Interferon in lupus: A complex pathway for ‘1,000 different diseases’
The possible FDA approval of a drug targeting the interferon pathway in lupus is the culmination of decades of work by clinicians and researchers alike. Yet, given the bewildering complexity of both interferon and lupus, this moment is likely only the midpoint of a much longer narrative.
“It is difficult to talk about the role of interferon in lupus without getting into the weeds,” Timothy B. Niewold, MD, Judith and Stewart Colton Professor of Medicine and Pathology and director, Colton Center for Autoimmunity at NYU School of Medicine, told Healio Rheumatology.
One reason for this is the multiplicity of lupus, which can affect the joints, skin, kidneys, blood cells, brain, heart and lungs. “Lupus can be a thousand different diseases,” Richard A. Furie, MD, chief of the division of rheumatology at Northwell Health, and professor of medicine at the Zucker School of Medicine at Hofstra/Northwell, said in an interview.
The other issue is that there are three types of interferon overall, with type 1 interferon being comprised of five subtypes: alpha, beta, omega, epsilon and gamma. A clear picture of how each type and subtype manifests in all of the aforementioned systems remains elusive.
The clinical implications of all of this at the patient level are clear, according to Niewold. “Despite what we know about the interferon signature in lupus, it seems unlikely that we will just be able to find a drug that works for everybody,” he said.
In the absence of such a silver bullet, experts have had to make do with piecemeal advances occurring over the course of decades. Key studies published in 1979 and 2003 moved the field forward, but experts continue to wrestle with those results today.
“We want to understand what role type 1 interferons play in autoimmunity, tissue damage, joint damage and other relevant complications, but I am not even sure any of us can sum it up very well,” Mary K. Crow, MD, physician-in-chief emerita at the Hospital for Special Surgery and director of the Autoimmunity and Inflammation Research Program at the HSS Research Institute, said in an interview.
Against these odds, the clinical and research communities arrived at anifrolumab (AstraZeneca), a drug that blocks all type 1 interferons. The drug has shown efficacy in skin and joint parameters and, as a result, may garner FDA approval in the near future. More recently, strategies targeting plasmacytoid dendritic cells, which are significant producers of interferon, are emerging.
These advances in interferon biology are extremely timely as research has demonstrated that interferon is playing a central role in COVID-19, according to Healio Rheumatology chief medical editor Leonard H. Calabrese, DO, director of the RJ Fasenmyer Center for Clinical Immunology at the Cleveland Clinic.
“We have to recognize that, biologically, interferon is a primordial defense against viral infections,” he said. “COVID-19 is a disease that links infection and the immune response. So, as it turns out, a lot of what happens in COVID-19 has to do with interferon response.”
Increased understanding of the interferon pathway may improve the lives of patients with lupus, yield information critical to battling COVID and offer hope to patients with other connective tissue diseases. Healio Rheumatology takes a closer look.
Interferon: Then and Now
The story of interferon in lupus goes back to a 1979 paper from experts at the NIH that appeared in TheNew England Journal of Medicine. Furie explained why the paper is still relevant. “They found that 70% of patients with active lupus had elevated levels of interferon,” he said. “What we are finding these days is that 70% of patients have an interferon signature. We have come full circle.”
The 2003 publication of a paper by Baechler and colleagues in the Proceedings of the National Academy of Sciences of the United States of America described an interferon gene expression signature, another leap forward for the field. The signature implicated HLA Class II DRB and DQB alleles, including DRB1*1501/DQB1*0602, and DRB1*0301/DQB1*0201 and early components of the complement cascade, such as C1q and C4. The researchers added that patients with this signature may experience more severe disease with involvement in the kidneys, hematopoietic cells and/or the central nervous system. They concluded that this information may aid clinicians in treating these patients in a more targeted way.
“The burning clinical question, generated around 20 years ago, was whether inhibition of the interferon pathway could reduce disease activity in lupus,” Furie said.
Answering that question has not been easy. Genentech developed rontalizumab, which failed in a phase 2 trial, while MedImmune developed sifalimumab, which made it as far as a successful phase 2 trial, according to Furie. “A shortcoming perhaps was that each of those drugs only bound interferon alpha and left all other type 1 subtypes to signal through the type 1 receptor, known as the Interferon Alpha Receptor or IFNAR,” he said.
