PAUL J. RUTGEERTS, MD, PhD, FRCP, AGAF: I am here to discuss this very difficult topic because, of course, it is a big puzzle, and we have only a few pieces of the puzzle, and I think experience will show in the coming years how we have to deal with this problem of failure. But I would like to give first some definitions, and I think we need to differentiate between primary failure and secondary failure to biologic therapy.

Primary failure is a patient who does not respond to a loading dose of a biologic agent when he or she receives it for the first time. The question then arises whether this patient, in fact, is a nonresponder to all drugs targeting this specific pathway, meaning if you give 1 anti-TNF agent, you can give maybe a second dose. If the patient does not respond to those 2 doses, is this a patient who will never respond to any other anti-TNF therapy?

This is different, of course, from the secondary failures. Secondary failure is a patient who has a response to biologic therapy at onset, but then loses response or becomes intolerant to the molecule. It is clear that this patient probably may respond to dose adjustment, or that this patient is likely to benefit from other molecules targeting the same pathway. There is a big difference between a primary failure and a secondary failure. For the secondary failures, I have to emphasize that this problem probably occurs more frequently when the first biologic agent elicits higher levels of antibodies.

I would like to tackle the second situation first, the loss of response or the intolerance, and I would like to give some recommendations on how to deal with the loss of response. I think you can prevent already in the long-term, a bit, the problem of loss of response by choosing a drug with low immunogenicity. That is certainly a possibility, because most of the losses of response are due to immunogenicity and antibodies to the drug.

Second, you can optimize treatment with the same agent and increase the dose, decrease treatment interval, optimize it, and try to pursue the treatment as far as possible. I think this is critical because once you stop a drug, once you discontinue a drug, you will probably never come back to that drug, and it is important to get the best out of a drug for your patients.

If this fails, if this therapy adjustment fails, then I think you need to switch to induction with another anti-TNF agent. I do not think you can give immediately a maintenance dose; I think you need to re-induce with another agent.

If this fails, then you are in a difficult situation, of course. I think it might be possible to switch to a biologic agent targeting another pathway, no longer the TNF pathway, but looking at adhesion molecules, interleukin (IL)-12, but as you know, there are no medications available at present. Of course, at the end, you have to consider other options like increase or start of steroids, and surgery, but I would say that this is, of course, not in the big interest of your patient.

This is a study that is going to be proposed or presented tomorrow at the meeting, and in fact, we have now more than 600 patients with lumenal and fistulizing disease treated with infliximab for Crohn's disease. We have a complete follow-up in these patients, and it is almost 4 years, a median follow-up of 4 years now. I do not want to walk you through the slides, but I would like to emphasize that, in fact, with, let's say, interventions or adapting therapy or changing the interval of the dose for infliximab here, we were able to maintain a good response and a good quality of life in our patients in more than 60% of the patients. This is quite a striking result; in more than 600 patients, complete follow-up, and we achieved this excellent result.

You see that you cannot do that without sweating, without adjusting your therapy, without managing your patients. In the left panel here, you see that patients who are treated episodically, or are switched from episodic to scheduled therapy, need much more interventions than the patient who is treated in the right manner from the onset. With a patient who is treated with scheduled maintenance every 8 weeks, you see that you need much less interventions than if a patient is treated episodically. It is critical also to treat your patient from the onset in the right manner.

The second point is, and this is also part of this study, what do you do when the patient fails anyway? You see that 64% of our patients were treated then with adalimumab, were switched to adalimumab; another anti-TNF, 7%; other biologics, of course, in investigational protocols, 14%. But you see that, in fact, it is exposing the patients to surgery and also other therapies, which mostly are steroids. If you fail with your drug, eventually, either you switch or you will go to the toxic therapies or the therapies that really are not that beneficial to your patient.

What data are there? This is data of CHARM, and I think it is very important data on switch of agents. You know the protocol of CHARM; I want to walk you through in a short time. You see that there is first an induction with 80 mg and 40 mg, so 80 mg at time point zero, and at week 2, 40 mg; and at week 4, patients were randomized to receive either placebo or adalimumab every other week 40 mg, or every week 40 mg. The end points were at week 26 and at week 56.

