Annals of Medicine 2023, VOL. 55, NO. 1, 1335–1345 https://doi.org/10.1080/07853890.2023.2192957 Systematic Review Criteria used to define tumor necrosis factor-alpha inhibitors failure in patients with moderate-to-severe psoriasis: a systematic literature review Isabel Belinchon Romeroa,b, Almudena Mateu Puchadesc, Miguel Ribera Pibernatd,e, Diana. P. Ruiz Genaof, Pablo de la Cueva Dobaog and Jose Manuel Carrascosae,h a Department of Dermatology, Hospital General Universitario Dr. Balmis-ISABIAL, Alicante, Spain; bDepartment of Medicine, Universidad Miguel Hernández de Elche, Alicante, Spain; cDepartment of Dermatology, Hospital Universitario Doctor Peset, Valencia, Spain; dDepartment of Dermatology, Hospital Universitario Parc Taulí, Sabadell, Spain; eDepartment of Medicine, Universitat Autònoma de Barcelona, Barcelona, Spain; fDepartment of Dermatology, Hospital Universitario Fundación Alcorcón, Alcorcón, Spain; g Department of Dermatology, Hospital Universitario Infanta Leonor, Madrid, Spain; hDepartment of Dermatology, Hospital Universitari Germans Trias i Pujol, Badalona, Spain ABSTRACT ARTICLE HISTORY Background: Determining tumor necrosis factor-alpha inhibitors (anti-TNF-α) failure is still a Received 20 September challenge in the management of moderate-to-severe psoriasis. Thus, our comprehensive systematic 2022 literature review aimed to gather information on the criteria used to define anti-TNF-α failure. Revised 2 January 2023 Accepted 14 March 2023 We also aimed to discover the main reasons for anti-TNF-α failure and define subsequently administered treatments. KEYWORDS Materials and methods: We conducted a systematic review following review and reporting Psoriasis; guidelines (Cochrane and PRISMA). International (Medline/PubMed and Cochrane Library) and tumor necrosis factor Spanish databases (MEDES, IBECS), and gray literature were consulted to identify publications alpha (TNF-α); issued until April 2021 in English or Spanish. treatment failure; failure criteria Results: Our search yielded 58 publications. Of these, 37 (63.8%) described the criteria used to define anti-TNF-α primary or secondary failure. Criteria varied across studies, although around 60% considered Psoriasis Area and Severity Index (PASI)-50 criteria. Nineteen (32.8%) reported the reasons for treatment failure, including the lack or loss of efficacy and safety-related problems, mainly infections. Finally, 29 (50%) publications outlined the treatments administered after anti-TNF-α: 62.5% reported a switch to another anti-TNF-α and 37.5% to interleukin (IL)-inhibitors. Conclusion: Our findings suggest a need to standardize the management of anti-TNF-α failure and reflect the incorporation of new targets, such as IL-inhibitors, in the treatment sequence. KEY MESSAGES • In the treatment of psoriasis, the primary and secondary anti-TNF-α failure criteria differ widely in the scientific literature. • The strictest efficacy criteria for defining anti-TNF-α failure, or those recommended by guidelines such as PASI75, were underused both in clinical trials and observational studies. • Most studies failed to consider patient-reported outcomes in assessing psoriasis treatment efficacy, which contrasts with recent recommendations on the inclusion of patient-reported HRQoL as a supporting criterion when considering clinical outcomes. Introduction IL-17 [secukinumab, ixekizumab, brodalumab, bimeki- zumab], IL-23 [guselkumab, tildrakizumab, risankizumab]) Therapeutic options for moderate-to-severe psoriasis [1–4]. Choosing the appropriate treatment option for have dramatically increased in recent years due to the each patient can be challenging [3,5]. More recently, incorporation of new biologic agents. These include the availability of anti-TNF-α biosimilars has provided a tumor necrosis factor-alpha inhibitors (anti-TNF-α) less costly alternative to their precursors, facilitating the (etanercept, adalimumab, infliximab, certolizumab) and access to anti-TNF-α as a first-line biologic treatment the interleukins (IL)-inhibitors (IL-12/23 [ustekinumab], for moderate-to-severe plaque psoriasis [6,7]. CONTACT Diana. P. Ruiz Genao diana@aedv.es Department of Dermatology, Hospital Universitario Fundación Alcorcón, Calle Budapest, 1, Alcorcón, Madrid 28922, Spain Supplemental data for this article can be accessed online at https://doi.org/10.1080/07853890.2023.2192957. © 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unre- stricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. 1336 I. BELINCHON ROMERO ET AL. Notwithstanding, observations show that around checklist for reporting [12] (see Supplementary Tables 20% of patients discontinue the first course of 1 and 2 for completed PRISMA checklists). This sys- anti-TNF-α and treatment discontinuation rates increase tematic approach minimized the risks of publication with time. The main reason for patients’ discontinua- bias, ensuring the robustness of our results. To this tion seems to be a failure to respond or a loss of end, we consulted international (Medline/PubMed and response to anti-TNF-α therapy [8]. Predicting or deter- Cochrane Library) and Spanish national databases mining patients’ response to anti-TNF-α remains chal- (Medicina en Español [MEDES], Índice Bibliográfico lenging in routine practice. This difficulty resides in Español en Ciencias de la Salud [IBECS]). In addition, the variety of clinical criteria followed, the multiple we undertook a broad search in the gray literature. causes for treatment failure [9], and the lack of specific The search targeted a wide range of publications, guidelines or