Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

MEDICAL FRONTIERS - A report from recent peer-reviewed publications and medical conferences

May 2011

The ability of highly targeted biologic agents to halt psoriasis and other inflammatory processes, first achieved with tumour necrosis factor alpha (TNF-a) inhibitors, has been a revolutionary advance for intractable cases. In psoriasis, the risks associated with biologics, which have dissimilar activities in tissues other than the skin, create a challenging apple-to-oranges comparison for safety. It is not enough to list risks; rather, risks must be compared for frequency and potential severity.

Focus on this issue has been renewed with evidence that IL-12/IL-23 inhibitors modestly but measurably increase risk of cardiovascular (CV) events. These concerns about CV are not new. A signal for an increased rate of CV events in active treatment arms of both ustekinumab and briakinumab, which have been the most widely studied IL-12/IL-23 inhibitors in clinical trials, was first detected in phase II trials, rates of major adverse CV events (MACE) were higher, particularly early after initiation of therapy, in those on active treatment vs. placebo. Although the differences were not statistically significant, they were consistent across studies, and no MACE was observed in patients on placebo.

MACE

According to data presented at several international dermatology meetings over the past 18 months, including the 18th European Academy of Dermatology and Venereology Annual Congress in 2009 and the 2011 American Academy of Dermatology (AAD) meeting, MACE—which includes myocardial infarction (MI), stroke and sudden cardiac death—has occurred in rates ranging from 0.44 to 1.44 per 100 patient-years (PYs) of follow-up on ustekinumab. Although there is evidence from a large, prospective, population-controlled study that psoriasis is independently associated with increased CV risk—calculated as a 10% greater rate of MACE in a 60-year-old individual with mild-to-moderate psoriasis (Gelfand et al. JAMA 2006;1735-41)—the effect of ustekinumab was observed above and beyond this effect as both the active and placebo groups in the clinical trials already had psoriasis.

Dr. Craig Leonardi, Clinical Assistant Professor of Dermatology, St. Louis University, Missouri, characterized the persistent increase in MACE in the clinical trials as an “imbalance.” As a co-investigator in ustekinumab trials, Dr. Leonardi, who expressed these concerns at the 2011 Skin Disease Education Foundation conference, suggested that the increased MACE risk does not preclude appropriate use of this agent, but it emphasizes that all of the effects of IL-12/IL-23 inhibitors, the newest biologics in psoriasis, are not yet well understood.

Some of the strongest corroborating evidence that IL-12/IL-23 inhibitors alter CV function and increase risk of events comes from the similarity of risk observed in independent placebo-controlled trials with briakinumab, which has now been withdrawn from clinical investigation. In pooled data, the MACE rate in the placebo-controlled period was 1.33 per 100 PYs for briakinumab vs. 1.23 for ustekinumab. The 95% confidence intervals of approximately 0.5 to 3.0 were also nearly identical for the 2 agents. Moreover, the MACE rate in the pooled placebo arms used for comparison were near zero in all cases (Figure 1).

Figure 1.

In a meta-analysis of phase II and III ustekinumab data presented at the 2011 AAD, Dr. Kenneth B. Gordon, Clinical Associate Professor of Dermatology, University of Chicago, Illinois, also reported a signal of CV events, particularly early after therapy onset. He also noted that the difference for MACE in the group receiving the anti-IL-12/IL-23 therapy did not reach statistical significance, but he emphasized that the studies were not powered to show significance for uncommon events. Importantly, the magnitude of events, if confirmed in larger patient experience appears to be at least as great as that of the uncommon events associated with TNF-a inhibitors such as adalimumab or infliximab.

When compared to the clinical data collected so far with IL-12/IL-23 inhibitors, the absence of any comparable signal for increased CV risk on anti-TNF-a inhibitors is striking because of much larger experience and greater ability to detect even rare events. In a recent publication comparing rates of both major and minor adverse events between available TNF-a inhibitors (Langley et al. Br J Dermatol 2010;162:1349-58) (Table 1), even though it was not part of the study design, no increase in MACE risk was observed for adalimumab, infliximab or etanercept. There was a very slight increase in the risk of congestive heart failure (CHF), which was 0.08 events/100 PYs for adalimumab, 0.12 for etanercept, and not evaluated for infliximab. However, CHF is not mechanistically related to MACE and rates were a fraction of those observed for MACE on
agents.

Table 1.

<img4866|center>

Uncommon Events

The relative risk of uncommon events is of critical importance when considering biologics for inflammatory diseases such as psoriasis because many of the events associated with TNF-a inhibitors that have received the most attention, such as tuberculosis (TB), were not recognized until very large numbers of patients were exposed. Specifically, the comparative pooled data with TNF-a inhibitors suggest TB occurs at a rate of approximately 0.2/100 PYs in patients receiving either adalimumab or infliximab, which is an incidence far below the MACE rates observed to date in patients receiving an anti-IL-12/IL-23 inhibitor with a much more limited follow-up.

