BJD

M E D I CA L DE R M A T O L O G Y British Journal of Dermatology

Safety, efficacy and drug survival of biologics and

biosimilars for moderate-to-severe plaque psoriasis*
A. Egeberg iD ,1 M.B. Ottosen,1 R. Gniadecki,2 S. Broesby-Olsen,3 T.N. Dam,4 L.E. Bryld,5 M.K. Rasmussen6 and
L. Skov1
1
Department of Dermatology and Allergy, Herlev and Gentofte Hospital, University of Copenhagen, Hellerup, Denmark
2
Department of Dermatology, Bispebjerg Hospital, Copenhagen, Denmark
3
Department of Dermatology and Allergy Centre, Odense University Hospital, Odense, Denmark
4
Skin Clinic, Nykøbing Falster, Denmark
5
Department of Dermatology, Roskilde Hospital, Roskilde, Denmark
6
Department of Dermatology, Aarhus University Hospital, Aarhus, Denmark

Linked Comment: Vender. Br J Dermatol 2018; 178:328–329.

Summary

Correspondence
Alexander Egeberg.
E-mail: alexander.egeberg@gmail.com
Accepted for publication
27 October 2017
Funding sources
None.
Conflicts of interest
A.E. has received research funding from Pfizer and
Eli Lilly, and honoraria as consultant and/or
speaker from Pfizer, Eli Lilly, Novartis, Galderma
and Janssen Pharmaceuticals. R.G. reports carrying
out clinical trials for AbbVie, MSD, Celgene,
Novartis, Janssen and Pfizer and has had paid con-
sultancies and lectures from AbbVie, MSD, Janssen
and Pfizer. S.B.-O. has served on advisory boards
with Celgene. T.N.D. is consultant and/or speaker
for Abbott, Janssen Cilag, MSD, LEO Pharma,
Novo Nordisk and Pfizer. M.K.R. has been a con-
sultant or served on advisory boards with Janssen
Cilag and Novartis, and has had paid consultancies
and lectures from AbbVie and LEO Pharma. He
reports carrying out clinical trials for Eli Lilly.
L.S. has been a paid speaker for Pfizer, AbbVie,
Eli Lilly, Novartis and LEO Pharma and has been
a consultant or served on advisory boards with Pfi-
zer, AbbVie, Janssen Cilag, Novartis, Eli Lilly,
LEO Pharma and Sanofi. She has served as an
investigator for Pfizer, AbbVie, Eli Lilly, Novartis,
Amgen, Regeneron and LEO Pharma and received
research and educational grants from Pfizer, Abb-
Vie, Novartis, Sanofi, Janssen Cilag and LEO
Pharma. The authors do not have equity in phar-
maceutical companies.
Background Real-life data on newer biological and biosimilar agents for moderate-
to-severe psoriasis are lacking.
Objectives To examine safety, efficacy and time to discontinuation (drug survival)
of biologics (adalimumab, etanercept, infliximab, secukinumab and ustekinumab)
and compare originators with biosimilars (i.e. Enbrel with Benepali, and Remi-
cade with Remsima).
Methods The DERMBIO registry contains data on all Danish patients with moder-
ate-to-severe plaque psoriasis treated with biologics. We examined patients
treated between 1 January 2007 and 31 March 2017. We used Kaplan–Meier
survival curves and Cox regression to examine drug survival patterns.
Results A total of 3495 treatment series (2161 patients) were included (adalimumab
n = 1332; etanercept n = 579; infliximab n = 333; ustekinumab n = 1055 and
secukinumab n = 196). Secukinumab had the highest number of PASI 100 (100%
improvement from baseline Psoriasis Area and Severity Index) respondents, but also
the lowest drug survival among all the biologics. Ustekinumab had the highest drug
survival overall. There were no significant differences in discontinuation risk
between originator and biosimilar versions of infliximab or etanercept. Treatment
with higher than approved dosages was frequent for all drugs except for adali-
mumab and secukinumab. Adverse events (predominantly infections) were most
frequent for secukinumab compared with the other agents.
Conclusions Ustekinumab was associated with the highest drug survival, and secuk-
inumab with the lowest, although most patients on secukinumab were non-
na€ıve. Switching from originator to biosimilar had no significant impact on drug
survival, and the safety profiles were comparable. Adverse events occurred most
frequently with secukinumab. Future studies are warranted to assess the long-
term safety of novel biologics for psoriasis.
What’s already known about this topic?
• Previous studies have reported that long-term drug survival is dependent on the
patient’s sex, choice of drug and previous exposure to biologics.
•
Ustekinumab has been put forward as the biologic with the highest drug survival.

© 2017 British Association of Dermatologists British Journal of Dermatology (2018) 178, pp509–519 509
510 Safety, efficacy and drug survival of biologics and biosimilars, A. Egeberg et al.

*Plain language summary available online

What does this study add?
DOI 10.1111/bjd.16102
• Ustekinumab had higher drug survival than secukinumab, adalimumab, infliximab
and etanercept. Secukinumab appeared to have the lowest drug survival and highest
risk of adverse events.
• Results were comparable between originators and biosimilars.
• These findings may affect choice of therapy for candidates for biological therapy,
including both bio-na€ıve and currently well treated patients who are currently, as
well as those who have failed on one or more biological agents.

