Secondary Logo

Journal Logo

Original Articles: Gastroenterology: Inflammatory Bowel Disease

Predictors of Long-term Clinical and Endoscopic Remission in Children With Crohn Disease Treated With Infliximab

D’Arcangelo, Giulia; Oliva, Salvatore; Dilillo, Anna; Viola, Franca; Civitelli, Fortunata; Isoldi, Sara; Cucchiara, Salvatore; Aloi, Marina

Author Information
Journal of Pediatric Gastroenterology and Nutrition: June 2019 - Volume 68 - Issue 6 - p 841-846
doi: 10.1097/MPG.0000000000002262


What Is Known

  • Predicting relapses in patients with Crohn disease (CD) through noninvasive biomarkers could allow timely treatment modifications.
  • Mathematically weighted Pediatric Crohn disease Activity Index (wPCDAI) has comparable performance with pediatric CD activity index but superior feasibility.
  • Data on the ability of fecal calprotectin to predict relapse in pediatric Crohn disease are poor.

What Is New

  • Weighted pediatric CD activity index, fecal calprotectin, and C-reactive protein are poor predictors of clinical and endoscopic response to biological therapy in children with Crohn disease.
  • Postinduction fecal calprotectin plus wPCDAI can predict 1-year clinical and endoscopic outcomes.
  • Fecal calprotectin has a good correlation with endoscopic inflammation.

Mucosal healing (MH) is now recognized as a major goal of therapy in adult and pediatric Crohn disease (CD), although its detection requires a complete workout including endoscopy and imaging studies (1). Nevertheless, repeated colonoscopies are difficult to perform, particularly in children, and hardly accepted by both patients and parents. For this reason, clinical and laboratory predictors of MH are hardly warranted to select those patients at high risk of medical failure and needing strict disease monitoring, particularly those under biologics, due to the high cost of the therapy and the availability of new therapeutic options. Therapeutic drug monitoring (TDM) is able to predict long-term response to biological therapy. Higher drug concentrations during both induction and maintenance phases have been demonstrated to be related to favorable outcomes (2). The extensive use of TDM is, however, commonly limited by local resources, so it is mainly recommended for those patients losing response to biologics and not for routine disease monitoring (2). Among clinical monitoring tools, the recently developed weighted pediatric CD activity index (wPCDAI) is commonly used in routine clinical practice, due to a higher feasibility and a slightly better accuracy, compared to the original pediatric CD activity index (PCDAI) (3). Nevertheless, both clinical scores do not seem to correlate with mucosal inflammation (4). Among the laboratory markers, C-reactive protein (CRP) and fecal calprotectin (FC) are the most commonly used, although no data are available on the performance of those markers to predict endoscopic remission in children with CD.

These noninvasive markers, alone or combined, far from replacing endoscopy or imaging for the assessment of MH or bowel damage, could play an important role in monitoring disease course and response to therapy, possibly selecting those patients requiring a tight control and, eventually, endoscopy, while sparing this procedure to those who could hardly present endoscopic recurrence and are likely to respond to treatment.

We therefore aimed at investigating the ability of postinduction wPCDAI, CRP, and FC to predict a sustained steroid-free clinical and endoscopic remission in children with CD treated with infliximab (IFX), and their correlation with endoscopic and clinical remission at 1-year follow-up.


This was a prospective observational single-center study conducted from April 1, 2016 to April 1, 2017 at the Pediatric Gastroenterology and Liver Unit of the Sapienza University of Rome. The protocol was approved by the ethics committee of the Hospital. Parents and children older than 6 years gave written informed consent before any study-related procedure was performed. Inclusion criteria were age less than 18 years; a diagnosis of CD made according to standardized clinical, endoscopic, and histological criteria (5); treatment with IFX either as a first-line therapy at the diagnosis or as a result of a treatment escalation; and endoscopic evaluation at 0 and 48 weeks. Exclusion criteria included previous use of anti-tumor necrosis factor α; previous bowel surgeries; presence of any infection preventing the use of anti-tumor necrosis factor α agents; and/or needing specific treatment (latent tuberculosis, cytomegalovirus, and Clostridium difficile infection).

