GENEVESTIGATOR EXAMPLE STUDIES

Identification of new potential indications for Dupilumab

Jana Sponarova, Frank Staubli and Philip Zimmermann
© NEBION AG. December 1, 2016

ABSTRACT

Dupilumab is an antibody therapeutic directed against IL-4R-alpha. It is currently in clinical trials for the treatment of atopic dermatitis and asthma. We compared a published expression signature derived from Dupilumab treated cells with 3,230 curated expression profiles from GENEVESTIGATOR covering a wide range of diseases. Three monoclonal immunosuppressive antibodies - Brodalumab, Etanercept and Infliximab - were identified as having the molecular signatures most similar to the one of Dupilumab. These drugs are known to be effective against various autoimmune disorders. In a reverse approach, we searched for conditions exhibiting the opposite molecular signature to the one of Dupilumab. Not only were we able to confirm the primary indication, but we identified also several other disorders - acne, psoriasis, rosacea, skin transplantation, rheumatoid arthritis and inflammatory bowel diseases in which Dupilumab causes similar signatures. This example shows how GENEVESTIGATOR can effectively compare results from one study/compound with those of many other studies and can help identifying new potential indications for a given drug.

IDENTIFICATION OF CONDITIONS CAUSING A SIMILAR RESPONSE

Dupilumab is a human monoclonal antibody against IL-4 receptor alpha that inhibits both IL-4 and IL-13 signaling. This antibody is currently being tested as a potential therapic agent for atopic dermatitis and asthma. Recently, the first molecular signature of Dupilumab treatment on patients with moderate to severe atopic dermatitis was published (Hamilton JD et al, J Allergy Clin Immunol, 2014). We chose to investigate the gene expression effect of a 4-weeks therapy with 300 mg of Dupilumab / week in lesional skin biopsies relative to untreated lesions. Only the genes with > 2-fold change were selected. The thereby retrieved signature comprised 182 genes of which 53 were up-regulated and 129 were down-regulated (data available upon request) To identify conditions causing a similar molecular signature, we used GENEVESTIGATOR (Hruz et al, 2008) and selected a compendium of 49191 samples profiled on the Affymetrix Human U133 Plus 2.0 platform. We then used the Signature tool from the SIMILARITY SEARCH toolset and searched for the top-50 conditions causing the most similar signatures (Figure 1).



Figure 1. Identification of the top-50 conditions with the most similar molecular signature to 4 weeks of Dupilumab therapy of atopic dermatitis. The signature used for this analysis comprised fold-change values from the comparison of lesioned skin biopsies after vs. before treatment, in a linear scale. The image was cropped for basic illustration of the Signature tool.


Three other antibodies designed for the treatment of inflammatory diseases, namely Brodalumab (directed against the IL-17 receptor), Etanercept and Infliximab (both TNF-alpha inhibitors) were identified among the top conditions causing a similar expression signature for the 182 genes tested (blue arrows indicating their first appearance the list of Figure 1). Brodalumab in particular showed a very high response similarity. Other conditions causing a similar effect are Ciclosporin (T-cell immunosuppressant) and Betamethasone (glucocorticosteroid with anti-inflammatory and immunosuppressive properties), both highlighted with yellow arrows in Figure 1.

IDENTIFICATION OF NEW POTENTIAL INDICATIONS

The high response similarity identified for other immunosuppressive agents, as described above, suggests a possible broader therapeutic use for Dupilumab. To further investigate this, we used the Signature tool in opposite fashion to find the most different rather than the most similar conditions. Since cancer is known to cause big disturbances of the genome, we chose to screen through non-neoplastic conditions only, corresponding to 28,683 samples aggregated in 2,941 experimental comparisons (Figure 2).



Figure 2. Identification of the top-50 conditions having the most different molecular signature to 4 weeks of Dupilumab therapy of atopic dermatitis. The signature was entered as fold-change comparison of lesioned skin biopsies after vs. before treatment, in a linear scale. The image was cropped for basic illustration of the Signature tool.


For the given set of 182 genes, the top negatively correlated conditions comprised other atopic dermatitis studies (which confirms our hypothesis) and several other skin disorders (marked with blue frames and arrows in Figure 2) such as skin transplantation, thermal injury, psoriasis, rosacea, acne or skin irritation with nickel sulfate. Furthermore, diseases from other therapeutic areas, such as rheumatoid arthritis, Crohn'€™s disease, ulcerative colitis or Dengue hemorrhagic fever, were also identified (yellow arrows in Figure 2).
EXPRESSION IN DIFFERENT ANATOMICAL PARTS

To further study the biological function related to the Dupilumab signature, we clustered the corresponding genes by their expression in different tissues and cell types. This was done using the Hierarchical Clustering tool in GENEVESTIGATOR. To this end, we separated the genes from the Dupilumab signature into two groups, i.e. the up- and down- regulated genes, respectively, and used the matrix of non-neoplastic conditions to run a two-way clustering by genes and by anatomical parts. We then looked for clusters of genes having common expression in specific tissues (Figure 3).




Figure 3. Representative tissue expression clusters of genes down-regulated (left panel) or up-regulated (right panel) in response to Dupilumab treatment. The image was shortened vertically for illustration of the clusters.



Among the genes up-regulated in response to Dupilumab treatment, the largest cluster exhibited high expression in adipose tissues (see Figure 3). Genes from this cluster are all involved in fatty acid synthesis and lipid transport (e.g. ACACB, LEP, PLIN1 and PLIN4). Both IL-4 and IL-13, the primary targets of Dupilumab, are on the contrary known to suppress lipid production and thereby contribute to disruption of the epidermal barrier. Thus the gene-anatomy association points to the biological effect of the drug. Our analysis allowed a better understanding of the biological impact of the drug and also predicting its potential side effects in other tissues than the one of primary therapeutic interest.
CONCLUSIONS

Taking advantage of the deeply curated, fully integrated compendium of drugs and diseases in GENEVESTIGATOR, we were able to obtain two types of results: a) find other drugs causing similar transcriptional changes as Dupilumab, and b) find novel indications for this drug. We further refined our analysis to understand the spatial expression regulation of the Dupilumab signature. In summary, we found that Dupilumab caused similar molecular responses as those from Brodalumab, Etanercept and Infliximab. In terms of indication finding, besides confirming the primary indication (atopic dermatitis), we identified several other potential indications, including skin transplantation, thermal injury, psoriasis, rosacea, acne, skin irritation with nickel sulfate, as well as indications from other therapeutic areas such as rheumatoid arthritis, Crohn's disease, ulcerative colitis or Dengue hemorrhagic fever. The results obtained provide experimental evidence for the biological impact of Dupilumab on human cells, facilitating a broader intellectual property protection and helping designing clinical trials on an extended set of indications.

To try out GENEVESTIGATOR or to replicate/extend this analysis, please go here.

REFERENCES

Hamilton JD, Suarez-Farinas M, Dhingra N, Cardinale I, Li X, Kostic A, Ming JE, Radin AR, Krueger JG, Graham N, Yancopoulos GD, Pirozzi G, Guttman-Yassky E. (2014) Dupilumab improves the molecular signature in skin of patients with moderate-to-severe atopic dermatitis J Allergy Clin Immunol. 2014 Dec;134(6):1293-300.  [Abstract]

Hruz T, Laule O, Szabo G, Wessendorp F, Bleuler S, Oertle L, Widmayer P, Gruissem W and P Zimmermann (2008) Genevestigator V3: a reference expression database for the meta-analysis of transcriptomes. Advances in Bioinformatics 2008, 420747 [Full Text]