Over the past two decades, the development of biological
agents has had a profound impact on management of moder-
ate-to-severe plaque psoriasis and other inflammatory diseases
such as psoriatic arthritis (PsA).1 In clinical trials, drugs that
target interleukin (IL)-12/23 and IL-17 have shown remark-
able rates of skin clearance, with a much faster onset of action
compared with older biologics that inhibit tumour necrosis
factor (TNF)-a.2,3 However, use of biologics is associated
with a much higher cost than traditional treatment options,4
and these medications may lose their effectiveness after long-
term use.5 Drug survival is defined as the probability that
patients will remain on a given drug, whereas discontinuation
of therapy can occur for a number of reasons, the most com-
mon being lack of efficacy.6
The following biologics are approved for treatment of mod-
erate-to-severe psoriasis in Denmark: two monoclonal anti-
bodies and one receptor targeting TNF-a, i.e. adalimumab
(Humiraâ, AbbVie, North Chicago, IL, U.S.A.), infliximab
(Remicadeâ, Merck Sharp & Dohme, Darmstadt, Germany;
Remsimaâ, Celltrion Healthcare Co, Budapest, Hungary; and
InflectraTM, Hospira, Maidenhead, UK), etanercept (Enbrelâ,
Amgen, Thousand Oaks, CA, U.S.A. and Benepaliâ, Biogen,
Cambridge, MA, U.S.A.), one IL-12/23 inhibitor i.e. ustek-
inumab (Stelaraâ, Janssen, Titusville, NJ, U.S.A.) and two
monoclonal antibodies targeting IL-17A, i.e. secukinumab
(Cosentyxâ, Novartis, East Hanover, NJ, U.S.A.) and ixek-
izumab (Taltzâ, Eli Lilly, Indianapolis, IN, U.S.A.). For etaner-
cept (Enbrel), Benepali is the biosimilar; for infliximab
(Remicade), Remsima and Inflectra are the biosimilars. A
biosimilar is a biological agent that is similar but not identical
to the reference product (henceforth the ‘originator’).
The introduction of biosimilars has become a way to
decrease medical care costs and increase patient treatment
options.7 In May 2015, a national policy was put in place in
Denmark stating that patients should be started on, or
switched to (in the case of patients who are currently treated
and well already on this therapy) the cheapest version of the
biologic, that is, the biosimilar version.8
Previous studies have suggested that drug survival is higher
for ustekinumab, and that this drug has a more favourable safety
profile than etanercept, adalimumab and infliximab.6,9–11 How-
ever, real-life data on newer agents such as secukinumab, as
well as comparisons between originators and biosimilars, are
lacking. We therefore compared the safety, efficacy and drug
survival in a nationwide cohort of all Danish patients with
psoriasis treated with adalimumab, etanercept, infliximab,
secukinumab and ustekinumab.
Materials and methods
Data sources
Study approval was obtained from the Danish Data Protection
Agency (ref. HGH-2016-048, I-Suite: 04520). Review by an
ethics committee is not required for register studies in Den-
mark. The study conduct was in accordance with the Strength-
ening the Reporting of Observational Studies in Epidemiology
recommendations.12
The DERMBIO registry6,13–16 contains data on all patients
in Denmark with moderate-to-severe psoriasis treated with
biologics, biosimilars or novel small-molecule agents such as
apremilast. Registration and prospective data collection in
DERMBIO has been mandatory for all Danish dermatologists
since 2007. National guidelines for use and monitoring of
biological treatment of psoriasis in Denmark has been
described in detail elsewhere.16 Adalimumab, etanercept and
infliximab have all been commercially available in Denmark
since before 2007, whereas ustekinumab and secukinumab
were introduced in April 2009 and April 2015, respectively.
The biosimilar drugs Remsima and Benepali have been
available since March 2015 and March 2016, respectively.
Ixekizumab was not marketed until July 2016, and was
therefore not included in this study. No patients were trea-
ted with Inflectra during the study period.
Patient selection criteria
Patients were included if they had a minimum treatment
duration of 1 month with either adalimumab, etanercept,
infliximab, secukinumab or ustekinumab. We excluded
patients who were treated with these drugs as part of a
clinical trial.
As previously described,6 data in DERMBIO are set up as
treatment series/sequences of continuous treatment with the