Disease location was defined using Paris classification (6). Clinical evaluation was performed at weeks 0, 14, and 52 through the wPCDAI; the previously validated cut-off values for remission (<12.5 points), mild (12.5–40), moderate (>40–57.5), and severe disease (>57.5), were considered (3). Blood and stool samples were collected at the enrollment (within 3 days before endoscopy), at weeks 14 and 52. The normal range for CRP in our laboratory was <6 mg/L. FC analysis was performed using a commercially available ELISA technique (Calprest; Eurospital SpA, Trieste, Italy). Based on previous data, FC ≥250 μg/g was considered a positive test, suggesting active disease (7).

A complete ileocolonoscopy was performed at baseline and at week 48 and the Simple Endoscopic Score for Crohn disease (SES-CD) was used for grading the endoscopic findings (8). MH was defined as an SES-CD <3 (3–6 mild, 7–15 moderate, >15 severe endoscopic activity). Endoscopy reading was performed by a single endoscopist blinded to patient’ clinical evaluation and laboratory examinations (S.O.).

Outcome Measures

Steroid-free clinical remission, defined as a wPCDAI <12.5 and no need for corticosteroid treatment, and MH were evaluated at 48 weeks.

The primary outcome was the ability of 14-week wPCDAI score <12.5, CRP <6 mg/L, and FC <250 μg/g, as isolated or combined variables, to predict steroid-free clinical and endoscopic remission after 1 year of biological therapy.

As a secondary endpoint we sought to evaluate the concordance between wPCDAI, CRP, and FC with SES-CD at week 48.

Statistical Analysis

Data are reported as means (±standard deviation) or medians (interquartile range). Shapiro-Wilk test was used to assess the distribution of data. Continuous variables were compared by using the Student t test for normal variables or the Wilcoxon rank sum test for not normally distributed ones, and categorical data by using Fisher exact test. Univariable Cox regression analysis was used to evaluate the ability of postinduction wPCDAI, FC, and CRP to predict 1-year clinical and endoscopic severity. Multivariable Cox regression model was used in case of significant results on the univariate analysis. The association between endoscopic disease activity and wPCDAI, FC, and CRP was assessed by the determination of Pearson correlation coefficient (r) for parametric correlation and Spearman rank correlation for nonparametric correlations. Receiver operating characteristic curves analysis were used to determine the best threshold of FC to assess endoscopic healing (defined as SES-CD <3). The accuracy of FC was evaluated using the area under the curve of the receiver operating characteristic and was defined as follows: poor 0.6 to 0.7; fair 0.8 to 0.9; good 0.8 to 0.9; excellent 0.9 to 1.0. The test characteristics are presented as sensitivity (Se), specificity (Sp), positive predictive value, negative predictive value (NPV), and likelihood ratio (LR). Cohen kappa was calculated to determine the agreement between wPCDAI activity categories (remission, mild, moderate, and severe) and SES-CD. A 2-sided P value of <0.05 was considered statistically significant for all analyses. Statistical analysis was performed using GraphPad Prism and Instat software (GraphPad 6.07, San Diego, CA). SPSS was used for Shapiro-Wilk test (IBM Corp Released 2017. IBM SPSS Statistics for Windows, Version 25.0; IBM Corp, Armonk, NY).



Forty-four patients were initially screened, 1 refused biological therapy, and 2 refused to give written informed consent. Therefore, 41 children were recruited (41.5% girls). One was lost to follow-up due to lack of the second endoscopy. All patients maintained IFX therapy during the study period and none underwent surgery.

According to wPCDAI, 58.5% of patients (n = 24) had a moderate disease, 29% (n = 12) a severe and 12% (n = 5) a mild disease activity. Mean CRP level and SES-CD were 24.1 ± 29.2 mg/L and 13.5 ± 10.2, respectively.

Mean baseline FC was 556.9 ± 318 μg/g, and 6 out of 41 patients had normal FC despite a clinically and endoscopically active disease. No significant correlation was found between baseline SES-CD and wPCDAI (r = 0.32, P = 0.12), whereas both FC and CRP resulted significantly correlated with SES-CD (r = 0.27, P = 0.04; r = 0.45, P = 0.003, respectively).

Clinical, laboratory, and endoscopic characteristics of the population are shown in Table 1.

Demographics and baseline characteristics

Four patients (10%) needed at least 1 course of steroid during the study period; 7 (17%) added an immunomodulator and 13 patients (31%) needed treatment optimization either by a shortened interval regimen or a dose intensification.

