ARTICLE 1 Antitumor Necrosis Factor Agents to Treat Endoscopic INFLAMMATORY BOWEL DISEASE Postoperative Recurrence of Crohn’s Disease: A Nationwide Study With Propensity-Matched Score Analysis Fiorella Cañete1,2,3, Míriam Mañosa1,2, Isabel Pérez-Martínez4, Manuel Barreiro-de Acosta5, Ramiro C. González-Sueyro6, Pilar Nos2,7, Eva Iglesias-Flores8, Ana Gutiérrez9, Luis Bujanda2,10, Jordi Gordillo11, Raquel Ríos León12, María José Casanova2,13, Albert Villoria2,3,14, Iago Rodríguez-Lago15,16, Pilar López Serrano17, Antonio García-Herola18, Patricia Ramírez-de la Piscina19, Mercè Navarro-Llavat20, Carlos Taxonera21,22, Jesús Barrio23, Laura Ramos24, Pablo Navarro25, Olga Benítez-Leiva26, Margalida Calafat1,3 and Eugeni Domènech, MD, PhD1,2, on behalf of the INFLIRECU study INTRODUCTION: Patients with Crohn’s disease experiencing endoscopic postoperative recurrence (POR) may benefit from antitumor necrosis factor (TNF) agents but scarce data on this are available. Our aim was to assess the efficacy of anti-TNF in improving mucosal lesions in patients with endoscopic POR. METHODS: Multicenter, retrospective, study of patients with Crohn’s disease who underwent therapy with anti-TNF agents for endoscopic POR (Rutgeerts score > i1). Treatment outcomes were assessed by the findings in the last ileocolonoscopy performed after anti-TNF therapy was initiated. Endoscopic improvement and remission were defined as any reduction in the baseline Rutgeerts score and by a Rutgeerts score < i2, respectively. RESULTS: A total of 179 patients were included, 83 were treated with infliximab and 96 with adalimumab. Median time on anti-TNF therapy at the last endoscopic assessment was 31 months (interquartile range, 13–54). Endoscopic improvement was observed in 61%, including 42% who achieved endoscopic remission. Concomitant use of thiopurines and treatment with infliximab were associated with endoscopic improvement (odds ratio [OR] 2.15, 95% confidence interval [CI] 1.04–4.46; P 5 0.03, and OR 2.34, 95% CI 1.18–4.62; P < 0.01, respectively) and endoscopic remission (OR 3.16, 95% CI 1.65–6.05; P < 0.01, and OR 2.01, 95% CI 1.05–3.88; P 5 0.04, respectively) in the multivariable logistic regression analysis. These results were confirmed in a propensity-matched score analysis. DISCUSSION: In patients with endoscopic POR, anti-TNF agents improve mucosal lesions in almost two-thirds of the patients. In this setting, concomitant use of thiopurines and use of infliximab seem to be more effective in improving mucosal lesions. SUPPLEMENTARY MATERIAL accompanies this paper at http://links.lww.com/CTG/A339, http://links.lww.com/CTG/A340, and http://links.lww.com/CTG/A341 Clinical and Translational Gastroenterology 2020;11:e00218. https://doi.org/10.14309/ctg.0000000000000218 1 Gastroentorology Department, Hospital Universitari Germans Trias i Pujol, Badalona, Spain; 2CIBERehd, Madrid, Spain; 3Departament de Medicina, Universitat Autònoma de Barcelona, Barcelona, Spain; 4Gastroentorology Department, Hospital Universitario Central de Asturias, Oviedo, Spain; 5Gastroentorology Department, Hospital Clínico Universitario de Santiago, Santiago de Compostela, Spain; 6Gastroentorology Department, Hospital Universitari Clínic, Barcelona, Spain; 7Gastroentorology Department, Hospital Universitario y Politécnico La Fe, Valencia, Spain; 8UGC Digestivo, Instituto Maimónides de Investigación Biomédica de Córdoba (IMIBIC), Hospital Universitario Reina Sofía, Universidad de Córdoba, Córdoba, Spain; 9Gastroentorology Department, Hospital General Universitario de Alicante, Alicante, Spain; 10Gastroentorology Department, Hospital Universitario de Donostia, Instituto Biodonostia, Universidad del País Vasco (UPV/EHU), San Sebastián, Spain; 11Gastroentorology Department, Hospital de La Santa Creu i Sant Pau, Barcelona, Spain; 12Gastroentorology Department, Hospital Universitario Ramón y Cajal, Madrid, Spain; 13Gastroentorology Department, Hospital Universitario de La Princesa, Instituto de Investigación Sanitaria Princesa (IIS-IP), Madrid, Spain; 14Gastroentorology Department, Hospital Parc Taulí, Sabadell, Spain; 15Gastroentorology Department, Hospital de Galdakao, Galdakao, Spain; 16Gastroentorology Department, Biocruces Health Research Institute, Barakaldo, Vizcaya, Spain; 17Gastroentorology Department, Hospital Universitario Fundación Alcorcón, Alcorcón, Spain; 18Gastroentorology Department, Hospital Marina Baixa de Villajoyosa, Alicante, Spain; 19Gastroentorology Department, Hospital Universitario Araba, Vitoria, Spain; 20Gastroentorology Department, Hospital de Sant Joan Despí Moisès Broggi, Sant Joan Despí, Spain; 21 Gastroentorology Department, Hospital Clínico San Carlos, Madrid, Spain; 22Gastroentorology Department, Instituto de Investigación del Hospital Clínico San Carlos (IdISSC), Madrid, Spain; 23Gastroentorology Department, Hospital Universitario Río Hortega, Valladolid, Spain; 24Gastroentorology Department, Hospital Universitario de Canarias, La Laguna, Spain; 25Gastroentorology Department, Hospital Clínico Universitario de Valencia, Valencia, Spain; 26Gastroentorology Department, Hospital Universitari Mútua Terrassa, Terrassa, Spain. Correspondence: Eugeni Domènech, MD, PhD. E-mail: eugenidomenech@gmail.com. A complete list of the affiliations of the INFLIRECU study investigators are listed in the Appendix 1. Received February 21, 2020; accepted June 29, 2020; published online August 3, 2020 © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of The American College of Gastroenterology American College of Gastroenterology Clinical and Translational Gastroenterology 2 Cañete et al. INTRODUCTION METHODS Crohn’s disease (CD) is a chronic condition of unknown origin Study population which leads, in most patients, to cumulative transmural in- This is a retrospective, multicenter, observational study including testinal tissue damage that causes disease-related complica- INFLAMMATORY BOWEL DISEASE all those patients with CD who underwent an intestinal resection tions, including intestinal strictures and intra-abdominal with ileocolonic anastomosis and who were treated with anti-TNF penetrating complications. These complications often require agents (infliximab or adalimumab) after a diagnosis of endoscopic a surgical approach, and, despite the current availability of POR (defined by a Rutgeerts endoscopic score . i1) (3), regardless new, potent