recommendations on how to determine including original articles, systematic and narrative treatment failure or which biologic agent should be reviews, or clinical practice guidelines, published up chosen after anti-TNF-α failure [9,10]. until April 2021 in English or Spanish, focusing mainly Therefore, there is a clear need to standardize the on the criteria used to define anti-TNF-α failure in management of anti-TNF-α failure in patients with moderate-to-severe psoriasis patients. We also searched moderate-to-severe psoriasis. With this overarching for the reasons why anti-TNF-α was discontinued in objective in mind, we conducted the present system- addition to the therapeutic options used after atic literature review, primarily to gather information anti-TNF-α failure and the treatment recommendations on the criteria used to define treatment failure. for patients failing to respond to anti-TNF-α. Table 1 Additionally, we investigated the main reasons for provides details of the publication inclusion and exclu- anti-TNF-α treatment discontinuation, the therapeutic sion criteria, following the PICOTS (population, inter- options chosen after anti-TNF-α failure, the current vention, comparator, outcomes, time, and study design) recommendations on treatment choice after anti-TNF-α definition. failure, and the main factors to consider when choos- ing a new therapy after anti-TNF-α failure in Search strategies moderate-to-severe psoriasis patients. Taken together this information might provide us useful insights into To consult international and national databases, we current practices with a view to establishing the foun- used either free-text or Medical Subject Heading dations for future standardization. (MeSH) terms related to psoriasis, anti-TNF-α treatment, and treatment failure combined with Boolean opera- tors (AND/OR) (see Supplementary Tables 3 and 4 for further details). Furthermore, we manually searched Materials and methods the reference lists of relevant articles obtained. Besides We conducted a systematic literature review following the database search, a broad gray literature search the Cochrane methodology [11] and reported our find- was also performed. Supplementary Table 5 describes ings according to the Preferred Reporting Items for the sources and search strategies applied for the gray Systematic Reviews and Meta-Analyses (PRISMA) literature. Table 1. Eligibility criteria defined by PICOTS. Inclusion criteria Exclusion criteria Population Patients with moderate-to-severe psoriasis Patients with mild psoriasis or other dermatological conditions Intervention (treatment) Biological treatments based on anti-TNF-α Other systemic treatments or biological agents other than anti-TNF-α Comparator N/A Outcome Primary: Criteria used to define anti-TNF-α failure. Efficacy and safety results Secondary: Results obtained from biologic agents other Reasons for anti-TNF-α treatment discontinuation. than anti-TNF-α Therapeutic strategies after anti-TNF-α treatment failure. Current recommendations on choice of treatment after anti-TNF-α failure. Factors to consider when selecting a new therapy after anti-TNF-α failure Study Design Original articles (including clinical trials and observational studies), Opinion articles, letters to the editor, or reviews, meta-analyses, clinical practice guidelines, expert consensus conference communications Time Without limitation N/A (not applicable) Language Published in English or Spanish Anti-TNF-α: tumor necrosis factor alpha inhibitors; N/A: not applicable; PICOTS: population, intervention, comparator, outcomes, time, and study design. Annals of Medicine 1337 Two researchers independently screened each of of 1153 remained for title and abstract screening. Of the identified publications. Any discrepancies between these, 1037 were excluded in the first phase as they reviewers were resolved through consensus and, if did not meet the eligibility criteria. We finally assessed necessary, by consulting a third reviewer. After dupli- 116 full-text articles, considering the review objectives, cate removal by the EndNote reference manager (v. and identified 54 publications. Additionally, four fur- 9.X), publication screening was conducted in two ther publications were recovered through the manual phases. In the first phase, articles were screened by search of the reference lists of relevant articles (Figure title and abstract according to the inclusion and exclu- 1). Thus, we finally obtained a total of 58 publications, sion criteria. In the second phase, articles selected for of which 38 (65.5%) were observational studies, 16 full-text reading in the previous phase were revised, (27.6%) were of clinical trial design, three (5.2%) were considering the objectives of the review. clinical practice guidelines, while one (1.7%) corre- sponded to a narrative review. Of the clinical trials (n = 16), we reviewed the CONSORT criteria in eight Compliance with the reporting guidelines and reports (50%) that had a randomized controlled design, level of evidence of which six (75.0%) were compliant with the CONSORT Two researchers independently assessed the compli- reported items. Among the observational studies ance of the original studies included to the reporting (n = 38), 23 (60.5%) had a retrospective design, whereas guidelines and the level of evidence of the publica- the rest (n = 15; 39.5%) followed a prospective design. tions. Any discrepancies were resolved by consensus We applied the STROBE criteria in 37 publications and, if necessary, by consulting a third reviewer. To based on cohort observational designs