Digestive and Liver Disease 54 (2022) 76–83 Contents lists available at ScienceDirect Digestive and Liver Disease journal homepage: www.elsevier.com/locate/dld Alimentary Tract Treatment patterns and intensification within 5 year of follow-up of the first-line anti-TNFα used for the treatment of IBD: Results from the VERNE study G. Bastida a,∗, I. Marín-Jiménez b, A. Forés c, E. García-Planella d, F. Argüelles-Arias e, I. Tagarro f, A. Fernandez-Nistal f, C. Montoto f, J. Aparicio f, M. Aguas a, J. Santos-Fernández g, M.M. Boscá-Watts h, R. Ferreiro-Iglesias i, O. Merino j, X. Aldeguer k, X. Cortés l,m, B. Sicilia n, F. Mesonero o, M. Barreiro-de Acosta i a Department of Gastroenterology, Hospital La Fe, Valencia, Spain b Department of Gastroenterology, Hospital Gregorio Marañón and Instituto de Investigación Sanitaria Gregorio Marañón (IiSGM), Madrid, Spain c Hospital General Universitario de Castellón, Castellón, Spain d Hospital de la Santa Creu i Sant Pau, Barcelona, Spain e Hospital Universitario Virgen Macarena, Sevilla, Spain f Takeda Farmacéutica España S.A., Madrid, Spain g Department of Gastroenterology, Hospital Universitario Río Hortega, Valladolid, Spain h IBD Unit, Gastroenterology Department of the University Clinic Hospital of Valencia, Valencia, Spain i Department of Gastroenterology, Hospital Clínico Universitario de Santiago, Santiago de Compostela, Spain j Department of Gastroenterology, Hospital Universitario Cruces, Bilbao, Spain k Department of Gastroenterology, Hospital Dr Josep Trueta, Girona, Spain l IBD Unit, Gastroenterology Section, Internal Medicine Hospital of Sagunto, Sagunto, Spain m University of Cardenal Herrera-CEU, Castellón, Spain n Hospital Universitario de Burgos, Burgos, Spain o Gastroenterology Department, Hospital Ramón y Cajal, Madrid, Spain. a r t i c l e i n f o a b s t r a c t Article history: Background: Anti-TNFα represent one of the main treatment approaches for the management of in- Received 22 March 2021 flammatory bowel diseases (IBD). Therefore,the evaluation of their treatment patterns over time provides Accepted 6 June 2021 valuable insights about the clinical value of therapies and associated costs. Available online 6 July 2021 Aims: To assess the treatment patterns with the first anti-TNFα in IBD. Keywords: Anti-TNFα Methods: Retrospective, observational study. Inflammatory bowel disease Results: 310 IBD patients were analyzed along a 5-year follow-up period. 56.2% of Crohn’s disease (CD) Treatment discontinuation patients started with adalimumab (ADA), while 43.8% started with infliximab (IFX). 12.9% of ulcerative Treatment intensification colitis (UC) patients initiated with ADA, while 87.1% initiated with IFX. Treatment intensification was required in 28.9% of CD and 37.1% of UC patients. Median time to treatment intensification was shorter in UC than in CD (5.3 vs. 14.3 months; p = 0.028). Treatment discontinuation due to reasons other than remission were observed in 40.7% of CD and 40.5% of UC patients, although, in UC patients there was a trend to lower discontinuation rates with IFX (36.6%) than with ADA (66.7%). Loss of response accounted for approximately one-third of discontinuations, in both CD and UC. Conclusions: Around one-third of IBD biologic-naive patients treated with an anti-TNFα required treat- ment intensification (earlier in UC) and around 40% discontinued the anti-TNFα due to inappropriate disease control. © 2021 The Authors. Published by Elsevier Ltd on behalf of Editrice Gastroenterologica Italiana S.r.l. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) ∗ Corresponding author. E-mail address: guille.bastida@gmail.com (G. Bastida). https://doi.org/10.1016/j.dld.2021.06.005 1590-8658/© 2021 The Authors. Published by Elsevier Ltd on behalf of Editrice Gastroenterologica Italiana S.r.l. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) G. Bastida, I. Marín-Jiménez, A. Forés et al. Digestive and Liver Disease 54 (2022) 76–83 1. Introduction [19] and gave their written informed consent for study participa- tion. Patients were excluded if they were participating in a clinical Inflammatory bowel disease (IBD) comprises two different trial during the study reference period or if, according to investi- conditions, Crohn’s disease (CD) and ulcerative colitis (UC), gator’s criteria, were not able to understand and fill in the study characterized by chronic and relapsing inflammation of the gas- questionnaires or to give written informed consent. trointestinal tract. They are disorders produced by a dysregulated The study was reviewed and approved by the corresponding immune response against the commensal microbiota, probably as ethics committees. The results obtained in the VERNE study on a result of complex interactions among host genetics, immune dys- the impact of both comorbidities and extraintestinal manifestations regulation, microbiota behavior and environmental predisposing (EIMs) on the response to anti-TNFα therapy have been published factors. Associated symptoms, mainly abdominal pain, diarrhea, elsewhere [20]. rectal bleeding, fatigue and anemia, produce a profound impact on patients’ quality of life, and prolonged inflammation results in 2.2. Patients description damage to the gastrointestinal tract, sometimes leading to severe disability and complications [1–5]. Demographic characteristics (age, gender, ethnicity), smoking Anti-tumor necrosis factor-alpha (anti-TNFα ) therapies, such habit, and clinical variables (type of IBD, previous treatments, dis- as the monoclonal antibodies infliximab (IFX) and adalimumab ease location, disease behavior and extraintestinal manifestations) (ADA), have revolutionized the treatment of IBD. They are effec- were recorded. tive not only at inducing and maintaining clinical remission and mucosal healing but also in improving patients’ quality of life and 2.3. Clinical outcome evaluation reducing surgery and hospitalization rates [6]. Hence, anti-TNFα therapies represent one of the main treatment approaches for The following treatment-related parameters were recorded the management of IBD [7–10]. Unfortunately, up to 40% of IBD and analyzed: 1) time to treatment with the first anti-TNFα ; 2) patients do not respond to induction therapy (primary nonre- the frequency of use of different anti-TNFα treatments; 3) rates sponse [PNR]), and between 24% and 46% of successfully treated of anti-TNFα therapy intensification, time to anti-TNFα therapy patients lose control over time (loss of response [LOR]) [11,12]. As intensification (defined as the time elapsed from initial treatment a consequence, new therapeutic options have been developed and date until drug intensification date), and the type of anti-TNFα are available to patients. therapy intensification strategy (dose increase or dosing interval The study of treatment patterns over time is becoming in- shortening); 4) rates of anti-TNFα therapy discontinuation, time creasingly important, as these may provide valuable insights to to anti-TNFα therapy discontinuation (defined as the time elapsed clinicians and administrators about the real-world utility of ther- from initial treatment date until drug discontinuation date), and apies and their economic costs; therefore, knowledge regarding reasons for discontinuation (remission, PNR, LOR, partial response treatment patterns may be helpful to inform the selection of the (PR), side effects (SEs), other); 5) treatments received for IBD most appropriate therapy in a context where the appropriate after the discontinuation of the first biologic; and 6) concomitant choice is often far from obvious. treatments for IBD received during induction and maintenance of Although different real-world studies have comparatively as- biologic therapy. sessed the treatment patterns of various anti-TNFα therapies in IBD patients [13–18], only a few of them have been conducted over a longer period of time. These few studies [13,17] are based 2.4. Statistical analysis on analyses of health insurance claims databases, which means that the data sources used are not clinical but administrative, Baseline parameters were analyzed descriptively, calculating with the well-known limitations of such claims data (poor disease medians and interquartile ranges (IQRs) for continuous variables characterization, a lack of data on the clinical reasons leading to and percentages for categorical variables. Kaplan-Meier survival medication changes, the impossibility of registering doses and dose analyses were used to estimate times to treatment intensification adjustments, the need to use arbitrary definitions to draw con- and to treatment discontinuation, and a log-rank test was applied clusions on persistent gaps in treatment, or data quality strongly to compare survival curves. P < 0.05 was considered the level dependent on the accuracy of administrative coding tasks). of significance. All data analyses were performed using IBM SPSS As CD and UC are chronic, lifelong conditions, it is essential to Statistics 22.0 Statistical Package for Windows. characterize treatment patterns across extended time spans and, ideally, using clinical data sources to circumvent the limitations 3. Results associated with health insurance claims databases. The objectives of this study were to describe long-term treat- 3.1. Patient characteristics and follow-up times ment patterns with the first anti-TNFα in a cohort of biologic-naive IBD patients, and to describe the need for treatment intensification Initially, 357 patients were identified for inclusion, but 47 of and discontinuation rates. them were eventually excluded from the analysis (screening fail- ures), resulting in a total of 310 analyzed patients, 194 diagnosed 2. Methods with CD and 116 with UC. The clinical and demographic charac- teristics of the patients are shown in Table 1. The median (IQR) 2.1. Study population follow-up times after treatment administration were 59.8 (53.2, 65.8) and 59.8 (53.4, 65.1) months for CD and UC, respectively. VERNE was a retrospective (using hospital patient charts as the source of data), observational, multicenter study involving 24 ter- 3.2. Frequency of use and time to treatment initiation of first tiary hospitals in Spain (ClinicalTrials.gov Identifier: NCT02861118). anti-TNFα treatment The study consecutively included adult patients diagnosed with UC or CD who started their first biologic treatment with an anti- The median (IQR) times from the diagnosis of IBD to treatment TNFα between June 2011 and June 2013. All patients were pre- with the first anti-TNFα were 45.5 (11.6, 156.0) and 43.8 (10.8, scribed anti-TNFα treatment according to standard clinical practice 143.8) months for CD and UC, respectively. 