miRNA diagnostic biomarkers for severe drug rash
Patent Information
- Application Number
- JP2020114710
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2020-07-02
- Publication Date
- 2024-11-27
- Estimated Expiration
- 2040-07-02
AI Technical Summary
【0012】 本発明により見いだされたmiRNAの発現量を、単独、あるいは複数項目を測定することにより、重症薬疹疑いの患者と非薬剤性皮膚疾患患者を鑑別することが可能である。加えて、本発明は、重症薬疹を発症した患者の病勢変化や重症化リスクを高い精度で診断することができる。
Smart Images

Figure 0007590633000006 
Figure 0007590633000007 
Figure 0007590633000008
Abstract
Description
[Technical field]
[0001] The present invention relates to a method for assisting in the diagnosis of the onset and progression of severe drug rash. [Background technology]
[0002] Drug rash is a general term for skin reactions induced by drugs and their metabolites, and the most severe of these include Stevens-Johnson syndrome (SJS), toxic epidermal necrolysis (TEN), and drug-induced hypersensitivity syndrome (DIHS). Although the incidence rate of these severe drug rashes is low, they are highly lethal and can leave severe aftereffects even after recovery. Therefore, it is important to diagnose patients suspected of having severe drug rashes early and start appropriate treatment promptly. Therefore, the development of biomarkers that can diagnose SJS / TEN and DIHS early is desired.
[0003] Previous reports have strongly suggested a relationship between the onset of severe drug eruptions and abnormal activation of the immune system. Therefore, cytokines and chemokines acting on the immune system have been reported as blood protein biomarkers for severe drug eruptions. Representative examples include TARC (chemokine CCL17), granulysin (GNLY), FAS-L, interleukin 6, IP-10 (chemokine CXCL10), and combinations thereof (Non-Patent Documents 1, 2, 3). Of these, TARC has already been covered by insurance as a biomarker that reflects the severity of atopic dermatitis (Non-Patent Document 4).
[0004] MicroRNA (miRNA) is a functional small nucleic acid approximately 22 bases long, and is known to play a role in controlling gene expression in the body by RNA interference. It has been reported that various miRNAs control epithelial regeneration and angiogenesis in wound repair, and are also known to be deeply involved in the pathogenesis of skin diseases such as scleroderma, psoriasis, and skin cancer (Non-Patent Document 5). However, there have been few reports on the relationship between the expression level of miRNAs present in the blood and the onset or aggravation of severe drug rash. [Prior art documents] [Non-patent literature]
[0005] [Non-Patent Document 1] Komatsu-Fujii T et al. The thymus and activation-regulated chemokine (TARC) level in serum at an early stage of a drug eruption is a prognostic biomarker of severity of systemic inflammation, Allergology Int. 67, 90-95, 2018 [Non-Patent Document 2] Abe R, et al. Rapid immunochromatographic test for serum granulysin is useful for the prediction of Stevens-Johnson syndrome and toxic epidermal necrolysis. J Am Acad Dermatol. 2011, 65(1):65-8. [Non-Patent Document 3] Shiohara T, et al. Monitoring the acute response in severe hypersensitivity reactions to drugs. Curr Opin Allergy Clin Immunol. 2015, 15(4):294-9. [Non-Patent Document 4] Atopic Dermatitis Diagnostic Guidelines Creation Committee. Atopic Dermatitis Diagnostic Guidelines 2018. Journal of the Japan Dermatological Association. 2018, 128(12), 2431-2502. [Non-Patent Document 5] Gautam S et al. MicroRNAs as biological regulators in skin disorders. Biomed Pharmacother. 2018, 108:996-1004. Summary of the Invention [Problem to be solved by the invention]
[0006] The objective of the present invention is to provide a method for assisting in the diagnosis of Stevens-Johnson syndrome / toxic epidermal necrolysis (SJS / TEN) and drug-induced hypersensitivity syndrome (DIHS) with high accuracy, a method for objectively evaluating the efficacy of treatment for these conditions, and a method for assisting in the differentiation of mild drug rash cases that heal without becoming severe from cases that become severe. [Means for solving the problem]
[0007] We used the miRCURY LNA microRNA PCR method with LNA-enhanced probes / primers for 752 miRNAs to search for biomarker miRNAs that may be useful for diagnosing the progression or change in disease progression of Stevens-Johnson syndrome / toxic epidermal necrolysis (SJS / TEN) and drug-induced hypersensitivity syndrome (DIHS).First, using volcano plots, receiver operating characteristic curves (ROC curves), and multiple comparison statistical tests, we identified 17 miRNAs that showed significant changes in expression between the acute and recovery phases of severe drug rash from the 502 miRNAs for which quantitative values were available in the miRCURY LNA microRNA PCR measurement of the search serum sample set. Subsequently, in a validation study in which serum from a different set of samples was measured using the Taqman Advanced miRNA Assay, four miRNAs (hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, hsa-miR-222-3p) were identified that showed reproducible significant changes in expression between recovery phase cases and the acute phase of severe drug rash.
[0008] ROC curve analysis using the corrected cycle values (ΔCq values) of hsa-miR-205-5p, hsa-miR-21-5p, hsa-miR-21-3p, and hsa-miR-222-3p in the exploration group showed that hsa-miR-205-5p distinguished acute and convalescent cases of SJS / TEN and DIHS with high accuracy (area under the ROC curve: AUROC 0.85 or higher). In particular, the diagnostic ability of this miRNA molecule in SJS / TEN was significantly higher than that of the previously reported marker GNLY (AUROC 0.54). On the other hand, significant diagnostic ability was observed for hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p in diagnosing the disease progression of DIHS. Furthermore, the severe drug rash diagnostic model combining hsa-miR-205-5p, hsa-miR-21-5p and hsa-miR-21-3p improved the ability to diagnose the disease in SJS / TEN and DIHS, and also demonstrated the potential to predict the disease progression of severe erythema multiforme (EM major), which is known to progress to SJS / TEN, with higher accuracy than previously reported markers.
[0009] ROC curve analysis of the disease progression and severity diagnostic ability of the above four miRNAs in the validation group showed that all miRNAs showed high diagnostic ability in diagnosing the disease progression of severe drug rash. In addition, similar to the results obtained with the miRCURY LNA microRNA PCR method, the severe drug rash diagnostic model combining hsa-miR-205-5p, hsa-miR-21-5p, and hsa-miR-21-3p showed high disease progression diagnostic ability in all drug rash disease types analyzed. On the other hand, although all four miRNAs analyzed this time showed significant diagnostic ability in diagnosing the severity of either SJS / TEN or DIHS, or both disease types, differences were observed in the severity prediction ability of individual miRNA molecules between the disease types of severe drug rash. The analysis showed that hsa-miR-205-5p could be the most promising biomarker for predicting the severity of SJS / TEN and hsa-miR-21-3p could be the most promising biomarker for predicting the severity of DIHS. By combining the three miRNAs, it was possible to predict the progression of both SJS / TEN (AUROC 0.86) and DIHS (AUROC 0.93) without reducing the severity prediction ability of each miRNA biomarker, and the predictive accuracy was shown to be higher than that of the existing marker TARC (AUROC 0.83).