“There was a sense that interferon was a good therapeutic target to try,” Crow added. “But those early unsuccessful attempts did not take into account that type 1 is comprised of a whole family of interferons.”
The next logical step was to attack the receptor and block all five subtypes of type 1 interferon. “That is what anifrolumab does,” Furie said. “If you inhibit the type 1 receptor, you get greater interferon inhibition.”
Researchers explored this approach in the TULIP studies. The results exemplified the struggle experts have had in parsing out the role of interferon in lupus and designing successful clinical trials.
Tiptoe Through the TULIPs
The phase 3 TULIP-1 trial, published by Furie and colleagues in Lancet Rheumatology, included 457 patients from 18 countries who were randomly assigned anifrolumab 300 mg (n=180), anifrolumab 150 mg (n=93) or placebo (n=184).
Primary endpoint results indicated no difference between the anifrolumab 300 mg dose and placebo as assessed by Systemic Lupus Erythematosus Responder Index-4 (SRI-4) response. However, the researchers noted that anifrolumab was associated with improvements in key secondary parameters, including reduction in oral corticosteroid use, Cutaneous Lupus Erythematosus Disease Area and Severity Index (CLASI) response and British Isles Lupus Assessment Group (BILAG)–based Composite Lupus Assessment (BICLA) responses.
“Our results suggest that anifrolumab might have the potential to provide a treatment option for patients who have active SLE while receiving standard therapy,” they wrote.
The failure of the study to reach its primary endpoint is par for the course in lupus, according to Crow. “Clinical trial design and clinical outcome measures have been very challenging and have led to a lot of lupus trial failures,” she said.
Undaunted, the experts involved in the TULIP program moved forward. In the phase 3 TULIP-2 trial, published in TheNew England Journal of Medicine, Morand and colleagues used a secondary endpoint from TULIP-1 — BICLA response week 52 — as the primary endpoint. The study included 180 patients treated with anifrolumab and 182 who received placebo. Results indicated a significant improvement over placebo in the active therapy group as assessed by this newly defined primary endpoint (P = .001).
With very robust results from the phase 2 study in SLE, it was only natural to retain a similar study design and endpoints for the phase 3 program. However, TULIP-1 did not meet the primary endpoint, SRI, but results of BICLA demonstrated superiority of anifrolumab to placebo. Before unblinding, the endpoint of TULIP-2 was modified to BICLA from SRI. TULIP-2 was successful not only using BICLA as the primary endpoint but also with the original endpoint, the SRI. “Such results show how fickle and unpredictable lupus clinical trials can be,” Furie said.
The TULIP program has provided some “pretty big steps” for the field, Furie said. “One is that it is the proof of a biologic concept. Another is that we have shown positive results for a targeted therapy in lupus.”
The other key take-home, for Furie, is that “CLASI works, but our global disease and arthritis measures need refinement.” In the meantime, physicians treating lupus wait expectantly for the approval of anifrolumab, and the new therapeutic option it represents. “Drug development in lupus has been a long tunnel,” Furie said. “But the light is starting to shine through. This means better outcomes for our patients with lupus.”
Despite the shortcomings in musculoskeletal outcomes, Crow is optimistic. “There is hope that anifrolumab will be approved by the FDA this coming year, which is tremendously gratifying,” she said.
On a broader level, Crow called on the research community to scrutinize the TULIP findings and continue to refine clinical trial endpoints. “There are different approaches you can take,” she said. “One is to only look at skin, another is to only look at the joints in a given clinical trial.”
The prevailing wisdom at the moment is to incorporate both skin and joint parameters in a single study, according to Crow. Whether this will yield clearer, irrefutable outcomes remains to be seen. What also remains to be seen is how effective anifrolumab will be when used more widely in the clinic. “What role is it going to have in lupus therapy? I do not think we can really answer that yet,” she said. “Lupus is an incredibly complex disease that involves activation of the immune system over time, many antibodies, production of different cytokines and other responses. Taking all the studies together, it is still not clear that blocking the interferon pathway will be sufficient to bring the disease under control.”
With this in mind, the research community has continued to push for new ways of attacking the pathway.
Targeting Plasmacytoid Dendritic Cells
In a key study in the Journal of Clinical Investigation, Furie and colleagues investigated BIIB059, a humanized monoclonal antibody “that binds blood DC antigen 2 (BDCA2), a [plasmacytoid dendritic cell]-specific receptor that inhibits the production of IFN-I and other inflammatory mediators when ligated.”