anti-inflammatory biological drugs, a noteworthy of concurrent clinical symptoms. Patients were identified from the proportion of patients with CD must still undergo intestinal local IBD databases of each participating center. CD diagnosis was resection (1,2). Unfortunately, surgery is not curative and 70% based on the accepted criteria of the European Crohn’s and Colitis of patients go on to develop new intestinal lesions in the Organization (18). To be included in the study, patients had to have neoterminal ileum within the first year if no therapy is pre- undergone an endoscopic examination at least 6 months after scribed early after surgery (3), a process known as post- starting anti-TNF therapy for POR. Exclusion criteria were the operative recurrence (POR). presence of transient or definitive ostomy at the time anti-TNF Active smoking, penetrating disease behavior, perianal disease, therapy was started, a Rutgeerts endoscopic score , i2 at the last and previous intestinal resection have been identified as risk factors endoscopic examination before anti-TNF treatment, anti-TNF for POR (4,5). Nonetheless, these factors do not afford an adequate therapy indicated for primary prevention of POR, and less than 6 prediction of POR nor a stratification of patients for prevention months of treatment with anti-TNF therapy at the time of the last strategies, and patients may be overtreated or undertreated if a endoscopic assessment. The collected variables and the case record systematic policy is followed in this scenario. Therefore, the ap- database (19) are detailed in the Supplemental Digital Content 1 propriateness of preventive therapies is still a matter of debate, even (http://links.lww.com/CTG/A339). in patients at a high risk of POR (6). The fact that mucosal lesions The study was approved by the Ethics Committee of the co- (as assessed by ileocolonoscopy and called endoscopic POR) occur ordinating center (Hospital Universitari Germans Trias i Pujol, early after surgery and always precede the development of symp- Badalona, Catalonia, Spain). toms (clinical POR) (4), and the existence of a fine correlation between their severity and the likeliness of developing clinical POR Outcomes and definitions (3), are the main arguments for advising early endoscopic moni- Endoscopic response to treatment was assessed in those ileoco- toring in these patients, regardless of the fact that preventive lonoscopies performed during anti-TNF therapy or within the therapy has been initiated (7–9). In the absence of lesions or in first 3 months after anti-TNF discontinuation. Endoscopic re- cases of mild mucosal lesions, some guidelines recommend a mission was defined as a Rutgeerts score , i2, whereas endoscopic conservative attitude, together with a close monitoring with non- improvement was defined by any reduction in the baseline Rut- invasive diagnostic tools afterward (8,10). By contrast, if severe geerts score. Finally, we defined advanced endoscopic POR as a lesions are noticed in this endoscopic assessment, treatment esca- Rutgeerts endoscopic score of i3 or i4. lation is strongly recommended (8,9). We also assessed the induction of clinical remission in patients Among the many drugs that have been tested in this clinical with clinical POR at the beginning of anti-TNF therapy and the setting, only thiopurines and antitumor necrosis factor (TNF) development of clinical POR during anti-TNF therapy or within agents have proven to be efficient in preventing endoscopic and the 3 months after treatment discontinuation. For this purpose, clinical POR (11,12). The POCER study, a randomized clinical and given that no clinical score has been validated to date for trial (RCT) designed to compare early vs conventional endo- clinical POR and these patients may develop chronic diarrhea as a scopic monitoring, demonstrated that early treatment escalation consequence of ileocolic resections (i.e., intestinal bacterial in patients with severe lesions results in a better short-term out- overgrowth and bile salt malabsorption), we defined clinical POR come even if they were on preventive therapy with thiopurines as the presence of 2 of the following 3 criteria: weight loss, increase or adalimumab (13). It therefore seems reasonable that anti-TNF in stool frequency of at least 2 bowel movements/day, and new agents should be used in patients developing endoscopic onset of abdominal pain. POR while on thiopurines or even in those patients who did not undergo any preventive therapy. Nevertheless, in the setting of asymptomatic patients with endoscopic POR, only one Statistical analysis RCT comparing the usefulness of azathioprine and oral Categorical variables were expressed as frequencies and percent- aminosalicylates 5-ASA in the prevention of clinical POR has ages. Continuous variables were described as means 6 SD or as been published to date, and no other drugs have been evaluated by medians and interquartile range for cases with a skewed distribu- the means of RCTs for the treatment of confirmed endoscopic tion. Normal distribution was assessed using normal Q-Q plots. POR (14). For anti-TNF agents, only some retrospective series Statistical differences between groups were assessed using the x2 and small, open studies are available, reflecting the need for test for categorical variables, the Student t test, or the Fisher exact further data on this issue (15–17), and the number of patients test for continuous variables with normal distribution or the starting adalimumab therapy for endoscopic POR in the POCER Kruskal-Wallis test for variables with a non-normal distribution. study was rather small (13). Univariable and multivariable logistic regression models were The aim of this study was to describe the use of infliximab and performed applying the backward stepwise procedure to determine adalimumab for the treatment of endoscopic POR in clinical which variables were independent predictors of endoscopic im- practice, to assess and compare their effectiveness, and to identify provement and remission. Variables included in the multivariable the factors associated with treatment outcomes. analysis were those significantly associated with the end point or Clinical and Translational Gastroenterology VOLUME 11 | AUGUST 2020 www.clintranslgastro.com Anti-TNF Treatment for Postoperative CD 3 with a P value , 0.05 in the univariable analysis. Odds