77 G. Bastida, I. Marín-Jiménez, A. Forés et al. Digestive and Liver Disease 54 (2022) 76–83 Table 1 Subjects’ clinical characteristics and demographics. CD UC Total N % N % N % Sex Male 103 53.1 63 54.3 166 53.5 Female 90 46.4 47 40.5 137 44.2 Not available 1 0.5 6 5.2 7 2.3 Total 194 100.0 116 100.0 310 100.0 Age, 194 43.0 116 46.0 310 44.0 median (IQR) (34.0-51.5) (38.0-57.8) (36.0-53.5) Ethnicity Caucasian 189 97.4 108 93.1 297 95.8 Other 5 2.6 8 6.9 13 4.2 Total 194 100.0 116 100.0 310 100.0 Smoking habits Nonsmoker 86 44.3 70 60.3 156 50.3 Ex-smoker 52 26.8 37 31.9 89 28.7 Smoker 55 28.4 7 6.0 62 20.0 Not available 1 0.5 2 1.7 3 1.0 Total 194 100.0 116 100.0 310 100.0 Previous Aminosalicylates 105 54.1 106 91.4 211 68.1 treatments for Corticosteroids 148 76.3 105 90.5 253 81.6 IBD† Immunosuppressants 159 82 84 72.4 243 78.4 Antibiotics 79 40.7 27 23.3 106 34.2 Other previous Pain medications 53 27.3 28 24.1 81 26.1 treatments Antidiarrheals 14 7.2 9 7.8 23 7.4 Antidepressants 11 5.7 13 11.2 24 7.7 Location of L1= Terminal ileum 72 37.3 72 23.6 disease L2= Colon 30 15.5 9 8.0 30 9.8 L3= Ileum and colon 86 44.6 49 43.8 86 28.2 L4= Upper GI track 5 2.6 54 48.2 5 1.6 E1= Proctitis 193∗ 100.0 112∗ 100.0 9 3.0 E2= Left colitis 49 16.1 E3= Extensive colitis 54 17.7 Total 305 100.0 Disease B1= Inflammatory 86 44.6 behavior B2= Stenosis 40 20.7 B3= Fistulizing 34 17.6 (B1p+B2p+B3p) = perianal 33 17.1 Total 193∗ 100.0 Extraintestinal Yes 51 26.3 38 32.8 89 28.7 manifestations Total 194 100.0 116 100.0 310 100.0 † Patients could have taken more than one previous medication. ∗ Not available: 1 patient (CD); 4 patients (UC). Table 2 Biologic treatment intensification during follow-up: Regimens and time to intensification. IBD Crohnś disease Ulcerative colitis Total N % N % N % Patients under biologic treatment intensification 56 28.9 43 37.1 99 31.9 Total IBD patients 194 116 310 100.0 Regimens Dose increase 11 19.6 13 30.2 24 24.2 Interval shortening 22 39.3 10 23.3 32 32.3 Both regimens 23 41.1 20 46.5 43 43.4 ∗ One patient with missing treatment intensification date. IQR: Interquartile range. One hundred nine CD patients (56.2%) started anti-TNF treat- months for CD patients and 5.3 months for UC patients (log-rank ment with ADA, while 85 (43.8%) started with IFX. In patients with test, p = 0.028; Fig. 1a). UC, 15 (12.9%) initiated anti-TNFα therapy with ADA, while 101 Regardless of the type of IBD diagnosis, CD or UC, the time were initially treated with IFX (87.1%). to treatment intensification did not significantly differ between patients receiving ADA and those receiving IFX (Figs. 1b and 1c). 3.3. Anti-TNFα therapy intensification However, in UC patients, there was a trend toward a shorter median time to intensification with ADA than with IFX (2.4 vs. 6.3 Fifty-six CD patients (28.9%) and 43 UC patients (37.1%) re- months). quired anti-TNFα treatment intensification, with similar propor- tions of patients on ADA and IFX, respectively (29.4% and 28.2% in the CD subgroup, and 33.3% and 37.6% in the UC subgroup). 3.4. Anti-TNFα therapy discontinuation Strategies implemented for treatment intensification are displayed in Table 2. The most common intensification strategy, used in more Ninety-two CD patients (47.4%) and 65 UC patients (56.0%) dis- than 40% of patients with CD and UC who needed it, was a com- continued anti-TNFα treatment during the follow-up. Discontinua- bination of dose increase and a dosing interval shortening of the tion due to reasons other than remission was observed in 79 CD respective anti-TNFs. The median time to intensification was 14.3 patients (40.7%) and 47 UC patients (40.5%). 78 G. Bastida, I. Marín-Jiménez, A. Forés et al. Digestive and Liver Disease 54 (2022) 76–83 Fig. 1. A) Time to intensification (months) by the type of IBD; B) CD patients: Time to intensification (months) by anti-TNF treatment; C) UC patients: Time to intensification (months) by anti-TNF treatment Fig. 2. A) Time to discontinuation (months) due to reasons other than remission by IBD type; B) CD patients: Time to discontinuation (months) due to reasons other than remission by anti-TNF treatment; C) UC patients: Time to discontinuation (months) due to reasons other than remission by anti-TNF treatment The survival curves assessing time to discontinuation due to to discontinuation were observed between patients receiving IFX reasons other than remission did not show significant differences and those receiving ADA (Fig. 2b). Conversely, in UC patients, the between CD patients and UC patients (log-rank test p = 0.917; survival analyses showed a significantly longer time to treatment Fig. 2a). However, discontinuation rates seemed to differ within the discontinuation with IFX than with ADA (log-rank test p = 0.008; UC subpopulation between IFX-treated patients (36.6%) and ADA- Fig. 2c). Among UC patients who discontinued treatment, the me- treated patients (66.7%). dian time to discontinuation was 15.3 months for those receiv- No differences were observed in the proportion of patients who ing IFX and 11.8 months for those receiving ADA, while for CD discontinued IFX in the CD and UC subpopulations (37%). On the patients who discontinued treatment, median times to drug dis- contrary, 67% of CD patients and 44% of UC patients discontinued continuation were 23.0 months and 27.4 months for IFX and ADA, ADA. Within the CD population, no significant differences in time respectively. 