The randomized, double-blind, placebo-controlled trial involved 12 individuals with SLE and 54 controls. Results indicated that single doses of BIIB059 yielded a favorable safety and pharmacokinetic profile, with BDCA2 internalization on plasmacytoid dendritic cells. This, in turn, correlated with circulating BIIB059 levels and led to “robust target engagement and biological activity,” according to the researchers. They added that this approach may be particularly beneficial for patients with cutaneous manifestations of SLE.
“Plasmacytoid dendritic cells are the most robust producers of interferon,” Crow said. She noted that there is a “small population” of these cells in the blood, but when they are stimulated can make “lots and lots” of interferon.
Many experts, including Crow, believe that the approach of inhibiting plasmacytoid dendritic cells may ultimately be more effective than the mechanism of anifrolumab. “Anifrolumab is blocking the action of type 1 interferon, but whatever is driving production of interferon is still intact,” she said. “It follows that if you move upstream, you do not just block what it is doing, but you block its production. It might be at least an equal effect, but this is speculative at the moment.”
One way for researchers to explore these therapeutic options more effectively is to have biomarkers that more clearly define lupus disease activity.
While many experts in the rheumatology field believe that improved biomarkers will improve treatment outcomes across the board, Niewold lamented that the biomarkers in lupus “have not been great.” He added that the interferon signature has not lived up to expectations as a prediction tool. “Thinking about biomarkers, interferon measurement has been both too inclusive and too general,” he said.
Another issue is that there is uncertainty about which types of interferon are activated, and how they are activated, in the lupus signature, according to Niewold. “It is difficult to tell whether the signal is type 1 or type 2, and we have not been able to resolve that issue,” he said.
The presence of type 2 interferon in the lupus signature “could be leading us astray,” Niewold added. “This group of antiviral genes could be turned on for another reason.”
This means that the research community has a tough road ahead. “If we are really going to get an effective biomarker, we may have to get into that level of detail,” Niewold said.
There will be clinical challenges, as well, according to Niewold. “It is hard to detect circulating cytokines without a laborious assay,” he said.
While none of these endeavors will be easy, Niewold believes they will be worth the effort. “It could lead to more targets and more therapies,” he said, noting toll-like receptors and innate immune receptors as potential avenues of exploration that may result from efforts to clarify biomarkers in lupus.
As researchers dig deeper into the interferon-lupus connection in search of biomarkers and other relevant information, they may get help from an unexpected place: COVID-19.
Link to COVID-19
There are two key questions for the rheumatology community to consider when it comes to the interferon-COVID-19 connection, according to Calabrese. One is what the interferon association with COVID-19 can tell rheumatologists about lupus, and the other is what the interferon association with lupus can illuminate about COVID-19.
“There are two major chapters to this story,” Calabrese said. “The first is that in the last several months, numerous groups have demonstrated that there is a deficit of interferon response in many people who get COVID-19 and experience severe disease courses. Viruses, in general, have developed elaborate defenses against the interferon response and COVID-19 has done this exceptionally well.”
Calabrese suggested that given the role of interferon in defense against viral infections, this “makes a lot of sense evolutionarily.”
The second chapter was illuminated by Bastard and colleagues in a landmark study published in Science showing that around 13% of people with advanced COVID-19 who end up in the ICU or die from the infection make autoantibodies to interferon. “It is impressive because such naturally occurring autoantibodies are extremely rare in healthy individuals.”
On the other hand, the situation in lupus is much different, he noted. “For many years... we have known that a significant number of patients with lupus develop anti-interferon antibodies with with neutralizing capacity,” Calabrese said. “This has opened up a Pandora’s box for rheumatology and the role of interferon in both lupus and COVID-19.”
Given that interferon response is “essential” in fighting viruses like COVID-19, Calabrese raised the question of how patients taking anifrolumab and other interferon-mitigating therapies will fare if they acquire the infection. “This needs to be meticulously examined,” he said.
The next issue is to consider patients with lupus who have anti-interferon antibodies, and whether such patients have an altered disease course with COVID-19. “This has not been systematically examined but preliminary data from small studies of lupus patients in general, in the absence of comorbidities, appear to do relatively well. However, larger and more robust studies are urgently needed,” Calabrese said.
