The research road to streamlining the model of psoriasis pathogenesis.
Dr. KruegerDermatologists' appreciation of the central role that the interleukin (IL)-23/Th17 pathway plays in psoriasis has developed gradually, through research and serendipity, according to James Krueger, M.D., Ph.D., who spoke on the topic at the MauiDerm 2017 meeting.
"When I started researching psoriasis in the early 1990s, there was considerable debate about pathogenesis. But the dominant hypothesis was that keratinocytes were growing autonomously by overproduction of growth factors (transforming growth factor alpha) that would interact with overactive EGF receptors, producing a proliferative reaction." In this hypothesis, "A few immune cells came along for the ride," Dr. Krueger explained. He is D. Martin Carter Professor in Clinical Investigation at Rockefeller University.
Based on biopsies, "It's clear that psoriasis represents a big change in biology from background skin. There's a tremendous epidermal thickening reaction, on a bed of mononuclear inflammatory cells in the infiltrate." Immunohistochemical (Ki67) staining of hyperkeratotic skin invariably shows that virtually every basal cell is in cycle, versus very few basal cells in background skin. This growth activation is also associated with incomplete differentiation - this is a wound-healing program called regenerative maturation."
The second invariable feature in psoriasis is a large infiltrate of T cells - mostly CD4+ in the dermis, and CD8+ in the epidermis, Dr. Krueger says. Consistent overexpression of T cells led immunologists to theorize that psoriasis must involve an inductive reaction provoked by T cells with abundant high-affinity IL-2 receptors, he says.
The first model
"The experiment that started the current biologics revolution involved 10 patients at Rockefeller University whom we gave a fusion toxin called DAB389IL-2 (Ontak). This molecule binds to high affinity IL-2 receptors and when internalized into cells, releases enzymatic fragments that inhibit protein synthesis and lead to apoptotic cell death. Keratinocytes have no IL-2 receptors; therefore, they remain untouched."
After two months of repetitive intravenous injections, a subset of patients had clear skin.1 Biopsies, immunohistochemical stains and other analyses showed that in the patients who experienced major reductions of T cells in lesional skin,
“All of the pathological defining features of psoriasis turned off. This set up the first T-cell-centered pathogenic model of psoriasis,” Dr. Krueger says.
This model fueled the first large-scale trials of biologics in psoriasis, starting with abatacept (Orencia, Bristol-Myers Squibb), alefacept (Amevive, Astellas Pharma), and efalizumab (Raptiva, Genentech). The latter two earned U.S. Food and Drug Administration (FDA) approval in 2003. All three drugs could reverse disease-defining features, he says, "But Psoriasis Area and Severity Index (PASI) response rates were less than 50%. We had no idea about the characteristics of T cells, the pathogenic T-cell subsets, associated cytokines or cellular and molecular drivers" that drove pathogenic immunity.
Dr. Krueger says that the T-cell model omitted a key element of immune reactions: dendritic antigen-presenting cells.
"We know today that to activate a CD8+ or CD4+ T cell, that T cell needs to see antigens (intracellular and extracellular, respectively). Dendritic cells are also a highly abundant part of the psoriasis lesion. There's a huge increase, not only in the dermis, but some also invade the epidermis."
Dendritic cells in psoriasis lesions are marked by increased expression of TNF and the enzyme iNOS, which makes nitrous oxide.
"We called them TNF/iNOS-producing (TIP) dendritic cells, in parallel to a dendritic cell that had been discovered a year before in mice."2 These dendritic cells help eliminate certain types of bacteria, he says.
"As we continued to study the TIP dendritic cell, we found many other markers that distinguish it from background dendritic cells. Some were highly inflammatory secreted molecules including S100A12, IL-20 and most importantly IL-23."
IL-23 is a dimeric cytokine related to IL-12, Dr. Krueger says. Both possess a common p40 subunit but IL-23 has a unique p19 protein subunit that allows it to bind the IL-23 receptor. This stimulates a T-cell response that drives production of IL-17, a cytokine used to define a Th17 T cell.
"In contrast, binding the IL-12 receptor mainly induces a Th1 or interferon gamma (IFN γ)-dominated response," he says.
This finding inspired the idea that perhaps multiple different T-cell subsets, called polar T-cell subsets, were associated with inflammatory diseases, Dr. Krueger says.
"For a long time, we talked about inflammatory skin disease as being either Th1 or Th2 dominated, where Th2 cells make IL-4, IL-5 and IL-13 and are likely stimulated upstream by thymic stromal lymphopoietin (TSLP)." IL-23 drives Th17 cells, he says.
Connecting the dots
The first suggestion that psoriatic lesions express IL-23 came in 2004, when Dr. Krueger and colleagues observed that the p40 subunit associated with IL-12 was upregulated in psoriasis, but the other subunit, p35, was not. However, another subunit that associates with p40 is the p19 subunit of IL-23, and this was highly upregulated in psoriasis lesions, similar to the p40 subunit. This finding established that IL-23 was highly upregulated in psoriasis lesions, he says.