79 G. Bastida, I. Marín-Jiménez, A. Forés et al. Digestive and Liver Disease 54 (2022) 76–83 Table 3 Biologic treatment discontinuation during follow-up due to reasons other than remission. Crohnś disease IBD Ulcerative colitis Total N % N % N % Biologic treatment discontinution Due to causes other than remission 79 40.7% 47 40.5% 126 40.6% Total 194 100.0% 116 100.0% 310 100.0% Discontinuation treatment reason∗ PNR 9 11.4% 9 19.1% 18 14.3% LOR 26 32.9% 18 38.3% 44 34.9% PR 6 7.6% 2 4.3% 8 6.3% SEs 18 22.8% 10 21.3% 28 22.2% Other reasons 20 25.3% 8 17.0% 28 22.3% Total 79 100.0% 47 100.0% 126 100.0% PNR: Primary nonresponse; LOR: Loss of response; PR: Partial response; SE: Side effects. ∗ Percentages calculated over patients with treatment discontinuation. Table 4 Concomitant treatments during induction and maintenance phases∗ , $ Crohnś disease IBD Ulcerative colitisN Total N % N % N % Induction phase Corticosteroids 54 27.8 65 56.0 119 38.4 Immunosuppressants 121 62.4 65 56.0 186 60.0 Total 194 100.0 116 100.0 310 100.0 Maintenance phase Corticosteroids 58 29.9 45 39.1 103 33.3 Immunosuppressants 131 67.5 70 60.9 201 65.0 Total 194 100.0 115∗ 100.0 309 100.0 ∗ One UC patient with missing concomitant information during the maintenance phase. $ Treatments were not continuously administered during the study follow-up period. 3.4.1. Discontinuations due to PNR 3.6. Treatments for IBD after the discontinuation of the first Overall, 5.8% of patients discontinued the anti-TNFα due to PNR anti-TNFα (4.6% of CD patients and 7.8% of UC patients). PNR accounted for 14.3% of the discontinuations due to reasons other than remission Among patients who discontinued the first anti-TNFα treat- (11.4% in CD patients and 19.1% in UC patients) (Table 3). ment due to reasons other than remission, 76.2% received a second biologic therapy after the discontinuation of the first biologic treatment (72.2% in CD and 83.0% in UC). The second 3.4.2. Discontinuations due to LOR biologic was another anti-TNFα in 89.6% of patients (87.8% in CD LOR was the most common reason for treatment discontinu- and 92.3% in UC) and vedolizumab in 9.4% of patients (10.5% in ation. Overall, 14.2% of patients discontinued the anti-TNFα due CD and 9.0% in UC). Ustekinumab was the second biologic agent to LOR (13.4% of CD patients and 15.5% of UC patients). LOR ac- in 1.7% of patients with CD. counted for 34.9% of the discontinuations due to reasons other than remission (32.9% in CD patients and 38.3% in UC patients) 4. Discussion (Table 3). To the best of our knowledge this is the first long-term retro- spective medical-record based study assessing the association be- 3.4.3. Discontinuations due to safety reasons tween the comorbidity profile in patients with IBD and PNR or LOR Side effects led to anti-TNFα discontinuation in 9.0% of patients to anti-TNFα drugs (results that have been previously published), (9.3% of CD patients and 8.6% of UC patients), They were the reason and additionally describing the treatment patterns with the first for 22.2% of the discontinuations due to reasons other than remis- anti-TNFα in IBD patients, in a real-world hospital setting in Spain. sion (22.8% in CD patients and 21.3% in UC patients) (Table 3). Although our study population reflected mostly the overall IBD Overall, almost 50% of the patients discontinued the anti-TNFα population, we observed a higher proportion of CD patients in- treatment because of LOR or PNR (Table 3). cluded in this study. A plausible explanation for this could be that a higher proportion of UC patients can be satisfactorily controlled by conventional anti-inflammatory drugs, such as aminosalicylates 3.5. Concomitant treatments during the induction and maintenance [21], and hence, those patients were not eligible for the study. phases Likewise, UC is generally seen as a less progressive and severe dis- ease, which is why in many countries the rate of biologic treat- Concomitant treatments, defined as those medications admin- ment is lower in UC versus CD. istered at any time during follow-up, based on specific patient re- The median time between IBD diagnosis and first anti-TNFα quirements, are summarized in Table 4. A high proportion of IBD prescription, approximately 3.5 years, was longer than that re- subjects were concomitantly treated with immunosuppressants at ported in other published studies [22,23]. However, we have to both the induction phase (60.0%) and at any time along the main- consider that this study reflects the clinical practice at the time tenance phase (65.0%), with similar proportions for CD and UC pa- the study data were collected (first anti-TNFα administration tients. Conversely, corticosteroids were more frequently adminis- between 2011 and 2013), and that, in recent years, the IBD treat- tered to UC patients both during the induction phase (CD 27.8% vs ment paradigm has shifted towards an earlier use of biologics UC 56.0%) as well as at any time during the maintenance phase in the course of the disease, especially in IBD patients with a (CD 29.9% vs UC 39.1%). worse prognosis. In addition, an earlier treatment with anti-TNFα 80 G. Bastida, I. Marín-Jiménez, A. Forés et al. Digestive and Liver Disease 54 (2022) 76–83 therapies is currently favored by the availability of IFX and ADA IFX-treated