and excluded evaluate the compliance with the reporting guidelines, a case-control study. Among the observational studies we applied the 30-point Consolidated Standards of analyzed, 16 (43.2%) had good compliance with Reporting Trials (CONSORT) statement for randomized STROBE guidelines. Supplementary Table 7 provides a clinical trials [13] and the 22 essential points of the summary of all the studies included. Strengthening the Reporting of Observational Studies All the publications identified covered the main in Epidemiology (STROBE) declaration for observational objective, and the secondary objectives stated. The cohort studies [14]. Studies were considered compliant following paragraphs describe the results obtained for when over 70% of items were reported. To evaluate each objective. the level of evidence of publications, we used the Oxford Centre for Evidence-Based Medicine (OCEBM) Results by objective scales [15], which ranged from 1a (systematic reviews of randomized clinical trials) to 5 (expert opinion with- Main objective: criteria used to define anti-TNF-α out explicit critical appraisal). failure We retrieved 37 (63.8%) publications—twenty-two observational studies (59.5%), 14 clinical trials (37.8%), Data abstraction and one clinical practice guideline (2.7%)—dealing The variables recorded for the included articles com- with the current criteria used to define anti-TNF-α fail- prised: the first author; the year of publication; the ure. Of these, 35 (94.6%) covered criteria for primary period of the study; the type of publication; the char- failure or lack of response, whereas 18 (48.6%) dis- acteristics of the treatment and population; the out- cussed criteria for secondary failure. comes evaluated; and their compliance with the Regarding primary failure, most studies (n = 33; reporting guidelines and level of evidence (see 94.3%) defined it using clinical criteria such as Psoriasis Supplementary Table 6 for further details). Area and Severity Index (PASI) or Physician’s Global Assessment (PGA). Nonetheless, the standards of these criteria varied from study to study, reporting most of Results them (n = 21; 60%) a less stringent PASI, inability to achieve a 50% reduction in PASI or PASI50. Figure 2 Results overview describes the distribution of the studies according to We retrieved 1507 publications in international the criteria reported, whereas Supplementary Table 8 (Medline/PubMed [n = 651], Cochrane Library [n = 367]) shows the details of the criteria reported in each study. and Spanish databases (IBECS [n = 326] and MEDES Three of the studies (8.6%) included patient-reported [n = 157]) in addition to six publications from gray lit- health-related quality of life (HRQoL), measured by the erature sources. After the removal of duplicates, a total Dermatology Life Quality Index (DLQI), in combination 1338 I. BELINCHON ROMERO ET AL. Figure 1. PRISMA flow diagram. with clinical criteria to assess primary therapeutic fail- to estimate the distribution of patients among the ure. One of them defined treatment failure as not different causes for discontinuation. Accordingly, most improving DLQI by 50% or below the value of 10 and/ anti-TNF-α discontinuations were due to efficacy- or failure to achieve PASI50 [16] while the other two related reasons, mainly a lack or loss of efficacy or considered it as a 5-point reduction in DLQI together due to disease remission. The following reasons for discontinuation were safety-related and, more with the failure to achieve PASI50 or PASI75 [17,18]. specifically, adverse events occurrence, with infections Apart from the variability of criteria, there was no clear cited as the most commonly reported adverse event pattern between the criteria and the period in which causing treatment withdrawal. Finally, other reasons the studies were undertaken (Figure 3 shows the arti- for discontinuation that were not related to the cles grouped together by the standards of the criteria treatment itself were patients’ preferences (e.g. patients’ [from the least strict, i.e. failure to achieve PASI50, to willingness to be referred to a nearby hospital), lack the strictest, namely a PGA of 0 or 1 or a PASI90] and of adherence to treatment, comorbidities unrelated to the period in which they were conducted). psoriasis, or a lack of patients’ follow-up. Table 2 shows The standards of secondary failure (n = 18) criteria the reasons grouped into four categories: inefficacy also varied from study to study. Loss of efficacy after (lack or loss of efficacy), safety, remission, and other reaching PASI50 was the most frequent criteria used causes. (n = 11; 61.1%). Additionally, two publications (11.1%) did not specify the criteria used [19,20]. Therapeutic options used after anti-TNF-α failure in moderate-to-severe psoriasis patients. Twenty-nine publications (50%), including 17 (58.6%) observational Secondary objectives studies and 12 clinical trials (41.4%), met the last M a i n r e a s o n s f o r a n t i -T N F - α t r e a t m e n t objective: namely, they described the therapeutic discontinuation. We obtained 19 (32.8%) publications, options used after anti-TNF-α failure in moderate-to- specifically 17 (89.5%) observational studies and two severe psoriasis patients. Among these, we identified (10.5%) clinical trials, which dealt with the main a total of 56 treatment changes after anti-TNF-α failure: reasons for anti-TNF-α treatment discontinuation. Of most of these consisted in a shift to another anti-TNF-α these, sixteen also provided us detailed information (n = 35; 62.5%), while the others involved changes