British Journal of Dermatology (2018) 178, pp509–519 © 2017 British Association of Dermatologists
 Safety, efficacy and drug survival of biologics and biosimilars, A. Egeberg et al. 511
same biological drug. If treatment with one drug is discontin-
ued, a new treatment series will commence when treatment
with another drug (or the same one as previously used) is
started. In accordance with previous analyses,6 treatment
sequences were merged together if the same drug was used in
two consecutive series and the discontinuation was less than
1 month for adalimumab and etanercept, less than 2 months
for infliximab and secukinumab, and less than 4 months for
ustekinumab. In analyses of etanercept and infliximab, series
were joined if patients were switched between Enbrel and
Benepali, and Remicade and Remsima, respectively.
The majority of patients were treated in academic hospital
centres (85% of all treatment series), whereas a smaller pro-
portion was seen in private clinics, predominantly in the first
years of the study period. From the academic hospital centres,
100% of treatment series are reported, whereas the coverage
from the private clinics is estimated to be greater than 80%.6
To ensure the highest data integrity, and enable comparison
with previous reports,6 we only included data from hospital
clinics.
Statistical analysis
We present patient characteristics as frequencies with per-
centages for categorical variables and means with standard
deviations for continuous variables. Survival analyses were
performed using Cox regression in which sex, PsA and con-
comitant methotrexate use were chosen a priori and
included as covariates in the adjusted models.6,10 We used
Kaplan–Meier plots for descriptive (unadjusted) survival
curves. For analyses of the Psoriasis Area and Severity Index
(PASI) response probabilities, estimates were adjusted for
established predictors of therapeutic response, i.e. sex, base-
line PASI and body weight.17 Body weight was chosen over
body mass index, as dosage of ustekinumab is based on
body weight (≤ 100 kg or > 100 kg), as is infliximab
(5 mg kg 1).
In analyses of patients who were well treated on an origina-
tor who switched to a biosimilar, we compared the 2-year
retention rate for Remsima with a cohort of all patients receiv-
ing treatment with Remicade on 1 March 2013 (2 years
before Remsima was marketed). Similarly, we compared the
6-month retention rate for Benepali with a cohort of all
patients receiving Enbrel on 1 September 2015 (6 months
before Benepali was marketed). This methodology was chosen
to ensure comparability with a recent study by Glintborg
et al.18 of switching in patients with PsA, rheumatoid arthritis
and axial spondyloarthritis.
For analyses of the 2-year and 6-month retention rates,
respectively, we used Cox proportional hazards regression
with a robust variance calculation implemented to account for
repeated participants. We considered P < 0
05 as statistically
significant in 2-sided tests, and results are reported with 95%
confidence intervals (CIs), where applicable. All analyses were
performed with SAS statistical software version 9
4 (SAS Insti-
tute Inc., Cary, NC, U.S.A.).
© 2017 British Association of Dermatologists
Results
The study population comprised a total of 2161 patients with
moderate-to-severe plaque psoriasis receiving biological therapy
for at least 1 month, between 1 January 2007 and 31 March
2017. In total, there were 3495 treatment series including 1332
with adalimumab, 579 with etanercept (621 when stratified by
brand name; 566 with Enbrel and 55 with Benepali), 333 with
infliximab (356 when stratified by brand name; 266 with Remi-
cade and 90 with Remsima), 1055 with ustekinumab and 196
with secukinumab, respectively. Patient characteristics at the time
of first treatment series with the respective drugs are shown in
Table 1 and Table S1 (see Supporting Information).
Drug survival of adalimumab, etanercept, infliximab,
secukinumab and ustekinumab
Overall, ustekinumab was associated with the highest drug sur-
vival and lowest risk of drug discontinuation (Figs 1 and 2).
Drug survival appeared higher for bio-na€ıve patients than patients
who had previously failed on one or more biological agent. Fol-
lowing failure because of lack of efficacy on one TNF inhibitor,
risk of discontinuation (second treatment with a biological agent)
was significantly higher with etanercept than infliximab, adali-
mumab and ustekinumab, respectively, whereas the remaining
estimates were nonsignificant (Fig. 2). Among patients failing on
ustekinumab as the first-line therapy, the risk of drug discontinu-
ation was not significantly different between any of the respective
drugs. Additional adjustment for number of previous treatments
is shown in Figure S1 (see Supporting Information).
As dose escalation may prolong drug survival, we per-
formed a sensitivity analysis where patients were considered
to have failed treatment if their dose was increased above the
approved European Medicines Agency (EMA) label. In this
analysis, drug survival was still highest for ustekinumab, fol-
lowed by adalimumab (Fig. S2; see Supporting Information).
The ‘novelty factor’ may affect treatment decisions,19 and thus
patients who are currently well treated may theoretically be
switched when a new drug enters the market. Examinations
showed that 81
8% and 11
7% of non-na€ıve patients who
switched to secukinumab did so because of lack of efficacy
and adverse events, respectively, on the previous biological
agent (47
4% switched from ustekinumab). Of non-na€ıve
patients, 29
2% of those who failed on secukinumab were
switched back to the immediately preceding therapy.
Drug survival in patients who were well treated on an
originator and switched to a biosimilar
In analyses of switching between an originator and a biosimi-
lar, a total of 114, 34, 147 and 44 treatment series were
included for Remicade, Remsima, Enbrel and Benepali, respec-
tively. Of patients who were well treated and switched from
the originator, patients had been treated with Enbrel and Rem-
icade for a median of 2089 and 1589 days, respectively, when
the switch occurred.
British Journal of Dermatology (2018) 178, pp509–519
512 Safety, efficacy and drug survival of biologics and biosimilars, A. Egeberg et al.

Table 1 Patient characteristics at time of first treatment series with each drug in the DERMBIO registry

Adalimumab Etanercept Infliximab Secukinumab Ustekinumab

a
Patients 1215 517 296 195 1002
Sex
Women 439 (36
1) 199 (38
5) 117 (39
5) 84 (43
1) 378 (37
7)
Men 776 (63
9) 318 (61
5) 179 (60
5) 111 (56
9) 624 (62
3)
Age 45
4  14
0 46
5  15
5 47
7  14
0 48
7  13
8 45
0  14
4
Weight, kg 88
6  20
5 87
4  21
4 92
7  23
2 93
6  22
0 90
6  22
5
Disease duration, years 19
9  12
3 21
6  13
5 22
1  13
6 23
2  11
7 20
2  12
2
Psoriatic arthritis 486 (40
0) 219 (42
4) 144 (48
7) 98 (50
3) 226 (22
6)
Dermatology Life Quality Index score 9
7  7
6 8
5  7
3 9
1  7
7 9
1  7
4 10
3  8
0
PASI score at baseline 9
0  7
2 8
9  8
0 9
6  8
5 7
1  6
4 9
4  7
2
Comorbidities
0 773 (63
6) 317 (61
3) 192 (64
9) 150 (76
9) 734 (73
2)
1 273 (22
5) 124 (24
0) 65 (22
0) 21 (10
8) 167 (16
7)
2 105 (8
6) 49 (9
5) 24 (8
1) 16 (8
2) 64 (6
4)
≥3 64 (5
3) 27 (5
2) 15 (5
1) 8 (4
1) 37 (3
7)
Previous biological treatments
0 896 (73
7) 364 (70
4) 156 (52
7) 42 (21
5) 541 (54
0)
1 275 (22
6) 100 (19
3) 74 (25
0) 43 (22
1) 293 (29
2)
2 31 (2
6) 30 (5
8) 41 (13
9) 37 (19
0) 129 (12
9)
3 8 (0
7) 19 (3
7) 17 (5
7) 35 (18
0) 23 (2
3)
≥4 5 (0
4) 4 (0
8) 8 (2
7) 38 (19
5) 16 (1
6)
Methotrexate 184 (15
1) 44 (8
5) 93 (31
4) 16 (8
2) 111 (11
1)

Data are mean  SD or n (%). PASI, Psoriasis Area and Severity Index. aA patient can be treated multiple times with the same drug. Only the
first series in which the patient is exposed to the drug is shown.