[0010] The diagnostic ability of hsa-miR-205-5p, hsa-miR-21-5p, hsa-miR-21-3p, and hsa-miR-222-3p was evaluated for severe drug rash and other skin diseases. As a result, it was found that hsa-miR-205-5p showed high diagnostic ability (AUROC>0.85) in distinguishing SJS / TEN from atopic dermatitis, psoriasis, chickenpox, and toxic rash. On the other hand, hsa-miR-21-5p and hsa-miR-21-3p showed high diagnostic ability between SJS / TEN and autoimmune bullous diseases. All of these diagnostic abilities were shown to be significantly higher than existing markers. Next, the diagnostic ability of these miRNAs to distinguish DIHS from other skin diseases was evaluated. When used alone, hsa-miR-21-3p showed an extremely high diagnostic ability of over AUROC 0.90 in distinguishing DIHS from all control skin diseases except toxic rash. In addition, a similar evaluation was performed on a severe drug rash diagnostic model combining hsa-miR-205-5p, hsa-miR-21-5p, and hsa-miR-21-3p, and it showed good diagnostic ability for all skin diseases including toxic rash. Furthermore, by combining this diagnostic model with the previously reported marker TARC, the differential diagnostic ability was improved, and it was shown that DIHS could be distinguished from all control skin diseases with AUROC 0.89 or more. These results suggest that hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p alone, or a combination of these, may be useful biomarkers for diagnosing the progression of severe drug rash, predicting its severity, and differentiating it from other skin diseases.
[0011] The present invention has been completed based on these findings. The gist of the present invention is as follows. (1) A method for testing for severe drug rash, comprising measuring gene expression in a sample derived from a subject for at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p. (2) The method described in (1), wherein the miRNA whose expression is measured is at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p, and the measured value assists in diagnosing the progression of severe drug rash. (3) The method according to (1), wherein the miRNA whose expression is measured is at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p, and the measured value assists in predicting the severity of a severe drug rash. (4) The method according to (1), wherein the miRNA whose expression is measured is at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p, and the measured value assists in distinguishing between a severe drug rash and other skin diseases. (5) The method according to (2), wherein the miRNAs whose expression is to be measured are at least one combination of miRNAs selected from the group consisting of combinations of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p expressions. (6) The method according to (3), wherein the miRNAs whose expression is to be measured are at least one combination of miRNAs selected from the group consisting of combinations of expression of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p. (7) The method according to (4), wherein the miRNAs whose expression is to be measured are at least one combination of miRNAs selected from the group consisting of combinations of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p expression. (8) A kit for testing for severe drug rash, comprising a reagent capable of measuring gene expression in a sample derived from a subject for at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p. Effect of the Invention
[0012] By measuring the expression levels of the miRNAs discovered by the present invention, either alone or in combination, it is possible to distinguish between patients suspected of having a severe drug rash and patients with non-drug-induced skin diseases. In addition, the present invention can diagnose with high accuracy the progression of disease and the risk of aggravation in patients who have developed a severe drug rash. [Brief description of the drawings]
[0013] [Figure 1] Comparison of gene expression levels to search for miRNA biomarkers for severe drug rash. The search sample set was measured using the miRCURY LNA microRNA PCR method, and (A) the expression ratio of miRNAs between SJS / TEN acute-phase cases (n=9) and fully recovered cases (n=18), and (B) the volcano plot of the expression ratio of miRNAs between DIHS acute-phase cases (n=15) and fully recovered cases (n=18). In this plot, the vertical axis shows the p-value obtained from the Student's t-test between the two groups, and the horizontal axis shows the logarithmic expression ratio of acute phase cases to convalescent phase cases (log2(mean value of 2-ΔCq [acute phase cases] / mean value of 2-ΔCq [all convalescent phase cases]). The cutoff values in the search for biomarkers of severe drug rash using this volcano plot were set as an absolute value of the fold change in expression level of 1.5 or more (vertical dotted line) and a p-value of 0.05 or less (horizontal dotted line) based on statistical calculation. The miRNA molecules described in the claims are highlighted with a square frame. (C) A Venn diagram showing the number of identified candidate biomarker miRNAs for SJS / TEN and DIHS and their overlap. [Diagram 2]Comparison of expression levels of candidate biomarker miRNAs between patients in the acute phase of severe drug rash, those in the complete recovery phase, and healthy adults in the validation group. For four types of miRNAs that show particularly significant changes in expression in severe drug rash, the ΔCq values obtained by Taqman miRNA assay measurement in patients in the acute phase and complete recovery phase of each severe drug rash disease type, and healthy adults were compared using dot plots. The ΔCq value represents the relative expression level of the target miRNA to the external spike control miRNA (cel-miR-54-3p), and a higher value indicates a lower expression level. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001, One-way ANOVA and post-hoc Dunn's test. Recovery, complete recovery (n=32); Mild-entry, mild drug rash (n=51); SJS / TEN, Stevens-Johnson syndrome / toxic epidermal necrolysis acute stage (n=13); DIHS, drug-induced hypersensitivity syndrome (n=15); EM major, severe erythema multiforme (n=15); Healthy, healthy adults (n=30). The center line of each group indicates the median ΔCq value. [Diagram 3]Comparison of scores of the severe drug rash miRNA diagnostic model between patients in the acute phase of severe drug rash, those in the complete recovery phase, and healthy adults. A severe drug rash diagnostic model was constructed by combining the ΔCq values of hsa-miR-205-5p, hsa-miR-21-3p, and hsa-miR-21-5p obtained from the measurement of the validation sample set, which are candidate miRNAs for severe drug rash biomarkers, using multiple logistic regression analysis. The values obtained by substituting the ΔCq value of each miRNA into this model were plotted as diagnostic scores. Intergroup comparison of diagnostic scores was performed using one-way ANOVA and post-hoc Dunn's test. **, p < 0.01; ***, p < 0.001; ****, p < 0.0001, . Recovery, complete recovery stage (n=32); Mild-entry, mild drug rash (n=51); SJS / TEN, Stevens-Johnson syndrome / toxic epidermal necrolysis acute stage (n=13); DIHS, drug-induced hypersensitivity syndrome (n=15); EM major, severe erythema multiforme (n=15); Healthy, healthy adults (n=30). The center line of each group indicates the median diagnostic score. [Figure 4]Comparison of expression levels of various candidate biomarker miRNAs between severe acute phase drug rash cases and non-severe drug rash skin diseases. The ΔCq value used in this analysis represents the relative expression level of the target miRNA to the external spike control miRNA (cel-miR-54-3p), and a higher value indicates a lower expression level. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001, One-way ANOVA and post-hoc Dunn's test. Mild-entry, mild drug rash cases (n=51); SJS / TEN, Stevens-Johnson syndrome / toxic epidermal necrolysis acute phase cases (n=13); DIHS, drug-induced hypersensitivity syndrome (n=15); EM major, severe erythema multiforme (n=15); Atopic dermatitis, allergic dermatitis (n=16); Psoriasis, psoriasis (n=16); Autoimmune bullous disease, autoimmune bullous disease (n=16); Varicella, chickenpox (n=16); Toxicoderma, toxic rash (n=32); Healthy, healthy adults (n=30). The center line of each group indicates the median ΔCq value. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0014] Hereinafter, an embodiment of the present invention will be described in detail.
[0015] The present invention provides a method for testing for severe drug rash, which comprises measuring gene expression in a sample derived from a subject for at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p.
[0016] A drug rash is a rash that occurs in response to a drug, and is a type of side effect of medicines. In general, it occurs 1-2 weeks after taking the drug, and is easily diagnosed because it improves when the causative drug is discontinued. However, severe drug rash, which is a serious drug rash, often does not improve simply by discontinuing the causative drug, and if not treated appropriately, it can be life-threatening, so early diagnosis is important.
