New frontiers in personalized medicine in psoriasis

KEYWORDS:

• Personalized medicine
• psoriasis
• artificial intelligence
• biomarkers
• stratification medicine
• precision medicine
• biologicals

1. Introduction

Personalized medicine is the new frontier in psoriasis as it may allow a patient-tailored management with treatment optimization and reduction of side effects as well as health costs [1]. Thanks to the growing advances in understanding psoriasis pathogenesis, a great number of new molecules have been discovered and developed to increase clinical outcome and improve patients’ quality of life. The so-called target therapies have then revolutionized psoriasis treatment, ensuring selectivity, high efficacy and safety, and allowing reach clear or almost clear skin manifestations. To date, eleven biologics, belonging to four pharmacological classes according to the target cytokine (tumor-necrosis factor [TNF], interleukins [IL] 12/23, 17, and 23), are available for psoriasis treatment with different peculiarities. Despite the great therapeutic armamentarium, treatment selection is still based on clinical factors (patients and psoriasis’ characteristics) and not on genetic ones. Indeed, the vast spectrum of psoriasis phenotypes together with the occurrence of primary and secondary lack of efficacy also with the new biologics are expression of a highly patient-dependent treatment response thus suggesting the need to identify individual genotype [2]. Hence, the identification and the introduction in daily clinical practice of biomarkers predictive of response to biological treatments may help clinicians in the therapeutic choice, optimizing clinical outcomes and minimizing the risk of side effects and treatment failure with related huge health costs and reduced patients’ quality of life. Herein we present and discuss the recent advances in personalized biological treatment for psoriasis.

2. Personalized biological treatment for psoriasis

The pharmacogenetics and pharmacogenomics have undergone a great push of development motivated from the need to provide innovative treatments especially for chronic diseases such as psoriasis [1,3]. Concerning biologicals, the recent advances mainly regard anti-TNF and IL12/23, while little evidence exists for anti-IL17 and no data are available for anti-IL23 [1,3–5].

2.1. Biomarkers predictive of response to anti-TNF

Response to anti-TNF was evaluated through the assessment of several genes among which the HLA:Cw6, the TNF cytokine, and its receptor were prominent [3–5]. The presence of the HLA:C:06:02 was traditionally associated with favorable outcome, although more recent studies questioned the association [1,3]. Probably, the great heterogeneity of genetic analyses as well as the differences in allele frequencies in ethnic subgroups may explain these discrepancies [3]. Similarly, polymorphisms (single nucleotide polymorphisms [SNPs]) in the TNF gene such as rs1800629, rs1799964, rs1799724, rs361520 were investigated with contrasting results [1]. As regards TNF receptor, the role of the TNF receptor superfamily member 1B (TNFRSF1B) was investigated in several studies in relation to different anti-TNF agents [3]. In detail, the TNFRSF1B rs1061622-TT was associated with a better response to all anti-TNF and statistically significantly with etanercept [3]. By contrast, the rs1061622-G polymorphism was associated with a worst outcome [3].

2.2. Biomarkers predictive of response to anti-IL 12/23

Ustekinumab has been among the most studied drugs under the pharmacogenetic aspects, probably due to the long-term presence on the market and its consequent high number of prescriptions worldwide [1,3–5]. The HLA:C:06:02 status has been thoroughly investigated in the influence of response to ustekinumab [3–5]. In detail, several studies confirmed the presence of such allele as positively associated with a great clinical and fast response to treatment, independently from ethnicity. Indeed, the same results were obtained from different studies performed on Caucasian, American, and Chinese subjects [1]. Anyway, a recent meta-analysis questioned the importance of the results given that high, although slightly different, response rates were observed in both patient groups, i.e. those HLA:C:06POS and NEG [6]. Hence, this allele should not be considered alone in the treatment choice, rather a panel of biomarkers could be more reliable [3,6]. Indeed, polymorphisms on the IL17 isoforms, IL12, and IL23R genes were assessed without reaching statistically significant results [1].

2.3. Biomarkers predictive of response to anti-IL 17

Studies on the research for biomarkers of response to anti-IL17 have focused on secukinumab and ixekizumab, while no data are available to date on brodalumab and bimekizumab [4,5]. The SUPREME study and its extension investigated the impact of the HLA-C*06:02 allele on the efficacy of secukinumab: no statistically significant differences have been found between those harboring the polymorphism and those not [7,8]. Indeed, the percentage of patients reaching PASI75 were comparable between the two groups (80.4% vs 79.7%; 62.0% vs 61.0%, respectively) [7,8]. Finally, studies on the polymorphisms of IL17 failed to retrieve promising results for both secukinumab and ixekizumab [5].

2.4. Biomarkers predictive of response to anti-IL 23

Given the recent introduction of anti-IL23, no data concerning pharmacogenetic studies are currently available.