There were no significant differences in risk of discontinua-
tion between Remsima compared with Remicade [crude hazard
ratio (HR) 1
64, 95% CI 0
69–3
89, P = 0
264] or Benepali
compared with Enbrel (crude HR 0
46, 95% CI 0
11–1
98,
P = 0
297) (Fig. 3). Adjustment for differences in sex,
methotrexate use, and PsA did not significantly alter the results
for Remsima compared with Remicade (adjusted HR 1
45, 95%
CI 0
61–3
45, P = 0
403) or Benepali compared with Enbrel
(adjusted HR 0
50, 95% CI 0
11–2
02, P = 0
317).
Onset of action, dosing patterns and efficacy of biologics
Among patients achieving a PASI 75 (≥ 75% improvement
from baseline PASI), PASI 90 (≥ 90% improvement) or PASI
100 (100% improvement) response, the time from start of
therapy to response was shortest for secukinumab. Between-
drug comparisons of the chance of achieving these responses
are shown in Table 2.
Contrary to clinical trials, patients in daily clinical practice
do not undergo washout periods prior to initiation of biologi-
cal therapy. Absolute PASI reductions may thus be more
appropriate for assessment of treatment efficacy than relative
reductions (e.g. PASI 75).20 Consequently, we assessed the
absolute PASI reductions, that is, the proportion of patients
(with a baseline PASI > 5) who obtained a PASI ≤ 5, ≤ 2, or
≤ 1, respectively, during therapy (Table S2; see Supporting
Information). Secukinumab had the highest proportion of
patients achieving a PASI of ≤ 2 or ≤ 1, respectively (but not
≤ 5), among bio-na€ıve patients, whereas the highest propor-
tions for non-na€ıve patients were seen with adalimumab.
Using ustekinumab as the reference, adjusted analyses yielded
the highest HRs for achievement of these responses with secuk-
inumab in bio-na€ıve patients, but not in non-na€ıve patients. Effi-
cacy up to 52 weeks of therapy is shown in Figures S3–S5 (see
Supporting Information). Effects on the Dermatology Life Quality
Index are shown in Figure S6 (see Supporting Information).
Cause-specific discontinuation and availability of PASI data are
shown in Figures S7 and S8 (see Supporting Information).
During the first 24 weeks of therapy (including the induc-
tion dose), 3
5%, 39
0%, 22
7%, 0
0% and 20
0%, of patients
were treated with a higher dose of adalimumab, etanercept,
infliximab, secukinumab and ustekinumab, respectively, than
the EMA-label dose. During maintenance therapy (weeks 25–
52), the EMA-label dose was exceeded in 0
9%, 35
1%,
56
7%, 0
0% and 46
2% of patients, on adalimumab, etaner-
cept, infliximab, secukinumab and ustekinumab, respectively
(Table S3; see Supporting Information).
Safety of biologics and biosimilars
For adalimumab, etanercept, infliximab and ustekinumab,
adverse events were comparable with previously published
observational studies for biological registers (Table 3).21 The
highest rate of infections occurred with secukinumab (inci-
dence rate 19
1 per 100 person-years). Four cases of cardio-
vascular events were reported during treatment with

British Journal of Dermatology (2018) 178, pp509–519 © 2017 British Association of Dermatologists
 Safety, efficacy and drug survival of biologics and biosimilars, A. Egeberg et al. 513

(a) (b)

(c) (d)

(e) (f)

Fig 1. Kaplan–Meier plots of drug survival for the biological agents used to treat psoriasis. Drug survival rates for all treatment series showing
discontinuation because of (a) any cause or because of (b) lack of efficacy. Drug survival rates for bio-na€ıve patients showing discontinuation
because of (c) any cause other than lack of efficacy or because of (d) lack of efficacy. Drug survival rates for patients previously exposed to
biologics showing discontinuation because of (e) any cause other than lack of efficacy or because of (f) lack of efficacy.

secukinumab (2% of treatment series), yielding a higher inci-
dence rate (3
1 per 100 person-years) than the other biologics
(0
1–0
5 per 100 person-years).
Recently, concern was raised regarding the relationship between
IL-17 inhibition and inflammatory bowel disease (IBD).22,23 In
DERMBIO, IBD is not coded as a separate adverse event. However,
to assess the potential association, we searched the adverse event
free-text field for any suspected or confirmed cases of new-onset,
or exacerbation of, IBD (i.e. ulcerative colitis or Crohn disease)
during biological therapy. A total of two cases were reported
(0
01% of all adverse events); one during treatment with adali-
mumab and one during treatment with infliximab, respectively.
Incidence of adverse events was not higher for biosimilars than
originators (Table S4; see Supporting Information).

© 2017 British Association of Dermatologists British Journal of Dermatology (2018) 178, pp509–519
514 Safety, efficacy and drug survival of biologics and biosimilars, A. Egeberg et al.

(a) (b)

(c) (d)

(e)

Fig 2. Comparison of risk of drug discontinuation for all treatment series discontinued between 1 January 2007 and 31 March 2017. Forest plots
of adjusted (sex, psoriatic arthritis, and concomitant methotrexate) hazard ratios for risk of (a) all-cause drug discontinuation in all treatment
series; (b) all-cause drug discontinuation in bio-na€ıve patients; (c) all-cause drug discontinuation in patients previously exposed to biologics; (d)
discontinuation because of lack of efficacy of a second biologic following initial failure on one tumour-necrosis factor inhibitor because of lack of
efficacy; (e) discontinuation because of lack of efficacy of a second biologic following initial failure on one interleukin-12/23 inhibitor because of
lack of efficacy. Significant estimates with hazard ratios (HRs) greater than 1 favours the drug listed on the right, and vice versa [e.g. ustekinumab
is favoured over infliximab in (a)]. CI, confidence interval.

treatment with secukinumab. To our knowledge, this is the first

Discussion study to compare drug survival between originator and biosimi-
We provide long-term safety, efficacy, and drug survival data on lar versions of etanercept and infliximab among patients with
anti-IL-12/23 and anti-TNF agents for moderate-to-severe pso- psoriasis. As in previous studies, ustekinumab had higher drug
riasis, and data on outcomes from more than 2 years of real-life survival than adalimumab, infliximab and etanercept.

