This uncommissioned review article was subject to full peer-review.
Review article: causative factors and the clinical management of patients with Crohn’s disease who lose response to anti-TNF-α therapy
Article first published online: 3 MAY 2011
© 2011 Blackwell Publishing Ltd
Alimentary Pharmacology & Therapeutics
Volume 34, Issue 1, pages 1–10, July 2011
How to Cite
Danese, S., Fiorino, G. and Reinisch, W. (2011), Review article: causative factors and the clinical management of patients with Crohn’s disease who lose response to anti-TNF-α therapy. Alimentary Pharmacology & Therapeutics, 34: 1–10. doi: 10.1111/j.1365-2036.2011.04679.x
- Issue published online: 2 JUN 2011
- Article first published online: 3 MAY 2011
- Publication data Submitted 10 June 2010 First decision 12 July 2010 Resubmitted 11 April 2011 Accepted 12 April 2011 EV Pub Online 3 May 2011
Aliment Pharmacol Ther 2011; 34: 1–10
Background The addition of antitumour necrosis factor-α (TNF-α) agents to the therapeutic armamentarium against Crohn’s disease has been a revolution in its management. However, approximately 25 to 40% of patients who initially benefit from anti-TNF-α treatment develop intolerable adverse events or loose their response during maintenance therapy.
Aim To summarise the current knowledge on the mechanisms underlying loss of response in these patients and the therapeutic strategies available to counteract this clinical challenge.
Method A literature search using PubMed, MedLine and Embase databases has been performed.
Results Anti-infliximab antibodies formation and autoantibodies (ANA, anti-DNA and other autoantibodies) have been associated with loss of response. Individual differences in drug metabolism may contribute to loss of response. Smoking may be a risk factor for loss of response. Dose escalation, reduction of infusion intervals and switch to other anti-TNF-α agents are effective as rescue strategies.
Conclusions Loss of response appears to result from different causes not fully established by now. Optimization of therapies, or switch to other anti-TNF-α, are currently the best studied strategies in case of loss of response, and can be successful in 40–60% of patients who lose response.
Tumour necrosis factor-α (TNF-α) is a key cytokine in the inflammatory process. Anti-TNF-α agents have revolutionised the treatment of patients with chronic inflammatory diseases such as Crohn’s disease (CD), achieving better outcomes for patients than previously obtained with conventional drugs, such as mesalazine (mesalamine), steroids or thiopurines. Anti-TNF-α agents effectively induce and maintain long-term remission in patients with CD.1 In addition, efficacy has been demonstrated for the treatment of patients with ulcerative colitis (UC), the other major entity of inflammatory bowel disease (IBD).2
Thus far, two anti-TNF-α antibodies have been approved in the EU for the treatment of moderate-to-severe active CD in patients who have not responded to a corticosteroid and/or an immunosuppressant or who are intolerant to or have medical contraindications for the standard therapeutics: infliximab, a chimeric monoclonal antibody and adalimumab, a fully humanised monoclonal antibody. In the US, additionally a pegylated Fab’ fragment of a humanised TNF inhibitor monoclonal antibody, certolizumab pegol, has been approved for administration to patients with CD who have not responded adequately to standard therapeutics.3
In randomised controlled trials between 25% and 40% of patients who initially benefit from treatment with an anti-TNF-α agent went on to develop intolerable adverse events or were losing their response during the scheduled maintenance treatment.4–7 The latter mostly occurred within the first 6 months after initiation of treatment. These secondary nonresponders should be differentiated from primary nonresponders, a subset of patients who do not respond to the initial induction therapy with an anti-TNF-α therapy. The reasons underlying loss of-response are not completely understood. The contribution of several factors have been investigated thus far, including the development of neutralising antibodies, the immune status of the patient, genetic factors, alterations in metabolism of the drug and concomitant medications that may interact with the activity of anti-TNF-α antibodies.
In this review, we discuss the reasons underlying this secondary LoR as well as the treatment options available for such patients. This review summarises currently available data on loss of-response to anti-TNF-α therapy and investigates possible underlying mechanisms.
A literature review was performed by searching for the terms ‘loss of response’, ‘infliximab’, ‘adalimumab’, ‘certolizumab’, ‘anti TNF-α’, ‘biologics’, combined with ‘Crohn’s disease’, ‘ulcerative colitis’, AND lose OR lost OR loss OR ‘dose escalation’ OR intensification in the PubMed, Medline, Cochrane and EMBASE databases. All relevant articles in English published until December 2009 were reviewed.
