Efficacy and Safety of Adalimumab in Pediatric Ulcerative Colitis: A Real-life Experience from the SIGENP-IBD Registry

Abstract

Objectives: 

The aim of this study was to evaluate the effectiveness and safety of adalimumab (ADA) in children with ulcerative colitis (UC) previously treated with infliximab (IFX).

Methods: 

Retrospective study including children with UC from a national registry who received ADA therapy. The primary endpoint was the rate of corticosteroid-free remission at week 52. Secondary outcomes were the rate of sustained clinical remission, primary nonresponse, and loss of response at weeks 12, 30, and 52 and rate of mucosal healing and side effects at week 52.

Results: 

Thirty-two children received ADA (median age 10 ± 4 years). Median disease duration before ADA therapy was 27 months. All patients received previous IFX (43% intolerant, 50% nonresponders [37.5% primary, 42.5% secondary nonresponders], 6.7% positive anti-IFX antibodies). Fifty-two weeks after ADA initiation, 13 patients (41%) were in corticosteroid-free remission. Mucosal healing occurred in 9 patients (28%) at 52 weeks. The cumulative probability of a clinical relapse-free course was 69%, 59%, and 53% at 12, 30, and 52 weeks, respectively. Ten patients (31%) had a primary failure and 5 (15%) a loss of response to ADA. No significant differences in efficacy were reported between not-responders and intolerant to IFX (P = 1.0). Overall, 19 patient (59%) maintained ADA during 52-week follow-up. Seven patients (22%) experienced an adverse event, no serious side effects were observed and none resulted in ADA discontinuation.

Conclusions: 

Based on our data, ADA seems to be effective in children with UC, allowing to recover a significant percentage of patients intolerant or not-responding to IFX. The safety profile was good.

What Is Known

• Adalimumab is effective and safe for inducing and maintaining remission in adult patients with moderate to severe active ulcerative colitis.
• Data on the effectiveness of adalimumab in pediatric ulcerative colitis are lacking, although its common use in clinical practice as an off-label prescription

What Is New

• Adalimumab seems to be effective in children with ulcerative colitis who previously failed or were intolerant to infliximab therapy.
• No serious adverse events were reported, confirming a good safety profile of the drug.

Pediatric ulcerative colitis (UC) is a chronic inflammatory disorder of the large bowel, characterized by unpredictable phases of relapses and remission. Although its etiology is still unknown, several data suggest an abnormal interplay of the immune system with the commensal microbiota in a genetically susceptible host ⁽¹⁾. A key role of the tumor necrosis factor (TNF)-alpha in determining the uncontrolled inflammatory response causing UC has been suggested based on empirical data ^(2,3). Indeed, the efficacy of anti-TNF-alpha therapies has been widely demonstrated in adult-onset disease, and afterward in children with UC. In adults, the Active ulcerative Colitis Trial I (ACTI) and ACTII trials clearly showed the effectiveness of infliximab (IFX) in achieving clinical remission, response, and mucosal healing (MH) ⁽⁴⁾. Even higher remission rates have been subsequently demonstrated in the T72 clinical trial, evaluating children with moderate-to-severe UC ⁽⁵⁾.

Adalimumab (ADA), a fully human monoclonal antibody specific for human TNF-alpha, is less immunogenic than the chimeric antibody IFX and has been initially proven to be as effective as IFX in Crohn disease ^(6,7).Thereafter, the pivotal ULTRA (Ulcerative colitis Long-Term Remission and maintenance with ADA) 1 and 2 clinical trials showed the effectiveness and safety of ADA 160/80/40/40 mg compared to placebo in inducing and maintaining remission at 8 and up to 52 weeks (18.5% vs 9.2%, P = 0.031, and 17.3% vs 8.5%, P = 0.004, respectively) in patients with moderately to severely active UC despite conventional treatment ^(8–10). Based on these studies, ADA was approved worldwide for the treatment of UC. The open-label extension study, ULTRA 3, confirmed a favorable long-term safety profile of ADA therapy ⁽¹¹⁾. After those trials, other studies and real-life experiences were published in adults, confirming a good efficacy of ADA in UC ^(12–21). Recently, a study conducted in a large cohort of adults with UC treated with ADA reported a rate of clinical response, clinical, and endoscopic remission at 12 weeks of 51%, 26%, and 14%, respectively, with a higher 1-year clinical and endoscopic remission in biologics-naïve patients compared to those who had already failed an anti-TNF (65% vs 49% and 50% vs 35%, respectively) ⁽²¹⁾. Data on the effectiveness of ADA in pediatric UC are sparse and limited to few case series, although ADA is commonly used in clinical practice as an off-label prescription. Therefore, the aim of this study was to evaluate the efficacy and safety of ADA in a large cohort of pediatric patients with UC.