[0017] Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN), which are typical examples of severe drug eruptions, are intractable diseases with high mortality rates and sequelae such as blindness and respiratory disorders. SJS and TEN are considered to be in the same spectrum of pathologies, and although the symptoms of TEN are similar to those of SJS, they are more severe, and TEN often progresses from SJS. SJS is characterized by widespread mucosal lesions at the mucocutaneous junctions of the lips, conjunctiva, and vulva, accompanied by fever, and erosions and blisters due to necrotic damage of the epidermis accompanied by generalized erythema of the skin. The area of the lesions is less than 10% of the body surface area. On the other hand, TEN is characterized by widespread erythema and significant necrotic damage of the epidermis, such as blisters, epidermal peeling, and erosions over 10% or more of the body, high fever, and mucosal rash, and is considered to be the most severe skin disorder caused by drugs. The incidence rates of SJS and TEN are said to be 1-6 and 0.4-1.3 cases per 1 million population per year, respectively.
[0018] Drug-induced hypersensitivity syndrome (DIHS) is another severe form of drug rash, comparable to SJS / TEN. Erythema papule and erythema multiforme are seen all over the body, accompanied by a high fever of 38 degrees or higher, and progress to erythroderma. Usually, mucosal rash is absent or mild, but oral mucosal erosion is sometimes observed. Generalized lymphadenopathy, organ dysfunction including liver dysfunction, and peripheral white blood cell abnormalities (leukocytosis, eosinophilia, appearance of atypical lymphocytes) are observed. It is characterized by reactivation of human herpesvirus (HHV)-6 during the course of the disease, and if lymphadenopathy or HHV-6 reactivation cannot be demonstrated, it is diagnosed as atypical DIHS. Unlike normal drug eruptions, they do not occur immediately after administration of the causative drug, but often occur more than two weeks later, and symptoms continue even after the causative drug is discontinued, often taking more than a month to improve.
[0019] It is important to diagnose these severe drug eruptions early and start appropriate treatment promptly, but because their initial symptoms are similar to those of mild drug eruptions, it is currently difficult to identify or predict early on which patients will develop severe cases. In addition, in order to provide appropriate treatment to patients who develop severe cases, it is important to understand the disease progression (activity of the condition).
[0020] TARC (chemokine CCL17), known as a biomarker for evaluating the severity of atopic dermatitis, has been proposed to verify DIHS, and granulysin and Fas ligand have been proposed as blood protein markers for severe drug eruptions to detect SJS / TEN. On the other hand, in this invention, we focus on miRNAs, which have not been used in the diagnosis of severe drug eruptions using blood samples, and demonstrate that their expression levels (ΔCq values), either alone or in combination, can be used to diagnose the disease progression and predict the progression of severe drug eruptions.
[0021] hsa-miR-205-5p (miRBase ID: MIMAT0000266) is a miRNA that has been reported to be highly expressed in skin tissue. It has been reported that the expression of this gene is essential for the formation of stratified epithelium and the proliferation of hair follicle stem cells during the neonatal period (Nat Cell Biol. 2013 Oct;15(10):1153-63). It is also known that this gene contributes to re-epithelialization in wound healing of the skin (Biochim Biophys Acta. 2016 Aug;1862(8):1443-52). Furthermore, it has been reported that this molecule plays a role in suppressing the metastasis and growth of cancer cells by suppressing the expression of genes involved in the epithelial-mesenchymal transition of skin cancer (Mol Carcinog. 2016 Aug;55(8):1229-42). The base sequence of hsa-miR-205-5p is shown in SEQ ID NO:1.
[0022] hsa-miR-21-3p (miRBase ID: MIMAT0004494) is known to be expressed in various cells, but in human normal tissues, it is highly expressed in fascia, arteries, nerves, and arachnoid membrane. Expression of this miRNA in skin has also been confirmed. This miRNA is present at the 3' end of the miR-21 precursor sequence encoded by the MIR21 gene, and is produced through processing by multiple enzymes. This miRNA has been reported to have the function of suppressing skin inflammation caused by ultraviolet rays (EMBO Mol Med. 2016 Aug 1;8(8):919-36.). On the other hand, this miRNA is also known to contribute to wound healing through improving angiogenesis and fibroblast migration (Theranostics. 2018 Jan 1;8(1):169-184.). The base sequence of hsa-miR-21-3p is shown in SEQ ID NO: 2.
[0023] hsa-miR-21-5p (miRBase ID: MIMAT0000076) has been reported to be highly expressed in the lung, thyroid, skin, spleen, etc. This miRNA is encoded by the same gene (MIR21) as hsa-miR-21-3p, exists at the 5' end of the transcript, and is produced through processing by multiple enzymes. This molecule functions as an oncogene in various solid cancer cells such as breast cancer, colon cancer, and glioma (Int J Biol Sci. 2011;7(5):685-90). On the other hand, this miRNA has also been reported to negatively regulate the activation of genes involved in signal transduction via the T cell receptor in T cells by suppressing their expression (Biochimie. 2014 Dec;107 Pt B:319-26). It has also been reported that this miRNA is upregulated in keratinocytes by fibrosis stimulation, improves cell migration, and promotes skin wound repair (Int J Biol Sci. 2011;7(5):685-90.). The base sequence of hsa-miR-21-5p is shown in SEQ ID NO:3.
[0024] hsa-miR-222-3p (miRBase ID: MIMAT0000279) has been confirmed to be highly expressed in normal tissues such as the prostate, bladder, skin, thyroid, and lung. This miRNA has been shown to be highly expressed in skin tissues of psoriasis and hypertrophic scars compared to normal skin tissue (Medicine (Baltimore). 2015 Feb;94(7):e458., J Dermatol Sci. 2010 Jun;58(3):177-85.). It has also been reported that this gene controls the expression of matrix degrading enzymes in keratinocytes (J Dermatol Sci. 2010 Jun;58(3):177-85.). Furthermore, it has been shown that this gene suppresses osteoclast formation (Int J Mol Sci. 2016 Feb; 17(2): 240.). In addition, this miRNA is known to contribute to the promotion of cell proliferation and cell migration / invasiveness in various cancer cells and to be involved in their progression (PLoS One. 2014 Jan 31;9(1):e87563., Technol Cancer Res Treat. Jan-Dec 2019;18:1533033819892256., Cell Signal. 2019 Nov;63:109386.). The base sequence of hsa-miR-222-3p is shown in SEQ ID NO:4.
[0025] These miRNAs are novel and useful biomarkers for severe drug rash.
[0026] In the present invention, the subject is a mammal suspected of developing a severe drug rash, but may be any mammal thought to be at risk of developing the disease. The mammal is typically a human. Examples of samples derived from a subject include cells, tissues, body fluids, etc. obtained from the subject, specifically, blood (e.g., whole blood, serum, plasma, plasma exchange fluid, etc.) of the subject. Whole blood, serum, or plasma obtained in a normal blood test (clinical test) may be used as the blood sample.
[0027] In the method of the present invention, expression in a sample derived from a subject can be measured by measuring the amount of the miRNA present in the sample. The measurement method is not particularly limited and may be a known method, such as Northern blot, RT-PCR, real-time PCR, digital PCR, miRNA microarray, and small RNA sequencing.
[0028] In order to measure the expression of the miRNA, it is preferable to use a nucleic acid probe capable of specifically hybridizing with the miRNA (when measuring by Northern blot method). Alternatively, at least one pair of nucleic acid primers capable of specifically amplifying cDNA synthesized using the miRNA as a template may be used (when measuring by RT-PCR method). Furthermore, a combination of the above-mentioned nucleic acid probe and primer set may be used (when measuring by real-time PCR method or digital PCR method). The nucleic acid probe and nucleic acid primer can be designed based on the genetic information of the miRNA (mentioned above). The nucleic acid probe is usually suitable for about 15 to 1500 bases. The nucleic acid probe may be labeled with a radioactive element, a fluorescent dye, an enzyme, or the like. The nucleic acid primer is usually suitable for about 15 to 30 bases. The nucleic acid primer may be labeled with a radioactive element, a fluorescent dye, an enzyme, or the like.