What you see here are the data on clinical remission with adalimumab, and this by previous anti-TNF use. The patients have first received infliximab. The patients have responded to infliximab but have discontinued infliximab. It is not clear in this study what all the reasons were for this continuation, but what is clear here is that patients who have seen already an anti-TNF agent are doing not so well. They are doing less well than patients who are naive to anti-TNF agents. You see for instance here, and this is the data for remission, 42% versus 50% for, sorry, this is at week 26, 42% versus 33%, and this is 16% in patients who have seen previously anti-TNF. This is patients who were naive to anti-TNF, you see that the rates are higher, and also in the placebo group.

The same is true at week 56, higher response rates or remission rates in patients who have not seen another anti-TNF in comparison with patients who have already experienced anti-TNF therapy.

The same was found in the PRECiSE 2 study. PRECiSE 2, which is a certolizumab pegol study, is quite a comparable protocol. Patients also received open label induction. At week 6, they were randomized either to placebo or to certolizumab pegol, and the endpoint was at week 26, and in fact, they were evaluated by response.

Let us look at the clinical response and clinical remission at week 26, and also look at the status of exposure to infliximab. You see again that infliximab-naive patients do better than patients who have already experienced an anti-TNF. You see that the rates, in fact, for clinical response and also for remission are clearly lower, meaning that there is a certain resistance that has built up in these patients, and it is not clear what the exact reason is for this.

Of course, the GAIN (Gauging Adalimumab Efficacy in Infliximab Nonresponders) study has tried to answer the question of whether you can switch from one anti-TNF to another, and the paper has been published by Bill recently in the Annals of Internal Medicine. I will walk you through this study. The patients, 325 subjects, in fact, were randomized to either placebo or to induction with adalimumab 160 mg at week zero, and 80 mg at week 2. The primary endpoint was at week 4 and patients were followed up, and all patients had already failed therapy with infliximab.

When we look at the outcome data, you see that remission is achieved in the adalimumab-treated patients in 21% versus 7%, 100-point response is achieved in 38% versus 25%, and response rate is a 70-point response rate in 52% versus 34%. You see that the data are interesting, that this policy works; all of the remission rates seem a bit lower than you would expect in patients who are naive to anti-TNF therapy.

However, the treatment effect is quick. You see that by week 2 already, by week 1, separation is already beginning, but by week 2, the maximal effect is achieved, and you see that in the patient then, that there is no increased effect anymore beyond week 2.

But what you see in this very important data, is that the benefit for the patient is extremely rapid. Here, you see data for day 1 through day 7, and this is for all the components of the Crohn's Disease Activity Index (CDAI), and you see that the components of the CDAI decrease quickly and that already by day 4 to day 7, you achieve statistical significantly difference with patients treated with placebo. You can rescue patients with adalimumab if they lose response to infliximab. The effect is quick although only the proportion of patients that reach remission is lower than in naive patients treated with adalimumab.

The GAIN study, again, looked at remission with adalimumab by baseline antibodies to infliximab (ATI). If you look at the ATI-positive patients and compare with the ATI-negative patients, there is no difference in remission rate, as far as the antibody to infliximab status is concerned.

Looking at other aspects of intolerance to infliximab, so loss of response, intolerance, and a combination of loss of response and intolerance, you see that the effect of adalimumab is observed in all the groups of patients, although the patients who have loss of response and intolerance have a bit lower response rate than patients who have only intolerance or loss of response.

Looking at the quality of life, this is important in the GAIN study, you see that with placebo treatment, the patients have an improvement of quality of life with 15 points. This is not a relevant increase; 16 is the line of relevance, I would say, a relevant improvement. With adalimumab, there is an improvement with more than 30 points, and this is certainly an important increase in quality of life. The CDAI components improve quickly, and the quality of life also improves in the patients.

Let's switch gears to the second problem or the first problem we discussed, and that is how to minimize the rate of primary nonresponders. This is certainly a difficult problem, but I think you can minimize this by choosing your patients correctly at the start. I think you need to use the drug only for the appropriate indications. So, use it only correctly according to the label, and do not use it off label or do not use it in patients who really have symptoms that you know are not easily controlled with an anti-TNF agent.

Second, make sure that the patient has active disease. I think we use a lot of colonoscopy to assess the activity of the disease, and also use magnetic resonance imaging prior to starting patients with fistulizing disease on an anti-TNF agent. It is important also to exclude other reasons than active disease as the cause of symptoms such as short bowel syndrome, intercurrent infection, and underlying irritable bowel syndrome. You need to exclude these, and in this response, C-reactive protein can be very useful.