British Journal of Dermatology (2018) 178, pp509–519 © 2017 British Association of Dermatologists
 Safety, efficacy and drug survival of biologics and biosimilars, A. Egeberg et al. 515
(a)
Fig 3. Kaplan–Meier plots of crude drug survival rates among patients switching from an originator to the corresponding biosimilar agent. (a)
Benepali compared with a comparison cohort of Enbrel-treated patients; (b) Remsima compared with a comparison cohort of Remicade-treated
patients.
Secukinumab was associated with the lowest drug survival
of all investigated compounds. Although the observation time
for secukinumab was only 2 years, a strikingly high propor-
tion of secukinumab-treated patients discontinued their ther-
apy during follow-up. This finding is consistent with a recent
single-centre report of 90 patients treated with secukinumab,
in which 30% of patients discontinued therapy before
32 weeks of treatment.24 Consistent with clinical trial data,2,25
treatment with secukinumab yielded the highest proportion of
PASI 100 respondents and displayed the most rapid onset of
action. The same held true for absolute PASI reductions in
bio-na€ıve, but not non-na€ıve patients. Previously, one system-
atic review found that in clinical trials, infliximab had the fast-
est onset of action, followed by ustekinumab, adalimumab,
and etanercept, in that order.26 For secukinumab, the low
drug survival and the lower long-term efficacy compared with
data from clinical trials is noticeable.
The development of IL-17 inhibitors have been met with
great anticipation, and PASI 90 has been suggested as the new
gold standard for satisfactory treatment response.27 Although
speculative, clinicians’ high expectations of the performance of
secukinumab, especially in patients with concurrent PsA, could
lead to an earlier switching to well-established PsA-approved
therapies such as adalimumab if these predictions are not met.
Moreover, the high proportion of patients returning to their
previous therapy upon failing on secukinumab could suggest
that these were controlled with an acceptable, but not com-
plete, skin clearance and consequently switched to secukinumab
in an attempt to obtain a lower absolute PASI score.
Up-dosing, i.e. treatment with higher than approved
dosages, was frequent for etanercept, infliximab and ustek-
inumab, which may reflect the physicians’ attempts to main-
tain patients’ response to treatment by increasing the dose or
shortening the interval between administrations. Although
ustekinumab showed the longest overall drug survival in this
study, the frequent dose escalation (either higher dosages or
© 2017 British Association of Dermatologists
(b)
shortening of the interval between injections) for this drug
(which has flat pricing in Denmark) warrants attention.
No patients were treated with higher than approved dosages
of secukinumab, which may reflect that use of this drug is rel-
atively new to physicians. However, the markedly shorter drug
survival of secukinumab is unlikely to be explained by lack of
dose escalation. Indeed, less than 4% of adalimumab-treated
patients received off-label dosing, and analysis of in-label dos-
ing regimens showed higher drug survival for adalimumab
(and ustekinumab) compared with secukinumab.
During our study, we identified no new safety signals for
adalimumab, etanercept, infliximab and ustekinumab com-
pared with previous reports.6,10,28 However, contrasting previ-
ous findings,21 infliximab had a lower incidence of infections,
which may reflect that patients with a high risk of infections
are not started on this therapy.
Secukinumab had the highest rate of infections; furthermore,
2% of patients treated with this drug suffered a cardiovascular
event yielding a conspicuously elevated incidence rate
(Table S3) compared with findings from the secukinumab
phase III clinical trial programme,29 although the absolute num-
bers were very low (n = 4). No confirmed or suspected cases of
IBD were reported during treatment with secukinumab.
Our study results support previous findings of higher drug
survival of ustekinumab than anti-TNF agents. The high drop-
out rate for secukinumab (predominantly because of lack of
efficacy) is noticeable, and future replication studies from
other countries are warranted. As many non-na€ıve patients
were treated with secukinumab as their third or fourth biolog-
ical therapy, these may comprise a group of patients who are
particularly difficult to treat. Nevertheless, the low drug sur-
vival for secukinumab was seen even in bio-na€ıve patients,
and the cause may therefore in part be as a result of the anti-
body itself. However, only a limited number of secukinumab-
treated patients were bio-na€ıve (n = 42), and the results
should thus be interpreted with caution.
British Journal of Dermatology (2018) 178, pp509–519
 Table 2 Treatment response, Psoriasis Area and Severity Index (PASI) reduction, within first 52 weeks

PASI responsea in first year Time to responsea in weeks, mean  SD PASI responsea
Previously
Previously exposed to
exposed to Bio-na€ıve, biologics,
biologics, adjusted hazard adjusted hazard
Bio-na€ıve, % (n/N) % (n/N) Bio-na€ıve Previously exposed to biologics ratio (95% CI)b P-value ratio (95% CI)b P-value

British Journal of Dermatology (2018) 178, pp509–519

PASI 75
Adalimumab 65
9 (493/748) 55
9 (214/383) 18
0  10
3 20
3  10
8 1
16 (1
01–1
34) 0
0432 1
22 (1
01–1
46) 0
0369
Etanercept 46
2 (144/312) 30
1 (65/216) 21
3  11
6 22
9  13
3 0
64 (0
52–0
78) < 0
0001 0
50 (0
38–0
65) < 0
0001
Infliximab 62
2 (84/135) 38
7 (60/155) 19
1  11
8 21
0  12
7 0
90 (0
70–1
16) 0
4256 0
70 (0
53–0
94) 0
0158
Secukinumab 60
0 (21/35) 53
8 (77/143) 13
0  3
7 16
7  8
6 2
54 (1
61–4
03) < 0
0001 1
43 (1
10–1
86) 0
0078
Ustekinumab 72
5 (337/465) 55
7 (268/481) 19
7  9
3 21
6  10
6 (reference) (reference)
PASI 90
Adalimumab 54
8 (410/748) 41
0 (157/383) 20
5  11
4 23
2  12
3 1
08 (0
93–1
27) 0
3160 1
11 (0
89–1
37) 0
3579
Etanercept 26
3 (82/312) 18
1 (39/216) 24
8  12
5 22
8  13
1 0
41 (0
32–0
53) < 0
0001 0
42 (0
30–0
60) < 0
0001
Infliximab 47
4 (64/135) 25
8 (40/155) 19
2  11
6 21
5  12
7 0
89 (0
67–1
18) 0
4208 0
65 (0
46–0
92) 0
0142
Secukinumab 57
1 (20/35) 41
2 (59/143) 14
3 7
1 18
8 10
0 3
24 (2
01–5
20) 1
48 (1
10–1
99) 0
0094
516 Safety, efficacy and drug survival of biologics and biosimilars, A. Egeberg et al.

