Safety of biologics for the treatment of psoriasis

Safety of biologics for the treatment of psoriasis


Hello dear colleagues, my name is Professor
Kristian Reich. I’m a dermatologist and immunologist practicing
in Hamburg, Germany. The topic I want to discuss with you today
is the safety of biologic therapies in the treatment of psoriasis. The targeted therapies, the biologics, as
you know, have revolutionized the way we can manage patients with psoriasis. It has been one of the most amazing developments
in the last 15 to 20 years. We have seen several generations of biologics
being developed for psoriasis, including the TNF antagonists (and there are several of
them); including ustekinumab, which theoretically blocks IL-12 and IL-23; and, more recently,
the IL-17 inhibitors; and many of you know that antibodies that specifically block IL-23
are at the horizon. So we have a number of different classes of
targeted therapies for the treatment of psoriasis. Of course, due to the fact that they have
been around for many, many years already, the amount of safety data we have for the
TNF antagonists—the body of safety evidence if you will—is larger than for ustekinumab
and of course larger than for the most recent class, the anti-IL-17A blockers. Now having said this, additional safety data
for the TNF antagonists comes from their use in other indications, including rheumatological
conditions, but also inflammatory bowel disease. Before I jump into the subject, let me make
it clear that safety data comes from different sources, and whenever you look at safety data,
I would advise you to ask the question, “What is the source of the safety data I’m looking
at?” because this has implications for the interpretation of the data. Number one, of course, we do have safety data
from the clinical trials. The safety data is usually high quality but,
and this is a limitation, is not really real-world data due to the inclusion and exclusion criteria
that are active and applied in clinical trials. In rheumatoid arthritis, there are registries
running already for several years. It has been estimated that up to 60–80%
of the patients enrolled into the registry, so meaning real-world patients could not have
been participating in a clinical trial in RA So just an anecdotal flashlight on how big
the problem of patient selection could actually be Another source is so-called post-marketing
surveillance data. These of course are real-world patients, real-world
safety data, but the quality is usually so low that interpretation is made very difficult. The best of both worlds are the registries,
where you have real-world patients, where for example, in Germany and in other countries,
every patient started on systemic therapy at a site, participating in the registry,
is actually very carefully assessed, is documented almost identically to the documentation in
a clinical trial. So you have the real-world patient, but you
also have the high quality. This is why we are specifically interested
to see safety data from the registries. There are two recent publications I want to
talk about, and they actually come from two different sources. Number one: there is a recent publication
in the Journal of the American Academy of Dermatology on the long-term safety of secukinumab,
a member of the most recent class of targeted therapies, a blocker of IL-17A. This is clinical trial data, so we have to
consider the limitations coming from patient inclusion/exclusion, and the way long-term
safety data is usually given is event rates per 100 patient-years of exposure. If you have a “one”, this means if you
treat 100 patients for 1 year, you see one event—I think this is important to understand. Why is this important to understand? Because during long-term clinical trials,
we normally have no longer a comparison—the placebo arm usually stops at Week 12 or Week
16. So you depend in your comparison on comparing
it to, when it comes to clinical trials, to safety data generated with other drugs in
other clinical trials. In registries, because registries include
all systemic therapies, you can more easily compare safety data across different treatments. Because clinical trials have been running
all ready for many, many years in psoriasis, we actually have a plethora of safety data
for all the targeted therapies and secukinumab is just the most recent example. On the other hand, we do have a list of what
I would call safety events of interest. We look at adverse events, we look at serious
adverse events, but specifically we look at serious or severe infectious adverse events;
we look at malignancies because the drugs that we use interfere with the immune system;
we look at major adverse cardiovascular events, because our psoriasis patients do have an
increased risk, and we want to see whether this risk is modulated, either in a good way
or in a bad way. When you look at these numbers, we have become
familiar that, for example when it comes to serious infections, when it comes to targeted
therapies, usually you see event rates in the range of 1.2 to 1.8 per 100 patient-years
of exposure. The good news is, this is a low absolute number—this
means if you have 100 patients on adalimumab, for example, for a year, you can expect to
see one serious infection. So the absolute numbers we are talking about
are low. Based on previous findings with other targeted
therapies, we haven’t seen a big significant difference between the different TNF antagonists
and, for example, ustekinumab. Usually infliximab tends to be more on the
higher end, ustekinumab and etanercept tend to be more on the lower end. But, as I said, there’s no significant difference. Major adverse cardiovascular events probably
in the range of 0.5 to 0.6 per 100 patient-years of patient exposure, malignancies—without
non-melanoma skin cancer—probably in the range of 0.5 to 0.7. So again, talking low absolute numbers, and
until now no real difference, no significant difference, between the different targeted
therapies. When I now look at the novel publications
coming from the pooled clinical trials done in psoriasis with secukinumab, it’s amazing
to see that the numbers for these events are actually pretty much the same. So my general reading would be that with all
the limitations—different clinical trials maybe having different inclusion/exclusion