to Annals of Medicine 1339 Figure 2. Distribution of studies according to the criteria applied for anti-TNF-α primary and secondary failure. Anti-TNF-α: tumor necrosis factor inhibitors; PASI: Psoriasis Area Severity Index; PGA: Physician’s Global Assessment; †Non-objective clinical criteria or non-specified criteria. an IL inhibitor (n = 21; 37.5%) mainly to an IL-12/IL-23 Discussion inhibitor (n = 10; 47.6%), followed by an IL-17 inhibitor (n = 8; 38.1%) or an IL-23 inhibitor (n = 3; 14.3%). When Anti-TNFs have become the predominant biological we considered the three anti-TNF-α evaluated— option to treat moderate-to-severe psoriasis patients etanercept, infliximab, and adalimumab—separately, due to the availability of biosimilars. However, deter- more studies showed a change of adalimumab to mining anti-TNF-α failure to respond to these treat- other therapeutic targets rather than other agents: for ments and deciding the best alternative is one of the adalimumab, 50% (n/N = 9/18) of changes were to an most challenging aspects of psoriasis management. IL inhibitor, compared with 31.6% (n/N = 6/19) for Within this scenario, our systematic literature review etanercept and 37.5% (n/N = 6/16) for infliximab. See aimed to cover key questions in the management of Figure 4 for further details. Likewise, Supplementary anti-TNF-α failure and, primarily, to know about the Figure 1 reflects the treatment changes after anti- criteria used to define treatment failure and whether TNF-α failure reported over time, showing the other these criteria are uniform among studies. therapeutic targets to be adopted in the last ten years, such as IL-12/IL-23 inhibitors. The primary and secondary failure criteria differed widely among the studies appraised; however, failure Current recommendations on the choice of treatment to achieve PASI50 was the most commonly cited cri- and the factors to consider when choosing a new teria. The predominance of PASI50 in the studies therapy after anti-TNF-α failure. The recommendations implies that stricter criteria such as the PASI75, rec- on these issues were scarce and unspecific. Only four ommended by the European consensus in 2011 [24] publications provided recommendations on treatment as the standard of response assessment for moderate choices after anti-TNF-α failure, of which three (75%) to severe psoriasis, were underused. Our findings also were clinical practice guidelines from European settings showed that less than 20% of studies regard the strict- [10,21,22] and one was a narrative review (25%) [23]. est criteria: PASI90 or PGA 0 or 1, both for primary Of these, the French Society of Dermatology [21] was and secondary treatment failure. Both criteria are the only one proposing a treatment algorithm for equivalent and show a ‘clear’ or ‘almost clear’ disease patients with moderate-to-severe psoriasis and without status and have traditionally been considered too strin- significant comorbidities based on expert opinion. Moreover, one clinical practice guideline released by gent to define treatment response. In this respect, the the British Association of Dermatologists [10] contained European EuroGuiDerm Guidelines [25] consider these specific recommendations regarding which factors stricter criteria (PASI90 or PGA 0 or 1) as the new should be considered when choosing a new therapy standards given new biologic treatments enable after anti-TNF-α failure. patients to achieve them at higher rates [26,27]. In 1340 I. BELINCHON ROMERO ET AL. Figure 3. Criteria defining anti-TNF-α failure reported over time. The ordinate axis shows articles grouped by the standards of exigence of the criteria [from the strictest: failure to achieve PGA of 0 or 1 or a PASI90 to the least strict PASI50], whereas the abscissa axis shows the period in which they were conducted. Band colors represent the treatments for which the primary criteria for failure were described in the studies: green = etanercept; yellow = infliximab; red = adalimumab; blue = combination of anti-TNF-α. Abbreviations: ADA (adalimumab); anti-TNF-α (tumor necrosis factor inhibitors); ETA (etanercept); INF (infliximab); PASI (Psoriasis Area Severity Index); PGA (Physician’s Global Assessment). †Includes patient-reported health-related quality of life (HRQoL) criterion too; ‡We did not include the IL-17 inhibitor, bimekizumab, since it had not yet been approved at the time of the study. particular, the Spanish Psoriasis Group (GPS) of the ≤ 5, which indicates a better HRQoL status [24,31]. Spanish Academy of Dermatology and Venereology The recent EuroGuiDerm Guidelines support the use (AEDV) [28] has recently recommended PASI90 as the of HRQoL through DLQI or the Skindex-29 or Skindex-17 clinically appropriate objective for biologic treatment instruments, together with an objective assessment of and PASI100 as the optimal objective. It is also note- treatment response [25]. In this regard, it is noteworthy worthy that all the studies considered a relative PASI that only three publications [16–18] in our review con- or PASI response (i.e. a PASI reduction between base- sidered patient-reported HRQoL as a therapeutic failure line and a specified treatment period by a percentage) criterion in combination with clinical criteria. rather than the absolute PASI value. Absolute PASI has Another important question raised in our review also begun to be recognized as a feasible alternative relates to the most common reasons for anti-TNF-α outcome for psoriasis, especially in routine practice discontinuation. As expected, efficacy-related reasons [29], as it is independent of the