242 Original article
Therapeutic drug monitoring of infliximab in inflammatory bowel disease patients in a teaching hospital setting: results of a prospective cohort study Andrea Warmana, Jan Willem A. Straathofb and Luc J.J. Derijksa Objective Therapeutic drug monitoring (TDM) of infliximab (IFX) is not routinely implemented in our clinical practice. We therefore carried out a prospective cohort study measuring IFX trough levels in our total inflammatory bowel disease (IBD) population in relation to remission. Methods Patient demographics, and medication and clinical history were collected from the electronic hospital information system. Blood was drawn at one time point for the determination of IFX trough levels and antibodies to IFX (ATI). Disease activity indices [Crohn’s disease activity index (CDAI) and the Truelove–Witts disease activity index (TWDAI) for Crohn’s disease and ulcerative colitis, respectively] and quality-of-life scores (Visual Analog Scale) were obtained. Results We included 107 patients. IFX levels varied from less than 0.02 to 21.9 μg/ml. The median IFX level was 2.8 μg/ml [interquartile range (IQR) 1.37–5.13]. The IFX level was associated significantly with remission (P = 0.007). The median IFX level was 3.9 μg/ml (IQR 1.9–6.53) in patients in remission and 2.1 μg/ml in patients with active disease (IQR 0.77–4.38) (P = 0.074). Receiver operating charecteristic curve analysis indicated a cutoff value of 2.18 μg/ml for CD and 6.26 μg/ml for UC. Eleven patients (10.3%) had developed ATI. The appearance of ATI was associated with
Introduction Ulcerative colitis (UC) and Crohn’s disease (CD), together known as inflammatory bowel disease (IBD), are both characterized by a diffuse inflammation of the bowel. Tumor necrosis factor-α (TNF-α) plays an important role in the pathogenesis of this inflammation. Infliximab (IFX) is the first chimeric mouse–human immunoglobulin G1 monoclonal antibody against TNFα and is approved for the treatment of UC and CD. It is effective in ∼ 60% of the IBD patients. Although a majority of the patients show clinical response, there are also nonresponders and in some initial responders, remission is not maintained. Detectable IFX trough levels are associated with remission, lower C-reactive protein, and endoscopic healing. Furthermore, detectable antibodies to IFX (ATI) are associated with relapse and a higher risk of infusion reactions. Therefore, the presence of IFX and the absence of ATI may play a key role in achieving remission [1–7]. 0954-691X Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
the disappearance of IFX [relative risk: 2.2 (95% confidence interval: 1.368–3.610) P < 0.0001], but not with relapse. The presence of ATI induced more infusion reactions [relative risk: 11.7 (95% confidence interval: 2.74–49.60) P < 0.001]. Conclusion TDM of IFX in IBD outpatients in a teaching hospital setting showed large interindividual differences in IFX trough levels. Despite this, we still found a significant association between remission and IFX trough levels. We determined cutoff values for both IBD modalities. IFX trough levels were not detectable in a significant proportion of IBD patients; TDM is indicated to identify this group of patients. Eur J Gastroenterol Hepatol 27:242–248 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. European Journal of Gastroenterology & Hepatology 2015, 27:242–248 Keywords: antibodies to infliximab, infliximab, therapeutic drug monitoring, trough level Departments of aClinical Pharmacy and bGastroenterology & Hepatology, Máxima Medical Center, Veldhoven, The Netherlands Correspondence to Andrea Warman, PharmD, Department of Clinical Pharmacy, Máxima Medical Center, PO Box 7777, 5500 MB Veldhoven, The Netherlands Tel: + 31 40 8889020; fax: + 31 40 8889030; e-mail: [email protected] Received 5 October 2014 Accepted 8 December 2014
Nevertheless, therapeutic drug monitoring (TDM) of IFX is not yet common practice. In clinical practice, the strategy in nonresponders often involves increasing the dose or reducing the interval before measuring IFX trough levels or ATI levels. To date, studies were mostly designed to measure IFX and ATI immediately after the start of IFX maintenance treatment. The aim of the present study was to provide more insight into IFX trough levels and ATI in an IBD population in which dose adjustments had already been performed on the basis of doctors’ view for the greater part. The primary goal of this study was to investigate whether there is still a relationship between IFX trough levels and remission in our IBD cohort.
Methods Patients
This single-center, prospective, observational study was carried out in Máxima Medical Center (MMC) in DOI: 10.1097/MEG.0000000000000279
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
TDM of infliximab in IBD patients Warman et al. 243
Veldhoven, the Netherlands, from October 2012 to May 2013. The protocol was cleared from approval by the local Medical Ethics Review Committee and the study was carried out in accordance with the Helsinki Declaration and Good Clinical Practice. CD and UC patients 18 years and older on maintenance IFX treatment were included, irrespective of the duration of IFX use. Patients were initially started on IFX 5 mg/kg every 8 weeks. In time, dose and interval were adjusted depending on the efficacy and side effects. Patients were allowed to use concomitant immunosuppressive therapy. Patient characteristics, comedication, and IFX dose were collected from the electronic hospital information system (EHIS-Chipsoft). Study design
Patients were admitted at day care to receive their IFX infusion. Blood for the determination of laboratory parameters was drawn before infusion. Blood samples were collected and stored at − 20°C. IFX trough level and the free fraction of ATI were determined by Sanquin (Amsterdam, the Netherlands) using a home-made enzyme-linked immunosorbent assay. The limit of detection was 0.01 μg/ml. The lowest level that could be quantified was 0.03 μg/ml and the highest level of quantification was 250 μg/ml. The precision as a percentage coefficient of variance was less than or equal to 18%. The lower level of quantification for ATI was 12 AE/ml. C-reactive protein, hemoglobin, erythrocyte sedimentation rate, alanine aminotransferase, alkaline phosphatase, creatinine, and estimated glomerular filtration rate were determined by the in-house laboratory of MMC. A combined questionnaire was developed to calculate the Crohn’s Disease Activity Index (CDAI) and the Truelove–Witts Disease Activity Index (TWDAI) [8,9]. Remission was achieved if the CDAI score was below 150 points and the TWDAI was below 6 points. The questionnaire was filled in by the patient during IFX infusion. Physical well-being such as the presence and severity of abdominal pain, stool, and complications (fistula, fissures, arthralgia/itis, and erythema nodosum) was noted. Patients were asked to score quality-of-life as a Visual Analog Scale (VAS) score. Statistical analysis
Statistical analysis was carried out using SPSS, version 19.0 (IBM, Armonk, New York, USA). Descriptive data were analyzed using frequency distributions for categorical data and calculation of median and range or interquartile range (IQR) for continuous variables. Univariate analysis with an unpaired t-test or the Mann–Whitney test was used for continuous variables. Categorical data were analyzed in a 2 × 2 contingency table. The odds ratio was calculated and significance was determined using Pearson’s χ2 or Fisher’s exact test. Correlations
were assessed using Spearman’s rank correlation test. A receiver operating characteristic (ROC) curve was performed, using remission as a discrete variable, to calculate sensitivity (Se), specificity (Sp), and area under the ROC curve and to determine predictive cutoff values with the associated P-value. P-values were calculated two tailed and P-value less than 0.05 was considered statistically significant.