ratios (ORs) median time from the index ileocolonoscopy to the beginning of with 95% confidence intervals (CIs) are reported. anti-TNF therapy was 2 months (IQR, 1–5), without differences In a secondary analysis and considering the difficulty of between infliximab and adalimumab-treated patients (2 [IQR, quantifying the association of infliximab or adalimumab with 1–6] vs 2 [IQR, 1–5] months, respectively; P 5 0.08). As men- INFLAMMATORY BOWEL DISEASE endoscopic improvement and remission in an unbiased manner, tioned previously, 53 patients had been exposed to anti-TNF a propensity score analysis was also performed with a logistic before the index surgery; of them, 24 (45%) were further treated regression model. The dependent variable was receipt treatment with the same anti-TNF for endoscopic POR, particularly among with infliximab, and a total of 17 covariates were selected (see those previously exposed to infliximab (67% vs 31%; P 5 0.01). Table, Supplemental Digital Content 2, http://links.lww.com/ Almost all the patients followed a conventional induction and CTG/A340). According to the propensity score, patients were maintenance regimen schedule for each anti-TNF drug (5 mg/kg selected by 1:1 matching without replacement using the nearest at weeks 0, 2, 6 and every 8 weeks for infliximab and 160 mg neighbor method. A caliper width of 0.2 standardized differences followed by 80 mg and 40 mg every 2 weeks, for adalimumab). was used for matching. The balance of the baseline characteristics’ The use of concomitant immunosuppressants was common distribution between the 2 groups was evaluated using the abso- among both infliximab and adalimumab-treated patients. lute standardized differences. Differences were considered sta- The median time of follow-up while on anti-TNF therapy was tistically significant at P , 0.05. All analyses were performed with 51 months. One-third of patients were dose escalated, almost in STATA V.13.0 (College Station, TX). half of them because of a lack of endoscopic improvement or biological activity and a large proportion for the development of RESULTS symptoms attributed to disease activity. Moreover, one-third Baseline characteristics discontinued the initial anti-TNF treatment, mainly because of A total of 179 patients were included, 96 (54%) of whom were clinical, biological, or endoscopic worsening or the development undergoing treatment with adalimumab and 83 (46%) with of drug-related adverse events. Once again, none of these features infliximab because of endoscopic POR as conventionally defined differed between infliximab and adalimumab-treated patients. by a Rutgeerts endoscopic score . i1. Table 1 summarizes the baseline characteristics of the included patients, and Supple- Treatment outcomes mental Digital Content 2 (see Table, http://links.lww.com/CTG/ During the follow-up, all the patients underwent at least one en- A340) details the baseline characteristics of patients in the pro- doscopic assessment (55% one, 31% two, and 14% three or more pensity score-matched cohorts. Of note, almost two-thirds had colonoscopies). The first endoscopic assessment was performed been exposed to thiopurines and one-third to anti-TNFs before after a median of 16 months (interquartile range [IQR], 11–31) the index surgery. As expected, 76% of patients had at least one from the initiation of anti-TNF therapy (no differences between well-established risk factor for POR (active smoking, penetrating treatment groups), whereas the last endoscopic assessment was disease behavior, previous intestinal resections, or history of performed after a median of 31 months (IQR, 13–54). At the last perianal disease) and more than one-third had more than one risk endoscopic assessment performed while on anti-TNF therapy, 109 factor, active smoking at surgery being the most common one. of 179 patients (61%) had achieved endoscopic improvement, in- Regarding CD treatment, 28% of the patients did not follow any cluding 65 patients (42%) who achieved endoscopic remission. preventive therapy for POR after the index surgery, 16% followed Infliximab-treated patients showed significantly higher rates of long-term mesalazine or a short course of metronidazole, whereas endoscopic response (70% vs 53%; P 5 0.02) and endoscopic re- 56% were on thiopurines (alone or together with a 3-month mission (57% vs 29%; P , 0.01) (Figure 1). course of metronidazole). All these demographic and clinical Among the 39 patients who also had clinical POR at the be- characteristics were evenly distributed between the 2 treatment ginning of anti-TNF therapy, 23 (59%) achieved clinical re- groups. mission at the end of the follow-up (61% with infliximab and 57% The first endoscopic assessment after the index surgery was with adalimumab; P 5 0.80). On the other hand, among the 140 performed after a median of 16 months (IQR, 8–56); however, the patients who were asymptomatic despite having mucosal lesions index ileocolonoscopy demonstrating endoscopic POR before at the beginning of anti-TNF therapy, 15 (11%) met the criteria starting anti-TNF therapy was performed after a median of 41 for clinical POR (12% with infliximab and 9% with adalimumab) months (IQR, 13–78) from surgery. Of note, 63% of the cohort after a median time of 31 months (IQR, 7–44). The cumulative had advanced endoscopic POR at the baseline endoscopy (Rut- probability of remaining free of clinical POR was 96%, 91%, and geerts score . i2). Furthermore, 39 patients (22%) met the ar- 86% at 1, 3, and 5 years, respectively. No differences were ob- bitrarily predefined criteria for clinical POR at the beginning of served between infliximab and adalimumab-treated patients. the anti-TNF therapy. Despite the presence of mucosal in- flammatory lesions, not all patients exhibited biological evidence Factors associated with treatment outcomes of disease activity; C-reactive protein measurements were avail- The results of the univariable and multivariable factors associated able in all the patients at the time anti-TNF therapy was started, with endoscopic improvement and endoscopic remission are and only 31% of them showed levels .5 mg/L (Table 2). shown in Table 4. Infliximab therapy (as opposed to adalimumab) for endoscopic POR (OR 3.16 [95% CI 1.65–6.05]; P , 0.01), Main features of anti-TNF therapy concomitant thiopurines (OR 2.15 [95% CI 1.04–4.46]; P 5 0.03), Tables 2 and 