patients and ADA-treated CD patients ; however, biosimilars [24]. in patients with UC, the time to treatment discontinuation Although our study results showed no difference in the drug of was significantly longer among patients treated with IFX than choice in CD patients (ADA or IFX), we observed a more frequent ADA. use of IFX in UC patients, that suggests a preference towards IFX in The results regarding the discontinuation rates and time to dis- UC, especially in more severe UC patients. However, these results continuation with ADA in UC patients suggest that perhaps the may also reflect to some extent the different time frames of the standard dose regimen approved for ADA in this indication might marketing authorization for IFX and ADA for UC in Spain. Whereas not be optimal. In fact, the SERENE studies were designed to de- this was granted in 2007 for IFX, ADA was approved for UC only in termine whether an intensified regimen with ADA might improve 2012, which means ADA only became available for UC during the the efficacy results in patients with IBD. In SERENE-UC, clinical study enrollment period. remission at week 52 was numerically higher in patients receiv- We observed that approximately one-third of biologic-naive pa- ing ADA every week compared with ADA every other week, al- tients with IBD who initiated anti-TNFα therapy (either IFX or though the difference was not statistically significant [41]. Simi- ADA) required treatment intensification and that the time to treat- larly, in SERENE-CD, there were no significant differences between ment intensification was significantly shorter for UC compared the standard and an intensified induction dosing scheme concern- to CD patients. An earlier need for treatment intensification in ing the rates of clinical remission at week 4 and of endoscopic re- UC patients has also been reported in other retrospective studies sponse at week 12 [42]. [25–27]. Some reasons have been postulated to explain these dif- In any case, in our study, ADA results in the UC cohort should ferences between UC and CD patients, such as a higher inflamma- be interpreted cautiously due to the limited number of patients in- tory burden in UC than in CD, a diverse pharmacokinetic profile cluded in this subgroup [20]. These results are somewhat in accor- depending on the type of IBD with a faster clearance in UC pa- dance with those of other published retrospective analyses where tients, or different serum anti-TNFα levels between UC patients UC patients treated with ADA showed higher, albeit not statistically and CD patients [25]. In any case, the shorter time to treatment significant, treatment discontinuation rates than those treated with intensification in UC patients observed in our study appears to be IFX [18,43-46]. driven by a shorter time to intensification with ADA compared Interestingly, although in the majority of patients the reason to IFX in this condition. Although the difference in the time to for treatment discontinuation was a lack of or insufficient efficacy treatment intensification between ADA- and IFX-treated patients (PNR, LOR, or PR), almost 90% of these patients were switched to in UC did not reach statistical significance, the survival curves a second anti-TNFα . A meta-analysis has shown that, after the fail- for ADA and IFX showed markedly different trends along the first ure of an anti-TNFα , the probability of remission using a second 20 months after therapy initiation, consistent with a shorter me- anti-TNFα substantially decreases, and the efficacy of the second dian time to treatment intensification with ADA (2.4 months) than anti-TNFα clearly depends on the reason for switching treatment, with IFX (6.3 months). Possibly, the limited size of the UC cohort with the highest remission rates for patients with drug intoler- treated with ADA might have hampered the statistical power to ance (61%) and the lowest for patients with PNR (30%) or LOR compare time to treatment intensification between both anti-TNFα (45%) [47]. Therefore, a sensible approach after the failure of the drugs in UC. It is interesting to note that, possibly, IFX prefer- first anti-TNFα could be to switch to another biologic with a dif- ence for the management of UC may be related, at least in part, ferent mode of action [48], especially in IBD patients with PNR or with the longer time to intensification with this anti- TNFα . In LOR [49,50]. Nevertheless, therapeutic drug level monitoring, when any case, the observed trends are aligned with those reported in a available, may be useful in guiding decisions related to the type of recent retrospective claims database study; treatment intensifica- biologic to use after the failure of the first anti-TNFα . tion was performed earlier for patients on ADA than for those on In any case, we should consider that our data on the biologic IFX [28]. selected after the failure of the first anti-TNFα are at a large ex- The observed early need of treatment intensification with ADA tent conditioned by the marketing approval dates of the newer bi- in UC patients has important practical implications. Given the ologics (in Spain vedolizumab became available for CD and UC in dosing scheme with ADA, this means that a large proportion of 2015, ustekinumab for CD in 2017), as well as by the commercial- biologic-naive UC patients will require to be