Predicting 1-Year Outcomes

Forty patients underwent endoscopy at week 48 (97.5%). Mean SES-CD was significantly reduced as compared to baseline (13.5 ± 10.2 vs 6.88 ± 7.34, P = 0.002); MH was achieved by 16 patients (39%).

Mean 48-week-wPCDAI, CRP, and FC were also significantly reduced when compared to baseline (17.13 ± 15.15 [P < 0.0001]; 5.15 ± 8.4 [P = 0.0057]; and 176.4 ± 94.62 [P < 0.0001]).

Twenty-one patients (51%) had a wPCDAI <12.5 at 14 weeks: 8 of whom (36%) achieved MH at 1 year versus 7 of 20 (35%) with a 14-week wPCDAI >12.5 (P = 1). A normalization of CRP after induction was observed in 28 patients (68%): 13 (46%) achieved MH at 1 year, compared to 3 of 13 (23%) patients with elevated postinduction CRP values (P = 0.18).

Fourteen-week FC was available in 37 of 41 children enrolled (90%). FC was <250 μg/g in 11 patients (30%): 8 of them (73%) showed MH at 1 year, compared to 11 of 26 (42%) with persistent elevated FC after induction (P = 0.2).

Similarly, postinduction CRP, wPCDAI, and FC were not able to predict 48-week steroid-free clinical remission (15/28 [53%] vs 6/13 [46%], P = 0.74; 13/21 [62%] vs 8/20 [40%], P = 0.21; 6/11 [54%] vs 15/26 [58%], P = 1, respectively). All these results are summarized in Figure 1.

Percentage of patients in clinical remission (A) and in endoscopic remission (B) at 1 year stratified by14 -week wPCDAI, FC and CRP. CRP = C-reactive protein; FC = fecal calprotectin; ePCDAI = weighted pediatric CD activity index.

The univariate analysis confirmed the above mentioned results showing no sufficient evidence of a significant relationship between 14-week wPCDAI, FC, and CRP with 1-year endoscopic remission (hazard ratio [HR] 1.40 [95% confidence interval (CI) 0.52–3.76], P = 0.72; HR 3.43 [95% CI 0.88–13.41], P = 0.08; HR 0.94 [0.51–1.71, P = 0.84]), and steroid-free clinical remission (Table 2).

Univariate analysis: 14-week weighted pediatric Crohn disease activity index, fecal calprotectin, and C-reactive protein as predictors of 1-year endoscopic remission (Simple Endoscopic Score for Crohn Disease <3), and steroid-free clinical remission (weighted pediatric CD activity index <12.5)

When associating different variables for predicting 1-year outcomes, 14-week wPCDAI <12.5 associated with FC normalization resulted significantly related to 1-year clinical (HR 4.81 [95% CI 1.76–20.45], P = 0.05) and endoscopic remission (HR 5.51 [95% CI 1.83–26.9], P = 0.03). Table 3 shows the Se, Sp, positive predictive value, NPV, and LR of 14-week single and combined variables against 1-year SES-CD.

Likelihood ratio of 14-week weighted pediatric CD activity index, fecal calprotectin, and C-reactive protein against 1-year Simple Endoscopic Score for Crohn Disease

Correlation Between Weighted Pediatric Crohn Disease Activity Index, C-Reactive Protein, Fecal Calprotectin, and Simple Endoscopic Score for Crohn Disease at 1 Year

One-year SES-CD moderately correlated with FC (Pearson rank correlation coefficient r = 0.52; P = 0.001). A low positive correlation was found between SES-CD and wPCDAI (r = 0.31; P = 0.05), whereas no sufficient evidence of a linear correlation was found for CRP. The area under the curve of the FC levels for MH was 0.863 (95% CI 0.734–0.991, P = 0.005), and the FC cut-off value was 120.5 μg/g, with 83% Se, 75.5% Sp and 3.88 LR (Supplemental Fig. 1, Supplemental Digital Content,