  < 0
0001
Ustekinumab 59
8 (278/465) 39
7 (191/481) 22
1  10
1 23
7  11
1 (reference) (reference)
PASI 100
Adalimumab 43
2 (323/748) 32
6 (125/383) 22
5  12
3 23
4  12
2 1
18 (0
98–1
42) 0
0781 1
06 (0
83–1
34) 0
6408
Etanercept 17
9 (56/312) 13
4 (29/216) 23
5  12
0 23
3  13
7 0
43 (0
31–0
58) < 0
0001 0
41 (0
28–0
61) < 0
0001
Infliximab 33
3 (45/135) 20
0 (31/155) 21
2  11
9 21
3  12
7 0
96 (0
69–1
34) 0
8070 0
66 (0
44–0
98) 0
0371
Secukinumab 51
4 (18/35) 35
0 (50/143) 15
8  8
0 18
5  10
2 3
89 (2
33–6
46) < 0
0001 1
48 (1
07–2
05) 0
0170
Ustekinumab 41
7 (194/465) 31
8 (153/481) 23
9  10
9 24
5  11
9 (reference) (reference)

n, number of series where the response was obtained; N, total number of series analysed. aAmong patients achieving a PASI 75 (≥ 75% improvement from baseline PASI), PASI 90 (≥ 90% improve-
ment) or PASI 100 (100% improvement) response within the first year of treatment; badjusted for sex, body weight and baseline PASI score.

© 2017 British Association of Dermatologists

 Safety, efficacy and drug survival of biologics and biosimilars, A. Egeberg et al. 517

Table 3 Absolute numbers and incidence rates of adverse events per 100 treatment-years

Any adverse event Infection NMSC Cardiovascular Neurological Other/unspecified

Adalimumab
Total number of events, n 1221 651 17 19 44 490
Treatment series with event, 793 (59
5) 390 (29
3) 13 (1
0) 18 (1
4) 40 (3
0) 332 (24
9)
n (%)
IR (95% CI) 32
6 (30
8–34
4) 17
4 (16
1–18
7) 0
5 (0
3–0
7) 0
5 (0
3–0
8) 1
2 (0
9–1
6) 13
1 (12
0–14
3)
Etanercept
Total number of events, n 413 193 6 5 12 197
Treatment series with event, 291 (50
3) 118 (20
4) 6 (1
0) 5 (0
9) 12 (2
1) 150 (25
9)
n (%)
IR (95% CI) 29
1 (26
5–32
1) 13
6 (11
8–15
7) 0
4 (0
2–0
9) 0
4 (0
1–0
8) 0
8 (0
5–1
5) 13
9 (12
1–16
0)
Infliximab
Total number of events, n 223 87 3 1 6 126
Treatment series with event, 164 (49
2) 65 (19
5) 2 (0
6) 1 (0
3) 6 (1
8) 90 (27
0)
n (%)
IR (95% CI) 31
0 (27
2–35
3) 12
1 (9
8–14
9) 0
4 (0
1–1
3) 0
1 (0
0–1
0) 0
8 (0
4–1
9) 17
5 (14
7–20
9)
Secukinumab
Total number of events, n 52 25 0 4 4 19
Treatment series with event, 50 (25
5) 23 (11
7) 0 (0
0) 4 (2
0) 4 (2
0) 19 (9
7)
n (%)
IR (95% CI) 39
7 (30
2–52
0) 19
1 (12
9–28
2) Not applicable 3
1 (1
1–8
1) 3
1 (1
1–8
1) 14
5 (9
2–22
7)
Ustekinumab
Total number of events, n 612 325 11 10 25 241
Treatment series with event, 463 (43
9) 235 (22
3) 7 (0
7) 10 (0
9) 25 (2
4) 186 (17
6)
n (%)
IR (95% CI) 24
3 (22
4–26
3) 12
9 (11
6–14
4) 0
4 (0
2–0
8) 0
4 (0
2–0
7) 1
0 (0
7–1
5) 9
6 (8
4–10
8)

NMSC, nonmelanoma skin cancer; IR, incidence rate per 100 treatment-years; CI, confidence interval.

Although speculative, one potential explanation may be that
antibodies may display differences in their binding affinity,
for example for IL-17, and a stronger binding affinity is gen-
erally related to less disassociation from the antibody. Secuk-
inumab, which is a medium affinity antibody, is given once a
week for the first 4 weeks and thereafter once a month. The
high loading dose can lead to a rapid and convincing
response; however, over time the maintenance dose may be
too low, leading to relapse at some threshold. Notably,
although our long-term efficacy data for secukinumab contra-
dict findings from randomized clinical trials,30 the overall sim-
ilar findings between the present study and previous drug
survival studies published using the DERMBIO registry6,13 sup-
port the robustness of our analyses.
In the case of anti-TNF agents, significant differences in
drug survival rates were seen in this and in other stud-
ies.6,9,10,13 A number of potential explanations have been put
forward, including differences in bioavailability, binding affin-
ity of the drug and the presence or absence of antidrug anti-
bodies.10 As anti-IL-17 agents may differ in their binding
affinity to IL-17, the degree of dissociation from these anti-
bodies may vary. Consequently, it remains unclear whether
the low secukinumab drug survival also holds true for other
biologics targeting IL-17.
We found no significant differences in drug survival for
biosimilars compared with originators. Adverse events were
also comparable with originators as seen in clinical trials
and observational studies in rheumatology.18,31 In a recent
study from the DANBIO registry18 of rheumatology patients,
Remsima was associated with a slightly (3
4%) lower reten-
tion rate compared with Remicade (HR 1
31 95% CI 1
02–
1
68, P = 0
03). Among a cohort of 30 Italian patients with
psoriasis switching from Remicade to Remsima, no signifi-
cant difference in PASI was observed over a median of
23 weeks (range 13–33 weeks) compared with the pre-
switch score.32
We provide up to 2 years of data on biosimilars and
secukinumab. However, we cannot refute that longer fol-
low-up of treatment with these drugs may provide different
results, as patient characteristics and prescription patterns
could vary, and their efficacy might change with prolonged
exposure. Switch from other compounds to a biosimilar
(e.g. from adalimumab to biosimilar infliximab) because of
price considerations is unlikely to have affected our results,
as strict adherence to guidelines regarding the initiation and
switching (including the order in which the drugs should
be used) is enforced on a national level. These guidelines
have been updated regularly, and may potentially bias inter-
pretation, as this could have led to different prescription
patterns during follow-up. Importantly, as previously
argued,19 biological registries are valuable tools to examine
adverse events and monitor safety signals of rare outcomes,
but between-drug comparisons should be cautioned. Indeed,
patient characteristics, underlying considerations for choice