dosed with weekly ization of some biosimilars; furthermore, we cannot discard that ADA , within only a few months after treatment initiation. Actu- the availability of those biosimilars might have influenced on a ally, a lower response to ADA has been also observed in pivotal UC therapy switching habit. trials with the drug, such as the ULTRA studies [29–33]. This study has some key strengths. It provides an estimation Approximately 40% of the study population discontinued anti- of the times to either therapy intensification or discontinuation TNFα treatment over time due to reasons other than remission. in biologic-naive patients with IBD, who are prescribed the most The discontinuation rates for CD and UC patients, as well as the frequently used biologic treatments (anti-TNF), based on clinical time to discontinuation, were similar. These findings are consistent rather than on administrative data. In addition, this study was con- with the results from other observational studies [13,14,34,35]. Al- ducted in a relatively large sample of patients and, importantly, though in our study the percentages of CD patients who discon- over a long follow-up period (i.e. 5 years). The major limitations of tinued either IFX or ADA were comparable, in patients with UC, this study are its observational and retrospective design. The time discontinuation rates among ADA-treated patients were markedly period covered by our study, with anti-TNFα treatments starting higher than those observed in IFX-treated patients. The high pro- between 2011 and 2013, poses another limitation, as it is possi- portion of patients who discontinued treatment with the first anti- ble that some treatment patterns may have changed to some ex- TNFα clearly shows how heterogeneous and hard to treat IBD is, as tent over time. Therefore, larger prospective controlled studies are well as the relevance of identifying reliable predictive factors of ef- needed to confirm our results. ficacy, in order to further advance towards a personalized medicine In conclusion, our study results suggest that, despite the ex- approach. In this regard the HLA-DQA1∗ 05 allele is emerging as tended use of anti-TNFα drugs as a first-line treatment for CD and a potential marker to select appropriate candidates for anti-TNFα UC following conventional therapy failure, this therapeutic option therapies [36]. is associated with relatively high rates of treatment intensification Consistent with several previous studies[37–40], the time to and discontinuation. Therefore, a substantial proportion of IBD pa- treatment discontinuation did not significantly differ between tients could benefit from alternative first-line biologic therapies. To 81 G. Bastida, I. Marín-Jiménez, A. Forés et al. Digestive and Liver Disease 54 (2022) 76–83 select the most appropriate IBD treatment it would be essential to [2] Crumbock SC, Loeb SJ, Fick DM. Physical activity, stress, disease activ- identify biomarkers able to predict the long-term success with the ity, and quality of life in adults with Crohn disease. Gastroenterol Nurs 2009;32:188–95. different available biologic agents. [3] Irvine EJ. Quality of life of patients with ulcerative colitis: past, present, and future. Inflamm Bowel Dis 2008;14:554–65. Data availability statement [4] Juan J, Estiarte R, Colome E, Artes M, Jimenez FJ, Alonso J. Burden of illness of Crohn’s disease in Spain. Dig Liver Dis 2003;35:853–61. [5] Sin AT, Damman JL, Ziring DA, et al. Out-of-pocket cost burden in pediatric The data that support the findings of this study are available inflammatory bowel disease: a cross-sectional cohort analysis. Inflamm Bowel from the sponsor upon request Dis 2015;21:1368–77. [6] Lopetuso LR, Gerardi V, Papa V, et al. Can we predict the efficacy of Anti-TNF-α agents? Int J Mole Sci 2017;18. Conflict of Interest [7] Ford AC, Sandborn WJ, Khan KJ, Hanauer SB, Talley NJ, Moayyedi P. Efficacy of biological therapies in inflammatory bowel disease: systematic review and meta-analysis. Am J Gastroenterol 2011;106:644–59 quiz 60. Guillermo Bastida has received a speaker honorarium from [8] Martinez-Montiel MP, Casis-Herce B, Gomez-Gomez GJ, et al. Pharmacologic Abvvie, Pfizer, Janssen, FAES, Takeda, Tillots and Abbott. Also, therapy for inflammatory bowel disease refractory to steroids. Clin Exp Gas- Guillermo Bastida has participated in Scientific Advisory Comitees troenterol 2015;8:257–69. of Takeda, Janssen and Abvvie. Ignacio Marín-Jiménez has served [9] Ordas I, Feagan BG, Sandborn WJ. Early use of immunosuppressives or TNF antagonists for the treatment of Crohn’s disease: time for a change. Gut as a consultant, advisory member, speaker, or has received re- 2011;60:1754–63. search funding from MSD, Abbvie, Takeda, Tillots, Ferring, Falk- [10] Zallot C, Peyrin-Biroulet L. Deep remission in inflammatory bowel disease: Pharma, Faes Farma, UCB Pharma, Otsuka Pharmaceutical, Shire, looking beyond symptoms. Curr Gastroenterol Rep 2013;15:315. [11] Allez M, Karmiris K, Louis E, et al. Report of the ECCO pathogenesis workshop Gebro Pharma, and Chiesi. Ana Forés has nothing to declare. Es- on anti-TNF therapy failures in inflammatory bowel diseases: definitions, fre- ther García-Planella has served as a speaker or received research quency and pharmacological aspects. J Crohns Colitis 2010;4:355–66. or educational funding or advisory