Supplemental Table 1 (Supplemental Digital Content, shows the correlation between endoscopic activity as measured by SES-CD and the corresponding wPCDAI, FC, and CRP. Only mean FC values were able to discriminate patient subgroups based on their SES-CD, but not to differentiate those with an endoscopic moderate (SES-CD 7–15) to severe activity (SES-CD >15) (P = 0.05 inactive to mild disease; P = 0.03 mild to moderate, and 0.07 moderate to severe disease), whereas no significant differences were found based on disease location. When evaluating the concordance between disease activity subgroups (remission, mild, moderate, and severe) measured by the wPCDAI and SES-CD, excellent and good result were found for severe disease (kappa = 0.87) and remission (kappa 0.76), whereas the concordance resulted faire and poor for mild (kappa 0.24) and moderate disease (kappa 0.07) (Supplemental Fig. 2, Supplemental Digital Content,


To the best of our knowledge, this is the first pediatric study trying to identify early clinical and laboratory markers for predicting clinical and endoscopic response in children with CD on biological therapy. In our study, the normalization of the wPCDAI along with FC after induction with IFX was able to predict steroid free clinical and endoscopic remission after 1 year of therapy. Although the limitations of patient-reported symptoms and clinical indexes in predicting disease course in CD are widely known, our findings suggest a role of the wPCDAI, when combined with an objective measure of inflammation, as FC, in predicting therapeutic response. In a recent prospective study of 184 patients with CD, investigating the role of noninvasive clinical and biological predictive markers of response to biosimilar IFX, postinduction clinical remission (CDAI <150) was associated with clinical remission both at week 30 (P < 0.001) and at week 54 (P < 0.001) (9). Other studies failed to demonstrate the utility of clinical indexes in predicting disease course. The CDAI, as well as the CRP, were unable to detect postoperative recurrence in a prospective randomized controlled trial on 135 adults patients with CD (10). In our study, postinduction FC alone was not significantly related to MH at 1 year. Data on the performance of FC in predicting disease relapse in CD are not clear. There is increasing evidence of the good performance of fecal markers in assessing disease activity; a pooled sensitivity and specificity of 78% and 73% was calculated for FC in predicting inflammatory bowel disease relapse in a meta-analysis of 6 prospective studies (11). Contrarily to CDAI and CRP, in the study of postoperative recurrence on CD (11), FC levels 6 months after surgery predicted maintenance of remission or recurrence at 18 months (NPV 71%). Molander et al (12) found, in 24 patients with CD treated with IFX that postinduction FC <100 mg/g, predicted sustained clinical remission at week 52. In our study, given the relatively small sample size, we cannot exclude a type 2 error, although our data are in keeping with other adult results (13).

No sufficient evidence of a significant correlation between postinduction CRP and 1-year outcomes emerged from our analysis. In line with our findings, the study of Gonczi et al (9), found normal CRP at week 14 not to correlate with clinical remission at 14, 30, and 54 weeks. Contrarily, postinduction CRP normalization has been reported to be predictive of improved endoscopic scores and MH at week 52 in patients on adalimumab (14). In a prospective study in 718 adults with CD treated with IFX, CRP was highly predictive for mucosal improvement after IFX therapy (15). Those conflicting results may be explained by several factors: the low specificity of pathological CRP levels, particularly in children, that can raise because of other inflammatory conditions beyond the colon, and genetic factors which, along with age, sex, and body mass index, have been demonstrated to influence CRP production (16,17).