baseline values and were the most frequent. The most-reported included also correlates with clinically meaningful measures and inefficacy in terms of both lack and loss of efficacy. HRQoL [30]. On that basis, the GPS of the AEDV also However, some studies include positive causes, such considered a PASI ≤ 3 as a clinically appropriate objec- as remission of psoriasis symptoms, which accounted tive for biologic treatment [28]. Some guidelines also for up to 24% of the reasons for discontinuation [32]. recommend applying HRQoL as a supporting criterion The second most frequently reported reasons for treat- to modify the treatment when the therapeutic PASI ment withdrawal were safety-related, specifically, infec- response improves between 50% and 75%. Accordingly, tions. In this regard, there is conflicting evidence, with therapy should be changed if the DLQI indicates a some studies showing an increase in infections with poorer HRQoL (DLQI > 5) but can be continued if DLQI anti-TNF-α [33] and others suggesting that the risk of Annals of Medicine 1341 infectious episodes for biologics does not exceed the accessibility (economic), unrelated remission; §Safety: adverse effects 14.3%– Range 0.5%– 7.4%– 6.4%– 93.6% 24.2% 43.7% 56.3% risk observed for conventional treatments [32,34]. (%) – – Apart from tolerability, our results showed that con- tinuance of the anti-TNF-α treatment might also be Reeck.2014 34 (77.5) 14 (31.1) influenced by other factors unrelated to the drug itself. van der 3 (6.7) 116 45 – These include patient preferences, lack of patients’ follow-up, comorbidities, psoriatic arthritis, or lack of patients’ adherence. In this respect, prior literature Roche. (73.4) (18.1) (13.5) 2018 305† 423 224 55 41 – indicates that discontinuation or switching of anti-TNF-α and deaths; ¶Others: preferences (request to change hospital, patients who do not attend medical visits, patient decision), development of psoriatic arthritis, lack of follow-up, Anti-TNF-α (tumor necrosis factor inhibitors); †Numbers and percentage refer to treatment courses or series instead of patients; ‡Efficacy: lack and loss of efficacy or disease agents for nonmedical reasons (unrelated to clinical (70.6) Puig. 2019 (7.4) (22) 185 efficacy or tolerability) might result in worse clinical 68 48 15 – 5 outcomes and increased health-care resource utiliza- Menter. Ortonne. Piaserico. tion [35,36]. (66.7) (33.3) 2014 4389 105‡ 70 35 – – With respect to the biologic agents chosen after anti-TNF-α failure, our findings reflect a change in (91.8) treatment sequence in the last ten years, incorporating 2011 (8.2) 282 195 179 16 – – new targets, mainly IL-inhibitors. Among these, the IL-12/IL-23 inhibitor was predominantly chosen, fol- (45.9) (10.5) (48.7) 2016 754 294 135 143 31 lowed by IL-17 inhibitors. These changes in the chosen – treatment are not surprising as they reflect the chrono- Gerdes. Gniadecki. Lunder. logical treatment approval sequence [37] and are sup- (63.5) (17.6) 2019 359† (19) 893 228 63 68 – ported by the latest clinical evidence [38]. Another comorbidities, self-injection problems, swallowing problems, pregnancy or intention, lack of adherence, vaccination, surgery. aspect to consider is that for the three anti-TNF-α— 183 (75.6) 7 (43.7) 30 (12.4) 3 (6.4) 9 (56.3) 29 (12.0) 2011 etanercept, infliximab, and adalimumab—considered 747 242 – separately, we observed that a comparatively higher proportion of patients receiving adalimumab changed 2018 472 16 to non-anti-TNF-α targeted therapies. These differences – – might be because the studies were conducted in dif- Bayaraa. Clemmensen. Dávila-Seijo. Espósito Fonseca. ferent time periods, and adalimumab was the last (93.6) 2014 35 47 44 – – anti-TNF-α to receive approval, i.e. it was approved after etanercept and infliximab. The fact that adalim- (71.9) (16.3) (11.8) 2013 650 178 128 umab was the last anti-TNF-α to be approved increases 29 21 – the likelihood of a change from adalimumab to other agents such as ustekinumab rather than to another 422 (43.3) 236 (24.2) 132 (13.6) 184 (18.9) 2016 1298 974† anti-TNF-α. This change also indicates a paradigm shift in favor of the IL-inhibitors, which have been shown Table 2. Reasons for anti-TNF-α treatment discontinuation. more effective in terms of PASI75 and PASI90 scores 21 (72.4) than anti-TNF-α agents [39]. This fact also reflects phy- 5 (17.3) 3 (10.3) 2010 179 29 – sicians’ treatment preferences as they prioritize efficacy outcomes, such as the proportion of patients achieving a certain outcome (e.g. PASI90), considering these to 20 (17.1) 3 (42.9) 60 (28.9) 36 (30.8) 3 (42.8) 16 (7.7) 1 (0.5) (62.9) 2019 211 126 131 be the most important factors for decision-making in lieu of characteristics like dosing attributes or route 68 (59.1) 1 (14.3) Ayala. of administration [40–42]. However, it is important to 2014 38 – 7 keep in mind that incorporating individual patient preferences and characteristics in the selection of the Arnold. 