PATIENTS AND METHODS

Data of all children with UC treated with ADA after IFX failure/intolerance, with a minimum follow-up of 6 months, enrolled and stored in the SIGENP-IBD registry from January 1, 2009 to January 5, 2017 (the data retrieval date) were used for this study. A written and signed informed consent was obtained from all patients and their parents and the institutional review board of each hospital approved the data collection. The methodology of SIGENP IBD registry has been previously described in detail ⁽²²⁾. The study end date was identified as the date of the most recent clinic visit prior to January 5, 2017. The diagnosis of UC was made according to Porto criteria and based on clinical history, physical examination, endoscopic, histologic, and radiologic findings ⁽²³⁾. Patients with any other cause of colitis (infection, eosinophilic colitis, immunodeficiency) were excluded. Clinical data for this study included demographic features (age, sex), disease location and duration, previous therapy, and extraintestinal manifestations. All patients were previously treated with IFX and reasons for IFX discontinuation were evaluated. Infliximab failure was defined by the treating physician as a disease relapse after IFX optimization (by increased dose and/or shortening infusion intervals). The choice of treating patients with ADA was at the discretion and expertise of the treating physician. An endoscopic evaluation was available for all patients at the diagnosis for the definition of the disease location, which was reported according to the Paris classification ⁽²⁴⁾: proctitis (E1, disease limited to the rectum), left-sided colitis (E2, inflammation of the portion of the colorectum distal to the splenic flexure), extensive colitis (E3, involvement distal to the hepatic flexure), and pancolitis (E4). Disease activity at the diagnosis and during follow-up was defined by the Pediatric Ulcerative Colitis Activity Index (PUCAI) ⁽²⁵⁾. Clinical remission was defined as a PUCAI <10. Laboratory tests included full blood count, C-reactive protein (CRP), erythrocyte sedimentation rate, perinuclear anti-neutrophil cytoplasmic antibodies, nutritional (albumin), pancreatic and liver parameters, and fecal calprotectin (FC) were recorded at the diagnosis and every 6 months during follow-up. MH was evaluated at 0 and 12 months by endoscopy or FC where endoscopy was unavailable. The grade of endoscopic activity was determined according to the Mayo score ⁽²⁶⁾ as follows: 0, normal or inactive; 1, mild (erythema, absent vascular pattern, mild friability); 2, moderate (marked erythema, absent vascular pattern, friability, erosions); and 3, severe (spontaneous bleeding, ulceration). Patients were classified on the basis of the maximum Mayo score recorded in any area of the colon. MH was defined as a value of 0 in Mayo score. In all remaining Mayo scores (1, 2, and 3), MH was deemed not to have been achieved. An FC level of <250 μg/g was used as a surrogate marker of MH in patients who did not undergo an endoscopic evaluation at 52 weeks. This cut-off has already been demonstrated to have a sensitivity and a specificity of 0.80 and 0.82, with a positive and negative likelihood ratio of 4.17 and 0.22, respectively, compared to MH as defined by endoscopic scores in a systematic review and meta-analysis ⁽²⁷⁾. Corticosteroid (CS)-free remission, defined as clinical remission (PUCAI <10) free from any intravenous or oral CS treatment, and MH at 52 weeks were the primary outcomes evaluated. As secondary outcomes patients were also evaluated for episodes of acute severe colitis, need for surgery, rate of continuous clinical response and remission, primary nonresponse and loss of response and therapy-related adverse events at a 12-month follow-up.