3 summarize the main clinical and therapeutic and the presence of clinical POR at the start of anti-TNF therapy characteristics at the beginning of anti-TNF therapy and during (OR 3.31 [95% CI 1.51–7.28]; P 5 0.03) were the factors in- follow-up, respectively. The median interval from surgery to the dependently associated with endoscopic remission in the multi- beginning of anti-TNF therapy was 44 months, with more than variable analysis. Moreover, endoscopic improvement was 77% of patients starting beyond the first year after surgery. The independently associated with having advanced endoscopic POR American College of Gastroenterology Clinical and Translational Gastroenterology 4 Cañete et al. Table 1. Baseline characteristics of patients including clinical features at index surgery Overall (n 5 179) Infliximab (n 5 83) Adalimumab (n 5 96) P Value INFLAMMATORY BOWEL DISEASE Male gender 98 (55) 51 (61) 47 (49) 0.09 Familial with inflammatory 18 (10) 8 (10) 10 (11) 0.85 bowel disease Active smoking at CD diagnosis 103 (58) 53 (64) 50 (52) 0.11 Disease location Ileal 97 (54) 47 (57) 50 (52) 0.54 Ileocolonic 82 (46) 36 (43) 46 (48) 0.54 Upper gastrointestinal involvement 5 (3) 2 (2) 3 (3) 0.77 Disease behavior Inflammatory 14 (8) 6 (7) 8 (8) 0.78 Stricturing 78 (44) 38 (46) 40 (42) 0.58 Penetrating 87 (49) 39 (47) 48 (50) 0.69 Extraintestinal manifestations 50 (28) 19 (23) 31 (32) 0.16 Rheumatologic 40 (22) 15 (18) 25 (26) 0.20 Cutaneous 13 (7) 5 (6) 8 (8) 0.55 Ocular 5 (3) 4 (5) 1 (1) 0.13 Thrombotic 1 (1) 1 (1) 0 (0) 0.28 Age group at diagnosis ,18 yr 19 (11) 7 (8) 12 (13) 0.38 18–40 yr 130 (73) 60 (72) 70 (73) 0.92 .40 yr 30 (17) 16 (19) 14 (15) 0.40 Indication of index surgery Intestinal stenosis 92 (51) 48 (58) 44 (46) 0.11 Intra-abdominal penetrating complication 67 (37) 25 (30) 42 (44) 0.06 Refractoriness to medical therapy 20 (11) 10 (12) 10 (10) 0.73 Thiopurines exposure before 100 (56) 50 (60) 50 (52) 0.27 index surgery Anti-TNF exposure before index surgery 53 (30) 21 (25) 32 (33) 0.24 Active smoking at index surgery 86 (48) 44 (53) 42 (44) 0.22 Heavy smokers (.10 cigarettes/d) 54 (75) 27 (77) 27 (73) 0.68 Penetrating behavior 87 (49) 39 (47) 48 (50) 0.69 Perianal disease before index surgery 30 (17) 16 (19) 14 (15) 0.40 Intestinal resections before 30 (17) 14 (17) 16 (17) 0.97 index surgery Median (range) 1 (1–3) 1 (1–2) 1 (1–3) At least 1 risk factor for POR 136 (76) 63 (76) 73 (76) 0.75 .1 known risk factor for POR 66 (37) 31 (37) 35 (36) 0.90 Primary prevention for POR after index surgery None 51 (28) 20 (24) 31 (32) 0.23 Mesalazine 11 (6) 2 (2) 9 (9) 0.06 Metronidazole 17 (10) 10 (6) 7 (4) 0.58 Thiopurines 81 (45) 40 (48) 41 (43) 0.46 Thiopurines 1 metronidazole 19 (11) 11 (13) 8 (8) 0.29 CD, Crohn’s disease; POR, postoperative recurrence; TNF, tumor necrosis factor. Clinical and Translational Gastroenterology VOLUME 11 | AUGUST 2020 www.clintranslgastro.com Anti-TNF Treatment for Postoperative CD 5 Table 2. Main features at the beginning of anti-TNF therapy for endoscopic POR Overall (n 5 179) Infliximab (n 5 83) Adalimumab (n 5 96) P Value INFLAMMATORY BOWEL DISEASE Time from index surgery to index 41 (13–78) 40 (18–77) 46 (12–81) 0.85 ileocolonoscopy (mo) Rutgeerts score at the beginning of anti-TNF Rutgeerts i2 65 (36) 30 (36) 35 (36) 0.96 Rutgeerts i3 47 (26) 23 (28) 24 (25) 0.68 Rutgeerts i4 67 (37) 30 (36) 37 (39) 0.74 Median time (IQR) from surgery 44 (18–87) 41 (20–84) 47 (15–96) 0.82 to anti-TNF (mo) Clinical POR at anti-TNF start 39 (22) 18 (22) 21 (22) 0.98 Exposure to the same anti-TNF prior 24 (45) 14 (67) 10 (31) 0.01 to index surgery Anti-TNF induction schedule 176 (98) 82 (99) 94 (98) 0.65 Concomitant immunosuppressants Thiopurines 124 (69) 62 (75) 62 (65) 0.14 Methotrexate 11 (6) 6 (7) 5 (5) 0.57 C-reactive protein .5 mg/L 55 (31) 24 (29) 31 (33) 0.63 Hemoglobin , 12 g/dL (men) 23 (13) 11 (13) 12 (13) 0.88 or ,11 g/dL (women) Faecal calprotectin .100 mg/kga 36 (65) 17 (71) 19 (61) 0.46 Bold indicates statistical significance with a P value less than 0.05. IQR, interquartile range; POR, postoperative recurrence; TNF, tumor necrosis factor. a Only available for 55 patients. (OR 2.63 [95% CI 1.37–5.07]; P , 0.01), concomitant use of in this setting (14–16,22). Thiopurines, although useful (14), are thiopurines (OR 2.34 [95% CI 1.18–4.62]; P 5 0.01), and inflix- hampered by a high rate of intolerance and their slow active imab therapy for endoscopic POR (OR 2.01 [95% CI 1.05–3.88]; P mechanism, and they may not be the most suitable option in 5 0.04). In a secondary analysis, and to establish the association patients with already existing clinical POR. Therefore, biological of infliximab with the outcome in an unbiased manner, a pro- agents seem to be the best potential choice in this clinical scenario. pensity score analysis was performed. Propensity score yielded 65 In previous small series, the reported rate of endoscopic remission matched pairs of patients from both groups (infliximab and with infliximab ranged from 47% to 52% (15,17). To the best of adalimumab). The distribution of created propensity score is our knowledge, this is the largest cohort assessing the usefulness presented in Supplemental Digital Content 3 (see Figure, http:// of anti-TNF agents as rescue therapy for endoscopic POR in CD, links.lww.com/CTG/A341). Infliximab therapy for POR was and we observed that anti-TNFs are able to improve or revert significantly associated with endoscopic improvement and re- recurrent mucosal lesions in more than half of the patients. mission (OR 2.2 [95% CI 1.1–4.5]; P 5 0.03, and OR 2.2 [95% CI As expected, most of our patients were at a high risk of POR. 1.1–4.6]; P 5 0.03, respectively). Although smoking is probably the greatest and most repeatedly found risk factor for POR, a great proportion of patients keep on DISCUSSION smoking after surgery, reflecting that beyond the prescription of Current guidelines on the postoperative management of CD efficient drug therapies, there is still room for improvement in recommend the primary prevention of POR in all patients except preventive measures by physicians and nurses in this particularly those with pure fibrotic and short ileal stenosis (8,9). However, risky clinical scenario. We also observed that half of the patients the need for systematic primary prophylaxis for POR is under in our cohort were treated with anti-TNF