2 (1.7) 2016† 696 117 optimal treatment favors greater adherence and sat- isfaction with the treatment [43,44]. In fact, it has discontinuations discontinuations previously been shown that demographic and socio- discontinuations n Remission-related‡ discontinued Efficacy-related‡ economic characteristics, as well as the psychosocial Other reasons¶ Safety-related§ anti-TNF-α Patients who Patients on Study, year burden of the disease, influence treatment therapy n (%) n (%) decision-making [44–46]. (%) N (%) 1342 I. BELINCHON ROMERO ET AL. Figure 4. Therapeutic options after Anti-TNF-α loss of response: frequency of changes after etanercept, infliximab, and adali- mumab failure. ADA: adalimumab; anti-TNF-α: tumor necrosis factor inhibitors; ETA: etanercept; IL: interleukin; INF: infliximab. Limitations and strengths information from trials with biosimilars in our review, we decided not to include them; nevertheless, our We wish to describe certain limitations and strengths findings should also be extrapolated to these mole- related to the design of the current review. With cules. Likewise, psoriasis treatment options have varied respect to the limitations, we acknowledge that we greatly over the years, meaning that the older studies appraised a wide range of publications, including those included did not contemplate all the therapeutic with low reporting quality or a low level of evidence, options currently available. However, including studies such as narrative reviews. We recognize that the latter from different periods could also be considered a might introduce the author’s bias, which might likewise strength of our review as this has allowed us to affect the results and conclusions of our study. observe the changing trends in the management of Additionally, the inclusion criteria were limited to psoriasis patients after anti-TNF-α failure. Other English and Spanish publications, possibly resulting in strengths include the fact that we have conducted our the omission of critical publications in other languages. review according to a systematic approach following Also, we are aware that treatment options for psoriasis a strict search strategy, article selection, and data continue to increase rapidly and, therefore, the validity extraction requirements. This systematic approach of our results could be limited in time. For example, ensured the robustness of our results as we minimized the IL-17 inhibitor, bimekizumab, was approved at the the publication bias. In addition, we did not restrict time of the review, so we could not include it in the the time or location of the publications, so the con- search. On the other hand, the articles reviewed cor- clusions drawn are not confined to a particular pop- respond to studies addressed to evaluate original mol- ulation or subgroup of psoriasis patients and could ecules and no biosimilars were included. This is therefore be extrapolated to the general moderate-to-se- because for the approval of biosimilars, extrapolation vere psoriasis population. of data is allowed, so that in some cases biosimilars are approved for the treatment of psoriasis without having been directly studied in comparative trials for Conclusions this pathology but for others [47]. In fact, the European Medicines Agency has recently declared that biosimi- Our findings suggest the need to standardize the cur- lars are interchangeable with their reference products rent management of anti-TNF-α failure in psoriasis and can be substituted for each other once they are patients. The appraised evidence illustrates significant approved in the EU [48]. To avoid including partial heterogeneity in the criteria used to define anti-TNF-α Annals of Medicine 1343 failure. Among the various criteria, PASI50 was chosen Janssen Pharmaceuticals Inc, Leo-Pharma, Lilly, Novartis, in a high proportion of studies and was applied indis- Pfizer-Wyeth, Pierre-Fabre, Sandoz and UCB. DPRG has received consulting fees; payment or honoraria tinctly of the study period, even though the stricter for lectures, presentations, speaker’s bureaus, manuscript PASI75 has only been recommended in recent years writing or educational events; payment for expert testimony; and new biologic treatments allow patients to achieve support for attending meetings and/or travel; participation PASI75 and even PASI90 or PGA 0 or 1. Another finding on a Data Safety Monitoring Board or Advisory Board from is that most studies failed to consider patient-reported Pfizer, Novartis, Lilly, Amgen, Janssen, UCB, Leo Pharma, outcomes in assessing psoriasis treatment, which con- Almirall S.A. and AbbVie. PCD has received consulting fees; payment or honoraria trasts with recent recommendations on the inclusion for lectures, presentations, speaker’s bureaus, manuscript of patient-reported HRQoL as a supporting criterion writing or educational events; support for attending meet- when considering clinical outcomes. The reasons for ings and/or travel; participation on a Data Safety Monitoring anti-TNF-α discontinuation also varied widely among Board or Advisory Board from AbbVie, Pfizer, UCB, Leo studies, with efficacy or safety-related issues being the Pharma, Almirall S.A., Lilly, Novartis and Janssen, and pay- ment for expert testimony from AbbVie. most common. However, our results also suggest that JMC acted as a consultant and/or speaker for and/or other factors, such as patients’ treatment preferences, participated in clinical trials sponsored by companies that should be considered. Finally, although the recommen- manufacture drugs used for the treatment of psoriasis, dations on which treatment to choose after anti-TNF-α including Janssen Pharmaceuticals Inc, Almirall S.A., Lilly, failure are scarce, our findings reflect a pattern in the AbbVie, Novartis, Celgene, Biogen Amgen, Leo-Pharma, last ten years, with greater incorporation of new tar- Pfizer-Wyeth, UCB, Amgen, Sandoz and Biogen. gets, mainly IL-inhibitors such as IL-12/IL-23, IL-17, and IL-23 inhibitors in the treatment sequence for Funding moderate-to-severe psoriasis patients. This project was supported by the Spanish Academy of Dermatology and Venereology (Academia Española de Acknowledgements Dermatología y Venereología [AEDV]) under an