Finally, I think you should avoid using biologics in patients with symptomatic strictures; consider surgery instead. You will never be successful in patients with symptomatic strictures.

How to deal with primary failures to anti-TNF therapy: if you have selected your patient well and still the patient does not respond to induction, or the patient does not respond to administrations of the drug, what happens? I think that dose increase is rarely effective unless the initial dose was calculated based on weight, and the body weight was very low.

In the Targan study, a second open-label infusion with 10 mg/kg gave only a response rate in 16% of the patients. Adding a second induction, for instance, will not help you.

Another point is can we switch to another anti-TNF agent? Is this effective? This has never been tested properly in a primary failure population, but I must say that in the real population, there is a lot of switching between agents, and that some patients are treated when they have failed one agent with another agent, and some patients do respond. Then switch to another class of agents such as selective adhesion molecules, we have touched on this, or anti-IL-12 therapies, or investigational therapies, and again, consider these other therapies like steroids and surgery.

The switch to another class of agents, is that effective? This is also a very important question.

This is the ENCORE (Efficacy of Natalizumab in Crohn's Disease Response and Remission) study; you have heard about this study certainly during the meeting. The patients, 509 patients, were randomized either to placebo or to natalizumab 300 mg, and they were induced at week 0 and week 4.

The endpoints were at week 8 and week 12, and it was a composite endpoint in the sense that it was sustained response at weeks 8 and 12 with a reduction of the CDAI of at least 70 points.

What you see here is that the response rates also with the natalizumab in patients who have already seen infliximab prior, are lower than the patients who were naive; this is the patients who have already seen treatment. You see that the data are showing that the responses are lower.

I have to go back to this slide because this is the overview slide where you look at the response and remission overall; this is the slide where you look at the sustained response through week 8 and 12 for placebo and for natalizumab in patients who have already been exposed to infliximab, which is much lower than the patients who are naive to anti-TNF therapy.

I would like to conclude that true primary failures on a biologic agent should probably be treated with a drug targeting a different molecular pathway, although the data are very sparse. I think if a patient exhibits a primary response to a biologic, we should try to pursue this therapy as far as possible by adjusting therapy if response to the agent weakens. I think third, true loss of response or intolerance to a biologic can be solved by the use of a different biologic agent targeting the same molecular pathway, and this is important.

There appears to be no cross-intolerance among different anti-TNF antibodies. The kind of response rates may be decreased in patients previously exposed to a TNF antagonist when they are treated with the second anti-TNF agent, but also when they are treated with a drug targeting another pathway.

I would like to document this with a case study.

This is a case of a 39-year-old female with a history of Crohn's colitis and ileitis diagnosed in 1993 at the age of 25 years, and she was treated with repeated courses of steroids.

In 1995, azathioprine was added when the patient developed Addison's disease, and this is due to the steroids, of course, or withdrawal of steroids. In 1998, the patient had repeated flares. She was treated with steroids, with antibiotics, and the azathioprine was increased to 200 mg/day.

In January 1999, she was started on infliximab 5 mg/kg with an excellent response. From 1999 to 2004, she got infliximab episodically for flares. The intervals between infusions ranged between 12 and 20 weeks, and she did not have any important problems with intolerance or with infusion reactions. In June 2005, the patient had a healthy daughter by cesarean section.

In January 2006, the patient had relapse and was restarted on infliximab. In February of 2006, the patient suffered the infusion reaction despite prophylaxis with hydrocortisone, and in April 2006, she suffered a new severe infusion reaction despite maximal prophylaxis with slow infusion rate, and all the measures you take for avoiding reactions. She had symptomatic relief for only 2 weeks. She was started on prednisone 40 mg/day.

In May 2006, she underwent a colonoscopy, and this revealed left-sided colitis, spared transverse colon, and also in the right colon and also in the terminal ileum. Maybe I could ask the panel how they would deal with this patient at the present time.

PAUL J. RUTGEERTS, MD, PhD, FRCP, AGAF: Do you think there is still a possibility to rescue her with the infliximab, Bill?

WILLIAM J. SANDBORN, MD: It might be possible, but I am not interested in bothering. She has already had all this episodic dosing; she has had an infusion reaction. If you had one agent, you would want to squeeze the last drop of juice out of the lemon, or the orange, shall we say, but not now; I want to be done with all the problems.