Results Patient characteristics
A total of 113 patients with CD or UC were treated with IFX during the inclusion period. In two patients, a blood sample could not be drawn, one patient refused to provide a blood sample, and in three patients, insufficient blood samples were taken. Thus, 107 patients were included. Patient characteristics are shown in Table 1. IFX trough level and ATI
Figure 1 presents the variation in the IFX trough levels, each bar representing a patient. The level varied from undetectable (< 0.02 μg/ml) to 21.90 μg/ml. The median trough level in the total IBD population was 2.8 μg/ml (IQR 1.37–5.13) and did not differ between CD and UC (respectively, 2.80 and 2.83 μg/ml, P = 0.933). The IFX trough level was lower than 1 μg/ml in 18.7% of the patients. Eleven patients developed ATI (10.3%), shown as dark gray bars in Fig. 1; six of these patients were diagnosed with CD and five patients were diagnosed with UC. Detectable ATI were associated with the absence of IFX trough level [relative risk (RR): 2.2 (95% confidence interval (CI): 1.368–3.610) P < 0.0001]. Eleven patients developed infusion reactions during the current infusion or in the past. Reactions as elevated temperature, transpiration, itch, airless and headache were mentioned. ATI were associated with the risk of infusion reactions Table 1
Patient characteristics
Sex [n (%)] Male Female Age (years) Median (IQR) Indication [n (%)] Crohn’s disease Ulcerative colitis Dose (mg) Median (IQR) Interval (weeks) Median (IQR) Duration infliximab therapy (months) Median (IQR) Comedication [n (%)] Immunosuppressant (mainly azathioprine and mercaptopurine) Corticosteroids 5-Aminosalicylate IQR, interquartile range.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
49 (45.8) 58 (54.2) 40.5 (30.81–53.38) 61 (57) 46 (43) 400 (300–400) 8 (7–8) 40.6 (22.4–63.97) 51 (47.7) 11 (10.3) 11 (10.3)
244 European Journal of Gastroenterology & Hepatology 2015, Vol 27 No 3
Fig. 1
100
Infliximab trough level (μg/ml)
Infliximab trough level, without antibodies against infliximab (89.7%) Infliximab trough level, with antibodies against infliximab (10.3%)
10
1
0.1
0.01 Infliximab trough levels ranging from high to low, and the appearance of antibodies against infliximab. Each bar represents an infliximab trough level of a patient.
[RR: 11.7 (95% CI: 2.744–49.599) P < 0.0001] and the need for premedication [RR: 6.0 (95% CI: 1.618–22.22) P = 0.003]. Remission
Overall, 53.9% of the IBD patients were in remission. Differentiating between CD and UC, respectively, 75.4 and 26.7% were in remission. The median CDAI and TDWAI scores were, respectively, 89 (IQR 39.50–146.00) and 8 (IQR 6.50–9.00). IFX trough level was associated significantly with remission in the IBD population (P = 0.007). When separated in terms of the two IBD modalities, this association remained significant for CD (P = 0.010), but not for UC (P = 0.051). Undetectable ATI were not associated with remission (P = 0.182). IBD patients in remission showed a median IFX trough level of 3.9 μg/ml (IQR 1.90–6.53) and patients with active disease showed a median of 2.12 μg/ml (IQR 0.77–4.38) (P = 0.074). On differentiation, the median IFX trough level for CD in remission was 3.35 μg/ml (IQR 1.83–6.00) and the median IFX trough level for CD not in remission was 1.52 µg/ml (IQR 0.44–2.47) (P = 0.017). For UC, the difference between the medians was not significant: 5.44 µg/ml (IQR 2.22–7.89) versus 2.66 µg/ml (IQR 1.18–4.59) (P = 0.445). Logistic regression analysis showed a 20% higher chance of remission with every increase of 1 μg/ml IFX trough level [RR: 1.207 (95% CI: 1.044–1.396) P = 0.011]. Figures 2 and 3 show the ROC curves with the corresponding Se and Sp for UC and CD. ROC analysis was carried out to determine the cutoff value for both
indications. CD showed a cutoff value of 2.18 µg/ml (Se 67.4% and Sp 78.6%). The cutoff level of UC was determined to be 6.26 µg/ml (Se 50% and Sp 87.9%). The association between different cutoff values and remission was calculated with the corresponding P-values and odds ratios, shown in Tables 2 and 3. Influence of patient characteristics on remission, IFX trough level, and ATI
Older age was associated with active disease (P = 0.001). Patients with CD were more likely to be in remission than UC [RR: 0.120 (95% CI: 0.050–0.290) P < 0.001]. Remission was independent of sex, dose regime, or duration of IFX use. Concomitant use of immunosuppressants, corticosteroids, and 5-aminosalicylates was not associated with remission. Of all laboratory parameters, only erythrocyte sedimentation rate was associated significantly with remission (P = 0.044). IFX trough levels were not significantly different between the sexes or indications (respectively, P = 0.308 and 0.960), neither was there an association between trough level and age. The IFX level was not influenced by the use of immunosuppressants, corticosteroids, and 5-aminosalicylates. The IFX trough level was associated with the interval of administration, but not with the dose [correlation coefficient, respectively, − 0.427 (P < 0.0001) and 0.189 (P = 0.051)]. Despite the association between IFX and remission, there was no relation between the IFX trough level and physical discomfort or the VAS score, nor was there a relation between remission and the VAS score. ATI were not influenced by the concomitant use of immunosuppressants or any of the patient characteristics.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
TDM of infliximab in IBD patients Warman et al. 245
Fig. 3
1.0
1.0
0.8
0.8 Sensitivity
Sensitivity
Fig. 2
0.6
0.4
0.6 0.4 0.2
0.2 0.0 0.0
0.0 0.0
0.2
0.4
0.6
0.8
1.0
IFX level (μg/ml)
IFX level (μg/ml)
0.5 1.0 1.5 2.0 3.0 4.0 5.0 6.0 7.0
Sp (%)
P-value
100 100 88 83 58 58 50 50 25
18 21 30 39 55 68 80 86 97
0.171 0.164 0.466 0.287 0.445 0.086 0.060 0.016 0.069
RR (95% CI) 0.818 0.788 2.174 2.857 1.680 3.220 3.714 5.600 5.000
(0.647–0.961) (0.660–0.940) (0.401–11.781) (0.535–15.264) (0.441–6.394) (0.821–12.632) (0.910–15.154) (1.277–24.564) (0.925–27.041)
Table corresponding to Fig. 2. Area under curve: 0.692 [P = 0.051 (95% CI: 0.517–0.867)] ulcerative colitis. CI, confidence interval; IFX, infliximab; RR, relative risk; Se, sensitivity; Sp, specificity.
Multivariate analyses showed no confounding factors in terms of patient characteristics on the association between IFX trough level and remission.
Discussion TDM of IFX is not common care in the daily practice of a gastroenterologist treating IBD. Lack of efficiency or the appearance of side effects may often be encountered by dose or interval adjustments before turning to TDM. Therefore, these trough levels are not based on a standard regime, as we see in patients newly started on IFX, but might be influenced by the adjustments. This study provided insights into IFX trough levels in our IBD cohort in which dose adjustments had already been performed and whether there is still an association with remission. IFX trough levels showed large interindividual differences and a high proportion of patients had IFX trough levels below 1 μg/ml (18.7%). We found a significant association between remission and IFX trough level
0.8
1.0
Association between IFX trough level and remission with the corresponding sensitivity and specificity for Crohn’s disease
Table 3
Se (%)
0.6
Receiver operating characteristic (ROC) curve – cutoff value Crohn’s disease.
Association between IFX trough level and remission with the corresponding sensitivity and specificity for ulcerative colitis
Table 2
0.4
1 – Specificity
1 − Specificity Receiver operating characteristic (ROC) curve – cutoff value ulcerative colitis.