agents for endoscopic debate, with some authors posing the use of an endoscopic-driven POR despite preventive therapy with thiopurines. This might strategy in which only those patients showing mucosal lesions explain that in a high proportion of our patients, anti-TNF agents would be treated (6). With this strategy, up to 55% of patients were initiated beyond the first year after the index surgery; in fact, without primary prevention will need rescue therapy within the it has been recently reported that there is a steady risk of endo- first year after surgery. On the other hand, even in patients who scopic POR over time among those patients on preventive thio- start thiopurines early after surgery, endoscopic POR occurs in up purines with a first ileocolonoscopy without mucosal lesions (23). to 40% within 1 year (11,20,21). Beyond the early postoperative In fact, almost half of those patients who started anti-TNF beyond management of CD, it is a fact that approximately half of these 18 months after surgery had an initial ileocolonoscopy with a patients develop early mucosal lesions that put them at risk of Rutgeerts score , i3. Anyway, preventive thiopurine therapy for clinical POR. Mesalazine has demonstrated a very limited efficacy POR, and exposure to thiopurines or anti-TNF agents before the American College of Gastroenterology Clinical and Translational Gastroenterology 6 Cañete et al. Table 3. Follow-up features and outcomes of anti-TNF therapy for endoscopic postoperative recurrence Overall (n 5 179) Infliximab (n 5 83) Adalimumab (n 5 96) P Value INFLAMMATORY BOWEL DISEASE Median (IQR) follow-up time on anti-TNF (mo) 51 (30–79) 58 (30–83) 47 (27–64) 0.08 Median (IQR) time between anti-TNF start 16 (11–31) 17 (11–29) 14 (10–32) 0.32 and first endoscopic assessment (mo) Time between anti-TNF start and first 99 (55) 45 (54) 54 (56) 0.79 endoscopic assessment , 18 mo Median (IQR) time between anti-TNF start 31 (13–54) 32 (16–58) 31 (12–47) 0.24 and last endoscopic assessment (mo) Anti-TNF dosemescalation 66 (37) 27 (33) 39 (41) 0.26 Reasons for anti-TNF dose escalation (could be more than one) Lack of endoscopic improvement 30 (45) 7 (26) 23 (59) 0.01 Digestive symptoms 42 (64) 19 (70) 23 (59) 0.34 Raised inflammatory biomarkers 10 (15) 3 (11) 7 (18) 0.35 Anti-TNF discontinuation 53 (30) 25 (30) 28 (29) 0.89 Main reasons for anti-TNF discontinuation Clinical, biological or endoscopic 23 (49) 11 (48) 12 (50) 0.88 worsening Adverse events 14 (30) 7 (30) 7 (29) 0.92 Other 10 (21) 5 (22) 5 (20) 0.94 Any C-reactive protein .5 mg/L a 21 (12) 11 (13) 10 (11) 0.57 Any fecal calprotectin .100 mg/kgb 29 (30) 13 (33) 16 (29) 0.68 Bold indicates statistical significance with a P value less than 0.05. IQR, interquartile range; TNF, tumor necrosis factor. a Available for 176 patients. b Available for 96 patients. index surgery, were not associated with a lower efficacy of anti- escalation for POR in Spain is advised only for advanced endo- TNF therapy once POR occurred. Another reason for the delay in scopic POR (8); for that matter, almost two-thirds of our patients starting anti-TNF agents could have been the fact that treatment had advanced endoscopic POR at the index ileocolonoscopy, Figure 1. Proportion of treatment success per type of antitumor necrosis factor. Clinical and Translational Gastroenterology VOLUME 11 | AUGUST 2020 www.clintranslgastro.com Anti-TNF Treatment for Postoperative CD 7 Table 4. Factors associated with treatment outcomes Endoscopic remission Endoscopic improvement INFLAMMATORY BOWEL DISEASE Univariable analysis Multivariable analysis Univariable analysis Multivariable analysis OR (95% CI) P Value OR (95% CI) P Value OR (95% CI) P Value OR (95% CI) P Value Age at surgery 1.02 (0.99–1.05) 0.05 1.02 (0.99–1.04) 0.16 Female gender 0.90 (0.29–1.64) 0.75 0.70 (0.38–1.29) 0.26 Penetrating behavior 0.89 (0.44–1.45) 0.46 0.69 (0.38–1.26) 0.22 Perianal disease before surgery 1.26 (0.57–2.78) 0.56 1.35 (0.59–3.08) 0.48 Extraintestinal manifestations 2.36 (0.74–7.53) 0.14 1.53 (0.77–3.05) 0.23 Active smokers at index surgery 0.91 (0.50–1.65) 0.75 1.33 (0.97–1.83) 0.08 Previous intestinal resections 0.91 (0.41–2.02) 0.82 0.96 (0.43–2.13) 0.91 Anti-TNF exposure before a surgery 1.22 (0.64–2.33) 0.55 1.96 (0.98–3.93) 0.06 Advanced endoscopic POR 0.68 (0.37–1.28) 0.24 2.38 (1.27–4.45) ,0.01 2.63 (1.37–5.07) ,0.01 Postoperative prevention with 1.95 (1.06–3.6) 0.03 1.63 (0.89–2.98) 0.12 thiopurines Concomitant thiopurines 2.21 (1.12–4.37) 0.02 2.15 (1.04–4.46) 0.03 2.26 (1.18–4.32) 0.01 2.34 (1.18–4.62) 0.01 At least 1 risk factor for POR 0.75 (0.39–1.46) 0.40 0.60 (0.29–1.25) 0.17 Clinical POR at anti-TNF start 2.79 (1.34–5.80) ,0.01 3.31 (1.51–7.28) 0.03 2.18 (0.98–4.81) 0.06 Infliximab therapy for POR 3.17 (1.71–5.88) ,0.01 3.16 (1.65–6.05) ,0.01 2.05 (1.10–3.79) 0.02 2.01 (1.05–3.88) 0.04 Time from index surgery to 0.99 (0.99–1.00) 0.96 1.00 (0.99–1.01) 0.99 anti-TNF starta Time from CD diagnosis to 1.00 (0.99–1.00) 0.65 1.00 (0.99–1.00) 0.72 index surgerya Endoscopic POR .12 mo 0.85 (0.42–1.73) 0.65 0.92 (0.44–1.89) 0.81 from surgery Bold indicates statistical significance with a P value less than 0.05. CD, Crohn’s disease; CI, confidence interval; OR, odds ratio; POR, postoperative recurrence; TNF, tumor necrosis factor. a Time expressed in months. whereas most of early endoscopic POR correspond to i2 in advantage (if any) for the adalimumab-treated patients who might reported series (13,20). benefit the most from a “switch strategy,” although we lack the One of the most interesting findings of this study are those information on immunogenicity data. To confirm the results of the factors associated with a higher efficacy of anti-TNF for im- logistic regression analysis, we performed a propensity score proving or reverting postoperative mucosal lesions. We found analysis using 16 covariates for matching and treatment with that concomitant use of thiopurines increased the efficacy of infliximab as the dependent variable, including 130 matched pa- anti-TNF agents. Two RCTs comparing, respectively, infliximab tients (65 treated with infliximab and 65 with adalimumab), and and adalimumab in monotherapy or in combination with aza- the results remained the same. Therefore, it seems that at least in thioprine for the treatment of active luminal CD showed that the postoperative setting, infliximab is superior