fees from MSD, Abbvie, Jansen, [12] Ben-Horin S, Chowers Y. Review article: loss of response to anti-TNF treat- ments in Crohn’s disease. Aliment Pharmacol Ther 2011;33:987–95. Ferring, Shire, Tillots, Faes. Federico Argüelles-Arias has served as [13] Bots SJA, Hoekman DR, Benninga MA, Ponsioen CY, D’Haens GR, Lowenberg M. a speaker, a consultant and as an advisory member for or have Patterns of anti-TNF use and associated treatment outcomes in inflammatory received research funding from Janssen, MSD, Abbvie, Pfizer, Kern bowel disease patients: results from an analysis of Dutch health insurance claims data. Neth J Med 2017;75:432–42. Pharma, Biogen, Sandoz, Takeda, Ferring, Faes Farma, Shire Phar- [14] Brady JE, Stott-Miller M, Mu G, Perera S. Treatment patterns and sequencing maceuticals, Dr. Falk Pharma, Tillotts Pharma, Gebro Pharma, Am- in patients with inflammatory bowel disease. Clin Ther 2018;40:1509–21 e5. gen and Vifor Pharma. Ignacio Tagarro, Alonso Fernandez-Nistal, [15] Chen C, Hartzema AG, Xiao H, et al. Real-world pattern of biologic use in patients with inflammatory bowel disease: treatment persistence, switching, Carmen Montoto and Jesús Aparicio are full employees of Takeda and importance of concurrent immunosuppressive therapy. Inflamm Bowel Dis Farmacéutica España. Mariam Aguas has served as a speaker for 2019;25:1417–27. MSD, Abbvie, Janssen, Takeda and Tillots, and received educational [16] Doecke JD, Hartnell F, Bampton P, et al. Infliximab vs. adalimumab in Crohn’s grants from Janssen, MSD and Abbvie. Javier Santos-Fernández disease: results from 327 patients in an Australian and New Zealand observa- tional cohort study. Aliment Pharmacol Ther 2017;45:542–52. has nothing to declare. Marta Maia Bosca-Watts declares educa- [17] Jung YS, Han M, Park S, Cheon JH. Biologic use patterns and predictors for tional activities, research projects, scientific meetings and advisory non-persistence and switching of biologics in patients with inflammatory boards sponsored by MSD, Ferring, Abbvie, Janssen and Takeda. bowel disease: a nationwide population-based study. Dig Dis Sci 2019. [18] Null KD, Xu Y, Pasquale MK, et al. Ulcerative colitis treatment patterns Rocío Ferreiro-Iglesias has served as a speaker for or has received and cost of care. Value in health. J Int Soc Pharmacoecon Outcomes Res research funding from Takeda, MSD, Abbvie, Janssen, Palex, Shire 2017;20:752–61. Pharmaceuticals, TillottsPharma and Casenrecordati. Olga Merino [19] Gomollón F, Dignass A, Annese V, et al. 3rd European evidence-based consen- sus on the diagnosis and management of Crohn’s disease 2016: part 1: diag- has nothing to declare. Xavier Aldeguer has nothing to declare. nosis and medical management. J Crohns Colitis 2017;11:3–25. Xavier Cortes has nothing to declare. Beatriz Sicilia has nothing to [20] Marin-Jimenez I, Bastida G, Forés A, et al. Impact of comorbidities on anti- declare. Francisco Mesonero has served as a speaker or consultant TNFα response and relapse in patients with inflammatory bowel disease: the VERNE study. 2020;7:e0 0 0351. from MSD, Takeda, Janssen, Abbvie and Ferring. Manuel Barreiro- [21] Williams C, Panaccione R, Ghosh S, Rioux K. Optimizing clinical use of de Acosta has served as a speaker, a consultant and advisory board mesalazine (5-aminosalicylic acid) in inflammatory bowel disease. Therap Adv member for, or has received research funding from, MSD, Abbvie, Gastroenterol 2011;4:237–48. [22] Targownik LE, Tennakoon A, Leung S, Lix LM, Singh H, Bernstein CN. Tempo- Janssen, Pfizer, Kern Pharma, Biogen, Takeda, Ferring, Faes Farma, ral trends in initiation of therapy with tumor necrosis factor antagonists for Shire Pharmaceuticals, Dr. Falk Pharma, Tillotts Pharma, Chiesi, Ge- patients with inflammatory bowel disease: a population-based analysis. Clin bro Pharma, Otsuka Pharmaceutical and Vifor Pharma. Gastroenterol Hepatol 2017;15:1061–70 e1. [23] Torres P, Canete F, Nunez L, et al. Spacing the administration interval of an- ti-TNF agents: a valid strategy for patients with inflammatory bowel disease? Funding Dig Dis Sci 2019. [24] Berg DR, Colombel JF, Ungaro R. The role of early biologic therapy in inflam- The VERNE study was funded by Takeda Farmacéutica España matory bowel disease. Inflamm Bowel Dis 2019;25:1896–905. [25] Magro F, Rodrigues-Pinto E, Lopes S, Macedo G. Earlier need of infliximab S.A. intensification in ulcerative colitis than in Crohn’s disease. J Crohns Colitis 2014;8:1331–2. Acknowledgments [26] Taxonera C, Olivares D, Mendoza JL, Diaz-Rubio M, Rey E. Need for infliximab dose intensification in Crohn’s disease and ulcerative colitis. World J Gastroen- terol 2014;20:9170–7. The authors thank Juan Luis Sanz and Susana Vara (Apices, [27] O’Donnell S, Stempak JM, Steinhart AH, Silverberg MS. Higher rates of dose Madrid, Spain) for their participation in the study coordination optimisation for infliximab responders in ulcerative colitis than in Crohn’s dis- ease. J Crohns Colitis 2015;9:830–6. and statistical analysis and Fernando Rico-Villademoros (Apices, [28] Kunovszki P, Judit Szanto K, Gimesi-Orszagh J, et al. Epidemiological Madrid, Spain) for providing medical writing assistance; their par- data and utilization patterns of anti-TNF alpha therapy in the Hungarian ticipation has been funded by Takeda Farmacéutica España S.A. We ulcerative colitis