0.2
0.5 1.0 1.5 2.0 3.0 4.0 5.0 6.0 7.0
Se (%)
Sp (%)
P-value
88 86 84 70 54 47 33 25 17
25 43 50 64 79 86 93 93 93
0.202 0.021 0.011 0.023 0.062 0.056 0.084 0.258 0.664
RR (95% CI) 3.040 4.625 5.143 4.154 4.217 5.217 6.276 4.469 2.528
(0.687–13.461) (1.181–18.119) (1.368–19.330) (1.164–14.825) (1.029–17.276) (1.040–26.165) (0.745–52.896) (0.523–38.219) (0.283–22.567)
Table corresponding to Fig. 3. Area under curve: 0.731 [P = 0.010 (95% CI: 0.579–0.882)] Crohn’s disease. CI, confidence interval; IFX, infliximab; RR, relative risk; Se, sensitivity; Sp, specificity.
(P = 0.007) in the total IBD population. On comparing the medians, IBD patients in remission showed a higher IFX trough level (3.9 vs. 2.1 μg/ml, not significant). On differentiating the two IBD modalities, the difference in the median IFX trough level between remission and not remission became significant for CD. We carried out ROC analyses to determine the cutoff value for CD and UC. We found a cutoff value of 2.18 μg/ ml for CD and 6.26 μg/ml for UC. Sp and Se for the cutoff value of CD showed a moderate Se and good Sp. The Se for the cutoff value of UC was poor: 50%. This means that there was a relatively high probability of finding false-negative UC patients (high IFX trough level, but not in remission). The literature describes several cutoff values. Steenholdt et al. [10] found good Se and Sp (Se 86%, Sp 85%, AUCROC 0.93, not significant) with a therapeutic cutoff value of 0.5 μg/ml in CD. Paul and colleagues and Vande Casteele and colleagues found cutoff values of, respectively, 2 μg/ml (Se 76%, Sp 82.3%, AUC-ROC 0.68,
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
246 European Journal of Gastroenterology & Hepatology 2015, Vol 27 No 3
P = 0.017) and 3 μg/ml in IBD patients. Furthermore, Vande Casteele and colleagues also reported a therapeutic upper limit of 7 μg/ml. Dose reduction above 7 μg/ml resulted in lower drug exposure and lower IFX cost while maintaining disease control [11,12]. We did not find more discomfort or adverse events when the IFX trough level was above 7 µg/ml, nor did we find a significantly higher number of patients in remission. The presence of ATI may lead to an absence in detectable IFX trough level by increasing drug clearance. Indeed, we did find a strong association between the disappearance of IFX and the presence of ATI. This finding, however, did not result in a significant association between ATI and remission. A meta-analysis by Lee et al. [13] reported the same conclusion. However, a meta-analysis by Nanda et al. [14] did find a significantly higher risk on loss of response because of detectable ATI [RR: 3.2 (95% CI: 2.0–4.9, P < 0.0001)]. Infusion reactions were also associated with detectable ATI. This was confirmed by the meta-analysis of O’Meara and colleagues. They reported that IBD patients treated with IFX who developed ATI had a two-fold risk of acute infusion reactions and a six-fold higher risk of serious acute infusion reactions [15]. The presence of ATI may be influenced by the immune status of the patient, dosing regime, and the use of concomitant immunosuppressant therapy [5]. Episodic IFX treatment is associated with a significantly higher risk of developing ATI compared with maintenance therapy [7,13,16]. At present, it is advised to administer IFX as maintenance therapy, as we did in our study. The concomitant use of immunosuppressants was not associated with less formation of ATI in our study, nor in the study of Maser et al. [7], although there was a tendency toward significance (10 vs. 26% P = 0.11) However, this seems to be power related. Lee et al. [13] pooled 11 studies and found a 50% risk reduction of developing ATI when IFX was combined with immunosuppressants (P < 0.00001). O’Meara et al. [15] reported a modest risk reduction of ATI development by pooling four studies [RR 0.6 (95% CI: 0.4–0.9, P = 0.02)]. A recent study of Vande Casteele and colleagues showed that ATI formation may be transient and therefore may not always lead to worse clinical outcome. The addition of an immunosuppressant or intensification of the IFX therapy might lead to the disappearance of ATI. Whether this strategy can be applied in the case of detectable ATI should be investigated further in the near future [17,18]. Differences between our findings and those of other studies in the cutoff levels, Se/Sp, and the influence of ATI may be partly explained by the differences in study designs. We included IBD outpatients admitted at day care for their regular IFX infusion. This was a relatively healthy group of IBD patients, and none of them had experienced an acute exacerbation of their disease.
Besides, the majority of studies included IBD patients who had just entered the IFX maintenance phase, which resulted in a fairly homogenous group of patients. We measured IFX trough levels and ATI in a relatively heterogeneous group of IBD patients in whom dose adjustments had already been performed. Time on IFX varied between 3 and 161 months. This might be considered as a weakness because it is unclear as to how reliable these results may be for patients just started on IFX. However, this was not the main aim of this study; we aimed to determine whether there was still an association between remission and IFX trough level in our IBD cohort, resembling daily practice. TDM is not yet a common practice, and is often warranted after dose adjustments have already been made. We conclude that there is still an association and we could determine the cutoff value for IFX trough levels for both CD and UC. We did find a huge difference in the cutoff values between UC and CD. Our analysis showed that patients with CD were more likely to be in remission than patients with UC. This might be explained by differences in disease pathology and therefore response to IFX. Perhaps, a higher cutoff value is required for UC. Arias et al. [19] found a cutoff IFX value of 7.19 µg/ml for a sustained benefit in newly diagnosed UC patients. Data on IFX trough levels in relation to response and remission are still scarce in UC. The method used to determine remission may also play a significant role in the different cutoff values that we found. Disease activity indices were scored with CDAI and TWDAI [8,9]. CDAI is a wellknown disease activity index for CD and has been used in many studies. We found a cutoff value for CD that was comparable with the literature. The method used to score disease activity for UC is more diverse and differs between studies. We found few UC patients in remission using a TWDAI score; this might have been because of the sensitivity of this index. ROC analysis did show a cutoff value (sum of Se and 1 − specificity), but graphically, there was no clear cutoff at all. Other indices, such as the Mayo Endoscopic Scoring of Ulcerative Colitis (MAYO) or the Simple Clinical Colitis Activity Index (SCCAI), might be more sensitive and therefore may define remission more clearly. More research is needed to establish a sensitive method for the measurement of disease activity for UC and to determine a cutoff value. Although the importance of an adequate IFX trough level has been reported in many studies, as in ours, there is no consensus as yet how to act to a certain IFX trough level. Several treatment algorithms exist, but to our knowledge, these have not been validated as yet in daily practice [11,12,17]. Our data point to an IFX trough level of at least 2 μg/ml in CD. But how should we react on IFX trough levels below 2 μg/ml? The first question that should be answered is whether the patient is in remission or is still experiencing complaints. We might consider discontinuation of IFX therapy in CD patients in
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
TDM of infliximab in IBD patients Warman et al. 247
remission with IFX trough levels below 2 μg/ml. This may be complicated by unwillingness to stop, the chance of relapse, or the risk of ATI formation. If not in remission, the trough level can be optimized by intensifying the dose or reducing the interval, or even by switching to another monoclonal antibody against TNF-α. Both Afif and colleagues and Pariente and colleagues showed that dose intensification (not differentiating between interval shortening and dose escalation) or switching to another anti-TNF-α agent in patients with subtherapeutic IFX levels resulted in complete or at least a partial response. They differed in the use of the methods in terms of detectable ATI and therapeutic failure. Afif et al. [20] reported a greater response when switching to another TNF agent compared with dose intensification, whereas Pariente et al. [21] reported a good clinical response after dose intensification. Hibi et al. [22] achieved therapeutic IFX trough levels by shortening the dosing interval to 4 weeks, resulting in improved clinical response or even remission. Paul and colleagues increased the IFX dose from 5 to 10 mg/kg in IBD patients who experienced loss of response. A delta IFX more than 0.5 μg/ml was associated with mucosal healing (Se 0.88, Sp 0.77, P = 0.0001, AUC-ROC 0.89) [11]. To our knowledge, there are no studies that compare these two strategies or the difference in cost effectiveness. In daily practice, these are used separately on the basis of the preference of the individual specialist or even applied in combination to optimize the IFX trough level.
