months, and exclusion of all patients with only a history of disseminated or dermatomal herpes zoster or simplex was introduced in recent protocols. History of cancer other than nonmelanoma skin cancer and in situ cervix cancer was an exclusion criterion in all protocols. Most protocols excluded patients with history of cancer within 5 years with a trend to exclude patients with a history of cancer without any time limitations in the second period. In one recent anti‐IL‐17‐protocol, introduction of a new criterion excluding patients with basal cell carcinoma or actinic keratosis within 12 weeks was observed. Concerning cardiovascular diseases and other organopathies, the protocols from period I excluded patients with symptoms of severe, progressive or uncontrolled disease or history of hospital admission for cardiac disease or stroke within the last year. More restrictive criteria, e.g. systolic blood pressure over 160 mmHg or diastolic over 95 mmHg, were introduced in the second period. Patients with a history of clinically significant organopathy were completely excluded in the most recent protocols. A total of 36 EC concerned laboratory values. There was an increase in the mean number of EC over time. The most important change was observed in bilirubin range with exclusion of patients with bilirubin value higher than the upper limit of normal, haemoglobin less than 11 g dL−1, haematocrit lower than 30% and haemoglobin A1C higher than 9% during the second period. No significant changes were observed concerning washout periods, allergy, pregnancy and nursing. Our results show a trend towards an increase in the number of exclusion criteria and more restrictive wording over time in phase III placebo‐controlled trials of investigational agents in psoriasis. The aim of EC is to reduce the risk of adverse events for participants to maintain a positive benefit–risk ratio. However, the benefit–risk ratio from randomized controlled trials has been shown to be more positive than real‐life data from registries.5 6 7 8 This data shows the importance of post‐marketing observational studies and registries. Populations included in phase III trials should remain sufficiently diverse to be able to generalize the study results to the population of patients who will be exposed to the drug once it is marketed.9 One limitation of the study is that we could not have access to all original phase III protocols from all pharmaceutical companies. Recently, the concept of making all clinical study data public has emerged and has been enforced by the European Medicines Agency. This will hopefully help to increase transparency and improve the quality of future protocols. We have shown that only a minority of EC from original protocols is listed in publications. For the clinical practitioner there is limited transparency concerning the population excluded from phase III trials. Listing all IC/EC from original protocols in publications would help to improve transparency. In addition, standardization of the catalogue of inclusion/exclusion criteria in collaboration with investigators, industry scientists, patient representatives and drug approval agencies could help to better assess the safety and efficacy of new agents in the population for which they are intended to be used. Funding sources: no external funding. Conflicts of interest: N.K., M.P.K. and C.B. have no conflicts of interest. A.K. has been an investigator for AbbVie, Amgen, Celgene, Galderma, LEO Pharma, Lilly, Novartis, Pfizer and Roche. C.B.L. has been an investigator for AbbVie, AB Science, Amgen, Boehringer, Celgene, Eli Lilly, Janssen‐Cilag, LEO Pharma, Novartis and Pfizer. C.G. has been a consultant and investigator for AbbVie, Celgene, Janssen‐Cilag, Lilly, Pfizer and Sandoz. M.A.R. has been a speaker or advisory board member for AbbVie, Eli Lilly, Janssen‐Cilag, LEO Pharma, MSD, Novartis and Pfizer; and investigator for Amgen, Boehringer, Jansen‐Cilag, Lilly, Novartis and Pfizer. M.B.B. has been on the advisory board, been a consultant or investigator for AbbVie, Amgen, Celgene, Janssen‐Cilag, Kyowa, LEO Pharma, Lilly, MSD, Nordic, Novartis, Pfizer, Roche and Takeda. C.P. has been a consultant and investigator for AbbVie, Amgen, Boehringer, Celgene, Eli Lilly, Janssen‐Cilag, LEO Pharma, Novartis and Pfizer