PAUL J. RUTGEERTS, MD, PhD, FRCP, AGAF: Brian, what is your feeling?

BRIAN G. FEAGAN, MD: I think it is probably likely that she is going to require switching. I think it illustrates, the case illustrates the problem that even continuous therapy or continuous prophylaxis with an antimetabolite is not uniformly effective. In this meeting, we have seen data with certolizumab with intermittent therapy, and you would anticipate that about 12% of patients over 1 year would be sensitized. And so, I think this is a common scenario with intermittent therapy. It is not as common as intermittent therapy without an antimetabolite, but it is still occurs.

PAUL J. RUTGEERTS, MD, PhD, FRCP, AGAF: In May 2006, she was started on adalimumab 160 mg at day 0 and followed by 80 mg at week 2, and then 40 mg every other week. She had an immediate good response allowing tapering of the steroids, and in October 2006, we increased the dose to 40 mg weekly because she had a slight increase in symptoms. In November 2006, we carried out a repeat colonoscopy, which showed improvement of colitis, especially in the right colon but not a complete healing.

In January 2007, we deescalated adalimumab again to 40 mg every other week, and since then and up to now in May, early May 2007, the patient is completely symptom-free on adalimumab 40 mg every other week.

I want to share the endoscopic data with you; this is the right column prior to the start of adalimumab, so that was after discontinuation of infliximab. You see this is, I do not exactly know how long after it was, 4 or 5 months, and you see that the colon has much improved, although there are still a few small ulcers present, but there is a tremendous improvement of the colonic lesions with the treatment with adalimumab.

My question would be, Bill, to you first maybe: if you increase your dose to every week, are you sometimes deescalating again to every other week?

WILLIAM J. SANDBORN, MD: I will answer, and then I am anxious to hear your rationale because I always seem to be one step behind your clever ideas, and I would not have done this. The way that I usually do this, if the 160 mg, 80 mg at week 0 and 2 gives a blood level that you would see a steady state after 12 weeks or so with 40-mg weekly dosing. The way I interpret the original presentation, you treated the patient with 160 mg, 80 mg, they did very well, and then it looked like about 3 months later, there was a return of clinical symptoms.

You would taper the steroids, and by that point, the loading dose would have worn off as you go to 40-mg weekly dosing. I would have interpreted that as the patient responded to the equivalent of 40-mg weekly dosing, relapsed when your dose deescalated to 40 mg every other week, and you sort of see that all equilibrate by about 3 months, and then you dose escalated and the patient responded. I guess I would not have done that experiment again, so I am anxious to hear why did it, and I am intrigued that it seemed to work.

PAUL J. RUTGEERTS, MD, PhD, FRCP, AGAF: I think we are trying to have a sort of algorithm now where we keep our patients on an escalation dose every week for 3 months, and then we try to deescalate again. Our reasoning is that probably you will restore the trough levels of the drug, and then when you have good trough levels, you might be able to decrease. We will see whether it substantiates in the patients, in all the patients we do this.

WILLIAM J. SANDBORN, MD: It is really very interesting. Early on, before we sort of had the hang of all of this, we were loading with 80 mg/40 mg; this was several years ago. And then, if the patient was not better at 4 weeks, I was escalating to 40 mg. As all the trials came in, it became clear that I was dose escalating too early, and I did deescalate a number of those patients and it was fine.

But for the audience, I have been following Paul far behind for a long time and every time I come to Digestive Disease Week (DDW), he says something that I initially think is crazy and I finally learned after a few years, pay attention to this guy because 2 years later, he comes with a definitive study, and most of the time he is right. I do not know why this is, but...

PAUL J. RUTGEERTS, MD, PhD, FRCP, AGAF: Soon I will say something really crazy.

WILLIAM J. SANDBORN, MD: I will bet you that he is going to have a controlled trial in a year and it is going to be right.

BRIAN G. FEAGAN, MD: I think you should expand the rationale of why you think that...

PAUL J. RUTGEERTS, MD, PhD, FRCP, AGAF: The rationale.

BRIAN G. FEAGAN, MD: ...because, I mean, in the range of doses we are talking about, I do not think it is a safety issue, so is it a cost issue?