A. Warman (PharmD) contribution: researcher, writing protocol, training nurses, collecting data, statistical analysis, writing paper; J.W.A. Straathof (MD, PhD) contribution: providing information, commentator; L.J.J. Derijks (PharmD, PhD) contribution: commentator, guidance during research. Conflicts of interest
There are no conflicts of interest.
References 1
2
3
4
5
6
7
8
Conclusion
We found large interindividual differences in IFX trough levels in an IBD population in whom dose adjustments had already been performed. IFX trough levels were below 1 μg/ml in a significant proportion of IBD patients. Moreover, ∼ 10% of the patients developed ATI. TDM is indicated to identify this group of patients, but is not a common practice as yet. Remission was associated with the IFX trough level. However, this was clearer for CD than for UC. We found an IFX cutoff value of 2.18 μg/ml for CD and 6.26 μg/ml for UC. We question, however, the applicability of the cutoff value for UC. To date, there is no consensus as yet on therapeutic lower and upper IFX limits, or the strategy to be applied when therapy fails. Adequately powered future studies or combined studies should establish the therapeutic window of IFX and determine which strategy is the best in case of therapeutic failure.
9 10
11
12
13
14
15
16
Acknowledgements The authors thank Jeanne Dieleman, epidemiologist at the Máxima Medical Center, for statistical advice. They also thank the scientific commission of the Máxima Medical Center for providing the funds to determine IFX trough levels and antibodies against IFX at Sanquin, Amsterdam.
17
18
Hanauer SB, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, et al. Maintenance infliximab for Crohn’s disease: the ACCENT I randomised trial. The Lancet 2002; 359:1541–1549. Sands BE, Anderson FH, Bernstein CN, Chey WY, Feagan BG, Fedorak RN, et al. Infliximab maintenance therapy for fistulizing Crohn’s disease. N Engl J Med 2004; 350:876–885. Rutgeerts P, Sandborn WJ, Feagan BG, Reinisch W, Olson A, Johanns J, et al. Infliximab for induction and maintenance therapy for ulcerative colitis. N Engl J Med 2005; 353:2462–2476. Reinisch W, Sandborn WJ, Rutgeerts P, Feagan BG, Rachmilewitz D, Hanauer SB, et al. Long-term infliximab maintenance therapy for ulcerative colitis: the ACT-1 and -2 extension studies. Inflamm Bowel Dis 2012; 18:201–211. Colombel JF, Feagan BG, Sandborn WJ, van Assche G, Robinson AM. Therapeutic drug monitoring of biologics for inflammatory bowel disease. Inflamm Bowel Dis 2012; 18:349–358. Seow CH, Newman A, Irwin SP, Steinhart AH, Silverberg MS, Greenberg GR. Trough serum infliximab: a predictive factor of clinical outcome for infliximab treatment in acute ulcerative colitis. Gut 2010; 59:49–54. Maser EA, Villela R, Silverberg MS, Greenberg GR. Association of trough serum infliximab to clinical outcome after scheduled maintenance treatment for Crohn’s disease. Clin Gastroenterol Hepatol 2006; 4:1248–1254. Yoshida EM. The Crohn’s disease activity index, its derivatives and the inflammatory bowel disease questionnaire: a review of instruments to assess Crohn’s disease. Can J Gastroenterol 1999; 13:65–73. Truelove SC, Witts LJ. Cortisone in ulcerative colitis. BMJ 1955; 29:1041–1048. Steenholdt C, Bendtzen K, Brynskov J, Thomsen O, Ainsworth MA. Cut-off levels and diagnostic accuracy of infliximab trough levels and anti-infliximab antibodies in Crohn’s disease. Scand J Gastroenterol 2011; 46:310–318. Paul S, del Tedesco E, Marotte H, Rinaudo-Gaujous M, Moreau A, Phelip JM, et al. Therapeutic drug monitoring of infliximab and mucosal healing in inflammatory bowel disease: a prospective study. Inflamm Bowel Dis 2013; 19:2568–2576. Vande Casteele N, Gils A, Ballet V, Compernolle G, Peeters M, van Steen K, et al. Randomised controlled trial of drug level versus clinically based dosing of infliximab maintenance therapy in IBD: final results of the TAXIT study. U Eur Gastroenterol J 2013; 1 (Suppl 1):A1. Lee LY, Sanderson JD, Irving PM. Anti-infliximab antibodies in inflammatory bowel disease: prevalence, infusion reactions, immunosuppression and response, a meta-analysis. Eur J Gastroenterol Hepatol 2012; 24:1078–1085. Nanda K, Cheifetz A, Moss AC. Impact of antibodies to infliximab on clinical outcomes and serum infliximab levels in patients with inflammatory bowel disease (IBD): a meta-analysis. Am J Gastroenterol 2013; 108:40–47. O’Meara S, Nanda KS, Moss AC. Antibodies to infliximab and risk of infusion reactions in patients with inflammatory bowel disease: a systematic review and meta-analysis. Inflamm Bowel Dis 2014; 20:1–6. Hanauer SB, Wagner CL, Bala M, Mayer L, Travers S, Diamond RH, et al. Incidence and importance of antibody responses to infliximab after maintenance or episodic treatment in Crohn’s disease. Clin Gastroenterol Hepatol 2004; 2:542–553. Vande Casteele N, Gils A, Singh S, Ohrmund L, Hauenstein S, Rutgeerts P, et al. Antibody response to infliximab and its impact on pharmacokinetics can be transiënt. Am J Gastroenterol 2013; 108:962–971. Ben-Horin S, Waterman M, Kopylov U, Yavzori M, Picard O, Fudim E, et al. Addition of an immunomodulator to infliximab therapy eliminates antidrug antibodies in serum and restores clinical response of patients with inflammatory bowel disease. Clin Gastroenterol Hepatol 2013; 11:444–447.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
248 European Journal of Gastroenterology & Hepatology 2015, Vol 27 No 3
19
20
Arias MT, Vande Casteele N, Drobne D, Ferrante M, Cleynen I, Ballet V, et al. OP10. Importance of trough levels and antibodies on the long-term efficacy of infliximab therapy in ulcerative colitis. J Crohn’s Colitis 2012; 6 (Suppl 1): S5. Afif W, Loftus EV Jr, Faubion WA, Kane SV, Bruining DH, Hanson KA, Sandborn WJ. Clinical utility of measuring infliximab and human anti-chimeric antibody concentrations in patients with inflammatory bowel disease. Am J Gastroenterol 2010; 105:1133–1139.