The correlation between clinical, laboratory markers, and SES-CD after 1 year of biologic therapy was the secondary outcome of our study. We were able to prove a good correlation between wPCDAI and SES-CD for severely active disease and for disease in remission, while the correlation was fair or absent for mild to moderate endoscopic activity. In both adult and pediatric CD, the lack of correlation between symptoms and endoscopic appearance is widely known, and thus making clinical evaluation a controversial mean to judge therapeutic response (18,19). Clinical and endoscopic assessment of disease activity correlated in only 39% of cases (k = 0.13; P = 0.03) in a recent study in adult CD (20). Although our results cannot support the use of wPCDAI alone to monitor patients under IFX, at the same time it could suggest that those patients in complete remission based on their wPCDAI can probably postpone endoscopic monitoring. Previous studies in adults proved the lack of correlation between clinical disease activity and the severity of the mucosal lesions in CD (21,22). We found FC to be moderately related to MH in our population of patients. There are several data evaluating the correlation of FC with endoscopic appearance: this fecal noninvasive marker is now considered as a surrogate marker of MH in several studies mainly in UC (23,24). Data in CD are less clear. Previous adult reports indicate a correlation with endoscopic appearance from 0.32 to 0.87, based on the type of patients, setting, and so on (25). Data in children are sparse. Aomatsu et al (26), in a small study including 18 children with CD, reported a correlation coefficient of 0.76 of FC with SES-CD. Overall, a recent meta-analysis, showed a pooled sensitivity and specificity of FC for endoscopic activity in CD of 0.88 and 0.73, respectively (27). We found a cut-off value of 120.5 μg/g to have the best sensitivity and specificity and LR for MH. Moreover, we found FC to effectively discriminate disease severity as measured by SES-CD. These results, if confirmed by larger study and more significant results, could be helpful in decision-making before performing endoscopy to monitor disease activity in pediatric CD. Nevertheless, endoscopic assessment of MH still remains inevitable, also due to the emerging role of histological healing, especially in UC, as the ultimate goal, which was not evaluated in our study and will surely need further and larger investigation in CD setting.

A limitation of our study could be the absence of TDM, by measuring IFX trough levels and antidrug antibodies. Nevertheless, our objective was to test the predictive value of noninvasive and easily accessible markers that can be commonly used in routine clinical practice. Moreover, due to the relatively small sample size we may have overlooked some results, namely the role of CRP as a predictive factor of therapeutic response. A strength of this study is that our population was prospectively monitored with a timely and structured study design.

In conclusion, we have shown that in children with CD treated with IFX, wPCDAI combined with FC assessed at the end of the induction can be used as reliable long-term predictors of clinical and endoscopic therapeutic response. Conversely, the same variables used separately, along with CRP, cannot discriminate those patients who will not respond to therapy, supporting the importance of a multiparametric evaluation. FC accurately reflects mucosal status after 1 year of therapy, and correctly discriminates patients based on their endoscopic activity, while wPCDAI correctly identifies patients showing a remission or a severe disease activity, but it cannot replace an endoscopic evaluation for patients with mild-to-moderate disease activity. These results, because of the small sample size are not conclusive, but they should be considered for future larger prospective studies.