unrestricted grant from Almirall. The authors wish to thank Francisco Pérez-Sádaba, Pedro Salmerón, Marta Comellas, and Lucía Pérez-Carbonell at Outcomes’10 (Castellón, Spain) for their assistance with the systematic review and medical writing. ORCID Jose Manuel Carrascosa http://orcid.org/0000-0003- 4266-0771 Author contributions IBR, AMP, MRP, DPRG, PCD and JMC were involved in the conception of the work, the analysis and the interpretation Data availability statement of the results. All authors contributed to writing and critically reviewing the draft versions of the manuscript. All authors There is no raw data associated with this systematic gave final approval for the submission. review. The authors confirm that the data supporting the findings of this study are available within the article and supplementary materials. Disclosure statement IBR acted as a consultant and/or speaker for and/or partic- References ipated in clinical trials sponsored by companies that manu- facture drugs used for the treatment of psoriasis, including [1] Brownstone ND, Hong J, Mosca M, et al. Biologic treat- Janssen Pharmaceuticals Inc, Almirall SA, Lilly, AbbVie, ments of psoriasis: an update for the clinician. Biologics. Novartis, Celgene, Biogen Amgen, Leo-Pharma, Pfizer-Wyeth, 2021;15:39–51. MSD, and UCB. [2] Rønholt K, Iversen L. Old and new biological therapies AMP has received consulting fees; payment or honoraria for psoriasis. Int J Mol Sci. 2017;18(11):2297. for lectures, presentations, speaker’s bureaus, manuscript [3] Armstrong AW, Read C. Pathophysiology, clinical pre- writing or educational events; payment for expert testimony; sentation, and treatment of psoriasis: a review. JAMA. support for attending meetings and/or travel; participation 2020;323(19):1945–1960. on a Data Safety Monitoring Board or Advisory Board from [4] BIMZELX (bimekizumab). [Internet]. EU Summary of Lilly, Novartis, Janssen, UCB pharma, Pfizer, Leo pharma Product Characteristics, June 2022; [cited 2022 Jun 1]. and AbbVie. Available from: https://www.ema.europa.eu/en. MRP acted as a consultant and/or speaker for and/or [5] Amin M, No DJ, Egeberg A, et al. Choosing first-line participated in clinical trials sponsored by companies that biologic treatment for moderate-to-severe psoriasis: manufacture drugs used for the treatment of psoriasis, What does the evidence say? Am J Clin Dermatol. including AbbVie, Almirall S.A., Amgen, Biogen, Celgene, 2018;19(1):1–13. 1344 I. BELINCHON ROMERO ET AL. [6] Barker J, Girolomoni G, Egeberg A, et al. Anti-TNF efficacy and safety results from an open-label study. biosimilars in psoriasis: from scientific evidence to J Am Acad Dermatol. 2011;64(4):671–681. real-world experience. J Dermatolog Treat. [21] Amatore F, Villani AP, Tauber M, et al. French guidelines 2020;31(8):794–800. on the use of systemic treatments for moderate-to-se- [7] Gisondi P, Geat D, Conti A, et al. TNF-α inhibitors bio- vere psoriasis in adults. J Eur Acad Dermatol Venereol. similars as first line systemic treatment for 2019;33(3):464–483. moderate-to-severe chronic plaque psoriasis. Expert Rev [22] Puig L, Carrascosa JM, Carretero G, et al. Spanish Clin Immunol. 2020;16(6):591–598. evidence-based guidelines on the treatment of psori- [8] Warren RB, Smith CH, Yiu ZZN, et al. Differential drug asis with biologic agents, 2013. Part 1: on efficacy and survival of biologic therapies for the treatment of pso- choice of treatment. Spanish psoriasis group of the riasis: a prospective observational cohort study from Spanish academy of dermatology and venereology. the British association of dermatologists biologic inter- Actas Dermosifiliogr. 2013;104(8):694–709. ventions register (BADBIR). J Invest Dermatol. [23] Hu Y, Chen Z, Gong Y, et al. A review of switching 2015;135(11):2632–2640. biologic agents in the treatment of moderate-to-severe [9] Menter A, Strober BE, Kaplan DH, et al. Joint AAD-NPF plaque psoriasis. Clin Drug Investig. 2018;38(3):191–199. guidelines of care for the management and treatment [24] Mrowietz U, Kragballe K, Reich K, et al. Definition of psoriasis with biologics. J Am Acad Dermatol. of treatment goals for moderate to severe psoriasis: 2019;80(4):1029–1072. a European consensus. Arch Dermatol Res. [10] Smith CH, Jabbar-Lopez ZK, Yiu ZZ, et al. British 2011;303(1):1–10. association of dermatologists guidelines for biologic [25] Nast A, Smith C, Spuls PI, et al. EuroGuiDerm guideline therapy for psoriasis 2017. Br J Dermatol. on the systemic treatment of psoriasis vulgaris – part 2017;177(3):628–636. 1: treatment and monitoring recommendations. J Eur [11] The Cochrane Collaboration. Cochrane handbook for Acad Dermatol Venereol. 2020;34(11):2461–2498. systematic reviews of interventions; [cited 2021 Aug 3]. [26] Sbidian E, Chaimani A, Afach S, et al. Systemic phar- Available from: https://training.cochrane.org/handbook/ macological treatments for chronic plaque psoriasis: a archive/v5.1/. network meta-analysis. Cochrane Database Syst Rev. [12] Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020;1(1):Cd011535. 2020 statement: an updated guideline for reporting [27] Torres T, Puig L. Treatment goals for psoriasis: should systematic reviews. BMJ. 2021;372:n71. PASI 90 become the standard of care? Actas [13] Schulz KF, Altman DG, Moher D. CONSORT 2010 state- Dermosifiliogr. 