© 2017 British Association of Dermatologists British Journal of Dermatology (2018) 178, pp509–519
518 Safety, efficacy and drug survival of biologics and biosimilars, A. Egeberg et al.
of therapy and study results from a real-life setting may
differ considerably from those of randomized clinical tri-
als.16,19
In conclusion, among patients with moderate-to-severe
plaque psoriasis, treatment with ustekinumab was associated
with the highest drug survival, whereas the lowest was seen
with secukinumab. However, interpretation was limited by
the low number of treatment series with secukinumab
(n = 196), of which only 42 patients were bio-na€ıve. Con-
sequently, definitive conclusions regarding this drug based
on the present data should be cautioned. Switching from
originator to biosimilar infliximab or etanercept had no sig-
nificant impact on drug survival. Infections occurred more
frequently with secukinumab, as did cardiovascular events
(although the absolute number was low) and future studies
are warranted to assess the long-term safety of novel bio-
logics for psoriasis.
Acknowledgments
We acknowledge the substantial contribution of the academic
hospitals and private dermatology clinics and their physicians
that report data to DERMBIO. The registry is managed by an
independent administrator (Zitelab Ltd, Frederiksberg, Den-
mark). Cost of registry maintenance is covered by unrestricted
grants of equal size from the pharmaceutical companies that
manufacture the respective drugs recorded in DERMBIO. These
companies have no access to the data recorded in DERMBIO,
and had no involvement in the present study. The views and
opinions expressed in this article are those of the authors in
their private capacity and do not necessarily reflect those of
the DERMBIO steering committee or the council for the use of
expensive hospital medicines, RADS (Danish: R adet for Anven-
delse af Dyr Sygehusmedicin). Dr Lone Hvid, Odense Univer-
sity Hospital is thanked for help collecting data in DERMBIO
in Odense.
References
1 Nast A, Jacobs A, Rosumeck S et al. Efficacy and safety of sys-
temic long-term treatments for moderate-to-severe psoriasis: a
systematic review and meta-analysis. J Invest Dermatol 2015;
135:2641–8.
2 Langley RG, Elewski BE, Lebwohl M et al. Secukinumab in plaque
psoriasis – results of two phase 3 trials. N Engl J Med 2014;
371:326–38.
3 Puig L, Lopez A, Vilarrasa E et al. Efficacy of biologics in the treat-
ment of moderate-to-severe plaque psoriasis: a systematic review
and meta-analysis of randomized controlled trials with different
time points. J Eur Acad Dermatol Venereol 2014; 28:1633–53.
4 Ahn CS, Gustafson CJ, Sandoval LF et al. Cost effectiveness of bio-
logic therapies for plaque psoriasis. Am J Clin Dermatol 2013;
14:315–26.
5 Levin EC, Gupta R, Brown G et al. Biologic fatigue in psoriasis. J
Dermatolog Treat 2014; 25:78–82.
6 Gniadecki R, Bang B, Bryld LE et al. Comparison of long-term drug
survival and safety of biologic agents in patients with psoriasis
vulgaris. Br J Dermatol 2015; 172:244–52.
British Journal of Dermatology (2018) 178, pp509–519
7 Blauvelt A, Cohen AD, Puig L et al. Biosimilars for psoriasis: pre-
clinical analytical assessment to determine similarity. Br J Dermatol
2016; 174:282–6.
8 Danske Regioner. RADS recommendation for the use of biosimilar infliximab
and etanercept. Available at: http://www.regioner.dk/media/3823/
rads-notat-om-anvendelsen-af-biosimilaere-laegemidler-252880.
pdf (last accessed 29 December 2017).
9 Menter A, Papp KA, Gooderham M et al. Drug survival of biologic
therapy in a large, disease-based registry of patients with psoriasis:
results from the psoriasis longitudinal assessment and registry
(PSOLAR). J Eur Acad Dermatol Venereol 2016; 30:1148–58.
10 Warren RB, Smith CH, Yiu ZZ et al. Differential drug survival of
biologic therapies for the treatment of psoriasis: a prospective
observational cohort study from the British Association of Derma-
tologists Biologic Interventions Register (BADBIR). J Invest Dermatol
2015; 135:2632–40.
11 Carretero G, Ferrandiz C, Dauden E et al. Risk of adverse events in
psoriasis patients receiving classic systemic drugs and biologics in
a 5-year observational study of clinical practice: 2008-2013 results
of the Biobadaderm registry. J Eur Acad Dermatol Venereol 2015;
29:156–63.
12 von Elm E, Altman DG, Egger M et al. The Strengthening the
Reporting of Observational Studies in Epidemiology (STROBE)
statement: guidelines for reporting observational studies. Lancet
2007; 370:1453–7.
13 Gniadecki R, Kragballe K, Dam TN et al. Comparison of drug sur-
vival rates for adalimumab, etanercept and infliximab in patients
with psoriasis vulgaris. Br J Dermatol 2011; 164:1091–6.
14 Ahlehoff O, Skov L, Gislason G et al. Cardiovascular outcomes and
systemic anti-inflammatory drugs in patients with severe psoriasis:
5-year follow-up of a Danish nationwide cohort. J Eur Acad Dermatol
Venereol 2015; 29:1128–34.
15 Ahlehoff O, Skov L, Gislason G et al. Pharmacological undertreat-
ment of coronary risk factors in patients with psoriasis: observa-
tional study of the Danish nationwide registries. PLOS ONE 2012;
7:e36342.
16 Egeberg A, Iversen L, Gniadecki R et al. Characteristics of patients