to adalimumab for combination therapy increased the rate of mucosal healing after 6 improving recurrent lesions. months (24,25). Moreover, combination therapy is associated Finally, we are aware of some additional limitations of our study. with higher trough levels of anti-TNF, preventing underexposure First, the lack of therapeutic drug measurements. This might affect to the drug and potentially increasing their efficacy, particularly in several ways: treatment dosing was not changed because of when mucosal healing is the therapeutic end point (26). Hence, it therapeutic drug monitoring in most patients, and we do not know seems reasonable that in patients with endoscopic POR in whom whether those patients who discontinued therapy before surgery did anti-TNF will be started, this should be in combination with it because of primary or secondary loss of response, with or without thiopurines whenever possible, and regardless, the patients was dosing adjustments based on therapeutic drug monitoring findings. on preventive thiopurine therapy. Moreover, we did not have data on immunogenicity that might be We also found a greater efficacy of infliximab over adalimumab. particularly relevant in the case of reinitiation of infliximab, which Despite the lack of face-to-face studies, both drugs have shown might have favored adalimumab. However, recent studies showed similar results in inducing mucosal healing in the setting of RCTs in controversial results regarding the utility of trough drug levels to luminal CD (27). In our study, there was a significant higher predict the efficacy of anti-TNF in preventing endoscopic POR proportion of patients who had been previously exposed to the (28–30). Second, endoscopic images were not recorded, and no same anti-TNF in the infliximab group; this would have been an central reading was possible; therefore, the information regarding American College of Gastroenterology Clinical and Translational Gastroenterology 8 Cañete et al. the Rutgeerts endoscopic score relied on local endoscopic reports. Takeda, and Janssen. M.R. as served as a speaker, a consultant, and Finally, the time point for endoscopic assessment was not pre- advisory member for MSD, AbbVie and Janssen. E.D. has served as a established and the response to treatment was not homogenously speaker or has received research or educational funding or advisory assessed in all patients. Despite this, there were no differences be- fees from MSD, AbbVie, Takeda, Kern Pharma, Pfizer, Janssen, INFLAMMATORY BOWEL DISEASE tween the 2 treatment groups regarding the time to therapeutic Celgene, Adacyte Therapeutics, Otsuka Pharmaceuticals, Ferring, endoscopic assessment. Shire Pharmaceuticals, Tillots, Thermofisher, Grifols, and Gebro. All In summary, anti-TNF therapy for endoscopic POR resulted in the remaining authors declared no conflicts of interest. a complete reversal of mucosal lesions in half of the patients and clinical improvement in more than half of the symptomatic pa- ACKNOWLEDGMENTS tients. We also found that the combination therapy of infliximab Fiorella Cañete received a research grant from the Societat Cat- and thiopurines resulted in a higher rate of mucosal improvement. alana de Digestologia. CONFLICTS OF INTEREST Guarantor of the article: Eugeni Domènech, MD, PhD. Appendix 1. INFLIRECU study investigators’ complete affiliations Specific author contributions: F.C. and E.D. designed the study, performed statistical analyses, interpreted the results, and drafted the Alicia Algaba, Hospital Universitario de Fuenlabrada (Fuenlabrada, manuscript. Germán Cediel designed, performed, and helped in the Spain; and Instituto de Investigación Sanitaria Hospital La Paz (IdiPaz) interpretation of the statistical analysis. The remaining authors (Madrid, Spain); Manuel Barreiro-de Acosta, Hospital Clı́nico collected data and critically reviewed the manuscript. All authors are Universitario de Santiago (Santiago de Compostela, Spain); Jesús aware and agree to the content of the manuscript and accept their Barrio, Hospital Universitario Rı́o Hortega (Valladolid, Spain); Olga authorship. Benı́tez-Leiva, Hospital Universitari Mútua Terrassa (Terrassa, Financial support: None to report. Catalonia, Spain); Luis Bujanda, Hospital Universitario de Donostia, Potential competing interests: F.C. has served as a speaker or has Instituto Biodonostia (San Sebastián, Spain), and Universidad del Paı́s received educational grants from Takeda, Janssen, MSD, and Ferring. Vasco (UPV/EHU) (San Sebastián, Spain), and CIBERehd (Madrid, M.M. has served as a speaker, has received research or educational Spain); Margalida Calafat, Hospital Universitari Germans Trias i Pujol funding or advisory fees from Danone, Allergan, Almirall, MSD, (Badalona, Catalonia, Spain), and Universitat Autònoma de Barcelona AbbVie, Takeda, Janssen, and Ferring. R.C.G.-S. has served as a (Barcelona, Catalonia, Spain; Patricia Camo, Hospital General San speaker or has received educational grants from AbbVie, Takeda, and Jorge (Huesca, Spain); Fiorella Cañete, Hospital Universitari Germans Janssen. M.B.d.A. has served as a speaker, a consultant, and an ad- Trias i Pujol (Badalona, Catalonia, Spain), and CIBERehd (Madrid, visory member for or has received research funding from MSD, Spain), and Universitat Autònoma de Barcelona; Marı́a José Casanova, AbbVie, Hospira, Takeda, Janssen, Kern Pharma, Ferring, Fae- Hospital Universitario de La Princesa, Instituto de Investigación sFarma, ShirePharmaceuticals, Dr. FalkPharma, Chiesi, Gebro- Sanitaria Princesa (IIS-IP) (Madrid, Spain) and CIBERehd (Madrid, Pharma, OtsukaPharmaceuticals, and TillottsPharma; Pilar Nos. Spain); Germán Cediel, Heart Institute, Hospital Universitari Germans Educational activities, research projects, and scientific meetings Trias i Pujol (Badalona, Catalonia, Spain); Pedro G Delgado-Guillena, sponsored by AbbVie, MSD, Janssen, Takeda, Otsuka, Kern Pharma, Hospital General de Granollers (Granollers, Catalonia, Spain); Eugeni and Pfizer. E.I.