criteria, may having enrolled different patient populations—with all this limitation, it
appears to be that there is a certain pattern of the safety events we are interested in,
and that this pattern seems to be very similar between the targeted therapies we have already
been using for many years and the novel IL-17A blocker secukinumab. There is one exception, and this is Candida
infections. This is explainable because IL-17A is known to play a specific role in the defense against Candida. We know that patients that have a mutation
in the IL-17 pathway can develop chronic mucocutaneous candidiasis, so very severe chronic Candida
infections. Of course, we are treating psoriasis patients, so we are treating patients that have an elevated, a largely elevated IL-17A cytokine level, and with giving secukinumab,
we don’t want to block this pathway completely—our hope is that we will be able to normalize
the IL-17A pathway. But looking at the findings from the clinical
trials, there is a Candida signal that we have not been seeing with the other targeted
therapies; and in the clinical trials, secukinumab was compared to etanercept, so we can directly
refer to a TNF antagonist in their own clinical trial setting. The event rate per 100 patient-years of exposure
for the high dose (for the 300 mg once per month) with secukinumab is in the range of
3 to 3.5 per 100 patient-years of exposure while in placebo, in etanercept, it’s 1.-something. So there is a small signal indicating that
patients treated with IL-17A blockers—and this is very likely to be a class effect—do
have a slightly elevated risk for Candida infections. Usually these Candida infections are mild
to moderate; patients do not have to stop the drug. But there are rare cases, for example of a
Candida infection of the esophagus, and I think we all agree that we have very rarely
seen this with other therapies before. So, when a patient comes to you and is telling
you that he has just been released from the hospital because of a Candida esophagitis,
and the patient comes to you with moderate-to-severe psoriasis, this is probably not an ideal candidate
to be treated with an IL-17A blocker. The other paper I want to talk about is coming
from the German Psoriasis Registry PsoBest. So now we are changing sources, now we go
to registry data that was published in the Archives of Dermatological Research some months
ago. This includes several thousand patients being
treated with the different conventional therapies, including methotrexate, including the German
fumaric acid esters, including cyclosporin, and the different targeted therapies without
anti-IL-17A antibodies—because they were not available when the data cut for this safety
analysis were done—but including ustekinumab and including the different TNF antagonists. Now the very good news is that when you go
through the list of the safety events of interest, the event rate per 100 patient-years of exposure
for serious infections, for example …for major adverse cardiovascular events …for
malignancies, is actually in the range or even lower than the event rate we have seen
with clinical trials. So our big fear that the safety profile of
the targeted therapies is overestimated because of the selection going on in clinical trials
does not seem to be true based on this data coming from real-world data, coming from the German Patient Registry. They also compared patients treated with biologic
therapies and patients treated with conventional therapies and there was no significant difference
either, but we all know that there we are not talking about the same patients. In the vast majority of countries, patients
have to go through conventional therapies first, and only then they can be treated with
targeted therapies, so there will be a tendency that the more complicated, the more severely
affected patients, are the ones that end up being treated with a biologic. And in this publication, interestingly, they looked
at comorbidities being present in the different patient populations and they could actually
show that there are more comorbidities present in patients treated with biologics than in
patients treated with conventionals, and there are overall, as expected, more comorbidities
in psoriasis patients than what we have in the healthy German general population. So I’m not so sure whether you can directly
compare safety findings between conventionals and biologics. I see a certain risk, or a certain rationale,
that patients treated in registries with biologics are the more complicated psoriasis patients
and are probably the ones that are more prone to safety events. But having said this, in the safety data from
the German Psoriasis Registry, there were not significantly more safety events in patients
treated with biologics compared to patients treated with conventional therapies. And a last conclusion on the safety coming from
clinical trials and also registries is that the MACE rate—the number of major adverse
cardiovascular events—is actually low; and there is I would say increasing evidence (I’m
using very careful wording here), there is increasing evidence that by bringing the inflammation
down in psoriasis, you may eventually bring the otherwise elevated risk in an untreated
patient with psoriasis of major adverse cardiovascular events down. This will have to be followed up in the future;
there is no published trial that directly looked into this. But if you look at all these safety tables,
if you look at MACE rates in general psoriasis patients, this is the feeling you walk home
with: that there might be a protective effect, especially in those drugs that really, really
bring the inflammation down. So, overall, good news for dermatologists:
the safety data from the clinical trials appear to be reliable and are being validated in
the emerging registry data. We’re talking about a known profile, a known
pattern now, without significant differences between the different targeted therapies. I talked about Candida and IL-17A antagonists
and low absolute numbers. So for me this data is reassuring and showing
that we can really use these drugs in a very safe way. Thank you very much for your attention.

Leave a Reply

Your email address will not be published. Required fields are marked *