[0029] The expression of the miRNA to be measured may be one type or multiple types. By referring to multiple miRNA expression data, more accurate evaluation may be possible. Representative combinations include, but are not limited to, Tables 2 and 3 (hsa-miR-205-5p, hsa-miR-21-5p, hsa-miR-21-3p). In order to simultaneously detect the expression of multiple miRNAs, detection methods such as multiplex real-time PCR (using multiple types of fluorescent probes), DNA array (probes immobilized on a substrate) (Nat Rev Drug Discov. 2002, 1:951-960), and small RNA sequencing may be used.
[0030] The expression of at least one miRNA selected from a group consisting of miRNAs is measured in a sample derived from a subject, and if the expression level is high (e.g., the ΔCq value is low), it is determined that there is a high possibility of a severe drug rash developing, and if the level is low (e.g., the ΔCq value is high), it is determined that there is a low possibility of a severe drug rash developing.
[0031] Therefore, the method of the present invention can assist in the diagnosis of severe drug rash (determination of the presence or absence of the onset of severe drug rash).
[0032] As an example of the present invention, the diagnosis of severe drug rash can be performed according to the following criteria. The expression level of at least one of the above miRNAs in serum collected from a subject is measured, and if a value lower than a preset cutoff value or reference value for a ΔCq value obtained from expression analysis or a diagnostic score obtained by substituting each ΔCq value into a combination of multiple miRNAs or a diagnostic model is obtained, the subject is evaluated as having developed severe drug rash. This preset cutoff value can be appropriately set by a person skilled in the art. For example, the 95% confidence interval of the quantitative value of a healthy person who has not developed severe drug rash can be used as the reference value, or the cutoff value can be set from an ROC curve. Alternatively, if at least one of the expression levels of the above miRNAs shows a tendency to increase compared to past measurements, the possibility of development of severe drug rash is suspected.
[0033] The method of the present invention can also be used to diagnose the progression of severe drug eruptions. In this specification, "diagnosis of progression" refers to the distinction between severe drug eruptions such as SJS / TEN and DIHS and mild drug eruptions such as disseminated maculopapular rash and erythema multiforme. When the expression level of at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p in a sample derived from a subject is high, it can be determined that the subject is likely to develop or has developed a severe drug rash in the future.
[0034] The method of the present invention can also be used to diagnose the severity of severe drug rash. In this specification, "disease diagnosis" refers to the distinction between the acute phase and the recovery phase of severe drug rash such as SJS / TEN and DIHS. When the expression level of at least one protein selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p in a sample derived from a subject is high, it can be determined that there is a high possibility that the subject is in the acute phase of severe drug rash.
[0035] Hsa-miR-205-5p has particularly high detection performance for SJS / TEN, while hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p have particularly high detection performance for DIHS.
[0036] The method of the present invention can be used for the differential diagnosis of severe drug rash and control skin disease. In this specification, "differential diagnosis" means the discrimination between severe drug rash and control skin disease. When the expression level in a sample derived from a subject of at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p is higher than the cutoff value or the reference value, it can be determined that the subject is likely to be in the acute phase of severe drug rash such as SJS / TEN or DIHS, rather than atopic dermatitis, psoriasis, autoimmune blistering disease, chickenpox, or toxic rash. For the differential diagnosis of severe drug rash, a biomarker miRNA may be used alone, or a combination of multiple miRNAs may be used as a diagnostic model to improve the performance of distinguishing SJS / TEN and DIHS. As a representative example of the combination, hsa-miR-205-5p, hsa-miR-21-5p and hsa-miR-21-3p are described in the following examples, but the present invention is not limited to these combinations, and any combination using the four miRNAs shown in the present invention may be used. In addition, the existing marker TARC may be added to the combination of these miRNAs. The expression level of TARC in serum can be measured by immunoserological testing methods such as enzyme immunoassay and chemiluminescence enzyme immunoassay.
[0037] The method of the present invention can be used not only for diagnosing the disease progression of severe drug rash, but also for prognosis and confirmation of treatment effect. For example, the expression of at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p in a sample from a subject determined to be likely to be in the acute phase of severe drug rash is measured once or multiple times at different times, and if the expression level falls to a level close to a cutoff value or a reference value, it is determined that the treatment has recovered from the severe drug rash, and if the level is high or does not fall, it is determined that the treatment has not recovered from the severe drug rash or that the recovery is insufficient.
[0038] If it is determined that the subject is likely to develop or has already developed a severe drug rash, the suspected drug should be discontinued and appropriate treatment should be initiated according to the patient's disease severity and symptoms. Systemic administration of steroids (initial dose of 0.5-1 mg / kg / day in prednisolone equivalent, gradually tapered as appropriate) is an effective drug therapy for DIHS, and the dose should be reduced while paying attention to the recurrence of symptoms due to reactivation of HHV-6. In the case of SJS / TEN, systemic administration of steroids with different doses depending on the severity is recommended (in terms of prednisolone, 0.5-1 mg / kg / day for moderate cases, 1-2 mg / kg / day for severe cases, and for the most severe cases, starting with methylprednisolone 1 g / day for three days, then gradually tapering as appropriate depending on the symptoms).Other treatments include high-dose intravenous human immunoglobulin (IVIG) therapy (400 mg / kg / day for five consecutive days, generally only one course), plasma exchange therapy, ocular surface inflammation for ocular lesions (betamethasone or dexamethasone eye drops (approximately four times a day)), suppressing blepharohematoid adhesions to preserve the ocular surface epithelium, and preventing ocular surface infections.
[0039] The present invention also encompasses a method for treating a severe drug rash, comprising measuring the expression in a sample derived from a subject of at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p, and administering treatment to the subject if the subject is assessed to be highly likely to have developed a severe drug rash based on the measured value.
[0040] The present invention also provides a kit for testing for severe drug rash, comprising a reagent capable of measuring the expression in a sample derived from a subject of at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p.
[0041] As an example, the kit of the present invention includes a nucleic acid probe capable of specifically hybridizing with the miRNA as a reagent. The nucleic acid probe may be immobilized on a substrate. The kit may further include an instrument for collecting a biological sample, a blood coagulant, reagents for extracting RNA from a sample derived from a subject, reagents for detecting RNA, and an instruction manual. The instruction manual may include, in addition to the method for using the kit, evaluation and / or differentiation criteria for acute exacerbation of severe drug rash.
[0042] As yet another example, the kit of the present invention contains as a reagent at least one pair of nucleic acid primers capable of specifically amplifying cDNA synthesized using the miRNA as a template. The kit may further contain an instrument for collecting a sample from a subject, a blood coagulant, reagents for extracting RNA from a sample from a subject, a reagent for reverse transcribing the extracted RNA, a PCR reaction reagent for detecting the target cDNA, and an instruction manual. The instruction manual may contain, in addition to instructions for using the kit, criteria for evaluation and / or differentiation of cases of acute exacerbation of severe drug rash.