21
22
Pariente B, Pineton de Chambrun G, Krysiek R, Desroches M, Louis G, Cassan De C, et al. Trough levels and antibodies to infliximab may not predict response to intensification of infliximab therapy in patients with inflammatory bowel disease. Inflamm Bowel Dis 2012; 18:1199–1206. Hibi T, Sakuraba A, Watanabe M, Motoya S, Ito H, Motegi K, et al. Retrieval of serum infliximab level by shortening the maintenance infusion interval is correlated with clinical efficacy in Crohn’s disease. Inflamm Bowel Dis 2012; 18:1480–1487.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
Therapeutic drug monitoring of infliximab in inflammatory bowel disease patients in a teaching hospital setting: results of a prospective cohort study Andrea Warmana, Jan Willem A. Straathofb and Luc J.J. Derijksa Objective Therapeutic drug monitoring (TDM) of infliximab (IFX) is not routinely implemented in our clinical practice. We therefore carried out a prospective cohort study measuring IFX trough levels in our total inflammatory bowel disease (IBD) population in relation to remission. Methods Patient demographics, and medication and clinical history were collected from the electronic hospital information system. Blood was drawn at one time point for the determination of IFX trough levels and antibodies to IFX (ATI). Disease activity indices [Crohn’s disease activity index (CDAI) and the Truelove–Witts disease activity index (TWDAI) for Crohn’s disease and ulcerative colitis, respectively] and quality-of-life scores (Visual Analog Scale) were obtained. Results We included 107 patients. IFX levels varied from less than 0.02 to 21.9 μg/ml. The median IFX level was 2.8 μg/ml [interquartile range (IQR) 1.37–5.13]. The IFX level was associated significantly with remission (P = 0.007). The median IFX level was 3.9 μg/ml (IQR 1.9–6.53) in patients in remission and 2.1 μg/ml in patients with active disease (IQR 0.77–4.38) (P = 0.074). Receiver operating charecteristic curve analysis indicated a cutoff value of 2.18 μg/ml for CD and 6.26 μg/ml for UC. Eleven patients (10.3%) had developed ATI. The appearance of ATI was associated with
Introduction Ulcerative colitis (UC) and Crohn’s disease (CD), together known as inflammatory bowel disease (IBD), are both characterized by a diffuse inflammation of the bowel. Tumor necrosis factor-α (TNF-α) plays an important role in the pathogenesis of this inflammation. Infliximab (IFX) is the first chimeric mouse–human immunoglobulin G1 monoclonal antibody against TNFα and is approved for the treatment of UC and CD. It is effective in ∼ 60% of the IBD patients. Although a majority of the patients show clinical response, there are also nonresponders and in some initial responders, remission is not maintained. Detectable IFX trough levels are associated with remission, lower C-reactive protein, and endoscopic healing. Furthermore, detectable antibodies to IFX (ATI) are associated with relapse and a higher risk of infusion reactions. Therefore, the presence of IFX and the absence of ATI may play a key role in achieving remission [1–7]. 0954-691X Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
the disappearance of IFX [relative risk: 2.2 (95% confidence interval: 1.368–3.610) P < 0.0001], but not with relapse. The presence of ATI induced more infusion reactions [relative risk: 11.7 (95% confidence interval: 2.74–49.60) P < 0.001]. Conclusion TDM of IFX in IBD outpatients in a teaching hospital setting showed large interindividual differences in IFX trough levels. Despite this, we still found a significant association between remission and IFX trough levels. We determined cutoff values for both IBD modalities. IFX trough levels were not detectable in a significant proportion of IBD patients; TDM is indicated to identify this group of patients. Eur J Gastroenterol Hepatol 27:242–248 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. European Journal of Gastroenterology & Hepatology 2015, 27:242–248 Keywords: antibodies to infliximab, infliximab, therapeutic drug monitoring, trough level Departments of aClinical Pharmacy and bGastroenterology & Hepatology, Máxima Medical Center, Veldhoven, The Netherlands Correspondence to Andrea Warman, PharmD, Department of Clinical Pharmacy, Máxima Medical Center, PO Box 7777, 5500 MB Veldhoven, The Netherlands Tel: + 31 40 8889020; fax: + 31 40 8889030; e-mail: [email protected] Received 5 October 2014 Accepted 8 December 2014
Nevertheless, therapeutic drug monitoring (TDM) of IFX is not yet common practice. In clinical practice, the strategy in nonresponders often involves increasing the dose or reducing the interval before measuring IFX trough levels or ATI levels. To date, studies were mostly designed to measure IFX and ATI immediately after the start of IFX maintenance treatment. The aim of the present study was to provide more insight into IFX trough levels and ATI in an IBD population in which dose adjustments had already been performed on the basis of doctors’ view for the greater part. The primary goal of this study was to investigate whether there is still a relationship between IFX trough levels and remission in our IBD cohort.
Methods Patients
This single-center, prospective, observational study was carried out in Máxima Medical Center (MMC) in DOI: 10.1097/MEG.0000000000000279
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
TDM of infliximab in IBD patients Warman et al. 243
Veldhoven, the Netherlands, from October 2012 to May 2013. The protocol was cleared from approval by the local Medical Ethics Review Committee and the study was carried out in accordance with the Helsinki Declaration and Good Clinical Practice. CD and UC patients 18 years and older on maintenance IFX treatment were included, irrespective of the duration of IFX use. Patients were initially started on IFX 5 mg/kg every 8 weeks. In time, dose and interval were adjusted depending on the efficacy and side effects. Patients were allowed to use concomitant immunosuppressive therapy. Patient characteristics, comedication, and IFX dose were collected from the electronic hospital information system (EHIS-Chipsoft). Study design
Patients were admitted at day care to receive their IFX infusion. Blood for the determination of laboratory parameters was drawn before infusion. Blood samples were collected and stored at − 20°C. IFX trough level and the free fraction of ATI were determined by Sanquin (Amsterdam, the Netherlands) using a home-made enzyme-linked immunosorbent assay. The limit of detection was 0.01 μg/ml. The lowest level that could be quantified was 0.03 μg/ml and the highest level of quantification was 250 μg/ml. The precision as a percentage coefficient of variance was less than or equal to 18%. The lower level of quantification for ATI was 12 AE/ml. C-reactive protein, hemoglobin, erythrocyte sedimentation rate, alanine aminotransferase, alkaline phosphatase, creatinine, and estimated glomerular filtration rate were determined by the in-house laboratory of MMC. A combined questionnaire was developed to calculate the Crohn’s Disease Activity Index (CDAI) and the Truelove–Witts Disease Activity Index (TWDAI) [8,9]. Remission was achieved if the CDAI score was below 150 points and the TWDAI was below 6 points. The questionnaire was filled in by the patient during IFX infusion. Physical well-being such as the presence and severity of abdominal pain, stool, and complications (fistula, fissures, arthralgia/itis, and erythema nodosum) was noted. Patients were asked to score quality-of-life as a Visual Analog Scale (VAS) score. Statistical analysis
Statistical analysis was carried out using SPSS, version 19.0 (IBM, Armonk, New York, USA). Descriptive data were analyzed using frequency distributions for categorical data and calculation of median and range or interquartile range (IQR) for continuous variables. Univariate analysis with an unpaired t-test or the Mann–Whitney test was used for continuous variables. Categorical data were analyzed in a 2 × 2 contingency table. The odds ratio was calculated and significance was determined using Pearson’s χ2 or Fisher’s exact test. Correlations
were assessed using Spearman’s rank correlation test. A receiver operating characteristic (ROC) curve was performed, using remission as a discrete variable, to calculate sensitivity (Se), specificity (Sp), and area under the ROC curve and to determine predictive cutoff values with the associated P-value. P-values were calculated two tailed and P-value less than 0.05 was considered statistically significant.