-F. has served as a speaker or has received educational Domènech, Hospital Universitari Germans Trias i Pujol (Badalona, funding or advisory fees from Takeda, Janssen, MSD, and AbbVie. Catalonia, Spain), and CIBERehd (Madrid, Spain); Antonio Garcı́a- A.G. has served as a speaker, a consultant or advisory member for Herola, Hospital Marina Baixa de Villajoyosa (Alicante, Spain); MSD, AbbVie, Hospira, Pfizer, Takeda, Janssen, Ferring, Faes Farma, Ramiro C González-Sueyro, Hospital Universitari Clı́nic (Barcelona, Shire Pharmaceuticals, Dr. Falk Pharma, and Tillots Pharma. M.J.C. Catalonia, Spain); Jordi Gordillo, Hospital de La Santa Creu i Sant Pau has received education funding from Pfizer, Janssen, MSD, Ferring, (Barcelona, Catalonia, Spain); Ana Gutiérrez, Hospital General and AbbVie. M.N.-L. has served as a speaker, has received research or Universitario de Alicante (Alicante, Spain); Alejandro Hernández- educational funding or advisory fees from MSD, AbbVie, Takeda, Camba, Hospital Universitario Nuestra Señora de Candelaria (Santa Janssen, and Ferring. A.V. has served as a speaker, or has received Cruz de Tenerife, Spain); Esther Hinojosa, Hospital de Manises educational funding from MSD, AbbVie, Takeda, Tillots, and Shire (Valencia, Spain); Eva Iglesias-Flores, Instituto Maimónides de Pharmaceuticals. C.T. has served as a speaker, a consultant or advi- Investigación Biomédica de Córdoba (IMIBIC), and Hospital sory member for MSD, AbbVie, Hospira, Pfizer, Takeda, Janssen, Universitario Reina Sofı́a, and Universidad de Córdoba (Córdoba, Ferring, Faes Farma, Shire Pharmaceuticals, Dr. Falk Pharma, Gebro Spain); Pilar López Serrano, Hospital Universitario Fundación Pharma, and Tillots Pharma. M.V.D. has served as a speaker, has Alcorcón (Alcorcón, Spain); Mı́riam Mañosa, Hospital Universitari received research or educational funding from MSD, AbbVie, Germans Trias i Pujol (Badalona, Catalonia, Spain), and CIBERehd Takeda, Janssen, Kern Pharma, and Ferring. A.H.C. has served as a (Madrid, Spain); Marı́a Dolores Martı́n-Arranz, Hospital Universi- speaker or has received educational funding from AbbVie, Takeda, tario La Paz, Instituto de Investigación Sanitaria La Paz (IdiPaz) Pfizer, Janssen, Ferring, Shire Pharmaceuticals, and Kern Pharma. (Madrid, Spain); Carmen Muñoz-Villafranca, Hospital Universitario L.R. has served as a speaker, or has received educational funding from Basurto (Bilbao, Spain); Pablo Navarro, Hospital Clı́nico Universitario MSD, AbbVie, Takeda, Pfizer, Janssen, Ferring, and Shire Pharma- de Valencia (Valencia, Spain); Mercè Navarro-Llavat, Hospital de Sant ceuticals; Ramón Pajares has received fees as a speaker from Janssen Joan Despı́ Moisès Broggi (Sant Joan Despı́, Catalonia, Spain); Pilar and Salvat. M.D.M.-A. has received fees as a speaker, consultant, an Nos, Hospital Universitario y Politécnico La Fe (Valencia, Spain) and advisory member for, and has received research funding from MSD, CIBERehd (Madrid, Spain); Ramón Pajares, Hospital Universitario AbbVie, Hospira, Pfizer, Takeda, Janssen, Shire Pharmaceuticals, Infanta Sofı́a (San Sebastián de los Reyes, Spain); Isabel Pérez- Tillotts Pharma, and Faes Pharma. A.A. has served as a speaker, has Martı́nez, Hospital Universitario Central de Asturias (Oviedo, Spain); received research grant or advisory fees from MSD, Lilly, Roche, Marta Piqueras, Consorci Sanitari de Terrassa (Terrassa, Catalonia, Clinical and Translational Gastroenterology VOLUME 11 | AUGUST 2020 www.clintranslgastro.com Anti-TNF Treatment for Postoperative CD 9 Spain); Alejandra F. Pordomingo, Hospital Universitario de Salamanca 3. Rutgeerts P, Geboes K, Vantrappen G, et al. Predictability of the (Salamanca, Spain); Patricia Ramı́rez-de la Piscina, Hospital Universi- postoperative course of Crohn’s disease. Gastroenterology 1990;99: 956–63. tario Araba (Vitoria, Spain); Laura Ramos, Hospital Universitario de 4. Domènech E, Mañosa M, Lobatón T, et al. Optimizing post-operative Canarias (La Laguna, Spain); Raquel Rı́os-León, Hospital Universitario Crohn’s disease treatment. Ann Gastroenterol 2014;27:313–9. INFLAMMATORY BOWEL DISEASE Ramón y Cajal (Madrid, Spain); Montserrat Rivero, Hospital Uni- 5. Auzolle C, Nancey S, Tran-Minh ML, et al. Male gender, active smoking and previous intestinal resection are risk factors for post-operative versitario Marqués de Valdecilla and IDIVAL (Santander, Spain); Iago endoscopic recurrence in Crohn’s disease: Results from a prospective Rodrı́guez-Lago, Hospital de Galdakao (Galdakao, Spain), and Biocruces cohort study. Aliment Pharmacol Ther 2018;48:924–32. Health Research Institute, Barakaldo (Vizcaya, Spain); Empar Sáinz- 6. Ferrante M, Papamichael K, Duricova D, et al. Systematic versus endoscopy-driven treatment with azathioprine to prevent postoperative Arnau, Hospital Arnau de Vilanova (Lleida, Catalonia, Spain); Manuela ileal Crohn’s disease recurrence. J Crohns Colitis 2015;9:617–24. Sampedro, Hospital de Mataró (Mataró, Catalonia, Spain); Katja Serra, 7. Gionchetti P, Dignass A, Danese S, et al. 3rd European evidence-based Hospital Universitario de Bellvitge (L’Hospitalet de Llobregat, Catalonia, consensus on the diagnosis and management of Crohn’s disease 2016: Part 2: Surgical management and special situations. J Crohns Colitis 2017; Spain); Carlos Taxonera, Hospital Clı́nico San Carlos, Instituto de 11:135–49. Investigación del Hospital Clı́nico San Carlos (IdISSC) (Madrid, Spain); 8. Domènech E, López-Sanromán A, Nos P, et al. Recommendations of the Leyanira Torrealba, Hospital Univesitario Dr. Josep Trueta (Girona, Spanish Working Group on Crohn’s Disease and Ulcerative Colitis (GETECCU) on the monitoring, prevention and treatment of post- Catalonia, Spain); Paola Torres, Hospital Universitari Germans Trias i operative recurrence in Crohn’s disease. Gastroenterol Hepatol 2017;40: Pujol Badalona, Catalonia, Spain); Marina Valls-Gandı́a, Hospital 472–83. General Universitario de Castellón (Castellón, Spain); Manuel Van 9. Nguyen GC, Loftus EV Jr, Hirano I, et al. American Gastroenterological Association Institute guideline on the management of Crohn’s disease Domselaar, Hospital Universitario de Torrejón (Torrejón de Ardoz, after surgical resection. Gastroenterology 2017;152:271–5. Spain); Cristina Verdejo Gil, Hospital General Universitario de Ciudad 10. Guardiola J, Lobatón T, Cerrillo E, et al. Recommendations of the Spanish Real (Ciudad Real, Spain); Albert Villoria, Hospital Parc Taulı́ (Sabadell, Working Group on Crohn’s Disease and Ulcerative Colitis (GETECCU) Catalonia, Spain), and Universitat Autònoma de Barcelona (Barcelona, on the utility of the determination of faecal calprotectin in inflammatory bowel disease. Gastroenterol Hepatol 2018;41:514–29. Catalonia, Spain), and CIBERehd (Madrid, Spain). 