[0043] The kit of the present invention may also contain a standard miRNA as an external spike control used to correct the expression level of the target miRNA, a positive control miRNA, a buffer, a reaction stop solution, a washing solution, a reaction vessel, and the like. The kit of the present invention may further include an antibody or a nucleic acid aptamer capable of specifically recognizing TARC. The antibody and the aptamer may be immobilized on a microtiter plate, a magnetic bead, a cellulose membrane, or a substrate. Alternatively, the kit may include a nucleic acid probe capable of specifically hybridizing with TARC mRNA. The nucleic acid probe may be immobilized on a substrate. EXAMPLES
[0044] The present invention will now be described in more detail with reference to examples. Example 1 (1) Sample Drug rash and related skin disease samples used in the analysis were collected with the approval of the research ethics committees of each base hospital (Yokohama City University, Shimane University, Shimada Municipal Hospital, Iwata Municipal General Hospital, Nara Medical University, Niigata University), National Institute of Health Sciences, Kihara Foundation, Astellas Pharma, and Daiichi Sankyo. Meanwhile, healthy adult samples were collected with the approval of the research ethics committees of Kitasato University and the National Institute of Health Sciences.
[0045] Blood samples were collected from patients with suspected mild drug-induced drug eruptions (erythema multiforme, disseminated erythema-papular skin rash, eczematous skin rash, etc.), patients with severe drug eruptions such as Stevens-Johnson syndrome (SJS), toxic epidermal necrolysis (TEN), drug-induced hypersensitivity syndrome (DIHS), and erythema multiforme major (EM major) during the acute phase (near the exacerbation phase) and during the recovery phase at the above-mentioned base hospital with the consent of the patients. Blood samples were also collected from patients with skin diseases other than drug eruptions (atopic dermatitis, psoriasis, autoimmune blistering disease, chickenpox, and toxic rash (not caused by drugs)). Blood samples were collected using 7 mL blood collection tubes containing a blood coagulation promoter for serum collection. After blood collection, the samples were left at room temperature for 30 minutes and then centrifuged at 3,000 rpm for 10 minutes (15°C to 20°C) to collect the serum fraction. The serum used in this test was stored frozen at -80°C and was thawed again on ice immediately before each measurement test.
[0046] For the screening analysis of candidate miRNAs for severe drug rash biomarkers using miRCURY LNA microRNA PCR, 32 serum samples from mild drug rash, 29 acute-phase serum samples from severe drug rash, and 18 convalescent-phase serum samples were provided (exploratory sample set). The severe drug rash samples were broken down into 9 cases of SJS / TEN, 15 cases of DIHS, and 5 cases of severe erythema multiforme (EM major).
[0047] In a validation study of candidate biomarker miRNAs using the Taqman Advanced miRNA Assay, serum samples from 51 mild drug rash cases, 43 acute-phase severe drug rash cases (13 SJS / TEN cases, 15 DIHS cases, 15 EM major cases), and 32 convalescent-phase cases were used. In addition, serum samples from non-drug rash control skin diseases, including atopic dermatitis (16 cases), psoriasis (16 cases), autoimmune blistering disease (16 cases), chickenpox (16 cases), and toxic rash (32 cases), as well as serum samples from healthy adults (30 cases) were used (validation sample set).
[0048] (2) RNA extraction and cDNA synthesis Total RNA was extracted from serum samples using the miRNeasy (registered trademark) Mini Kit (Qiagen). cDNA for miRCURY LNA microRNA PCR measurement was synthesized using the Universal cDNA Synthesis Kit II (Qiagen). On the other hand, cDNA for Taqman Advanced miRNA Assay measurement was synthesized using the Taqman TM The Advanced miRNA cDNA synthesis kit (Thermo Fisher Scientific) was used. In preparing samples for the Taqman Advanced miRNA Assay, cel-miR-54-3p (0.2 fmol) was added to each serum sample as an exogenous spike control, followed by total RNA extraction and cDNA synthesis.
[0049] (3) Measurement and correction methods For comprehensive miRNA expression analysis, QIAGEN's miRCURY LNA microRNA PCR system was used. This system is a measurement method that specifically detects 752 types of miRNA using locked nucleic acid (RNA modified nucleic acid) primers. For analysis of the measurement data, GenEx software (Multid) was used to calculate ΔCq, defined by the following formula, using the following steps: [1] Interplate calibration (IPC), [2] Cut off (excluding Cq>37), [3] Call rate (excluding less than 10%), [4] Missing data (complementing non-detected samples with the maximum Cq value of each probe + 1), and [5] Normfinder Normalization. The ΔCq value represents the relative expression level of the target miRNA to the internal standard gene (reference probe), and the smaller this value, the higher the expression level. In the above formula, k is the total number of reference probes for each sample, GOI is the target miRNA, IPC is the interplate calibrator, m is the number of IPCs on each plate, and n is the total number of IPCs on all plates. Cq Ref is selected by Normfinder as the reference probe with the smallest variability. In this analysis, hsa-let-7d-3p (miRBase ID; MIMAT0004484, CUAUACGACCUGCUGCCUUUCU (sequence number 5)), hsa-miR-23b-3p (miRBase ID; MIMAT0000418, AUCACAUUGCCAGGGAUUACCAC (sequence number 6)), hsa-miR-24-3p (miRBase ID; MIMAT0000080, UGGCUCAGUUCAGCAGGAACAG (sequence number 7)), hsa-miR-30d-5p (miRBase ID; MIMAT0000245, UGUAAACAUCCCCGACUGGAAG (sequence number 8)), and hsa-miR-361-5p (miRBase ID; MIMAT0000703, UUAUCAGAAUCUCCAGGGGUAC (sequence number 9)) were used as reference probes.
[0050] On the other hand, in the validation study of the selected severe drug rash biomarker candidate miRNAs (hsa-miR-205-5p; MIMAT0000266, UCCUUCAUUCCACCGGAGUCUG (SEQ ID NO: 1), hsa-miR-21-3p; MIMAT0004494, CAACACCAGUCGAUGGGCUGU (SEQ ID NO: 2), hsa-miR-21-5p; MIMAT0000076, UAGCUUAUCAGACUGAUGUUGA (SEQ ID NO: 3), hsa-miR-222-3p; MIMAT0000279, AGCUACAUCUGGCUACUGGGU (SEQ ID NO: 4)), the Taqman Advanced miRNA Assay (Thermo Fisher Scientific, hsa-miR-205-5p; 477967_mir, hsa-miR-21-3p; 477973_mir, hsa-miR-21-5p; 477975_mir, hsa-miR-222-3p; 477982_mir, cel-miR-54-3p; 478410_mir) were used. The expression level (ΔCq value) of each miRNA in this measurement was calculated by subtracting Cq (external spike control; cel-miR-54-3p, MIMAT0000025, UACCCGUAAUCUUCAUAAUCCGAG (SEQ ID NO: 10)) from Cq (target miRNA). As with the analysis of the miRCURY LNA microRNA PCR system, the smaller the calculated ΔCq value, the higher the expression level of the target miRNA.
[0051] (4) Statistical calculation Volcano plot analysis for the search for biomarker miRNAs was performed using the EnhancedVolcano package of the statistical software R. The common logarithm of p-values calculated from Student's t-test between two groups was used to create the volcano plot. One-way ANOVA Dunn's test was used to compare miRNA expression levels between three or more groups. The diagnostic ability of each miRNA biomarker was evaluated using the area under the ROC curve (AUROC) value calculated from the receiver operating characteristic curve (ROC curve). These statistical calculations were performed using the medical statistics software GraphPad Prism8 (GraphPad Software). A combined diagnostic model aimed at improving the diagnostic ability of candidate miRNAs for severe drug rash biomarkers was constructed using multiple logistic regression analysis. The statistical software JMP12 (SAS Institute) was used to construct the diagnostic model.