2.5. The role of inflammatory/immunologic biomarkers in psoriasis management

Inflammatory molecules and their genes may play a key role in psoriasis management since they can be used to assess the disease activity and optimize treatment strategies. Indeed, it has been shown that psoriasis significantly correlates with distinctive serum cytokine levels, which can then be employed as biomarkers for the disease [9]. In detail, adiponectin was displayed to be significantly lower in psoriasis subjects than in controls; an opposite trend was observed concerning the level of Th1 (IL-2, IL-12, IFN-γ), macrophagic M1 (IL-1, IL-6, TNF-α), and Th17 (IL-6, IL-17) cytokines, confirming their prominent role in psoriasis pathogenesis [9]. Contrarily, the role of TGF-beta is still controversial [9]. Also, from the results of a meta-analysis by Bai et al., other inflammatory molecules, such as TNF-α, IFN-γ, sE-selectin, IL-2, IL-6, IL-8, IL-18, IL-22, fibrinogen, and C3, appeared to be elevated in psoriasis patients [10]. Similarly, a positive correlation between YKL-40, a mammalian chitinase 3-like protein associated with inflammatory diseases, and IL-17 levels, as well as PASI score was reported [11]. Although the number of studies investigating the role of cytokines and inflammatory molecules in measuring psoriasis activity and in influencing treatment management is growing, available evidence is scant and more standardized research is warranted.

3. Expert opinion

Although the number of available drugs is continuously growing, there still exist additional limitations such as predicting side-effect profile, immunogenicity, and primary lack or secondary loss of efficacy; hence, long-term safety and efficacy are sometimes impaired [2]. Indeed, notwithstanding the great advancement in understanding psoriasis pathogenesis, we are still far from a genetic profiling of individual patients given the high costs, time constraints, and the lack of highly specific biomarkers predictive of response to treatment. Overall, a standardized panel of biomarkers would be more reliable as compared to single target genes and probably the creation of a comprehensive algorithm that merges genotype and phenotype of each patient would be even more precise. In detail, the worldwide abundancy of data available from medical devices and electronic medical records has favored the development of a machine learning (ML) technology, which will have a prominent role in the creation of a personalized psoriasis bio-panel; also, such approach is meant to integrate patients’ genetic and phenotypic characteristics with comorbidities (that may contraindicate specific treatments), and previous failures of treatment (that may contribute to depict a specific psoriasis endotype) to aid physicians in guiding the best therapeutic strategy. Hence, the application of artificial intelligence (AI) may broaden the powerful therapeutic armamentarium expanding its application in a futuristic and personalized approach. The most discussed gene, which has been worldwide recognized as associated with psoriasis, is the HLA:Cw6, whose status has been evaluated in response to all the biologics available to date for the treatment of psoriasis. Contrasting results exist concerning its positive association with response to anti-TNF and anti -IL12/23, while it appears indifferent in the case of anti-IL17. Moreover, key cytokines involved in the psoriasis inflammatory pathway and target of biologicals have been taken into consideration in the genetic profiling of patients in several studies. Overall, serum Th1, Th17, and M1 cytokine levels were higher in patients with psoriasis than controls, while those of adiponectin were lower. Since adiponectin levels correlate with psoriasis, obesity, and metabolic syndrome, it may represent a promising innovative drug target. Anyway, the great heterogenicity of patients’ ethnicity and subpopulations as well as studies’ characteristics may have impacted on the variability of retrieved results. Thus, more research on the topic is warranted as a patient-tailored management would make biologicals more accessible, reduce health costs, and improve clinical results. This is particularly true in the modern era where biologics have positively revolutionized chronic inflammatory disease treatment but also, in selected cases, raised new clinical concerns such as paradoxical reactions (e.g. paradoxical psoriasis or eczematous eruptions, paradoxical arthritis, or hidradenitis suppurativa) [12,13]. Hence, indubitably, personalized and precision medicine should also try to focus the research on identifying biomarkers of possible paradoxical eruptions trying to prevent their development.

4. Conclusion

To date, although several studies have been conducted to identify biomarkers predictive of response to biological therapies, no consensus has been reached on the definition of a panel to be introduced in daily clinical practice. The application of AI in the creation of an algorithm that merges genotype and phenotype of a single psoriatic subject would represent the best way to approach holistically a patient, ensuring the most appropriate tailor-tailored management. Anyway, more research is needed on the topic.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Authors contributions

All authors contributed to the article conception and design. Material preparation, data collection, and analysis were performed by E Camela, L Potestio, and M Megna. The first draft of the manuscript was written by E Camela, L Potestio, and M Megna, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. All authors contributed equally to the manuscript and read and approved the final version of the manuscript.

Additional information

Funding

This paper was not funded.