STATISTICAL METHODS

All data were summarized and displayed as the mean ± SD for the continuous variables. Categorical data were expressed as frequencies and percentages. Comparison of groups was performed using Student t test for unpaired data in 2 group comparison and 1-way analysis of variance with Bonferroni test for multiple group comparison. Chi square test with Fisher correction was used to evaluate the differences for categorical variables wherever needed. A P value of 0.05 or less was considered significant. The Kaplan–Meier survival method was used to estimate the interval free from disease relapses and before ADA withdrawal at follow-up. Differences between curves were tested using the log-rank test. Logistic regression was performed for predictors of colectomy at 1 year. Odds ratios (OR) and their 95% confidence intervals (95% CI) were calculated. The Graphpad statistical package was used to perform all statistical evaluations (GraphPad Software, Inc, San Diego, CA, USA).

RESULTS

Demographics

Five hundred fourteen patients with UC were identified at the retrieval date, 32 (6%) of those received ADA at any time point from the diagnosis (median age at the time of ADA initiation 12.8 years, interquartile range 4–18, 53% girls). The mean disease duration before starting ADA was 2.3 ± 2.1 years. At baseline, 29 patients (91%) were receiving CS therapy at a mean dose of 20.62 ± 2.36 mg · kg⁻¹ · day⁻¹. All patients were previously exposed to IFX: 14 (43%) discontinued due to adverse events, 16 (50%) for IFX failure (6 primary nonresponse, 9 secondary nonresponse), 2 patients (7%) shifted to ADA due to high IFX antibodies titers (>10 ng/mL). Nineteen of 32 (59.3%) had a colonoscopy at the beginning of ADA therapy: 2 (10.5%) had Mayo 1, 11 (5%) Mayo 2, and 6 (31.5%) Mayo 3 score. Eleven patients had a basal FC assessment; the mean FC value at the introduction of ADA was 696.5 ± 315.81 μg/g. Two patients had no endoscopy/calprotectin at baseline. The demographic, clinical, and laboratory data at the diagnosis are listed in Table 1.

TABLE 1: Baseline characteristics of 32 children with ulcerative colitis treated with adalimumab

Outcomes

The median follow-up after ADA initiation was 16 months (interquartile range 12–22), the mean treatment duration with ADA was 23 ± 22 months. The mean ADA first dose was 117 ± 46, 69 ± 1 at the second administration and 41 ± 11 mg at the subsequent ones. Fifty-three percent (17/32), 47% (15/32), and 41% (13/32) were in steroid-free complete remission at 12, 30, and 52 weeks, respectively. Thirty-one percent of patients (10/32) showed a primary nonresponse to ADA and discontinued the drug by 12 weeks. The rate of subsequent treatment failure was 9% at week 30 and 6% at 52-week follow-up (Fig. 1). Mean PUCAI and CRP dropped from baseline to week 52 (P < 0.001). The drop was faster for PUCAI (baseline 31.3 ± 24; week 12: 15.2 ± 18.3; P < 0.05) and slower for CRP (baseline 1.8 ± 3.3 mg/dL; week 30 0.6 ± 1.2; P < 0.01) (Supplementary Fig. 1, Supplemental Digital Content 1, https://links.lww.com/MPG/B247).

FIGURE 1: Efficacy of adalimumab in 32 children with ulcerative colitis at 12, 30, and 52 weeks of treatment. LOR indicates loss of response.

Data on MH at 12 months were available in 22 of the 32 patients initially enrolled: 15 (68%) by endoscopy, 7 (32%) by FC. Among those patients evaluated, MH was reported in 8 of 22 (36%).