Incidence of LoR
Infliximab has been used since more than one decade and evidence strongly supports the importance for long-term scheduled regimen. In the past, infliximab was often used episodically or stopped after a short course of therapy.2, 8 Recently, new evidence on long-term safety of infliximab and adalimumab as well as the benefit to treat patients who were in remission after a year of treatment has changed our clinical practice towards long-term use.9, 10 On the other hand, scheduled treatment regimens with infliximab and adalimumab are associated with an increased incidence of LoR. In fact, this event is estimated to account for 13% of patients/year under scheduled treatment,11 although the incidence varies markedly in different studies. Several predictors for LoR have been suggested, although reproducibly still remains to be determined.
Mechanisms of LoR
Anti-infliximab antibodies. The mechanisms underlying LoR to an anti-TNF-α therapy by patients with CD are poorly understood. This limitation has prevented the delineation of robust strategies to manage this clinical challenge. However, reviewing the currently available literature indicates that several mechanisms are involved in the process of LoR, although no single one of them appears to completely explain the event. The mostly investigated process associated with LoR is the influence of immunogenicity.7 Up to 60% of patients who are undergoing an episodic treatment regimen with infliximab develop antibodies to infliximab (ATIs), predisposing those patients to acute infusion reactions, delayed serum sickness-like reactions and to a reduction in the duration of their response to treatment.7 LoR is common under scheduled treatment as well whereas ATI formation is less frequent in this situation, which indicates that LoR to infliximab is not solely to be explained by immunogenicity.12 Furthermore, although episodic treatment and associated disappearance of infliximab from the circulation appear to booster ATI formation and favour LoR, ATIs are not detected in approximately 25% of patients who lack detectable serum trough levels of infliximab.13 A variety of possible mechanisms have been postulated to underlie this observation, which are shown in Table 1.
|Factor||Mechanism of action||Role|
|ATI||IgG-mediated neutralisation of the anti-TNF-α therapy||Probable|
|Augmented clearance||Rapid elimination of drug from circulation leads to serum concentrations below its therapeutic level||Probable|
|Concomitant therapies||Interaction with drug metabolism||Probable|
|High baseline WBC count||Unknown||Possible|
|Anti-DNA, antihistone ANA||Unknown||To be confirmed|
|Low CRP at baseline||Reduced clearance of material derived from anti-TNF-α induced apoptosis||Not confirmed|
|Liver and renal function||Action on clearance of the drug||Not confirmed|
|Gender, age, BMI||Unknown||Unrelated|
Residual trough serum levels of infliximab are inversely correlated with the levels of immunoreactive TNF-α in the circulation. However, ATIs have been shown to diminish the biologically inactivating effects of systemic infliximab.14 Indeed, the neutralising activities exerted by ATI appear to be linked to their serum titres: although high titres are associated with LoR and infusion-mediated reactions, when the interval between infusions is decreased in patients with moderate titres, some will continue to maintain remission under infliximab, which points to the presence of circulating biologically active drug.14 Additionally, ATI impact the pharmacokinetics of infliximab by increasing its clearance. In a study in patients with ankylosing spondylitis, the clearance of infliximab was up to 75% greater in patients who had developed ATI compared with patients who were ATI negative.14
Other autoantibodies. In addition to ATI, auto-antibodies can play a role in LoR to infliximab. A small prospective study in patients with psoriasis investigated the role of antinuclear antibodies (ANA) and anti-dsDNA antibodies in patients who no longer responded to one or more of the three anti-TNF-α agents.15 Patients who had lost their response to the anti-TNF-α therapeutics had high titres of ANA and anti-dsDNA antibodies. In particular, the rate of positivity for those auto-antibodies was proportional to the number of different anti-TNF-α therapeutics the patients were administered. Up to 83% of patients who underwent three different anti-TNF-α had positivity for those auto-antibodies. One possible explanation for this observation is that the apoptotic effects of the drugs may increase the exposure of nuclear material to the immune system which appears particularly increased when coupled with low serum levels of C-reactive protein (CRP), a natural scavenger for nuclear detritus following apoptosis.15 A prospective cohort study showed that more than 50% of patients under prolonged therapy with infliximab develop auto-antibodies such as double-stranded (ds) DNA, single-stranded (ss) DNA and antihistone antibodies and that they generally persist up to 1 year after the last infusion.16 However, the clinical relevance of antinuclear antibodies remains elusive, albeit they appear to be associated with some infusion-related disease, especially skin disorders. In addition, ANA-positive patients have been reported to be more likely to develop significant ATI titres compared with patients who remain ANA negative, which poses another explanation for increased risk of LoR in ANA-positive patients although not shown in the respective study.16 More detailed studies are required to improve our understanding on how ANA and anti-dsDNA antibodies could be involved in the induction of LoR to anti-TNF-α and as to whether the observed associations are causatively linked or simply an epiphenomenon.
Trough serum concentration. More recent studies on trough serum concentration of anti-TNF-α agents entered the focus of interest in the research on mechanisms driving LoR. Clinical remission was achieved by 82% of patients with CD in whom there was a detectable serum trough level of infliximab as compared to 6% of patients in whom serum infliximab was undetectable (P < 0.001), including those without ATI. In addition to improved remission rates, detectable trough levels of infliximab in the serum were also associated with lower serum levels of CRP and a higher rate of endoscopic improvement.17 Similarly, in patients with UC administered infliximab, clinical remission was clearly associated with detectable trough levels of infliximab in the serum, as was endoscopic improvement and a decrease in the risk of subsequent colectomy (odds ratio 9.3; 95% confidence interval, 2.9–29.9; P < 0.001).18
However, when patients are administered the standard regimen of 5 mg/kg every 8 weeks, only 40% of patients have detectable preinfusion trough levels of infliximab in their serum at the scheduled therapy points.18 In addition, there is considerable inter-patient variability in the trough concentration of serum infliximab, pointing to broad variations in individual’s clearance rates. The standard regimen would therefore appear to be inadequate for a substantial proportion of patients. Pharmacokinetic studies have found that higher trough serum concentrations of infliximab can be achieved by either shortening the interval between infusions, or by increasing the dose of infliximab.19 In humans, adjusting the dose and/or the interval improves the pharmacokinetic profile of infliximab and thereby increases the proportion of patients who maintain a long-term clinical response.9 Recently, a retrospective study aimed to investigate whether shortening of infliximab infusions or dose increase in patients with Crohn’s disease may have differences in rescuing response resulted in no significant differences between strategies.20
More recently, serum albumin levels have been revealed as predictor for infliximab pharmacokinetics and clinical response in patients with moderate-to-severe active ulcerative colitis Patients with higher serum albumin concentrations maintained higher infliximab concentrations, lower clearance and longer half-life than patients with lower serum albumin. These observations are of major interest and may help to steer treatment with infliximab in ulcerative colitis. However, a potential mechanism behind these associations and its relevance for Crohn’s disease remains to be determined.21 In patients with ankylosing spondylitis liver enzymes or creatinine clearance do not affect the metabolism of infliximab.14 Body surface area and gender were significant covariates to the distribution of infliximab in the central compartment. No interaction has been found between nonsteroidal anti-inflammatory drugs (NSAID) or proton pump inhibitors (PPI).14 However, a correlation between higher white blood cell count at baseline and clearance of infliximab has been observed, probably because an activated reticular-endothelial system plays a key role in the catabolism of therapeutic IgG.22
As an alternative to the measurement of serum trough levels and ATIs, the role of functional infliximab circulating in the blood has been investigated in a small study on 33 patients with Crohn’s disease treated with infliximab.23 The authors suggested that levels of functional IFX and TNF-α-binding capacity (TNF-α-BC) as measured by fluid-phase radioimmunoassay can better predict primary failure of infliximab than ATI levels alone. Moreover, the infliximab concentration/TNF-α-BC ratio could be a useful tool to optimise therapy and to select patients for dose increase or shortening of intervals.23 Further studies are needed to confirm this observation and on how to make the method easily accessible to clinical practice.
Concomitant medications and metabolic factors. Evidence is accumulating that metabolic factors, as well as particular concomitant therapies, could influence the clearance of anti-TNF-α antibodies. In patients with rheumatoid arthritis (RA), the elimination rate of infliximab is lower in patients who are on concomitant methotrexate.24 Thus far, this observation has not been extended to patients with CD. On the other hand, when patients with CD discontinue concomitant azathioprine, trough serum concentrations of infliximab are decreasing compared with patients who continue a concomitant maintenance regimen with azathioprine.25
Smoking is a risk factor for relapses, surgery and need for steroids and immunosuppressants in Crohn’s disease.26 There is increasing evidence that smoking also affects the efficacy of anti-TNFs and increases the risk of LoR. A recent study on 30 patients treated with adalimumab suggested that smokers are significantly less likely to respond to treatment and that this effect correlates with the number of cigarettes/day.27
Concerning adalimumab, the induction regimen appears to impact long-term efficacy. Loftus et al. found that patients receiving an induction regimen of 160/80 mg were half as likely to receive weekly dosing because of LoR compared with 80/40 regimen. Regional variation in the United States population was also described. Gender and age did not impact on the risk of LoR.28
The role of pharmacogenomics in biological therapy for IBD is a recent concept currently evolving. The mechanisms of action of anti-TNFs are not completely elucidated29 and studies concerning this particular topic are at the very early stage. In RA, van Baarsen et al.30 found that thousands of genes are involved in the pharmacological response to TNF-blockers including infliximab, adalimumab and etanercept irrespective of clinical response by investigating peripheral blood cells. Down-regulated genes reflected several biological pathways such as inflammation, angiogenesis, B- and T-cell activation. The same group of authors do not support the notion that microarray analysis of pre-treatment biopsy specimens from synovia could be used in the context of personalised medicine to identify nonresponders to anti-TNF therapy. They argue with the impact of the varying presence of lymphoid aggregates in rheumatoid synovial tissue on results from studies about gene expression.31 Nevertheless, a recent gene array study on mucosal biopsies from patients with ulcerative colitis identified a predictive panel of genes for response to infliximab. The top five differentially expressed genes which differentiated responders from nonresponders with 95% sensitivity and 85% specificity were osteoprotegerin, stanniocalcin-1, prostaglandin-endoperoxide synthase 2, interleukin 13 receptor alpha 2 and interleukin 11.32 A similar approach on patients with Crohn’s disease allowed complete separation between colonic CD responders and nonresponders. The top five genes in that study were TNFAIP6, S100A8, interleukin 11, G0S2 and S100A9. A predictive gene panel for IFX responsiveness in ileal CD could not be derived. Interestingly, there was a huge overlap between the predictive genes identified in colonic CD and ulcerative colitis.33 Of note is another study showing that dysregulated antimicrobial peptides (AMPs) in IBD mucosa in general normalise in responders to infliximab, suggesting that dysregulation of AMPs is rather the consequence of inflammation than primary cause in IBD.34 All the above studies need to find further confirmation in large series but are pointing towards a direction of gene expression-based tailoring of anti-TNF-α treatment in IBD.
Management of LoR
Prevention. Some prevention strategies have been investigated to reduce the risk of LoR in scheduled maintenance therapy with infliximab. Combination with steroids and immunosuppressants can help to reduce ATI formation, but data are contrasting and currently do not allow clear indications.
Premedication with hydrocortisone has been studied in a prospective randomised placebo-controlled trial in 53 patients, showing that it can be useful in reducing ATI formation, although it does neither eliminate ATI formation nor infusion reactions.35 The use of concomitant azathioprine/mercaptopurine or methotrexate has also been shown to reduce ATI formation significantly.36 Recently, premedication with hydrocortisone has been compared with the concomitant use of azathioprine; however, no significant differences between both strategies in reducing the risk of LoR could be shown.37 Whereas van Assche et al. could not reveal differences in terms of LoR in patients who discontinued or continued azathioprine after 6 months of combined therapy infliximab-azathioprine,25 Oussalah et al. harvested different results.38 In the latter study, withdrawal of azathioprine during combined azathioprine-infliximab treatment before 27 months as well as the presence of signs of biological inflammation was found to be predictive for further loss of response to infliximab monotherapy. Indeed, in a large-scale clinical trial, SONIC (the Study of Biological and Immunomodulator Naïve Patients with CD), a higher proportion of patients remained in steroid-free remission in the combination arm of infliximab plus azathioprine compared with patients who were on infliximab alone.39 Whether this results from a negative effect of azathioprine on clearance of infliximab, a less immunogenic environment as a result of the combination therapy or is simply due to a bias resulting from an uneven distribution of patients with objective signs of inflammation at baseline remains to be clarified.39 At the moment, it seems that concomitant immunosuppression is the best strategy to increase therapeutic efficacy and reduce the risk of LoR.
Dose adjustment and switching to other anti-TNF. Pharmacokinetic studies and randomised controlled trials on infliximab suggest that a dose increase up to 10 mg/kg or shortening the intervals between infusions at 4 or 6 weeks can be an effective strategy to rescue response after LoR.8, 9, 20 Different strategies seem to be not significantly different, therefore infliximab can be administered at 5 mg/kg every 4 or 6 weeks, or at dose of 10 mg/kg every 8 weeks.20 Although neither adalimumab nor certolizumab pegol appear to be as immunogenic as infliximab, nonetheless patients administered these therapeutics demonstrate a similar frequency of LoR during maintenance therapy compared with patients administered infliximab. Despite differences in chemical structure that should reduce immunogenicity, the use of adalimumab can also lead to formation of antiadalimumab antibodies (AAA) and to the decrease of trough serum levels, similarly to infliximab.40 This formation is also associated with LoR, suggesting that immunogenicity seems not to depend solely upon the chimeric structure of infliximab and supports the concept that the underlying mechanism is related to inter-individual variations in drug clearance rather than immunogenicity.5, 6 When patients lose response to adalimumab, escalation of their dose to weekly instead of every 2 weeks (40 mg, subcutaneously) restores the response, with a drop in the CD activity index (CDAI) of >70 occurring in 76% of patients, and 45% of patients entering remission (CDAI < 150).41 Dose escalation also rescues response in patients administered 400 mg certolizumab pegol subcutaneously, with the dosing frequency being increased from every 4 to every 2 weeks. In the noncontrolled PRECiSE 4 clinical trial, this re-induction strategy resulted in a 29% remission rate in patients who had lost their response to certolizumab pegol during maintenance treatment, which was stable throughout a follow-up period of 12 months.42, 43 In summary, a substantial proportion of patients who stop responding to first-line anti-TNF-α treatment regain response after a dose intensification strategy with the same agent is implemented either by involving a reduction in interval or an increase in dose. A dose intensification strategy should therefore be considered prior to switching the patient to another agent (Figure 1). Indeed, dose escalation yields more quality-adjusted life-years than switching, although the costs associated with this approach seem to be higher.44
On the other hand, switching to an alternative anti-TNF-α drug is also an effective strategy and currently considered to pose the first treatment option in patients intolerant of their primary anti-TNF-α therapy. Indeed, a switch to a secondary anti-TNF-α antibody appears warranted even after futile dose adjustments of the primary anti-TNF-α compound have been attempted. Given that infliximab has been on the market for a significantly longer time than either adalimumab or certolizumab pegol, it is not surprising that the majority of the available data involves patients who have lost their response to infliximab and who were subsequently switched to one of the other two anti-TNF-α therapeutics. Post hoc analyses of trials involving patients who were intolerant of or losing response to infliximab and were subsequently enrolled into maintenance studies with adalimumab and certolizumab pegol indicate that a lower proportion of patients respond to the second-line anti-TNF-α therapy compared with patients naïve to anti-TNF-α therapeutics.5, 6
Thus far only one randomised, double-blind, placebo-controlled trial has reported on the efficacy of a second-line anti-TNF-α agent in patients who were intolerant of or lost response to a first-line anti-TNF-α agent. The GAIN (Gauging Adalimumab Efficacy in Infliximab NonResponders) trial found that adalimumab (160 mg at week 0 followed by 80 mg at week 2) induced remission at week 4 in 21% of infliximab nonresponders with moderate-to-severe CD compared with 7% of patients administered placebo (P < 0.001).45 This rate of remission was lower than the response rate reported for patients’ naïve to biological agents. In the CLASSIC I clinical trial, 36% of patients previously not exposed to biologics achieved remission at week 4 with an induction regimen of adalimumab 160 mg at week 0 followed by 80 mg at week 2.46
More support for switching to a second anti-TNF-α agent came from a systematic review of 15 studies, which identified 1810 patients who had discontinued infliximab and switched to adalimumab.47 Remission rates were highly variable across the 15 different studies, with short-term rates of between 41 and 83%, and rates of 19 to 68% at 12 months. Nonetheless, overall adalimumab was found to be effective in patients who have lost response to or were intolerant of infliximab.47 Recently, Afif et al. have showed that measurement of Human Anti-Chimeric Antibodies (HACAs) can be useful to select patients who need directly a switch from infliximab to adalimumab without passing through dose escalation or interval reduction. In fact, they found that 23% of patients (35 patients on 150) resulted positive to HACAs. In those patients, switching directly to adalimumab was associated with a complete or partial response in 92% of patients, whereas dose escalation had a response of 17%. On the other hand, in patients with subtherapeutic infliximab concentrations negative for HACAs, dose escalation was associated with complete or partial clinical response in 86% of patients whereas changing to another anti-TNF agent had a response of 33%.48 This study shows that measurement of HACA together with serum levels of IFX, when available, may give very useful additional information to make the best therapeutic choice in case of LoR.
The long-term clinical benefits of adalimumab in infliximab nonresponders were assessed in an observational study.40 Long-term clinical benefit was observed in 62% of patients for a median follow-up period of 20.4 months. However, the dose was stepped up to 40 mg/week in 65% of patients to maintain response, while 39% of the patients ceased therapy with adalimumab, mainly because of LoR. Importantly, patients who discontinued therapy had significantly lower trough serum concentrations than patients in whom the therapy was effective. Trough serum concentration was also affected by antibodies against adalimumab in 9% of the patients.40 This effect of antibodies to adalimumab mirrors the effects seen with ATI and is therefore an important indication of the similarity of the mechanisms at play in nonresponders. Thus, variations in clearance of adalimumab also appeared to be affected by immunogenicity.
Similar data have been obtained with certolizumab pegol from an open-label induction part of a clinical trial. Patients with moderate-to-severely active CD who had failed infliximab were administered certolizumab pegol (400 mg every 2 weeks) as a second-line treatment in the WELCOME (26-Week trial Evaluating the clinical benefit and tolerability of certoLizumab pegol induCtiOn and Maintenance in patients suffering from Crohn’s disease with prior LoR or intolErance to infliximab) trial. At week 6, 61% of patients had achieved a clinical response (decrease in CDAI score >100 points) and 39% had entered remission.49
In clinical routine, there is a subset of patients with CD who have received already all three of the anti-TNF- α agents as part of their overall treatment. A retrospective study identified 67 patients, 27 of whom had received adalimumab as their third-line treatment and 40 that received certolizumab pegol as their last-line anti-TNF-α treatment after failing infliximab and adalimumab. A clinical response was observed in 41 (61%) patients at week 6 and in 34 patients (51%) at week 20. The probability that treatment was ongoing at 3 months, 6 months and 9 months was 68%, 60% and 45% respectively.50 This study suggests that administration of a third anti-TNF-α therapy should be considered if there are no alternative therapy options remaining.
The second-line indication of the anti-TNF-α therapeutics may result in a situation where these agents are being prescribed earlier and more progressively in patients with CD, even prior to utilisation of immunosuppressants. Indeed, it may not be long before we start to see patients who have failed to respond to or have become intolerant of an anti-TNF-α therapy being scheduled maintenance therapies without previous exposure to immunosuppressants. Although such scenarios have not yet been investigated, it seems likely that thiopurines and/or methotrexate will be an appropriate subsequent treatment option.
Further nonanti-TNF-α therapies. The availability of biological therapeutics targeting alternate pathways than TNF-α is limited in IBD. Natalizumab is a humanised IgG4 monoclonal antibody that targets the α4 integrin, thereby blocking adhesion of leucocytes and their subsequent migration into the gut.1 It has been approved for the treatment of patients with CD in the US, but not in Europe. However, in the US natalizumab is only available through a restricted prescribing programme termed CD-TOUCH because of the risk of progressive multifocal leukoencephalopathy. The efficacy of natalizumab in patients who have previously failed to respond to an anti-TNF-α therapy has been investigated in a post hoc analysis of data from the ENACT (Evaluation of Natalizumab as Active Crohn’s Therapy) series of clinical trials. This analysis demonstrated that natalizumab poses a valid and efficacious treatment option in such patients.51
Ustekinumab is a human interleukin 12/23 monoclonal antibody. A placebo-controlled randomised study in patients with moderate-to-severe active CD showed that ustekinumab resulted superior to placebo at week 4 (P = 0.02), but not at week 8. However, in patients previously treated with infliximab, the response rate to ustekinumab remained significantly higher than placebo through week 8 (P < 0.05).52 Data from a large prospective randomised placebo-controlled phase II and III trial on ustekinumab in patients with previous primary or secondary failure to anti-TNFs will provide further insights.53
Finally, the safety and efficacy of extracorporeal photopheresis has been investigated in a prospective, uncontrolled pilot study involving 28 patients with moderate-to-severe CD who were refractory to or intolerant of immunosuppressants and/or anti-TNF-α medications. After 12 weeks, 50% of the patients had responded to the treatment and 25% had entered remission. Of particular interest, the response was similar among patients who were naïve to anti-TNF agents and patients who had previously been refractory to or were intolerant of an anti-TNF-α therapy.54
Patients who are undergoing maintenance treatment with an anti-TNF-α agent frequently lose their response during standard maintenance regimens. This largely appears to occur as a result of inter-individual variations in the pharmacokinetics of the agents. Optimisation of the dose administered, with particular regard to serum trough levels and functional circulating infliximab levels successfully restores response in the majority of patients and is therefore the option of choice, prior to considering switching to a second or third-line anti-TNFα therapy.
The serum levels of ATI have been proposed as a predictive factor for LoR. However, this is not a widely accepted hypothesis and their predictive potential needs to be confirmed in large-scale clinical trials. The role of other auto-antibodies, such as ANA, dsDNA, ssDNA, antihistone antibodies and other influencing factors are still to be investigated. Further studies on the role of activated genes in course of anti-TNF therapies are needed to select patients who would maintain prolonged response and to act with more effective therapeutic modifications in those who would be likely not to respond or lose response.
Concomitant medications may help in reducing the risk of LoR, but their efficacy seems to be limited overtime. For infliximab, thiopurines should be associated at least for 6 months, also to increase efficacy,25, 36, 39 although this could most likely only impact on ATI-derived LoR. Furthermore, the optimum timing of combining infliximab and thiopurines remains to be determined. More patients should be strongly requested to quit smoking, generally as soon as they start any therapy for CD, and not only before starting with infliximab. Concerning adalimumab and certolizumab pegol, evidence is needed to advise monotherapy vs. combination therapy with immunosuppressants. For adalimumab, evidence accumulates that the 160/80 induction regimen is preferable to reduce the risk of LoR.
Finally, in situations where a patient is no longer responding to a first-line anti-TNF-α therapy, it is essential that the clinician carefully re-evaluate objective signs of inflammation prior to making any further treatment decisions. Deterioration of clinical symptoms while undergoing a maintenance therapy regimen can occur as a result of many factors unrelated to the pharmacodynamics of the agent, including non-inflammatory intestinal impairments such as bacterial overgrowth, secondary lactose malabsorption, chologenic diarrhoea, steatorrhoea and drug-induced adverse events. In addition, the presence of complications that would merit surgical intervention, including the formation of enteric fistulas and strictures, must be assessed. It is important also to wait at least 12 weeks from the start of anti-TNF to evaluate eventual primary failures to therapy.3
Loss of response is an emerging challenge in the biological era, especially because no other valid therapeutic weapons are currently available in complicated moderate-to-severe disease, except for steroids. Data are still limited and further prospective trials on large populations of patients are strongly required. Optimising the available medications to prolong their use as long as possible and save further medical options is currently the best strategy to recommend in our clinical practice.
Declaration of personal interests: Silvio Danese has served as a speaker, consultant and advisory board member for Schering-Plough, Abbott Laboratories, UCB, Ferring, Cellerix, Millenium Takeda, Nycomed, Actelion, AstraZeneca, Novo Nordisk and Cosmo Pharmaceuticals. Gionata Fiorino has served as a consultant and advisory board member for Schering-Plough and Abbott Laboratories. Walter Reinisch has served as a speaker and advisory board member for Abbott, Aesca, AstraZeneca, Biogen, Centocor, Ferring, Elan, Essex, Genentech, MSD, Millenium, Novartis, Pharmacosmos, Schering-Plough, Shire, Takeda, UCB and Vifor. Declaration of funding interests: None.
- 19The relationship of serum infliximab concentrations to clinical improvement in rheumatoid arthritis: results from ATTRACT, a multicenter, randomised, double-blind, placebo-controlled trial. Arthritis Rheum 2002; 46: 1451–9., , , et al.
- 23Tumor necrosis factor-alpha binding capacity and anti-infliximab antibodies measured by fluid-phase radioimmunoassays as predictors of clinical efficacy of infliximab in Crohn’s disease. Am J Gastroenterol 2008; 103: 944–8., , .Direct Link:
- 28Patterns and predictors of dosage increase in patients treated with adalimumab for Crohn’s disease in the United States. J Crohns Colitis 2009; 3: S24., , , , .