"Then we traced IL-23 synthesis back to the TIP dendritic cells. Genome-wide association studies showed that the IL-23 receptor, the p40 subunit and the p19 gene had risk alleles for psoriasis, suggesting that the IL-23 pathway was very important."3
An early IL-17 antibody allowed Dr. Krueger and colleagues to show elevated Th17 T cells in psoriatic plaques. Messenger RNA measurements during cyclosporine treatment showed increases in production of IFN γ, IL-17 and IL-22. This suggested that psoriasis involves coactivation of Th1, Th17 and Th22 T cell subsets, he says.
By 2010, researchers had identified more than 4,000 dysregulated genes in psoriasis, Dr. Krueger says.
"All along, we had seen that interferon-stimulated pathways were highly abundant in psoriasis lesions. But we also knew that interferon pathways are highly abundant in atopic dermatitis (AD) and other inflammatory diseases. So we needed a somewhat different map to point out what is special about psoriasis," he explains.
Subsequent research showed that genes such as S100A8/9 (calgranulins) and S100A7 (psoriasin) were much more highly expressed in psoriasis than in AD. Immunohistochemistry showed that these proteins are highly abundant in keratinocytes.
"How does one get these proteins in keratinocytes? The first insight into Th17-type cytokines came with a simple experiment of adding IL-22 to human keratinocytes in a culture dish," he says.
The result was high-level production of psoriasin and calgranulin.4
"These two molecules were not induced by IFN γ. We began to see a molecular fingerprint of a psoriasis-associated Th17-type cytokine," he says.
Ultimately, "We put all this information together to say that there is a pathway from the TIP dendritic cell through IL-23 to the activation of type 17 T cells and production of IL-17 and IL-22, and then keratinocyte responses which are unique and have features of psoriasis.5"
But even five years ago, researchers believed that immune activation involved activation of three parallel immune axes:
· TIP DCs, making IL-12 and IL-23
· Th1, producing IFN γ
· Th17 and Th22
"The idea was that each of these immune cytokines has receptors on the keratinocytes and activates different transcription factor pathways, and the output is a series of inflammatory molecules that can feedback and amplify this reaction," he explains.
Testing the theory
The first test of this theory came with ustekinumab (Stelara, Janssen). In phase 3 trials, approximately 70% of ustekinumab-treated patients reached PASI 75.6,7 Dr. Krueger says that as expected, blocking IL-12 reduced IFN γ, and blocking IL-23 reduced Th17.
"IL-22 also decreases," he notes.
At a cellular and molecular level, "What starts as a typical psoriatic lesion that expresses keratin 16, by week 12 looks like the background non-lesional skin. Keratin 16 is turned off, and infiltrating T cells and dendritic cells are reduced to background levels," he explains.
In the process, he adds, blocking IL-12 and IL-23 ostensibly normalizes nearly 3,000 genes.
Elucidating the relative contributions of type 1, type 22 and type 17 T cells began by targeting interferon, Dr. Krueger says. He adds that he and his co-authors were "absolutely convinced" that giving patients massive doses of an IFN γ antibody (fontolizumab) would shut down psoriasis. But only one of 10 patients showed significant improvement.8 Unpublished studies showed that targeting IL-22 with antibodies also failed.
"In 2010, we really figured out what was making psoriasis tick." A phase 1 study first presented that year showed that extremely high intravenous doses of brodalimumab (Siliq, Valeant) produced clear skin or minimal disease in eight of eight patients in just six weeks, establishing IL-17 as a prime driver of psoriasis.9 Seven patients reached PASI 75, Dr. Krueger says, and biopsies revealed that within two weeks of starting therapy, "Keratin 16 and cell proliferation were turned off. Within six weeks, tissue structure normalized."
Genetic expression profiling showed that within two weeks, brodalimumab reduced the 10,000 gene (lesional) psoriasis transcriptome to 500 genes.
In developing ixekizumab (Taltz, Eli Lilly), which targets the IL-17 receptor, Dr. Krueger and colleagues used a variety of IL-17-induced genes (responsible for inflammatory and antimicrobial proteins and keratinocytes) to gauge how much antibody was needed to block the development of psoriasis. In a six-week, ascending-dose study versus placebo, "We found that for four different IL-17 sensors, 50 mg of antibody consistently turned off all these IL-17-induced products.10 At 150 mg, there was a faster effect, in two weeks. The surprise was that everybody who got this high dose (150 mg) of antibody maintained a PASI 75 response. That showed that IL-17 was much more important than we had heretofore appreciated."
Slightly upstream, he says, "IL-23 is a master regulator of the type 17 T-cell response. So the question is, can one target IL-23 in addition to IL-17 to approach psoriasis?"
Phase 2 studies of guselkumab (Janssen) and risankizumab (ABBV-066, Abbvie) showed PASI 75 responses in more than 90% of patients,11,12 with similar phase 3 results for guselkumab.
"With some of the agents, the PASI 100 rates have been remarkable: more than 60% of risankizumab-treated patients attained complete clearing" by week 2011, Dr. Krueger notes.