receiving ustekinumab compared with secukinumab for treatment
of moderate-to-severe plaque psoriasis – nationwide results from
the DERMBIO registry. J Eur Acad Dermatol Venereol 2017; 31:1183–7.
17 Zweegers J, Roosenboom B, van de Kerkhof PC et al. Frequency and
predictors of a high clinical response in patients with psoriasis on
biological therapy in daily practice: results from the prospective,
multicenter BioCAPTURE cohort. Br J Dermatol 2017; 176:786–93.
18 Glintborg B, Sorensen IJ, Loft AG et al. A nationwide non-medical
switch from originator infliximab to biosimilar CT-P13 in 802
patients with inflammatory arthritis: 1-year clinical outcomes from
the DANBIO registry. Ann Rheum Dis 2017; 76:1426–31.
19 Egeberg A, Nast A. A critical eye on registry data in psoriasis. Br J
Dermatol 2017; 177:245–6.
20 Puig L, Zachariae C, Berggren L et al. (P2115) Using absolute and
relative psoriasis area and severity indices for comparison of the
efficacy of ixekizumab to etanercept and placebo in patients with
moderate-to-severe plaque psoriasis. Presented at the Congress of
the European Academy of Dermatology and Venerology, Vienna,
Austria, 28 September 2016; Abstract P2115.
21 Belinchon I, Ramos JM, Carretero G et al. Adverse events associated
with discontinuation of the biologics/classic systemic treatments
of moderate-severe plaque psoriasis: data from the Spanish Biolog-
ics Registry, Biobadaderm. J Eur Acad Dermatol Venereol 2017;
31:17008.
22 Hohenberger M, Cardwell LA, Oussedik E, Feldman SR. Inter-
leukin-17 inhibition: role in psoriasis and inflammatory bowel
© 2017 British Association of Dermatologists
 Safety, efficacy and drug survival of biologics and biosimilars, A. Egeberg et al. 519
disease. J Dermatolog Treat 2017; DOI: 10.1080/09546634.2017.
1329511.
23 Egeberg A. Phase 3 trials of ixekizumab in moderate-to-severe pla-
que psoriasis. N Engl J Med 2016; 375:2101–2.
24 Huang YYM, Hsu S. Loss of efficacy of secukinumab for psoriasis
at 24 to 32 weeks: update and commentary. J Am Acad Dermatol
2017; 76:e221.
25 Thaci D, Blauvelt A, Reich K et al. Secukinumab is superior to
ustekinumab in clearing skin of subjects with moderate to severe
plaque psoriasis: CLEAR, a randomized controlled trial. J Am Acad
Dermatol 2015; 73:400–9.
26 Nast A, Sporbeck B, Rosumeck S et al. Which antipsoriatic drug
has the fastest onset of action? Systematic review on the rapidity
of the onset of action. J Invest Dermatol 2013; 133:1963–70.
27 Manalo IF, Gilbert KE, Wu JJ. Time to raise the bar to Psoriasis
Area Severity Index 90 and 100. J Drugs Dermatol 2015; 14:1086–8.
28 Papp K, Gottlieb AB, Naldi L et al. Safety surveillance for ustek-
inumab and other psoriasis treatments from the Psoriasis Longitu-
dinal Assessment and Registry (PSOLAR). J Drugs Dermatol 2015;
14:706–14.
29 van de Kerkhof PC, Griffiths CE, Reich K et al. Secukinumab long-
term safety experience: a pooled analysis of 10 phase II and III
clinical studies in patients with moderate to severe plaque psoria-
sis. J Am Acad Dermatol 2016; 75:83–98:e4.
30 Blauvelt A, Reich K, Tsai TF et al. Secukinumab is superior to
ustekinumab in clearing skin of subjects with moderate-to-severe
plaque psoriasis up to 1 year: results from the CLEAR study. J Am
Acad Dermatol 2017; 76(60–9):e9.
31 Jorgensen KK, Olsen IC, Goll GL et al. Switching from originator
infliximab to biosimilar CT-P13 compared with maintained treat-
ment with originator infliximab (NOR-SWITCH): a 52-week, ran-
domised, double-blind, non-inferiority trial. Lancet 2017;
389:2304–16.
32 Dapavo P, Vujic I, Fierro MT et al. The infliximab biosimilar in the
treatment of moderate to severe plaque psoriasis. J Am Acad Dermatol
2016; 75:736–9.
© 2017 British Association of Dermatologists
Supporting Information
Additional Supporting Information may be found in the online
version of this article at the publisher’s website:
Fig S1. Comparison of risk of drug discontinuation of all
treatment series discontinued between 1 January 2007 and 31
March 2017 (including adjustment for number of previous
treatments)
Fig S2. Kaplan–Meier plots of drug survival for in-label dos-
ing of the biological agents in psoriasis
Fig S3. Efficacy over time to week 52 (last-observation-car-
ried-forward)
Fig S4. Efficacy over time to week 52 (last-observation-car-
ried-forward) – bio-na€ıve patients
Fig S5. Efficacy over time to week 52 (last-observation-car-
ried-forward) – non-na€ıve patients
Fig S6. Dermatology Life Quality Index ≤ 1 over time to
week 52 (last-observation-carried-forward)
Fig S7. Overall and cause-specific discontinuation
Fig S8. Psoriasis Area and Severity Index data availability
Table S1 Patient characteristics at time of first treatment
with Enbrel, Benepali, Remicade or Remsima (2007–2017)
Table S2 Treatment response (absolute Psoriasis Area and
Severity Index) within first 52 weeks
Table S3 Biologic dosing patterns in new treatment series
and during maintenance therapy
Table S4 Absolute numbers and incidence rates of adverse
events per 100 treatment-years for biosimilars and originators.
Video S1 Author video.
Powerpoint S1. Journal Club Slide Set.
British Journal of Dermatology (2018) 178, pp509–519