[0052] (5) Selection of candidate miRNAs for severe drug rash biomarkers miRNAs with absolute values of the expression ratio between SJS / TEN (9 cases) and the recovery phase (18 cases) and between DIHS (15 cases) and the recovery phase (18 cases) calculated from the ΔCq values of 502 miRNAs for which measurement values were available by miRCURY LNA microRNA PCR, and with p-values of 0.05 or less calculated by Student's t-test between the two groups, were extracted using a volcano plot. Thirteen molecules were identified in the acute phase of SJS / TEN (Figure 1A) and 41 molecules in the acute phase of DIHS (Figure 1B) (selection condition 1, Figure 1C). To narrow down miRNAs that could be biomarkers for severe drug rash, we calculated the diagnostic ability of each severe drug rash disease type and recovery period for all miRNAs that showed significant changes in expression during the acute phase of severe drug rash, and extracted miRNA molecules with AUROC values of 0.75 or higher (6 types of SJS / TEN, 18 types of DIHS) (selection condition 2, Figure 1C). Furthermore, in a multiple comparison test (one-way ANOVA Dunn's test) comparing the expression levels of various miRNAs in recovery phase cases with SJS / TEN, DIHS, and EM major, we extracted miRNAs that showed statistically significant differences in any of the severe drug rash disease types and recovery period (5 types of SJS / TEN, 17 types of DIHS, selection condition 3, Figure 1C). Among the miRNAs that met the above-mentioned selection criteria 1-3, four promising molecules (hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, hsa-miR-222-3p) were selected as candidate miRNAs for severe drug rash biomarkers. In the exploratory sample set, these miRNA molecules were shown to be significantly more highly expressed in SJS / TEN, DIHS, or both compared to the recovery phase (Table 1). Hsa-miR-205-5p showed a particularly high expression change in SJS / TEN, while hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p showed relatively high expression change in DIHS (Table 1).
[0053] (6) Verification of expression changes of candidate miRNAs as biomarkers in patients with acute severe drug rash To demonstrate that the results of the miRCURY LNA microRNA PCR assay using the exploration specimen set are universal and not dependent on the measurement method, the expression levels (ΔCq values) of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p, and hsa-222-3p in the validation specimen set, which is a separate group from the exploration specimen set, were measured using Taqman Advanced miRNA Assay, and the expression ratios in the acute and recovery phases of severe drug rash were analyzed. As a result, for all severe drug rash biomarker candidate miRNAs, the expression level fluctuations were confirmed in the validation group, similar to the expression increase profile observed in the exploration group (Table 1). This suggests that the changes in the expression levels of these miRNAs molecules between severe drug rash and the recovery phase are universal and reproducible.
[0054] Next, the expression levels of each biomarker candidate miRNA in each severe drug rash disease type were compared with those in mild drug rash, recovery phase, and healthy adults by dot plot (Figure 2). When the median values for each group were calculated, hsa-miR-205-5p was most highly expressed in patients in the acute phase of SJS / TEN (Figure 2). It was revealed that hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p were most highly expressed in patients in the acute phase of DIHS (Figure 2). On the other hand, although overexpression of these miRNAs was observed in some patients in the acute phase of EM major, as in SJS / TEN and DIHS, the median values were shown to be similar to those in mild drug rash.
[0055] From the viewpoint of diagnosing the progression of severe drug rash, we analyzed miRNAs whose expression levels changed between the acute phase of SJS / TEN and mild drug rash. The only miRNA that showed a statistically significant change in expression level was hsa-miR-205-5p (Figure 2). On the other hand, the miRNAs that showed a statistically significant change in expression level between the acute phase of DIHS and mild drug rash were hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p (Figure 2). Therefore, it is considered that miRNAs that showed significant differences in each type of severe drug rash can be used to diagnose the progression of that disease type (Figure 2).
[0056] From the viewpoint of diagnosing the disease progression of severe drug rash and confirming the therapeutic effect, the difference in the expression level of each miRNA molecule between severe drug rash and the recovery phase and healthy adults was analyzed. As a result, significant differences were observed between all severe drug rash types and the recovery phase for hsa-miR-205-5p (Figure 2). On the other hand, more significant differences in expression levels were observed between DIHS acute phase patients and the recovery phase for hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p compared with other severe drug rash types (Figure 2). On the other hand, in the comparison between healthy adults and severe drug rash, significant differences in expression levels were observed between SJS / TEN and DIHS acute phase patients for hsa-miR-205-5p, and between DIHS acute phase patients for hsa-miR-21-3p and hsa-miR-222-3p and DIHS acute phase patients for hsa-miR-205-5p (Figure 2). These results suggest that the miRNAs can be used in tests aimed at diagnosing the progression of severe drug rash and confirming the effectiveness of treatment for severe drug rash.
[0057] (7) Biomarker potential of miRNAs as severe drug rash biomarkers Tables 2 and 3 show the AUROC values of four types of candidate biomarker miRNAs, previously reported severe drug rash markers (Granulysin (Table 2 only), TARC), and their combinations, between the acute phase of severe drug rash of each disease type and the recovery phase, mild drug rash, and healthy adults (Table 3 only) using the measurements of the exploration sample set and the validation sample set, respectively. When the disease diagnostic ability was evaluated using the exploration sample set, hsa-miR-205-5p showed a high AUROC value (0.93) between the recovery phase and the acute phase of SJS / TEN, which is poorly distinguished by existing markers such as Granulysin and TARC (Table 2). In addition, hsa-miR-205-5p, hsa-miR-21-3p, and hsa-miR-222-3p showed the same diagnostic ability (0.86-0.87) as existing markers (Granulysin: 0.84, TARC; 0.88) in diagnosing the progression of DIHS (Table 2). Furthermore, similar results were observed in the validation dataset (Table 3). In the validation dataset, the AUROC value of hsa-miR-205-5p in diagnosing the progression of SJS / TEN was 0.96, which was significantly high (Table 3). Furthermore, the AUROC values of the four candidate miRNAs for severe drug rash biomarkers in diagnosing the progression of DIHS were all 0.89 or higher (Table 3). These results indicate that hsa-miR-205-5p and all four miRNAs are useful biomarkers for diagnosing the progression of SJS / TEN and DIHS, respectively.
[0058] Next, the diagnostic ability of each miRNA for the progression of severe drug rash was evaluated by calculating the AUROC value between the acute phase of severe drug rash and mild drug rash. In the exploration group, hsa-miR-205-5p showed a moderate but significant diagnostic ability (AUROC 0.75) in diagnosing the progression of SJS / TEN (Table 2). In predicting the acute phase of DIHS, hsa-miR-21-5p and hsa-miR-222-3p were significant with an AUROC value of 0.70, but Granulysin and TARC showed high predictive ability of AUROC 0.85 or more, and no molecules with predictive ability exceeding this were found among the four miRNAs (Table 2). On the other hand, more promising results were observed for the above miRNAs in the validation group compared to the exploration group. As a result of the analysis, the AUROC value of hsa-miR-205-5p in the diagnosis of the worsening of SJS / TEN was high at 0.86 (Table 3). In addition, in the diagnosis of the worsening of DIHS, hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-3p showed AUROC values of 0.88 or more, which were higher than the existing marker TARC (Table 3). For EM major, no miRNA biomarkers predicting the worsening of EM were found in either the exploration group or the validation group with AUROC values of 0.8 or more (Tables 2 and 3). These results suggest that hsa-miR-205-5p can be a diagnostic biomarker for the worsening of SJS / TEN, and hsa-miR-21-5p and hsa-miR-222-3p can be a diagnostic biomarker for the worsening of DIHS.