No significant differences in terms of efficacy were found between nonresponders and intolerant to IFX to (P = 1.0). Sixteen patients (50%) reported disease exacerbations during follow-up, 10 of which (31%) in the first 12 weeks of therapy and 6 (19%) in the subsequent follow-up. The cumulative probability of disease relapses during follow-up was 31%, 41%, and 47% at 12, 30, and 52 weeks, respectively. Overall, 19 patients (59%) maintained ADA therapy during 52-week follow-up. Figure 2 shows time free from clinical relapses and the probability of remaining under ADA.

FIGURE 2: Cumulative risk of disease relapse (A) and of adalimumab discontinuation (B) at follow-up. ADA = adalimumab.

During follow-up, 4 of 32 patients (12.5%) needed surgery. At a univariate analysis, primary nonresponse to ADA therapy resulted significantly related to the surgical risk (OR 13.8, 95% CI 1.17–161.8, p 0.039) (Supplementary Table 1, Supplemental Digital Content 2, https://links.lww.com/MPG/B248). Six patients (19%) underwent drug optimization during follow-up, re-induction in 2 (160 and 80 mg) and reduction of intervals between doses (every week) in 4 patients. In all of them, a clinical remission after optimization was observed.

Twenty-nine (91%) patients were on concomitant medical therapy during ADA treatment: 17 (53%) mesalamine, 4 (12.5%) sulfasalazine, and 8 (25%) immunomodulators (4 methotrexate, 4 azathioprine). No significant differences in 52-week clinical remission were found among patients in combination therapy with immunomodulators compared to those in monotherapy (P = 0.62).

Six patients (19%) experienced an adverse event. No malignancies, serious side effects, and death were observed and none resulted in ADA discontinuation. Table 2 shows the adverse effects during ADA treatment. No correlation between previous IFX intolerance and ADA intolerance was found in this population.

TABLE 2: Adverse events in a population of 32 children under adalimumab therapy

DISCUSSION

Pediatric UC is commonly extensive already at the diagnosis with a frequent need of immunomodulatory or biological therapies ⁽²⁸⁾. However, among biological therapies available for adult inflammatory bowel disease, only IFX has been approved in children so far, with a very limited choice for both physicians and families. Several data in adults have clearly demonstrated the effectiveness of ADA in UC, and although no head-to-head trial has been conducted, 2 network meta-analyses have suggested a similar efficacy of different biologics (IFX, ADA, vedolizumab and golimumab) in treating adults with moderate to severe UC ^(29,30), whereas 1 recent systematic review with meta-analysis reported a slight inferiority of ADA, compared to IFX (OR 0.45, 95% credible interval [CrI] 0.25–0.82) for inducing MH in UC ⁽³¹⁾, in accordance with a previous systematic review and network meta-analysis by Danese et al ⁽³²⁾. These results explain the need for data on the effectiveness and safety of ADA in pediatric UC as well. This retrospective, registry-based study describes the real-life clinical experience on the use of ADA in pediatric patients affected with UC. Based on our results, ADA seems to be effective in inducing and maintaining steroid-free clinical remission in pediatric UC previously treated with IFX.

To the best of our knowledge, there is only a case series by Volonaki et al ⁽³³⁾, describing the outcomes of ADA in pediatric UC. Eleven patients previously failing IFX were retrospectively analyzed, reporting 55% to be in clinical remission at a 6-month follow-up. These data are in line with our clinical outcomes; indeed, >50% of our patients were in complete steroid-free clinical remission at 12 weeks and 41% maintained a sustained remission at 1-year follow-up. All of our patients had been previously treated with IFX, with approximately half of them having discontinued it for intolerance and half because of treatment failure. Based on our results, ADA allowed to recover 1 out of 2 of them in the short-term, and to maintain a sustained remission in a significant proportion at a long term. No differences in terms of efficacy were found based on the reason for IFX discontinuation, as already reported in previous studies in adults ⁽¹⁷⁾. One strength of our study is that we were able to investigate the efficacy of ADA on mucosal inflammation. We found a rate of MH of 28% at 1-year follow-up, in line with previous adult data reporting rates of MH between 14% and 50%, based on the setting of the study (real-life data vs randomized clinical trials) ^{(10,13,18,21)}. Because current treatment strategies identify resolution of patient’ disease outcomes (control of symptoms) and MH as major target goals ⁽³⁴⁾, based on our data ADA seems a valid therapeutic option in children with UC. Moreover, due to its registry-based nature, our study reflects the current clinical practices in the use of ADA in a general pediatric UC population with real-life results.