Results Patient characteristics
A total of 113 patients with CD or UC were treated with IFX during the inclusion period. In two patients, a blood sample could not be drawn, one patient refused to provide a blood sample, and in three patients, insufficient blood samples were taken. Thus, 107 patients were included. Patient characteristics are shown in Table 1. IFX trough level and ATI
Figure 1 presents the variation in the IFX trough levels, each bar representing a patient. The level varied from undetectable (< 0.02 μg/ml) to 21.90 μg/ml. The median trough level in the total IBD population was 2.8 μg/ml (IQR 1.37–5.13) and did not differ between CD and UC (respectively, 2.80 and 2.83 μg/ml, P = 0.933). The IFX trough level was lower than 1 μg/ml in 18.7% of the patients. Eleven patients developed ATI (10.3%), shown as dark gray bars in Fig. 1; six of these patients were diagnosed with CD and five patients were diagnosed with UC. Detectable ATI were associated with the absence of IFX trough level [relative risk (RR): 2.2 (95% confidence interval (CI): 1.368–3.610) P < 0.0001]. Eleven patients developed infusion reactions during the current infusion or in the past. Reactions as elevated temperature, transpiration, itch, airless and headache were mentioned. ATI were associated with the risk of infusion reactions Table 1
Patient characteristics
Sex [n (%)] Male Female Age (years) Median (IQR) Indication [n (%)] Crohn’s disease Ulcerative colitis Dose (mg) Median (IQR) Interval (weeks) Median (IQR) Duration infliximab therapy (months) Median (IQR) Comedication [n (%)] Immunosuppressant (mainly azathioprine and mercaptopurine) Corticosteroids 5-Aminosalicylate IQR, interquartile range.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
49 (45.8) 58 (54.2) 40.5 (30.81–53.38) 61 (57) 46 (43) 400 (300–400) 8 (7–8) 40.6 (22.4–63.97) 51 (47.7) 11 (10.3) 11 (10.3)
244 European Journal of Gastroenterology & Hepatology 2015, Vol 27 No 3
Fig. 1
100
Infliximab trough level (μg/ml)
Infliximab trough level, without antibodies against infliximab (89.7%) Infliximab trough level, with antibodies against infliximab (10.3%)
10
1
0.1
0.01 Infliximab trough levels ranging from high to low, and the appearance of antibodies against infliximab. Each bar represents an infliximab trough level of a patient.
[RR: 11.7 (95% CI: 2.744–49.599) P < 0.0001] and the need for premedication [RR: 6.0 (95% CI: 1.618–22.22) P = 0.003]. Remission
Overall, 53.9% of the IBD patients were in remission. Differentiating between CD and UC, respectively, 75.4 and 26.7% were in remission. The median CDAI and TDWAI scores were, respectively, 89 (IQR 39.50–146.00) and 8 (IQR 6.50–9.00). IFX trough level was associated significantly with remission in the IBD population (P = 0.007). When separated in terms of the two IBD modalities, this association remained significant for CD (P = 0.010), but not for UC (P = 0.051). Undetectable ATI were not associated with remission (P = 0.182). IBD patients in remission showed a median IFX trough level of 3.9 μg/ml (IQR 1.90–6.53) and patients with active disease showed a median of 2.12 μg/ml (IQR 0.77–4.38) (P = 0.074). On differentiation, the median IFX trough level for CD in remission was 3.35 μg/ml (IQR 1.83–6.00) and the median IFX trough level for CD not in remission was 1.52 µg/ml (IQR 0.44–2.47) (P = 0.017). For UC, the difference between the medians was not significant: 5.44 µg/ml (IQR 2.22–7.89) versus 2.66 µg/ml (IQR 1.18–4.59) (P = 0.445). Logistic regression analysis showed a 20% higher chance of remission with every increase of 1 μg/ml IFX trough level [RR: 1.207 (95% CI: 1.044–1.396) P = 0.011]. Figures 2 and 3 show the ROC curves with the corresponding Se and Sp for UC and CD. ROC analysis was carried out to determine the cutoff value for both
indications. CD showed a cutoff value of 2.18 µg/ml (Se 67.4% and Sp 78.6%). The cutoff level of UC was determined to be 6.26 µg/ml (Se 50% and Sp 87.9%). The association between different cutoff values and remission was calculated with the corresponding P-values and odds ratios, shown in Tables 2 and 3. Influence of patient characteristics on remission, IFX trough level, and ATI
Older age was associated with active disease (P = 0.001). Patients with CD were more likely to be in remission than UC [RR: 0.120 (95% CI: 0.050–0.290) P < 0.001]. Remission was independent of sex, dose regime, or duration of IFX use. Concomitant use of immunosuppressants, corticosteroids, and 5-aminosalicylates was not associated with remission. Of all laboratory parameters, only erythrocyte sedimentation rate was associated significantly with remission (P = 0.044). IFX trough levels were not significantly different between the sexes or indications (respectively, P = 0.308 and 0.960), neither was there an association between trough level and age. The IFX level was not influenced by the use of immunosuppressants, corticosteroids, and 5-aminosalicylates. The IFX trough level was associated with the interval of administration, but not with the dose [correlation coefficient, respectively, − 0.427 (P < 0.0001) and 0.189 (P = 0.051)]. Despite the association between IFX and remission, there was no relation between the IFX trough level and physical discomfort or the VAS score, nor was there a relation between remission and the VAS score. ATI were not influenced by the concomitant use of immunosuppressants or any of the patient characteristics.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
TDM of infliximab in IBD patients Warman et al. 245
Fig. 3
1.0
1.0
0.8
0.8 Sensitivity
Sensitivity
Fig. 2
0.6
0.4
0.6 0.4 0.2
0.2 0.0 0.0
0.0 0.0
0.2
0.4
0.6
0.8
1.0
IFX level (μg/ml)
IFX level (μg/ml)
0.5 1.0 1.5 2.0 3.0 4.0 5.0 6.0 7.0
Sp (%)
P-value
100 100 88 83 58 58 50 50 25
18 21 30 39 55 68 80 86 97
0.171 0.164 0.466 0.287 0.445 0.086 0.060 0.016 0.069
RR (95% CI) 0.818 0.788 2.174 2.857 1.680 3.220 3.714 5.600 5.000
(0.647–0.961) (0.660–0.940) (0.401–11.781) (0.535–15.264) (0.441–6.394) (0.821–12.632) (0.910–15.154) (1.277–24.564) (0.925–27.041)
Table corresponding to Fig. 2. Area under curve: 0.692 [P = 0.051 (95% CI: 0.517–0.867)] ulcerative colitis. CI, confidence interval; IFX, infliximab; RR, relative risk; Se, sensitivity; Sp, specificity.
Multivariate analyses showed no confounding factors in terms of patient characteristics on the association between IFX trough level and remission.
Discussion TDM of IFX is not common care in the daily practice of a gastroenterologist treating IBD. Lack of efficiency or the appearance of side effects may often be encountered by dose or interval adjustments before turning to TDM. Therefore, these trough levels are not based on a standard regime, as we see in patients newly started on IFX, but might be influenced by the adjustments. This study provided insights into IFX trough levels in our IBD cohort in which dose adjustments had already been performed and whether there is still an association with remission. IFX trough levels showed large interindividual differences and a high proportion of patients had IFX trough levels below 1 μg/ml (18.7%). We found a significant association between remission and IFX trough level
0.8
1.0
Association between IFX trough level and remission with the corresponding sensitivity and specificity for Crohn’s disease
Table 3
Se (%)
0.6
Receiver operating characteristic (ROC) curve – cutoff value Crohn’s disease.
Association between IFX trough level and remission with the corresponding sensitivity and specificity for ulcerative colitis
Table 2
0.4
1 – Specificity
1 − Specificity Receiver operating characteristic (ROC) curve – cutoff value ulcerative colitis.