Something else for the rheumatology research community to consider is that some patients with COVID-19 who experience severe symptoms demonstrate a delayed interferon response. Interferon production that is inappropriately timed or over-exuberant can contribute to the hyperinflammation of COVID-19, Calabrese noted. “You need a robust interferon response when you are first infected with COVID, in that first week,” he said. “Experimentally, we know that if the interferon response is delayed, it tends to track with patients who are experiencing hyperinflammation and poor virologic control.”
This is a complex biologic issue however because type 1 interferon is but one cytokine of many found to elevated in the hyperinflammatory stage of COVID-19 along with other mediators, including TNF, IL-6, CKCL10, and gamma interferon. “So far, evidence would suggest that initial response and defense and then viral load dictate how things go with COVID,” Niewold said. “But we still have to answer the question of whether interferon is important in later COVID or how important.”
Calabrese stressed that addressing all of these factors in “an integrated fashion” is critical to understanding not only the role of interferon in COVID, but in lupus and other rheumatologic diseases, as well. “As we put the pieces of the puzzle together, the COVID puzzle piece will be of great value across the board to unraveling lupus.”
That said, measuring interferon or antibodies to interferon is still not part of routine care for COVID patients, according to Calabrese. “We have to be very careful about it, particularly when we are talking about potentially giving unscreened convalescent plasma as a therapeutic approach,” he said, noting that convalescent plasma which could contain anti-interferon antibodies given to a COVID patient could have severe consequences.
The question of the potential utility of an agent such as anifrolumab in hyperinflammatory COVID-19 is not unreasonable given its anti-inflammatory properties and safety profile, noted Calabrese. “In COVID-19, interferon may play a role as either a therapeutic or a culprit, depending on when the patient is being treated,” he said. “As with many other biologic systems, research into interferon biology in COVID-19 is likely to have many positive spin-offs for patients with autoimmune diseases.”
Beyond Lupus and Into the Future
Rheumatologists of all stripes are watching how anifrolumab and other interferon-based therapies play out in lupus. “There is a fair amount of evidence that this whole family of cytokines is potentially relevant in scleroderma, dermatomyositis, Sjogren’s syndrome and other connective tissue diseases,” Crow said.
If anifrolumab gains FDA approval, Niewold believes that it will be used off-label in these and other conditions. “It is worthwhile to think about interferon-based therapies in those diseases,” he said. “Off-label use is not uncommon in rheumatology, but, of course, there are logistical hurdles when using them outside of a clinical trial. But there is something to be said for collecting experience with this approach.”
But this is not the only approach that should be in play, according to Crow. “We have to look at drivers and inducers of interferon, at the receptor it binds to, the signaling pathways downstream and we have to look at what is turned on by those signaling pathways,” she said.
In addition to targeting plasmacytoid dendritic cells, Crow believes that experts in lupus can borrow from other rheumatologic diseases to determine whether the janus kinase (JAK)- signal transducer and activator of transcription (STAT) pathway could bear fruit. “JAK-1 and [tyrosine kinase-2] may also be relevant molecules,” she said.
Niewold underscored this point. “Kinase inhibition made it to prime time in RA,” he said. “It might be achievable in lupus in the near term.”
In the meantime, physicians treating lupus wait expectantly for the approval of anifrolumab, and the new therapeutic option it represents. “This is a long tunnel,” Furie said. “Any progress is good progress.”
- Baechler EC, et al. PNAS. 2003;doi:10.1073/pnas.0337679100
- Bastard P, et al. Science. 2020;doi:10.1126/science.abd4585.
- Furie RA, et al. J Clin Invest. 2019;doi:10.1172/JCI124466.
- Furie RA, et al. Lancet Rheumatol. 2019;doi:10.1016/ S2665-9913(19)30076-1.
- Morand EF, et al. NEJM. 2019;doi:10.1056/NEJMoa1912196.
- For more information:
- Leonard A. Calabrese, DO, can be reached at 9500 Euclid Ave., Cleveland, Ohio 44195; email: email@example.com.
- Mary K. Crow, MD, can be reached at 535 E 70th St., New York, NY 10021; email: firstname.lastname@example.org.
- Richard A. Furie, MD, can be reached at 865 Northern Blvd., Suite 302 Great Neck, NY 11021; email: email@example.com.
- Timothy B. Niewold, MD, can be reached at 435 East 30th St., 5th floor; New York, NY 10016; email: firstname.lastname@example.org.