Of the serious but uncommon events associated with TNF-a inhibitors, only serious infections and non-melanoma skin cancer occur at a rate comparable to the MACE rates in the anti- IL-12/IL-23 trials. In the Langley paper, the incidence rates per 100 PYs of exposure for serious infections was 1.5 for adalimumab, 1.8 for infliximab and 1.4 for etanercept. The rates of non-melanoma skin cancer were 0.7, 1.68 and 1.02, respectively (Figure 2). Other risks, such as demyelination syndromes or lupus-like diseases, which have also only been detected because of several th
ve exposures, have permitted rare events to emerge.

Figure 2.

<img4869|center>

The risk of considering rare events out of context is the potential subjectivity of emphasizing 1 type of adverse event over others. Biologics alter fundamental molecular activities to achieve their potent anti-inflammatory effects, but not all of the activities of the biologic targets or their effect on downstream molecular events can be understood without a large and prolonged clinical experience. Relative to IL-12/ IL-23 inhibitors, the several thousand PYs of cumulative patient exposure with TNF-a inhibitors across multiple indications makes their safety easier to gauge. Many experts addressing relative risks of TNF-a and IL-12/ IL-23 inhibitors
he essential problem for comparing these 2 classes is that data for the latter remain incomplete.

Figure 3.

<img4870|center>

Even once the relative risks of biologic options are well established, the comparison of different serious events must be conducted cautiously. In addition to the relative incidence of uncommon but serious side effects, there is a need to consider their relative severity. Neither patients nor physicians may weigh a 1.0/100 PY incidence of a significant cancer the same as a 1.0/100 PY incidence of serious infection, which may be more readily identifiable with surveillance and more readily controlled. MACE, which may include irreversible cardiac damage from MI or even sudden cardiac death, is particularly problematic because of its sudden onset and potential for long-term sequelae.

Measuring Relative Efficacy

It is also important to recognize that biologics have been pursued for difficult-to-treat diseases for which patients have accepted a low risk of serious events in order to achieve relief unattainable by other means. As a result, relative efficacy remains an important part of treatment selection within the context of relative risks. In the placebo-controlled phase III PHOENIX 1 trial, 66% of patients on the highest dose of ustekinumab (90 mg) achieved a psoriasis area severity index (PASI) score of 75 at week 12 (Leonardi et al. Lancet 2008;371;1665-74). The rate for this end point was 80% in the similarly designed CHAMPION study on adalimumab (Saurat et al. Br J Dermatol 2008;158:558-66).

More importantly for the patient, PASI 100, which defines total skin clearance, was achieved in 35% at 52 weeks among those randomized to adalimumab in the REVEAL study (Menter et al. J Am Acad Dermatol 2008;58: 106-15). This level of benefit has now been sustained for 3 years of follow-up in the REVEAL population. At 160 weeks, 31% on those on therapy maintained a PASI 100 response (Gordon et al. Abstract 3367, AAD 2010 Winter Meeting). There are no PASI 100 data for ustekinumab at 52 weeks or for longer periods because the trials with the longest follow-up employed re-randomization schemes at intervals during over the course of the trials that prevent efforts to evaluate long-term response on uninterrupted therapy.

True rates of relative efficacy can be evaluated only in blinded and randomized studies, but it is critical to recognize that biologics are complex agents that require long follow-up and a determination of benefit that is carefully evaluated in the context of safety. The development of anti-IL-12/IL-23 agents is a potentially important addition to pharmacologic options for psoriasis and perhaps other inflammatory diseases. But several experts, including Dr. Leonardi, currently believe that they may not be appropriate for first-line therapy when other biologics are available. It is still unclear whether the relative safety of these agents is similar, better or not as good as that of TNF-a inhibitors. Most importantly, the relative risks of these agents are likely to be different from that of TNF-a inhibitors, making it important that clinicians and patients weigh options with a command of relative rates of these uncommon events and understand the severity and reversibility of these potential consequences if they occur.

Summary

The recent introduction of an anti-IL-12/IL-23 agent for the treatment of psoriasis has created a need to review relative risks of biologic agents and to develop objective methodology for weighing their clinical importance. A signal for increased CV risk from IL-12/IL-23 inhibitors does not preclude an important clinical role in patient management, but it does emphasize that long-term follow-up is needed to understand the full impact of biologics, which are known to have numerous direct and indirect effects on molecular signalling outside of their desired effect on the control of inflammatory processes. Long-term experience with TNF-a inhibitors has demonstrated that uncommon events may not be identified until relatively large numbers of patients have been exposed. However, their substantial benefit typically far outweighs relatively modest risks. The relative role of IL-12/IL-23 inhibitors may be better understood as their application in second-line therapy provides a larger experience.