0.2
0.5 1.0 1.5 2.0 3.0 4.0 5.0 6.0 7.0
Se (%)
Sp (%)
P-value
88 86 84 70 54 47 33 25 17
25 43 50 64 79 86 93 93 93
0.202 0.021 0.011 0.023 0.062 0.056 0.084 0.258 0.664
RR (95% CI) 3.040 4.625 5.143 4.154 4.217 5.217 6.276 4.469 2.528
(0.687–13.461) (1.181–18.119) (1.368–19.330) (1.164–14.825) (1.029–17.276) (1.040–26.165) (0.745–52.896) (0.523–38.219) (0.283–22.567)
Table corresponding to Fig. 3. Area under curve: 0.731 [P = 0.010 (95% CI: 0.579–0.882)] Crohn’s disease. CI, confidence interval; IFX, infliximab; RR, relative risk; Se, sensitivity; Sp, specificity.
(P = 0.007) in the total IBD population. On comparing the medians, IBD patients in remission showed a higher IFX trough level (3.9 vs. 2.1 μg/ml, not significant). On differentiating the two IBD modalities, the difference in the median IFX trough level between remission and not remission became significant for CD. We carried out ROC analyses to determine the cutoff value for CD and UC. We found a cutoff value of 2.18 μg/ ml for CD and 6.26 μg/ml for UC. Sp and Se for the cutoff value of CD showed a moderate Se and good Sp. The Se for the cutoff value of UC was poor: 50%. This means that there was a relatively high probability of finding false-negative UC patients (high IFX trough level, but not in remission). The literature describes several cutoff values. Steenholdt et al. [10] found good Se and Sp (Se 86%, Sp 85%, AUCROC 0.93, not significant) with a therapeutic cutoff value of 0.5 μg/ml in CD. Paul and colleagues and Vande Casteele and colleagues found cutoff values of, respectively, 2 μg/ml (Se 76%, Sp 82.3%, AUC-ROC 0.68,
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
246 European Journal of Gastroenterology & Hepatology 2015, Vol 27 No 3
P = 0.017) and 3 μg/ml in IBD patients. Furthermore, Vande Casteele and colleagues also reported a therapeutic upper limit of 7 μg/ml. Dose reduction above 7 μg/ml resulted in lower drug exposure and lower IFX cost while maintaining disease control [11,12]. We did not find more discomfort or adverse events when the IFX trough level was above 7 µg/ml, nor did we find a significantly higher number of patients in remission. The presence of ATI may lead to an absence in detectable IFX trough level by increasing drug clearance. Indeed, we did find a strong association between the disappearance of IFX and the presence of ATI. This finding, however, did not result in a significant association between ATI and remission. A meta-analysis by Lee et al. [13] reported the same conclusion. However, a meta-analysis by Nanda et al. [14] did find a significantly higher risk on loss of response because of detectable ATI [RR: 3.2 (95% CI: 2.0–4.9, P < 0.0001)]. Infusion reactions were also associated with detectable ATI. This was confirmed by the meta-analysis of O’Meara and colleagues. They reported that IBD patients treated with IFX who developed ATI had a two-fold risk of acute infusion reactions and a six-fold higher risk of serious acute infusion reactions [15]. The presence of ATI may be influenced by the immune status of the patient, dosing regime, and the use of concomitant immunosuppressant therapy [5]. Episodic IFX treatment is associated with a significantly higher risk of developing ATI compared with maintenance therapy [7,13,16]. At present, it is advised to administer IFX as maintenance therapy, as we did in our study. The concomitant use of immunosuppressants was not associated with less formation of ATI in our study, nor in the study of Maser et al. [7], although there was a tendency toward significance (10 vs. 26% P = 0.11) However, this seems to be power related. Lee et al. [13] pooled 11 studies and found a 50% risk reduction of developing ATI when IFX was combined with immunosuppressants (P < 0.00001). O’Meara et al. [15] reported a modest risk reduction of ATI development by pooling four studies [RR 0.6 (95% CI: 0.4–0.9, P = 0.02)]. A recent study of Vande Casteele and colleagues showed that ATI formation may be transient and therefore may not always lead to worse clinical outcome. The addition of an immunosuppressant or intensification of the IFX therapy might lead to the disappearance of ATI. Whether this strategy can be applied in the case of detectable ATI should be investigated further in the near future [17,18]. Differences between our findings and those of other studies in the cutoff levels, Se/Sp, and the influence of ATI may be partly explained by the differences in study designs. We included IBD outpatients admitted at day care for their regular IFX infusion. This was a relatively healthy group of IBD patients, and none of them had experienced an acute exacerbation of their disease.
Besides, the majority of studies included IBD patients who had just entered the IFX maintenance phase, which resulted in a fairly homogenous group of patients. We measured IFX trough levels and ATI in a relatively heterogeneous group of IBD patients in whom dose adjustments had already been performed. Time on IFX varied between 3 and 161 months. This might be considered as a weakness because it is unclear as to how reliable these results may be for patients just started on IFX. However, this was not the main aim of this study; we aimed to determine whether there was still an association between remission and IFX trough level in our IBD cohort, resembling daily practice. TDM is not yet a common practice, and is often warranted after dose adjustments have already been made. We conclude that there is still an association and we could determine the cutoff value for IFX trough levels for both CD and UC. We did find a huge difference in the cutoff values between UC and CD. Our analysis showed that patients with CD were more likely to be in remission than patients with UC. This might be explained by differences in disease pathology and therefore response to IFX. Perhaps, a higher cutoff value is required for UC. Arias et al. [19] found a cutoff IFX value of 7.19 µg/ml for a sustained benefit in newly diagnosed UC patients. Data on IFX trough levels in relation to response and remission are still scarce in UC. The method used to determine remission may also play a significant role in the different cutoff values that we found. Disease activity indices were scored with CDAI and TWDAI [8,9]. CDAI is a wellknown disease activity index for CD and has been used in many studies. We found a cutoff value for CD that was comparable with the literature. The method used to score disease activity for UC is more diverse and differs between studies. We found few UC patients in remission using a TWDAI score; this might have been because of the sensitivity of this index. ROC analysis did show a cutoff value (sum of Se and 1 − specificity), but graphically, there was no clear cutoff at all. Other indices, such as the Mayo Endoscopic Scoring of Ulcerative Colitis (MAYO) or the Simple Clinical Colitis Activity Index (SCCAI), might be more sensitive and therefore may define remission more clearly. More research is needed to establish a sensitive method for the measurement of disease activity for UC and to determine a cutoff value. Although the importance of an adequate IFX trough level has been reported in many studies, as in ours, there is no consensus as yet how to act to a certain IFX trough level. Several treatment algorithms exist, but to our knowledge, these have not been validated as yet in daily practice [11,12,17]. Our data point to an IFX trough level of at least 2 μg/ml in CD. But how should we react on IFX trough levels below 2 μg/ml? The first question that should be answered is whether the patient is in remission or is still experiencing complaints. We might consider discontinuation of IFX therapy in CD patients in
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
TDM of infliximab in IBD patients Warman et al. 247
remission with IFX trough levels below 2 μg/ml. This may be complicated by unwillingness to stop, the chance of relapse, or the risk of ATI formation. If not in remission, the trough level can be optimized by intensifying the dose or reducing the interval, or even by switching to another monoclonal antibody against TNF-α. Both Afif and colleagues and Pariente and colleagues showed that dose intensification (not differentiating between interval shortening and dose escalation) or switching to another anti-TNF-α agent in patients with subtherapeutic IFX levels resulted in complete or at least a partial response. They differed in the use of the methods in terms of detectable ATI and therapeutic failure. Afif et al. [20] reported a greater response when switching to another TNF agent compared with dose intensification, whereas Pariente et al. [21] reported a good clinical response after dose intensification. Hibi et al. [22] achieved therapeutic IFX trough levels by shortening the dosing interval to 4 weeks, resulting in improved clinical response or even remission. Paul and colleagues increased the IFX dose from 5 to 10 mg/kg in IBD patients who experienced loss of response. A delta IFX more than 0.5 μg/ml was associated with mucosal healing (Se 0.88, Sp 0.77, P = 0.0001, AUC-ROC 0.89) [11]. To our knowledge, there are no studies that compare these two strategies or the difference in cost effectiveness. In daily practice, these are used separately on the basis of the preference of the individual specialist or even applied in combination to optimize the IFX trough level.