These findings streamline the model of psoriasis pathogenesis, in that IL-23/Th17 appears to be the key axis, Dr. Krueger says. However, he added, the model fails to explain epidermal hyperplasia as a direct IL-17 response, as well as the recruitment of other polar T-cell subsets and the persistence of T-cell activation.
Researchers have found protein antigens (LL37 and ADAMTSL5) that very likely serve as autoantigens in psoriasis,13 Dr. Krueger says.
"They have both been shown to activate T cells to produce IL-17. They're expressed in psoriasis lesions and upregulated directly or indirectly by IL-17. We now have what might be called a waterfall plot of disease," he says.
It starts with activation of Th17, which he says produces cytokines beyond IL-17, including the interferon-like cytokines IL-26 and IL-29.
"We have a keratinocyte that has very abundant IL-17 receptors, and a huge outpouring of different inflammatory regulators in response to IL-17. One can explain every feature that we use to classify psoriasis under a microscope by some pathway that is stimulated initially by a Th17 T-cell cytokine. With this we can get chronicity, and autoantigens provide the ability to perpetuate the disease," he explains.
Immunohistochemically, Dr. Krueger says the high expression of psoriasis stems from the fact that IL-17 signals through the IL-17 receptor, which ultimately activates the transcription factors C/EBP β or δ.14 While all keratinocytes have IL-17 receptors, he adds, unpublished research reveals that the high concentration of C/EBP β or δ resides within granular layer keratinocytes. Gene profiling shows that suprabasal cells have much more C/EBP β than do basal keratinocytes, he added.
Staining of lesional skin for C/EBP reveals a heavy concentration, mainly in granular keratinocytes, Dr. Krueger says.
"The upper spinous and granular layer has very high expression of this transcription factor – this is exactly where we see high-level expression of molecules that are induced by IL-17 in psoriasis, such as beta defensin and lipocalin,” he says. “What we get at the end of this is psoriasis, with incredibly high expression of these antimicrobial peptides in granular and cornified layers of the epidermis. Psoriasis can be viewed as chronic activation of what is a natural immune response pathway that is geared to eliminate Candida albicans and possibly other extracellular pathogens, because we know IL-17 deficiency can produce cutaneous candidiasis. However, in psoriasis, an autoantigen - not a skin infection - drives the reactions."
Disclosures: Dr. Krueger has received research support, consulting or lecture fees from most pharmaceutical and biotechnology companies with a product or investigational agent for psoriasis. However, he has no patents or ownership of and receives no financial gain from any psoriasis drug or product.
1. Gottlieb SL, Gilleaudeau P, Johnson R, et al. Response of psoriasis to a lymphocyte-selective toxin (DAB389IL-2) suggests a primary immune, but not keratinocyte, pathogenic basis. Nat Med. 1995;1(5):442-7.
2. Serbina, NV, Salazar-Mather TP, Biron CA, Kuziel WA, Pamer EG. TNF/iNOS-producing dendritic cells mediate innate immune defense against bacterial infection. Immunity. 2003;19:59-70.
4. Wolk K, Witte E, Wallace E, et al. IL-22 regulates the expression of genes responsible for antimicrobial defense, cellular differentiation, and mobility in keratinocytes: a potential role in psoriasis. Eur J Immunol. 2006;36(5):1309-23.
5. Nestle FO, Kaplan DH, Barker J. Psoriasis. N Engl J Med. 2009;361(5):496-509.
6. Leonardi CL, Kimball AB, Papp KA, et al. Efficacy and safety of ustekinumab, a human interleukin-12/23 monoclonal antibody, in patients with psoriasis: 76-week results from a randomised, double-blind, placebo-controlled trial (PHOENIX 1). Lancet. 2008;371(9625):1665-74.
7. Papp KA, Langley RG, Lebwohl M, et al. Efficacy and safety of ustekinumab, a human interleukin-12/23 monoclonal antibody, in patients with psoriasis: 52-week results from a randomised, double-blind, placebo-controlled trial (PHOENIX 2).Lancet. 2008;371(9625):1675-84.
9. Papp KA, Reid C, Foley P, et al. Anti-IL-17 receptor antibody AMG 827 leads to rapid clinical response in subjects with moderate to severe psoriasis: results from a phase I, randomized, placebo-controlled trial.J Invest Dermatol. 2012;132(10):2466-9.
11. Callis-Duffin K, Gordon K, Wasfi Y, Shen YK. A phase 2 multicenter, randomized, placebo- and active-comparator-controlled, dose-ranging trial to evaluate guselkumab for the treatment of patients with moderate to severe plaque-type psoriasis (X-PLORE). Poster P8353. American Academy of Dermatology 72nd Annual Meeting. March 21-25, 2014. Denver.
12. Papp KA. Session F010. American Academy of Dermatology 73rd Annual Meeting. March 20-24, 2015. San Francisco.
13. Kim J, Krueger JG. Highly effective new treatments for psoriasis target the IL-23/Type 17 T cell autoimmune axis. Ann Rev Med. 2017;68:255-269.
14. Gaffen SL. Structure and signalling in the IL-17 receptor family.Nat Rev Immunol. 2009;9(8): 556-67.