population between 2012-2016. Expert Opin Biol Ther 2020;20:443–9. thank Ana Sánchez-Gabriel and Tamara Pozo (Takeda Farmacéutica [29] Colombel JF, Sandborn WJ, Ghosh S, et al. Four-year maintenance treatment España S.A., Madrid, Spain) for critically revising this manuscript. with adalimumab in patients with moderately to severely active ulcerative col- itis: data from ULTRA 1, 2, and 3. Am J Gastroenterol 2014;109:1771–80. References [30] Reinisch W, Sandborn WJ, Hommes DW, et al. Adalimumab for induction of clinical remission in moderately to severely active ulcerative colitis: results of [1] Busch K, da Silva SA, Holton M, Rabacow FM, Khalili H, Ludvigsson JF. Sick a randomised controlled trial. Gut 2011;60:780–7. leave and disability pension in inflammatory bowel disease: a systematic re- [31] Reinisch W, Sandborn WJ, Panaccione R, et al. 52-week efficacy of adali- view. J Crohns Colitis 2014;8:1362–77. mumab in patients with moderately to severely active ulcerative colitis who 82 G. Bastida, I. Marín-Jiménez, A. Forés et al. Digestive and Liver Disease 54 (2022) 76–83 failed corticosteroids and/or immunosuppressants. Inflamm Bowel Dis tive colitis: results from the SERENE-UC maintenance study. J Crohn’s Colitis 2013;19:1700–9. 2020;14:S001 -S. [32] Sandborn WJ, Colombel JF, D’Haens G, et al. One-year maintenance outcomes [42] Jr, r D‘Haens G SW, Loftus E Jr, Schreiber S, Peyrin-Biroulet L, Panaccione R, among patients with moderately-to-severely active ulcerative colitis who re- Panés J, et al. High versus standard adalimumab induction dosing regimens in sponded to induction therapy with adalimumab: subgroup analyses from UL- patients with moderately to severely active Crohn’s disease: results from the TRA 2. Aliment Pharmacol Ther 2013;37:204–13. SERENE-CD induction study. United Eu Gastroenterol J 2019;7. [33] Sandborn WJ, van Assche G, Reinisch W, et al. Adalimumab induces and main- [43] Gies N, Kroeker KI, Wong K, Fedorak RN. Treatment of ulcerative colitis with tains clinical remission in patients with moderate-to-severe ulcerative colitis. adalimumab or infliximab: long-term follow-up of a single-centre cohort. Ali- Gastroenterology 2012;142:257–65 e1-3. ment Pharmacol Ther 2010;32:522–8. [34] Casanova MJ, Chaparro M, Garcia-Sanchez V, et al. Evolution after anti-TNF [44] Gil-Candel M, Gascon-Canovas JJ, Urbieta-Sanz E, Rentero-Redondo L, Onte- discontinuation in patients with inflammatory bowel disease: a multicenter niente-Candela M, Iniesta-Navalon C. Comparison of drug survival between long-term follow-up study. Am J Gastroenterol 2017;112:120–31. infliximab and adalimumab in inflammatory bowel disease. Int J Clin Pharm [35] Targownik LE, Tennakoon A, Leung S, et al. Factors associated with discontinu- 2020. ation of anti-TNF inhibitors among persons with IBD: a population-based anal- [45] Han M, Jung YS, Cheon JH, Park S. Comparison of real-world outcomes of in- ysis. Inflamm Bowel Dis 2017;23:409–20. fliximab versus adalimumab in biologic-naive Korean patients with ulcerative [36] Guardiola J, Serra K, Rodríguez-Alonso L, et al. P711 carriage of the colitis: a population-based study. Yonsei Med J 2020;61:48–55. HLA-DQA1∗ 05 allele is associated with a high risk of loss of response to adal- [46] Pouillon L, Baumann C, Rousseau H, et al. Treatment persistence of infliximab imumab in patients with Crohn’s disease. J Crohn’s Colitis 2020;14:S574 -S. versus adalimumab in ulcerative colitis: a 16-year single-center experience. In- [37] Cosnes J, Sokol H, Bourrier A, et al. Adalimumab or infliximab as monother- flamm Bowel Dis 2019;25:945–54. apy, or in combination with an immunomodulator, in the treatment of Crohn’s [47] Gisbert JP, Marin AC, Chaparro M. Systematic review: factors associated with disease. Aliment Pharmacol Ther 2016;44:1102–13. relapse of inflammatory bowel disease after discontinuation of anti-TNF ther- [38] Kestens C, van Oijen MG, Mulder CL, et al. Adalimumab and infliximab apy. Aliment Pharmacol Ther 2015;42:391–405. are equally effective for Crohn’s disease in patients not previously treated [48] Holko P, Kawalec P, Pilc A. Cost-effectiveness analysis of Crohn’s disease treat- with anti-tumor necrosis factor-alpha agents. Clin Gastroenterol Hepatol ment with vedolizumab and ustekinumab after failure of tumor necrosis fac- 2013;11:826–31. tor-alpha antagonist. Pharmacoeconomics 2018;36:853–65. [39] Narula N, Kainz S, Petritsch W, et al. The efficacy and safety of either inflix- [49] Patel H, Latremouille-Viau D, Burne R, Shi S, Adsul S. Comparison of real-world imab or adalimumab in 362 patients with anti-TNF-alpha naive Crohn’s dis- treatment outcomes with vedolizumab versus infliximab in biologic-naive pa- ease. Aliment Pharmacol Ther 2016;44:170–80. tients with inflammatory bowel disease. Crohn’s & Colitis 2019;360:1. [40] Osterman MT, Haynes K, Delzell E, et al. Comparative effectiveness of in- [50] Schreiber S, Dignass A, Peyrin-Biroulet L, et al. Systematic review with meta– fliximab and adalimumab for Crohn’s disease. Clin Gastroenterol Hepatol analysis: real-world effectiveness and safety of vedolizumab in patients with 2014;12:811–17 e3. inflammatory bowel disease. J Gastroenterol 2018;53:1048–64. [41] Colombel JF, Panés J, D’Haens G, et al. OP01 Higher vs. standard adalimumab maintenance regimens in patients with moderately to severely active ulcera- 83