A. Warman (PharmD) contribution: researcher, writing protocol, training nurses, collecting data, statistical analysis, writing paper; J.W.A. Straathof (MD, PhD) contribution: providing information, commentator; L.J.J. Derijks (PharmD, PhD) contribution: commentator, guidance during research. Conflicts of interest
There are no conflicts of interest.
References 1
2
3
4
5
6
7
8
Conclusion
We found large interindividual differences in IFX trough levels in an IBD population in whom dose adjustments had already been performed. IFX trough levels were below 1 μg/ml in a significant proportion of IBD patients. Moreover, ∼ 10% of the patients developed ATI. TDM is indicated to identify this group of patients, but is not a common practice as yet. Remission was associated with the IFX trough level. However, this was clearer for CD than for UC. We found an IFX cutoff value of 2.18 μg/ml for CD and 6.26 μg/ml for UC. We question, however, the applicability of the cutoff value for UC. To date, there is no consensus as yet on therapeutic lower and upper IFX limits, or the strategy to be applied when therapy fails. Adequately powered future studies or combined studies should establish the therapeutic window of IFX and determine which strategy is the best in case of therapeutic failure.
9 10
11
12
13
14
15
16
Acknowledgements The authors thank Jeanne Dieleman, epidemiologist at the Máxima Medical Center, for statistical advice. They also thank the scientific commission of the Máxima Medical Center for providing the funds to determine IFX trough levels and antibodies against IFX at Sanquin, Amsterdam.
17
18
Hanauer SB, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, et al. Maintenance infliximab for Crohn’s disease: the ACCENT I randomised trial. The Lancet 2002; 359:1541–1549. Sands BE, Anderson FH, Bernstein CN, Chey WY, Feagan BG, Fedorak RN, et al. Infliximab maintenance therapy for fistulizing Crohn’s disease. N Engl J Med 2004; 350:876–885. Rutgeerts P, Sandborn WJ, Feagan BG, Reinisch W, Olson A, Johanns J, et al. Infliximab for induction and maintenance therapy for ulcerative colitis. N Engl J Med 2005; 353:2462–2476. Reinisch W, Sandborn WJ, Rutgeerts P, Feagan BG, Rachmilewitz D, Hanauer SB, et al. Long-term infliximab maintenance therapy for ulcerative colitis: the ACT-1 and -2 extension studies. Inflamm Bowel Dis 2012; 18:201–211. Colombel JF, Feagan BG, Sandborn WJ, van Assche G, Robinson AM. Therapeutic drug monitoring of biologics for inflammatory bowel disease. Inflamm Bowel Dis 2012; 18:349–358. Seow CH, Newman A, Irwin SP, Steinhart AH, Silverberg MS, Greenberg GR. Trough serum infliximab: a predictive factor of clinical outcome for infliximab treatment in acute ulcerative colitis. Gut 2010; 59:49–54. Maser EA, Villela R, Silverberg MS, Greenberg GR. Association of trough serum infliximab to clinical outcome after scheduled maintenance treatment for Crohn’s disease. Clin Gastroenterol Hepatol 2006; 4:1248–1254. Yoshida EM. The Crohn’s disease activity index, its derivatives and the inflammatory bowel disease questionnaire: a review of instruments to assess Crohn’s disease. Can J Gastroenterol 1999; 13:65–73. Truelove SC, Witts LJ. Cortisone in ulcerative colitis. BMJ 1955; 29:1041–1048. Steenholdt C, Bendtzen K, Brynskov J, Thomsen O, Ainsworth MA. Cut-off levels and diagnostic accuracy of infliximab trough levels and anti-infliximab antibodies in Crohn’s disease. Scand J Gastroenterol 2011; 46:310–318. Paul S, del Tedesco E, Marotte H, Rinaudo-Gaujous M, Moreau A, Phelip JM, et al. Therapeutic drug monitoring of infliximab and mucosal healing in inflammatory bowel disease: a prospective study. Inflamm Bowel Dis 2013; 19:2568–2576. Vande Casteele N, Gils A, Ballet V, Compernolle G, Peeters M, van Steen K, et al. Randomised controlled trial of drug level versus clinically based dosing of infliximab maintenance therapy in IBD: final results of the TAXIT study. U Eur Gastroenterol J 2013; 1 (Suppl 1):A1. Lee LY, Sanderson JD, Irving PM. Anti-infliximab antibodies in inflammatory bowel disease: prevalence, infusion reactions, immunosuppression and response, a meta-analysis. Eur J Gastroenterol Hepatol 2012; 24:1078–1085. Nanda K, Cheifetz A, Moss AC. Impact of antibodies to infliximab on clinical outcomes and serum infliximab levels in patients with inflammatory bowel disease (IBD): a meta-analysis. Am J Gastroenterol 2013; 108:40–47. O’Meara S, Nanda KS, Moss AC. Antibodies to infliximab and risk of infusion reactions in patients with inflammatory bowel disease: a systematic review and meta-analysis. Inflamm Bowel Dis 2014; 20:1–6. Hanauer SB, Wagner CL, Bala M, Mayer L, Travers S, Diamond RH, et al. Incidence and importance of antibody responses to infliximab after maintenance or episodic treatment in Crohn’s disease. Clin Gastroenterol Hepatol 2004; 2:542–553. Vande Casteele N, Gils A, Singh S, Ohrmund L, Hauenstein S, Rutgeerts P, et al. Antibody response to infliximab and its impact on pharmacokinetics can be transiënt. Am J Gastroenterol 2013; 108:962–971. Ben-Horin S, Waterman M, Kopylov U, Yavzori M, Picard O, Fudim E, et al. Addition of an immunomodulator to infliximab therapy eliminates antidrug antibodies in serum and restores clinical response of patients with inflammatory bowel disease. Clin Gastroenterol Hepatol 2013; 11:444–447.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
248 European Journal of Gastroenterology & Hepatology 2015, Vol 27 No 3
19
20
Arias MT, Vande Casteele N, Drobne D, Ferrante M, Cleynen I, Ballet V, et al. OP10. Importance of trough levels and antibodies on the long-term efficacy of infliximab therapy in ulcerative colitis. J Crohn’s Colitis 2012; 6 (Suppl 1): S5. Afif W, Loftus EV Jr, Faubion WA, Kane SV, Bruining DH, Hanson KA, Sandborn WJ. Clinical utility of measuring infliximab and human anti-chimeric antibody concentrations in patients with inflammatory bowel disease. Am J Gastroenterol 2010; 105:1133–1139.
21
22
Pariente B, Pineton de Chambrun G, Krysiek R, Desroches M, Louis G, Cassan De C, et al. Trough levels and antibodies to infliximab may not predict response to intensification of infliximab therapy in patients with inflammatory bowel disease. Inflamm Bowel Dis 2012; 18:1199–1206. Hibi T, Sakuraba A, Watanabe M, Motoya S, Ito H, Motegi K, et al. Retrieval of serum infliximab level by shortening the maintenance infusion interval is correlated with clinical efficacy in Crohn’s disease. Inflamm Bowel Dis 2012; 18:1480–1487.
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.