PAUL J. RUTGEERTS, MD, PhD, FRCP, AGAF: I think what we observe is that when we dose escalate with infliximab and we go to 10 mg/kg, and we do this 3 times in a row, we are mostly able to go back to 5 mg/kg every 8 weeks without any problem. I think we try to do this, and as I know, I have heard, I think that Humira is not, or adalimumab is not approved for weekly therapy in the United States.

MALE VOICE: I mean, that is not a major barrier often, but Maria, is there an immunologic reason why this is?

MARIA T. ABREU, MD: I cannot think of an immunologic reason, but really there is no problem in Minnesota getting it every week? I have actually found that it is a hurdle and like Paul, ...I think the issue is that I am at the point now where I have had to dose escalate to 40 mg every week, largely in patients who have already gone through infliximab, the more refractory patients where clearly Paul has shown us already that patients that have previously been exposed to infliximab are less likely to do well anyway regardless of what you do. I have found that patients are slipping before their next dose 2 weeks later. I actually have not, except for 2 people I can think of, dose deescalated them, and I really do not know yet whether that is an effective strategy. I wonder how many iterations of that do you have to go through just to decide that they need to be on every week, and of course, it is a bit of a battle with the insurance company.

WILLIAM J. SANDBORN, MD: Paul, just to, as I am thinking about this, actually, there is an exactly analogous situation with certolizumab. There was a poster that I think you and I and Brian were all on today where patients that had responded to certolizumab and stopped responding, when they got a single extra dose, they were getting 400 mg every 4 weeks, and they got 1 extra dose in between the 2, and at 6 months later, or even a year, you had about 30% to 40% of those patients recaptured.

BRIAN G. FEAGAN, MD: Yes.

WILLIAM J. SANDBORN, MD: It is essentially the same experiment, and I do not understand it.

BRIAN G. FEAGAN, MD: Bill, I think we have got to be careful. These are not controlled experiments.

PAUL J. RUTGEERTS, MD, PhD, FRCP, AGAF: But it is something you need to investigate because we have also patients, for instance, who fail on weekly therapy, and we have in some of these patients reinduced and we have restored the effect and went back to a weekly dose. I think everything is a question of trough levels, and some patients have a need for a trough level, and when you achieve the trough level, they will respond and keep responding. I would like to have your opinion also on maybe the mechanism, and Maria can maybe talk about that.

Why would patients who are failing an anti-TNF agent be also more resistant to biologics targeting another pathway? What would be the possibility there?

MARIA T. ABREU, MD: I think there are a couple. One possibility is that these patients are going to be patients who have had disease for longer, so things have to be corrected for length of having had their Crohn's disease. The second is what is it that you are actually treating? Is it that these are patients that do not have inflammation? You have to go back to the list that Paul put up of all the reasons for, in a way, a primary nonresponder.

One of the things that I think perhaps has contaminated some of the clinical trials of the second anti-TNFs is patients being enrolled into the clinical trial that essentially would be what we would feel are primary nonresponders, and why did the doctor do it? Because these people are desperate. You want to do something for a sick person, and you want to convince yourself, "aren't you sure you had one less bowel movement when we gave you infliximab the first time," and as long as they met even a low bar, you could not enroll them into the clinical trial.

Then the final and sort of more intriguing reason is, is there an escape pathway that can develop in certain patients in whom you have tonically inhibited TNF-alpha? I think that I find that the most intriguing explanation will be the hardest to get at, but I think that is probably going to be patients that have adequate trough levels of infliximab, yet are losing their response. Maybe there needs to be a reset of what a trough level needs to be in that more refractory group of patients, and maybe that is what we should be looking at. Maybe there is, for naive, the trough levels can be down here; for people who are really more refractory, the trough levels need to be up there.

WILLIAM J. SANDBORN, MD: I think right now, there is nothing else to do but the class, although if the patient has not had methotrexate, and there are quite a few patients that have not, I think that actually switching away from biologics and trying methotrexate is a useful strategy. Hopefully, in the near future, there might be the option of the alpha-4 integrin blocker and natalizumab, and I think that will be a good option for primary nonresponders. Then, anti-IL-12 antibodies and abatacept are in clinical trials, and those will be options maybe in the future as well.

PAUL J. RUTGEERTS, MD, PhD, FRCP, AGAF: This completes my case.

WILLIAM J. SANDBORN, MD: We come now to the final, more formal section. We want to hear from Maria Abreu, who will talk to you about using biologics as monotherapy in Crohn's disease, efficacy, and safety.