Overall, 47% of children failed their treatment with ADA during follow-up; it is worth to be noted that the majority of them were primary nonresponders and needed to discontinue treatment already in the first 12 weeks of therapy, whereas only a small percentage suspended thereafter. We could hypothesize that a significant proportion of patients who have already failed IFX may have different inflammatory pathways underlying their UC, explaining the high proportion of primary failures, whereas a secondary loss of response due to antibody production is less common ⁽³⁵⁾. Indeed, Steenholdt et al ⁽³⁶⁾ have reported, in an exploratory, hypothesis-generating study, a predominantly TNF-α-independent signaling pathway with a potential relevance of different cytokines (IL-6 and sTNF-R2) in driving the inflammatory response in IFX refractory Crohn's disease patients. Moreover, several experimental data have reported that an altered regulation of intestinal T cell function can result in chronic intestinal inflammation, due to either impaired regulatory T cell activity or excessive effector T cell function. Several of these inflammatory pathways could be TNF-α independent ⁽³⁷⁾.

We found 12.5% of our population to require colectomy at 1-year follow-up. This rate is consistent with previous data on disease course of pediatric UC, considering that all patients requiring colectomy failed 2 lines of biological therapies and had a mean disease duration at ADA initiation of 2.3 ± 2.1 years ^(28,38,39). Interestingly, we found a primary nonresponse do ADA to be the only predictive marker of a poor prognosis in this cohort of patients; these data may help in early identification of those patients who may benefit to a shift to a different biological strategy.

Overall, 59% of patients maintained ADA at the latest follow-up. This figure is similar to what is reported for IFX in pediatric UC ^(12,39,40). Most patients in our cohort were under concomitant therapies, 1/3 of them under immunomodulators. Consistently with other data ⁽⁴¹⁾, we did not find any significant difference for efficacy rates based on the use of a combination therapy, although we cannot exclude a 2-type error based on the small population of patients.

As for the safety aspects, the number of adverse events captured in our patients is tolerable (22%), with no serious side effect requiring drug withdrawal. A favorable safety profile is reported in real-life experiences and clinical trials in the adult population ^(11,41,42).

The main limitation of our study is the small population, because only 32 patients of 514 with UC underwent ADA therapy. These data could be explained as ADA is still an off-label therapy in pediatric UC. For this reason we may have overlooked some results. Secondly, immunogenicity, that is, occurrence of anti-ADA antibodies, was not evaluated. However, in clinical practice it is uncommon to determine serum trough levels and antibody formation at every visit, mainly due to financial reasons. Moreover, because this study examined real-life management, the use of concomitant treatment (included CS and immunomodulators) was not controlled and may have impacted on different patient’ outcomes. Lastly, MH was not assessed in all patients and in 1/3 of them it was evaluated by FC as a surrogate marker, and thus making the results not absolutely homogenous.

In conclusion, this study reflects the experience in daily clinical practice with ADA in a cohort of children with UC previously treated with IFX therapy. Overall, our data suggest that ADA could be effective in inducing clinical and endoscopic remission in pediatric UC, allowing to recover half of the patients who had failed or were intolerant to IFX. Interestingly, no difference in terms of efficacy based on the reason underlying IFX discontinuation was found, suggesting the use of ADA both in IFX intolerance and failure. Because the therapeutic armamentarium of pediatric UC is limited, and most new drugs already approved in adults ⁽⁴³⁾ are not available for the routinely use in children, our data could support a safe and effective use of ADA in children with UC. Nevertheless, larger and prospective data are needed for assessing the definite role of ADA in the management of pediatric UC.