1. Peyrin-Biroulet L, Sandborn W, Sands BE, et al. Selecting therapeutic targets in inflammatory bowel disease (STRIDE): determining therapeutic goals for treat-totarget. Am J Gastroenterol 2015; 110:1324–1338.
2. Heron V, Afif W. Update on therapeutic drug monitoring in Crohn's disease. Gastroenterol Clin North Am 2017; 46:645–659.
3. Turner D, Griffiths AM, Walters TD, et al. Mathematical weighting of the pediatric Crohn's disease activity index (PCDAI) and comparison with its other short versions. Inflamm Bowel Dis 2012; 18:55–62.
4. Turner D, Levine A, Walters TD, et al. Which PCDAI version best reflects intestinal inflammation in pediatric Crohn disease? J Pediatr Gastroenterol Nutr 2017; 64:254–260.
5. 2014; Levine A, Koletzko S, Turner D, et al. ESPGHAN revised porto criteria for the diagnosis of inflammatory bowel disease in children and adolescents. 58:795–806.
6. Levine A, Griffiths A, Markowitz J, et al. Pediatric modification of the Montreal classification for inflammatory bowel disease: the Paris classification. Inflamm Bowel Dis 2011; 17:1314–1321.
7. Lin JF, Chen JM, Zuo JH, et al. Meta-analysis: fecal calprotectin for assessment of inflammatory bowel disease activity. Inflamm Bowel Dis 2014; 20:1407–1415.
8. Daperno M, D’Haens G, Van Assche G, et al. Development and validation of a new, simplified endoscopic activity score for Crohn's disease: the SES-CD. Gastrointest Endosc 2004; 60:505–512.
9. Gonczi L, Vegh Z, Golovics PA, et al. Prediction of short- and medium-term efficacy of biosimilar infliximab therapy. do trough levels and antidrug antibody levels or clinical and biochemical markers play the more important role? J Crohns Colitis 2017; 11:697–705.
10. Wright EK, Kamm MA, De Cruz P, et al. Measurement of fecal calprotectin improves monitoring and detection of recurrence of Crohn's disease after surgery. Gastroenterology 2015; 148:938.e1–947.e1.
11. Mao R, Xiao YL, Gao X, et al. Fecal calprotectin in predicting relapse of inflammatory bowel diseases: a meta-analysis of prospective studies. Inflamm Bowel Dis 2012; 18:1894–1899.
12. Molander P, af Bjorkesten C-G, Mustonen H, et al. Fecal calprotectin concentration predicts outcome in inflammatory bowel disease after induction therapy with TNFα blocking agents. Inflamm Bowel Dis 2012; 18:2011–2017.
13. Costa F, Mumolo MG, Ceccarelli L, et al. Calprotectin is a stronger predictive marker of relapse in ulcerative colitis than in Crohn's disease. Gut 2005; 54:364–368.
14. Kiss LS, Szamosi T, Molnar T, et al. Early clinical remission and normalisation of CRP are the strongest predictors of efficacy, mucosal healing and dose escalation during the first year of adalimumab therapy in Crohn's disease. Aliment Pharmacol Ther 2011; 34:911–922.
15. Jürgens M, Mahachie John JM, Cleynen I, et al. Levels of C-reactive protein are associated with response to infliximab therapy in patients with Crohn's disease. Clin Gastroenterol Hepatol 2011; 9:421.e1–427.e1.
16. Willot S, Vermeire S, Ohresser M, et al. No association between C-reactive protein gene polymorphisms and decrease of C-reactive protein serum concentration after infliximab treatment in Crohn's disease. Pharmacogenet Genomics 2006; 16:37–42.
17. Panes J, Jairath V, Levesque BG. Advances in use of endoscopy, radiology, and biomarkers to monitor inflammatory bowel diseases. Gastroenterology 2017; 152:362–373.
18. Cellier C, Sahmoud T, Froguel E, et al. Correlations between clinical activity, endoscopic severity, and biological parameters in colonic or ileocolonic Crohn's disease. A prospective multicentre study of 121 cases. The Groupe d’Etudes Thérapeutiques des Affections Inflammatoires Digestives. Gut 1994; 35:231–235.
19. Modigliani R, Mary JY, Simon JF, et al. Clinical, biological, and endoscopic picture of attacks of Crohn's disease. Evolution on prednisolone. Groupe d’Etude Therapeutique des Affections Inflammatoires Digestives. Gastroenterology 1990; 98:811–818.
20. Laterza L, Piscaglia AC, Minordi LM, et al. Multiparametric evaluation predicts different mid-term outcomes in Crohn's disease. Dig Dis 2018; 36:184–193.
21. Peyrin-Biroulet L, Reinisch W, Colombel JF, et al. Clinical disease activity, C-reactive protein normalisation and mucosal healing in Crohn's disease in the SONIC trial. Gut 2014; 63:88–95.
22. Cellier C, Sahmoud T, Froguel E, et al. Correlations between clinical activity, endoscopic severity, and biological parameters in colonic or ileocolonic Crohn's disease. A prospective multicentre study of 121 cases. Gut 1994; 35:231–235.
23. Chen JM, Liu T, Gao S, et al. Efficacy of noninvasive evaluations in monitoring inflammatory bowel disease activity: a prospective study in China. World J Gastroenterol 2017; 23:8235–8247.
24. Jusué V, Chaparro M, Gisbert JP. Accuracy of fecal calprotectin for the prediction of endoscopic activity in patients with inflammatory bowel disease. Dig Liver Dis 2018; 50:353–359.
25. Boon GJ, Day AS, Mulder CJ, et al. Are faecal markers good indicators of mucosal healing in inflammatory bowel disease? World J Gastroenterol 2015; 21:11469–11480.
26. Aomatsu T, Yoden A, Matsumoto K, et al. Fecal calprotectin is a useful marker for disease activity in pediatric patients with inflammatory bowel disease. Dig Dis Sci 2011; 56:2372–2377.
27. Mosli MH, Zou G, Garg SK, et al. C-reactive protein, fecal calprotectin, and stool lactoferrin for detection of endoscopic activity in symptomatic inflammatory bowel disease patients: a systematic review and meta-analysis. Am J Gastroenterol 2015; 110:802–819.

biologics; inflammatory bowel disease; monitoring; pediatric

Supplemental Digital Content

Copyright © 2019 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition