Approaches to primary nonresponse to anti-TNF therapy
Despite the shift in treatment paradigms ushered in by the arrival of anti-TNF therapy for CD, it is important to remember that one-third of patients will not respond to anti-TNF therapy [Sandborn et al. 2007a,b; Hanauer et al. 2006a,b; Targan et al. 1997]. This proportion may decrease as we start to treat patients earlier and with combinations of effective therapy. Patients exposed to anti-TNF therapy and who have failed to respond are referred to as primary nonresponders. In clinical trials, this is defined as failure to achieve a 70- or 100-point reduction in the CDAI score. In clinical practice this may be established by the physician's clinical judgment. Most of the issues surrounding primary nonresponse surround IFX and CD; which is not surprising given that IFX has been available in clinical practice for longer than the other anti-TNF agents.
Patients with primary nonresponse to anti-TNF therapy should be evaluated to determine why they failed to respond. The most important practice point is to exclude the possibility of an underlying irreversible structural lesion, such as a fibrostenotic stricture, or a possible co-infection (e.g. with Clostridium difficile or cytomegalovirus) or absence of ‘true’ inflammation with symptoms driven by irritable bowel syndrome-like symptoms. If these CD complications or concomitant conditions have been ruled out, then it is generally accepted that primary nonresponse would result from a different underlying CD pathobiology in which TNF is not an important component of the inflammatory cascade. Therefore, choosing an agent with a different mechanism of action would be scientifically rational in this setting. There are data supporting the use of NTZ in this patient population. In subgroup analyses of pivotal trials, there were no statistical differences in the efficacy of NTZ for the induction or maintenance of response/remission in patients who were primary nonresponders to IFX as compared with the intent-to-treat population [Panaccione et al. 2004]. NTZ has been approved by the FDA as monotherapy for the treatment of patients with moderate-to-severe CD with biologic evidence of disease activity and who have failed to respond to conventional therapy and at least one anti-TNF agent. In the ENCORE (Efficacy of Natalizumab in Crohn’s Disease Response and remission) trial, which evaluated the efficacy of NTZ induction therapy in patients with CD, subgroup analysis demonstrated that the overall efficacy of NTZ was similar for patients who had been exposed to, or failed to respond to, previous anti-TNF therapy and patients who were anti-TNF-naive [Targan et al. 2007]. In patients who have never been exposed to MTX or the purine anti-metabolites, introducing these agents or optimizing their use may be considered. This likely represents a small minority of patients and there is no data to support this practice, but it may be rational.
Despite the rationale for switching mechanisms of action, many clinicians will attempt a second or even a third anti-TNF agent in primary nonresponders, or even a dose escalation (e.g. 10mg/kg of IFX). In the case of IFX, one could check for early antibody formation by testing for IFX levels and anti-IFX antibodies at week 6 prior to a third induction dose or dose escalation. There are very limited data to support this practice or switching to an alternate anti-TNF in primary nonresponders [Lofberg et al. 2008; Mozziconacci et al. 2008]. It has been clearly shown that the overall response and remission rates are lower when attempting to use a second or third anti-TNF in clinical trials and therefore the risk–benefit equation is shifted with decreasing benefits while assuming a similar risk. Clinicians should reconsider this strategy and wait for better evidence to be available. Most of this practice surrounds physicians' and patients' concerns about the issue of toxicity associated with NTZ, particularly with regards to the incidence of progressive multifocal leukoencephalopathy (PML) [Yousry et al. 2006]. Although PML is a serious and often life-threatening disease, it is estimated that the risk for PML in patients treated with NTZ is between approximately 1 in 5,000 and 1 in 10,000 patients, from the postmarketing reports and the number of cases treated in multiple sclerosis and CD. If this estimate is accurate, the risk of developing PML and dying is extremely low and does not make NTZ any less safe than the anti-TNF class of agents. Recent evidence in multiple sclerosis suggests that PML may be stabilized after its onset by plasma exchange to deplete NTZ and permit immune reconstitution.
Approaches to secondary nonresponse to anti-TNF therapy
In addition to primary nonresponders, there are secondary nonresponders who may initially respond to an anti-TNF agent and then lose response or become intolerant to their initial TNF inhibitor over time. Secondary nonresponders may represent 30–40% of patients during the first year of therapy [Colombel et al. 2007; Schreiber et al. 2007a,b,c; Hanauer et al. 2002]. Secondary loss of response may be due to disease-related factors or drug-related factors. Disease-related factors include the development of a complication of CD, such as a fibrostenotic stricture or abscess formation. Once again, disease-related factors and co-infection should be ruled out before deeming secondary nonresponse to be a drug-related phenomenon. Drug-related factors include the formation of neutralizing antibodies, which are associated with decreased serum IFX levels, altered clearance of drug, or possibly biologic escape mechanisms. It is important in this patient population to obtain drug and antibody levels if available (currently only available for IFX) and to then re-evaluate disease activity to ensure that symptoms are not arising due to irritable bowel syndrome, bile acid diarrhea, development of fibrostenotic strictures, or postsurgical causes. The majority of data available on the treatment of this patient population is, once again, focused on patients who developed secondary non response to IFX. If patients have developed intolerance to IFX, challenging them with a second anti-TNF agent is appropriate. Most of the experience with this approach is in switching to ADA. Sub-analysis of the CHARM trial demonstrated that patients who were previously exposed to IFX still responded to ADA therapy, but with response rates approximately 10% lower than those of patients who were naive to anti-TNF agents [Colombel et al. 2007]. Similar data exist for CZP from a sub-analysis of the PRECiSE 2/3 trials [Schreiber et al. 2007a,b,c]. The interdependence of this reduced response rate to duration of disease is not entirely clear.
The options for patients who are losing response to IFX are to ‘optimize’ the dosing of the drug by increasing the dose or shortening the dosing interval, or to switch from IFX to an alternative anti-TNF agent [DeSilva et al. 2008]. The measurement of antibodies to IFX and IFX trough levels is useful in guiding treatment strategy. In patients with no detectable anti-IFX antibodies and low serum trough levels of IFX, optimizing the dosing of IFX is recommended. A study conducted at the University of Pittsburgh has shown that many patients with CD receiving long-term IFX required an increase in dose or a decrease in dosing interval to regain lost response to the agent [Regueiro et al. 2007]. Concomitant immunosuppressant therapy did not prevent the need for escalated IFX dosing. In the ACCENT (A Crohn’s disease Clinical trial Evaluating IFX in a New Long-term Treatment regimen) I trial, 88% of patients with active CD who had initially responded to IFX but then lost response during maintenance therapy regained response by increasing the IFX dose to 10mg/kg [Rutgeerts et al. 2004]. The phenomenon of regaining IFX efficacy by dose escalation and shortening the dosing interval is in part explained by data from a study conducted by Maser and colleagues, which demonstrated that the major factor influencing the clinical efficacy of IFX during scheduled maintenance therapy was the serum trough level of this agent [Maser et al. 2006]. This study found that patients with higher trough levels of IFX have better outcomes and that the use of concomitant immunosuppressants did not alter outcomes. One must consider that these data are from an open-label study and therefore are inherently biased. In this series, it appears that shortening the interval may be more cost-effective than doubling the dose. In cases where there is a loss of response to ADA dosed at 40mg every other week, shortening the dosing interval to 40mg weekly has been shown to re-establish response in approximately 75% of patients [Sandborn et al. 2008a,b]. The optimal strategy for managing patients who lose response to CZP is unclear. Data from a small group of patients who lost response to CZP during PRECiSE 2 suggest that a single re-induction dose with CZP 400mg may re-establish response [Lichtenstein et al. 2008], but whether this strategy can be translated into clinical practice remains debatable.
The only randomized controlled trial specifically designed to address the issue of secondary nonresponse to IFX is the GAIN (Gauging Adalimumab efficacy in Infliximab Nonresponders) trial [Sanborn et al. 2007a,b,c]. In GAIN, 325 patients with moderate-to-severe CD who had previously been exposed to IFX and lost response and/or become IFX intolerant were randomized to receive either ADA 160mg and 80mg subcutaneously at weeks 0 and 2, respectively, or placebo. In the group receiving ADA, 21% of patients entered remission (CDAI<150), 52% had a CDAI decrease of 70 points or more, and 38% had a CDAI decrease of 100 points or more at week 4, compared with 7%, 34%, and 25% of patients in the placebo group, respectively (p<0.05). There was no difference in overall efficacy if the patients lost response to, or were intolerant to, IFX upon study entry. Although the 4-week remission rates reported in GAIN may appear low at first glance, recently presented data from an open-label extension of this trial demonstrated that, over time, remission and response rates increased during the maintenance phase with ADA therapy [Panaccione et al. 2008c]. At 6 and 12 months, 57% and 40%, respectively, of the week 4 responders to ADA were in clinical remission. Thus, 4 weeks may be too short a time point to assess induction of response in CD patients with prior exposure to IFX. Results from a similar open-label study of CZP in IFX nonresponders were also presented recently [Vermeire et al. 2008]. In WELCOME (26-Week open-label trial Evaluating the clinical benefit and tolerability of certoLizumab pegol induCtiOn and Maintenance in patients suffering from Crohn’s disease with prior loss of response or intolErance to IFX), secondary nonresponders to IFX were treated with CZP 400mg subcutaneously at weeks 0, 2, and 4. At week 6, 62.2% of patients achieved response (CDAI score reduction of 100 points or more) and 39.3% achieved remission. The long-term outcome of these patients has not yet been reported.
Primary and secondary nonresponse to anti-TNF therapy is a reality of treatment with this drug class. In patients with primary nonresponse, serious consideration should be given to switching to another class of biologic agent. The only option for switching to another class (i.e. a biologic agent with a different mechanism of action) at this time is NTZ. However, this decision should be made in concert with the patient, and many patients may choose to try a second anti-TNF agent. For patients who become intolerant to one anti-TNF agent, switching to another anti-TNF agent makes sense and is supported by clinical data. For patients who lose response to IFX, measurement of anti-IFX antibodies and IFX trough levels can be used to optimize the treatment strategy. In the presence of antibodies to IFX, the most robust (placebo-controlled) data at this time suggest a switch to ADA therapy.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3002580/#!po=28.3333