11. López-Sanromán A, Vera-Mendoza I, Domènech E, et al. Adalimumab vs azathioprine in the prevention of postoperative Crohn’s disease Study Highlights recurrence. A GETECCU randomised trial. J Crohns Colitis 2017;11: 1293–301. WHAT IS KNOWN 12. Cañete F, Mañosa M, Casanova MJ, et al. Adalimumab or infliximab for the prevention of early postoperative recurrence of Crohn disease: Results 3 Seventy percent of patients with CD develop POR within the from the ENEIDA registry. Inflamm Bowel Dis 2019;25:1862–70. 13. De Cruz P, Kamm MA, Hamilton AL, et al. Crohn’s disease management first year if no therapy is prescribed early after surgery. after intestinal resection: A randomised trial. Lancet 2015;385:1406–17. 3 Thiopurines and anti-TNF agents have proven to be efficient 14. Reinisch W, Angelberger S, Petritsch W, et al. Azathioprine versus mesalazine in preventing endoscopic and clinical POR. for prevention of postoperative clinical recurrence in patients with Crohn’s 3 Whether universal prevention or endoscopy-driven therapy is disease with endoscopic recurrence: Efficacy and safety results of a randomised, double-blind, double-dummy, multicentre trial. Gut 2010;59:752–9. the best strategy is still under debate. 3 No data are available on the efficacy of anti-TNF agents in the 15. Sorrentino D, Terrosu G, Paviotti A, et al. Early diagnosis and treatment of postoperative endoscopic recurrence of Crohn’s disease: Partial benefit setting of asymptomatic patients with endoscopic POR. by infliximab-a pilot study. Dig Dis Sci 2012;57:1341–8. 16. Yamamoto T, Umegae S, Matsumoto K. Impact of infliximab therapy WHAT IS NEW HERE after early endoscopic recurrence following ileocolonic resection of Crohn’s disease: A prospective pilot study. Inflamm Bowel Dis 2009; 3 Anti-TNF therapy for endoscopic POR achieves a complete 15(10):1460–6. reversal of mucosal lesions in almost half of the patients. 17. Assa A, Bronsky J, Kolho KL, et al. Anti-tnfa treatment after surgical 3 Combination therapy with thiopurines increases the likeliness resection for Crohn’s disease is effective despite previous pharmacodynamic failure. Inflamm Bowel Dis 2017;23:791–7. of improving mucosal lesions. 3 In the postoperative setting, infliximab seems to be superior to 18. Maaser C, Sturm A, Vavricka SR, et al. ECCO-ESGAR guideline for diagnostic assessment in IBD part 1: Initial diagnosis, monitoring of adalimumab for improving recurrent lesions. However, face- known IBD, detection of complications. J Crohns Colitis 2019;13:144–64. to-face prospective studies are needed in this setting. 19. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for TRANSLATIONAL IMPACT providing translational research informatics support. J Biomed Inform 2009;42:377–81. 3 Since this article is not a pre-clinical study we have not made a 20. Mañosa M, Cabré E, Bernal I, et al. Addition of metronidazole to translational impact. azathioprine for the prevention of postoperative recurrence of Crohn’s 3 In the instructions and guidelines of the original article, said “if disease: A randomized, double-blind, placebo-controlled trial. Inflamm Bowel Dis 2013;19:1889–95. your manuscript is pre-clinical, briefly explain how the results 21. Rivière P, Vermeire S, Irles-Depe M, et al. Rates of post-operative recurrence may impact future clinical practice.” of Crohn’s disease and effects of immunosuppressive and biologic therapies. Clin Gastroenterol Hepatol 2020:pii: S1542–3565(20)30441–9. doi: 10. 1016/j.cgh.2020.03.064. [Epub ahead of print April 6, 2020.] 22. Zabana Y, Mañosa M, Cabré E, et al. Addition of mesalazine for subclinical post-surgical endoscopic recurrence of Crohn’s disease despite REFERENCES preventive thiopurine therapy: A case-control study. J Gastroenterol 1. Bossuyt P, Debeuckelaerea C, Ferrante M, et al. The operative risk and Hepatol 2014;29:1413–7. natural history after the diagnosis of ileal penetrating Crohn’s disease. Eur 23. Mañosa M, Puig M, Torres PF, et al. Risk of long-term post-operative J Gastroenterol Hepatol 2018;30:539–45. recurrence (POR) in Crohn’s disease patients with a first postoperative 2. Domènech E, Mañosa M, Cabré E. An overview of the natural history of normal endoscopic assessment under thiopurine prevention. J Crohns inflammatory Bowel disease. Dig Dis 2014;32:320–7. Colitis 2019;13(suppl 1):S275. American College of Gastroenterology Clinical and Translational Gastroenterology 10 Cañete et al. 24. Colombel JF, Sandborn WJ, Reinisch W, et al. Infliximab, azathioprine, or follow-up and impact of infliximab trough levels and antibodies to combination therapy for Crohn’s disease. N Engl J Med 2010;362:1383–95. infliximab. PLoS One 2015;10:e0144900. 25. Watanabe K, Matsumoto T, Hisamatsu T, et al. Clinical and pharmacokinetic 29. Wright EK, Kamm MA, De Cruz P, et al. Anti-TNF therapeutic drug factors associated with adalimumab-induced mucosal healing in patients with monitoring in postoperative Crohn’s disease. J Crohns Colitis 2018;12:653–61. Crohn’s disease. Clin Gastroenterol Hepatol 2018;16:542–9. 30. Fay S, Ungar B, Paul S, et al. The association between drug levels and INFLAMMATORY BOWEL DISEASE 26. Gibson DJ, Ward MG, Rentsch C, et al. Review article: Determination of endoscopic recurrence in postoperative patients with Crohn’s disease treated the therapeutic range for therapeutic drug monitoring of adalimumab and with tumor necrosis factor inhibitors. Inflamm Bowel Dis 2017;23:1924–9. infliximab in patients with inflammatory bowel disease. Aliment Pharmacol Ther 2020;51(6):612–28. 27. Cholapranee A, Hazlewood GS, Kaplan GG, et al. Systematic review with meta-analysis: Comparative efficacy of biologics for induction and Open Access This is an open-access article distributed under the terms of the maintenance of mucosal healing in Crohn’s disease and ulcerative colitis Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 controlled trials. Aliment Pharmacol Ther 2017;45:1291–302. (CCBY-NC-ND), where it is permissible to download and share the work pro- 28. Sorrentino D, Marino M, Dassopoulos T, et al. Low dose infliximab for vided it is properly cited. The work cannot be changed in any way or used prevention of postoperative recurrence of Crohn’s disease: Long-term commercially without permission from the journal. Clinical and Translational Gastroenterology VOLUME 11 | AUGUST 2020 www.clintranslgastro.com