2015;106(3):155–157. ment: updated guidelines for reporting parallel group [28] Carrascosa JM, Puig L, Belinchón Romero I, et al. randomised trials. Int J Surg. 2011;9(8):672–677. Practical update of the recommendations published by [14] Vandenbroucke JP, von Elm E, Altman DG, et al. the psoriasis group of the Spanish academy of derma- Strengthening the reporting of observational studies tology and venereology (GPS) on the treatment of in epidemiology (STROBE): explanation and elaboration. psoriasis with biologic therapy. Part 1. Concepts and Ann Intern Med. 2007;147(8):W163–W194. general management of psoriasis with biologic therapy. [15] Group OLoEW. [Internet]. The Oxford 2011 levels of Actas Dermo-Sifiliográficas. 2021;113(3):261–277. evidence. 2011; [cited 2021 Aug 3]. Available from: [29] Carretero G, Puig L, Carrascosa JM, et al. Redefining https://www.cebm.ox.ac.uk/resources/levels-of-evidence/ the therapeutic objective in psoriatic patients candi- ocebm-levels-of-evidence. dates for biological therapy. J Dermatolog Treat. [16] Clemmensen A, Spon M, Skov L, et al. Responses to 2018;29(4):334–346. ustekinumab in the anti-TNF agent-naïve vs. anti-TNF [30] Gerdes S, Körber A, Biermann M, et al. Absolute and agent-exposed patients with psoriasis vulgaris. J Eur relative psoriasis area and severity index (PASI) treat- Acad Dermatol Venereol. 2011;25(9):1037–1040. ment goals and their association with health-related [17] Bewley A, Cerio R, Clement M, et al. Current application quality of life. J Dermatolog Treat. 2020;31(5):470–475. of national institute for health and clinical excellence [31] Pathirana D, Ormerod AD, Saiag P, et al. European (NICE) guidance in the management of patients with S3-guidelines on the systemic treatment of psoriasis severe psoriasis: a clinical audit against NICE guidance vulgaris. J Eur Acad Dermatol Venereol. 2009;23(Suppl in seven national health service specialist dermatology 2):1–70. units in England. Clin Exp Dermatol. 2011;36(6): [32] Dávila-Seijo P, Dauden E, Carretero G, et al. Survival of 602–606. classic and biological systemic drugs in psoriasis: results [18] Warren RB, Barker J, Finlay AY, et al. Secukinumab for of the BIOBADADERM registry and critical analysis. J patients failing previous tumour necrosis factor-α in- Eur Acad Dermatol Venereol. 2016;30(11):1942–1950. hibitor therapy: results of a randomized open-label [33] Kalb RE, Fiorentino DF, Lebwohl MG, et al. Risk of study (SIGNATURE). Br J Dermatol. 2020;183(1):60–70. serious infection with biologic and systemic treatment [19] Umezawa Y, Nobeyama Y, Hayashi M, et al. Drug sur- of psoriasis: results from the psoriasis longitudinal as- vival rates in patients with psoriasis after treatment sessment and registry (PSOLAR). JAMA Dermatol. with biologics. J Dermatol. 2013;40(12):1008–1013. 2015;151(9):961–969. [20] Strober BE, Poulin Y, Kerdel FA, et al. Switching to adali- [34] Naldi L. Infections and psoriasis treatment: more mumab for psoriasis patients with a suboptimal re- “Real-World” data needed with critical appraisal. J Invest sponse to etanercept, methotrexate, or phototherapy: Dermatol. 2017;137(2):271–274. Annals of Medicine 1345 [35] Wolf D, Skup M, Yang H, et al. Clinical outcomes asso- [42] Alcusky M, Lee S, Lau G, et al. Dermatologist and ciated with switching or discontinuation from anti-TNF patient preferences in choosing treatments for inhibitors for nonmedical reasons. Clin Ther. moderate to severe psoriasis. Dermatol Ther. 2017;39(4):849–862.e6. 2017;7(4):463–483. [36] Gibofsky A, Skup M, Mittal M, et al. Effects of [43] Umar N, Schaarschmidt M, Schmieder A, et al. Matching non-medical switching on outcomes among patients physicians’ treatment recommendations to patients’ prescribed tumor necrosis factor inhibitors. Curr Med treatment preferences is associated with improvement Res Opin. 2017;33(11):1945–1953. in treatment satisfaction. J Eur Acad Dermatol Venereol. [37] Reid C, Griffiths CEM. Psoriasis and treatment: past, 2013;27(6):763–770. present and future aspects. Acta Derm Venereol. [44] Scala E, Megna M, Amerio P, et al. Patients’ demograph- 2020;100(3):adv00032. ic and socioeconomic characteristics influence the ther- [38] Wan MT, Alvarez J, Shin DB, et al. Head-to-head trials apeutic decision-making process in psoriasis. PLoS One. of systemic psoriasis therapies: a systematic review of 2020;15(8):e0237267. study design and maximum acceptable treatment [45] Escala E, Kaczmarczyk R, Zink A, et al. Sociodemographic, differences. J Eur Acad Dermatol Venereol. 2019;33(1): clinical and therapeutic factors as predictors of life 42–55. quality impairment in psoriasis: a cross-sectional study [39] Ten Bergen LL, Petrovic A, Krogh Aarebrot A, et al. The in Italy. Dermatol Ther. 2022;35(8):e15622. TNF/IL-23/IL-17 axis-Head-to-head trials comparing dif- [46] Lesner K, Reich A, Szepietowski JC, et al. Determinants ferent biologics in psoriasis treatment. Scand J Immunol. of psychosocial health in psoriatic patients: a 2020;92(4):e12946. multi-national study. Ac ta Derm Venerol. [40] Schaarschmidt ML, Herr R, Gutknecht M, et al. Patients’ 2017;97(10):1182–1188. and physicians’ preferences for systemic psoriasis treat- [47] Carrascosa JM, Jacobs I, Petersel D, et al. Biosimilar ments: a nationwide comparative discrete choice ex- drugs for psoriasis: principles, present, and near future. periment (PsoCompare). Acta Derm Venerol. Dermatol Ther. 2018;8(2):173–194. 2018;98(2):200–205. [48] EMA. Statement on the scientific rationale supporting [41] Bolt T, Kobayashi H, Mahlich J. Patient and physician interchangeability of biosimilar medicines in the EU; preferences for therapy characteristics for psoriasis: a [cited 2022 Nov 14]. Available from: https://www.ema. discrete choice experiment in Japan. PharmacoEconomics europa.eu/en/news/biosimilar-medicines-can-be- Open. 2019;3(2):255–264. interchanged.