[0059] (8) Improving diagnostic accuracy by using miRNA combination models for severe drug rash Combining multiple candidate biomarker miRNAs is expected to improve diagnostic accuracy. Next, we used the ΔCq values of hsa-miR-205-5p, hsa-miR-21-3p, and hsa-miR-21-5p between the acute phase of all types of severe drug rash and mild drug rash in the exploratory sample set to create a severe drug rash diagnostic model [diagnostic score = 0.23486537 × ΔCq (hsa-miR-205-5p) + 0.34814324 × ΔCq (hsa-miR-21-3p) - 0.0088437 × ΔCq (hsa-miR-21-5p) - 3.1236303] by multiple logistic regression. Based on the diagnostic scores obtained from this model, a comparison was made between severe drug rash and mild drug rash (diagnosis of progression) and recovery phase (diagnosis of disease progression). A slight improvement in diagnostic ability was observed in distinguishing mild drug rash, while a diagnostic ability of AUROC value 0.89 or more was shown in distinguishing recovery phase from all severe drug rashes (Table 2). In addition, by adding the existing marker TARC to this diagnostic model, it was shown that disease progression diagnosis of all severe drug rash disease types is possible with AUROC value 0.9 or more (Table 2). Next, using the measurements from the validation sample set, a severe drug rash diagnostic model based on a combination of miRNAs was constructed using the same method as above [diagnostic score = 0.34042857 × ΔCq (hsa-miR-205-5p) + 2.65078716 × ΔCq (hsa-miR-21-3p) - 1.8524265 × ΔCq (hsa-miR-21-5p) - 18.211688]. When the diagnostic score values obtained from this diagnostic model were plotted, a significant decrease in the diagnostic score value was observed in the severe drug rash group such as SJS / TEN and DIHS compared with healthy adults, the recovery period, and even mild drug rash (Figure 3). Next, ROC curve analysis was performed between severe drug rash and healthy adults, mild drug rash, and the recovery period. As a result, by combining the three miRNAs, it was possible to distinguish all severe drug rash cases from healthy adults with an AUROC value of 0.9 or more, and in particular, for DIHS and EM major, the AUROC value was higher than that obtained with each miRNA alone (Table 3). Furthermore, it was found that this diagnostic model improves the diagnostic ability in diagnosing the disease progression and predicting the progression of DIHS compared to when each miRNA marker was used alone (Table 3). Furthermore, when the previously reported marker TARC was incorporated into this diagnostic model, it was revealed that it is possible to diagnose the disease progression and predict the progression of DIHS with an accuracy of AUROC value of 0.98 or more (Table 3). From the above, it was shown that the combination of severe drug rash biomarker miRNAs identified in this study, and the construction and use of diagnostic models that add TARC to these combinations, will lead to improved accuracy in diagnosing the disease progression and progression of severe drug rash. On the other hand, in the differential diagnosis between the two groups in which a decrease in diagnostic ability was confirmed in the miRNAs combination model, it was considered that it would be better to use the miRNA with the highest diagnostic ability alone.
[0060] (9) Comparison of expression levels of candidate biomarker miRNAs between severe drug rash and control skin diseases The serum expression levels (ΔCq values) of each biomarker candidate miRNA in severe drug rash and control skin disease cases were plotted, and the median expression levels between each group were statistically compared (Figure 4). Compared to control skin diseases, it was found that hsa-miR-205-5p tended to be highly expressed in the acute phase of SJS / TEN, and hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-5p tended to be highly expressed in the acute phase of DIHS (Figure 4). Statistical tests focusing on SJS / TEN showed that the expression level of hsa-miR-205-5p was significantly increased in the acute phase of SJS / TEN compared to all control skin disease patients except atopic dermatitis / autoimmune blistering disease (Figure 4). On the other hand, for hsa-miR-21-3p, hsa-miR-21-5p, and hsa-miR-222-5p, statistically significant differences in expression between the acute phase of SJS / TEN and autoimmune bullous disease and chickenpox (only hsa-miR-222-3p) were observed. Next, focusing on DIHS, statistically significant differences in expression were observed between hsa-miR-21-3p and all control skin diseases (Figure 4). On the other hand, significant differences in expression were observed between hsa-miR-21-5p and hsa-miR-222-5p and all skin diseases except atopic dermatitis (Figure 4). Almost no clear differences in expression were observed between the acute phase of EM major and control skin diseases.
[0061] (10) Differential diagnostic ability of miRNAs, candidate biomarkers for severe drug rash, from control skin diseases The diagnostic ability of the above four candidate biomarker miRNAs in distinguishing severe drug eruptions from other skin diseases was evaluated. First, the miRNA with the highest diagnostic ability in distinguishing SJS / TEN from control skin diseases was hsa-miR-205-5p. This molecule showed high diagnostic ability with AUROC values exceeding 0.85 in all control skin diseases except autoimmune bullous diseases (Table 4). In addition, hsa-miR-205-5p only had a moderate diagnostic ability (AUROC 0.77) in distinguishing SJS / TEN from autoimmune bullous diseases, whereas hsa-miR-21-3p (AUROC 0.90) and hsa-miR-21-5p (AUROC 0.89) showed good diagnostic ability (Table 4).
[0062] The miRNA that showed the most promising results in distinguishing DIHS from control skin diseases was hsa-miR-21-3p, with AUROC values of 0.9 or more between atopic dermatitis, psoriasis, autoimmune bullous diseases, and chickenpox (Table 4). The AUROC value of hsa-miR-21-3p between DIHS and toxic rash was 0.79, although its diagnostic ability was moderate. Furthermore, this miRNA molecule showed a higher diagnostic ability than the previously reported marker TARC in distinguishing DIHS from atopic dermatitis and autoimmune bullous diseases (Table 4). On the other hand, hsa-miR-21-5p and hsa-miR-222-5p showed a similar diagnostic ability profile to hsa-miR-21-3p, but their AUROC values were similar or slightly lower (Table 4).
[0063] To improve the ability to differentiate between severe drug rash and control skin diseases, we evaluated the diagnostic ability of a diagnostic model that combines three miRNAs or that adds existing markers to the combination. As a result, it was revealed that these diagnostic models are most useful in the differential diagnosis of DIHS from control skin diseases among all severe drug rash disease types (Table 4). This diagnostic model showed a high diagnostic ability of over AUROC value of 0.84 in differentiating DIHS from all control skin diseases. Furthermore, the addition of TARC to this diagnostic model further improved the differential diagnostic ability (Table 4).
[0064] Based on the above results, the miRNAs identified in this study, either alone or in combination, are novel biomarkers for severe drug rash that can specifically detect severe drug rash and assist in distinguishing it from skin diseases that are difficult to distinguish using previously reported markers alone.
[0065] Table 1. Expression ratios of candidate miRNAs for severe drug rash biomarkers in the exploration and validation groups JPEG0007590633000002.jpg38170In this analysis, the expression ratios of candidate miRNAs for severe drug rash biomarkers (median of acute phase severe drug rash cases / median of fully recovered phase cases) that were statistically significantly different in both the exploration sample set and the validation sample set were shown. FC; fold change (expression ratio). p-value; p-value obtained by one-way ANOVA Dunn's test (vs. fully recovered phase cases). NS; not significant.
[0066] Table 2. Diagnostic and severity prediction capabilities of candidate miRNAs for severe drug rash biomarkers in the exploratory group JPEG0007590633000003.jpg96170Statistically significant p-values are highlighted in bold and underlined. Area under the receiver operating characteristic curve (AUROC) values greater than 0.75 are highlighted in bold and underlined. NS; not significant. (i) Using the ΔCq values of hsa-miR-250-5p, hsa-miR-21-3, and hsa-miR-21-5p between the acute phase of severe drug rash and cases of mild drug rash, ROC analysis was performed using the score values obtained from the severe drug rash diagnostic model constructed by multiple logistic regression. (ii) ROC analysis was performed using the score values obtained from the severe drug rash diagnostic model constructed by multiple logistic regression using the delta Cq values of the three miRNAs used in the model in (i) between the severe acute phase and mild drug rash cases and the clinical test value of TARC, a previously reported drug rash marker. Samples for which clinical tests were unavailable (one case of mild drug rash) were excluded from the analysis. (iii) Samples for which quantitative results were unavailable (two convalescent cases) were excluded from the ROC analysis. (iv) Samples for which quantitative results were unavailable (one case of mild drug rash) were excluded from the ROC analysis.
[0067] Table 3. Diagnostic and severity prediction abilities of candidate miRNAs for severe drug rash biomarkers in the validation group JPEG0007590633000004.jpg55170Statistically significant p-values are highlighted in bold and underlined. Area under the receiver operating characteristic curve (AUROC) values greater than 0.75 are highlighted in bold and underlined. NS; not significant. (i) Using the ΔCq values of hsa-miR-250-5p, hsa-miR-21-3, and hsa-miR-21-5p between the acute phase of severe drug rash and cases of mild drug rash, ROC analysis was performed using the score values obtained from the severe drug rash diagnostic model constructed by multiple logistic regression. (ii) Samples for which quantitative results were unavailable (four cases of mild drug rash and all healthy adults) were excluded from the ROC analysis. (iii) ROC analysis was performed using the score obtained from the severe drug rash diagnostic model constructed by multiple logistic regression analysis using the delta Cq values of the three miRNAs used in the model in (i) between the severe acute phase and mild drug rash cases and the clinical test value of TARC, an existing drug rash marker. Cases for which clinical test results were unavailable (4 cases of mild drug rash and all healthy adults) were excluded from the analysis.
[0068] Table 4. Differential diagnostic ability of candidate miRNAs for severe drug rash biomarkers against control skin diseases JPEG0007590633000005.jpg58170Statistically significant p-values are highlighted in bold and underlined. Area under the receiver operating characteristic curve (AUROC) values greater than 0.75 are highlighted in bold and underlined. NS; not significant. (i) Subjects for whom TARC laboratory test values were unavailable (atopic dermatitis, 9 cases; psoriasis, 9 cases; chickenpox, 9 cases; toxic dermatitis, 4 cases) were excluded from the ROC analysis. (ii) Using each of the severe drug rash diagnostic models constructed in Table 3, the discriminatory diagnostic ability between each severe drug rash disease type and other control skin diseases was evaluated using ROC curves. [Industrial Applicability]
[0069] The present invention can be used for in vitro diagnostics, clinical tests, and the like. [Sequence List Free Text]
[0070] <Sequence number 1> The nucleotide sequence of hsa-miR-205-5p is shown. <Sequence number 2> The nucleotide sequence of hsa-miR-21-3p is shown. <Sequence number 3> The nucleotide sequence of hsa-miR-21-5p is shown. <Sequence number 4> The nucleotide sequence of hsa-miR-222-3p is shown. <Sequence number 5> The nucleotide sequence of hsa-let-7d-3p is shown. <Sequence number 6> The base sequence of hsa-miR-23b-3p is shown. <Sequence number 7> The nucleotide sequence of hsa-miR-24-3p is shown. <Sequence number 8> The nucleotide sequence of hsa-miR-30d-5p is shown. <Sequence number 9> The nucleotide sequence of hsa-miR-361-5p is shown. <Sequence number 10> The nucleotide sequence of cel-miR-54-3p is shown.
Claims
1. A method for testing for severe drug rash, comprising measuring gene expression in a sample derived from a subject for at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, and hsa-miR-222-3p.
2. The method according to claim 1, wherein the miRNA whose expression is measured is at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p and hsa-miR-222-3p, and the measured value assists in diagnosing the progression of severe drug rash.
3. The method according to claim 1, wherein the miRNA whose expression is measured is at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p and hsa-miR-222-3p, and the measured value assists in predicting the severity of a severe drug rash.
4. The method according to claim 1, wherein the miRNA whose expression is measured is at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p and hsa-miR-222-3p, and the measured value assists in distinguishing between a severe drug rash and other skin diseases.
5. The method according to claim 2, wherein the miRNAs whose expression is to be measured are at least one combination of miRNAs selected from the group consisting of the following combinations of miRNAs: (i) Combination of hsa-miR-205-5p and hsa-miR-21-3p (ii) combination of hsa-miR-205-5p and hsa-miR-21-5p (iii) combination of hsa-miR-205-5p and hsa-miR-222-3p (iv) combination of hsa-miR-21-3p and hsa-miR-21-5p (v) combination of hsa-miR-21-3p and hsa-miR-222-3p (vi) combination of hsa-miR-21-5p and hsa-miR-222-3p (vii) a combination of hsa-miR-205-5p, hsa-miR-21-3p and hsa-miR-21-5p (viii) combination of hsa-miR-205-5p, hsa-miR-21-3p, and hsa-miR-222-3p (ix) combination of hsa-miR-205-5p, hsa-miR-21-5p and hsa-miR-222-3p (x) combination of hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p (xi) combination of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p
6. The method according to claim 3, wherein the miRNAs whose expression is to be measured are at least one combination of miRNAs selected from the group consisting of the following combinations of miRNAs: (i) Combination of hsa-miR-205-5p and hsa-miR-21-3p (ii) combination of hsa-miR-205-5p and hsa-miR-21-5p (iii) combination of hsa-miR-205-5p and hsa-miR-222-3p (iv) combination of hsa-miR-21-3p and hsa-miR-21-5p (v) combination of hsa-miR-21-3p and hsa-miR-222-3p (vi) combination of hsa-miR-21-5p and hsa-miR-222-3p (vii) a combination of hsa-miR-205-5p, hsa-miR-21-3p and hsa-miR-21-5p (viii) combination of hsa-miR-205-5p, hsa-miR-21-3p, and hsa-miR-222-3p (ix) combination of hsa-miR-205-5p, hsa-miR-21-5p and hsa-miR-222-3p (x) combination of hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p (xi) combination of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p
7. The method according to claim 4, wherein the miRNAs whose expression is to be measured are at least one combination of miRNAs selected from the group consisting of the following combinations of miRNAs: (i) Combination of hsa-miR-205-5p and hsa-miR-21-3p (ii) combination of hsa-miR-205-5p and hsa-miR-21-5p (iii) combination of hsa-miR-205-5p and hsa-miR-222-3p (iv) combination of hsa-miR-21-3p and hsa-miR-21-5p (v) combination of hsa-miR-21-3p and hsa-miR-222-3p (vi) combination of hsa-miR-21-5p and hsa-miR-222-3p (vii) a combination of hsa-miR-205-5p, hsa-miR-21-3p and hsa-miR-21-5p (viii) combination of hsa-miR-205-5p, hsa-miR-21-3p, and hsa-miR-222-3p (ix) combination of hsa-miR-205-5p, hsa-miR-21-5p and hsa-miR-222-3p (x) combination of hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p (xi) combination of hsa-miR-205-5p, hsa-miR-21-3p, hsa-miR-21-5p and hsa-miR-222-3p
8. A kit for testing for severe drug rash, comprising a reagent capable of measuring gene expression in a sample derived from a subject for at least one miRNA selected from the group consisting of hsa-miR-205-5p, hsa-miR-21-3p, and hsa-miR-222-3p.
9. The kit described in claim 8, further comprising a reagent capable of measuring gene expression of hsa-miR-21-5p in a sample derived from a subject.
Citation Information
Patent Citations
Evaluation method of risk of onset of severe drug eruption accompanied by ophthalmic complications
JP2015107095A
MIR-21-3p inhibitors in skin disorders
US20170175112A1
