WO2020234072A1

WO2020234072A1 - Pcos diagnosis

Info

Publication number: WO2020234072A1
Application number: PCT/EP2020/063278
Authority: WO
Inventors: Francis De Zegher; Lourdes IBAÑEZ
Original assignee: Katholieke Universiteit Leuven; Hospital Sant Joan De Deu
Priority date: 2019-05-17
Filing date: 2020-05-13
Publication date: 2020-11-26

Abstract

The invention relates to the in vitro use of reagents for measuring the expression level of at least miR-451a in a body sample, in the diagnosis of PCOS (PolyCystic Ovary Syndrome) in an adolescent girl between 10 and 24 years.

Description

PCOS DIAGNOSIS

FIELD OF THE INVENTION

The invention relates to methods of diagnosing PolyCystic Ovary Syndrome (PCOS), methods of detecting responders and non-responders to a PCOS therapy and methods to monitor the efficacy of such therapy.

The invention further relates to miRNA based asays in performing the above methods.

BACKGROUND OF THE INVENTION

Polycystic ovary syndrome (PCOS) is a prevalent disorder in adolescent girls and young women, commonly presenting with hirsutism and menstrual irregularity [Ibanez et at. (2014) Nat Rev Endocrinol 10, 499-508; Ibanez et at. (2017) Horm Res Paediatr 88, 371-395], and frequently associated with comorbidities in adulthood, including subfertility, obesity, type 2 diabetes and cardiovascular disease. PCOS is thought to result from the interaction of many factors, the latest evidence pointing to a key role for hepato-visceral fat excess which is usually preceded by a mismatch between (reduced) prenatal and (augmented) postnatal weight gain [de Zegher et at. (2018) Trends Endocrinol Metab 29, 815-818]. The combination of lifestyle measures and medications that reduce hepato-visceral adiposity, for example, a low-dose combination of spironolactone, pioglitazone and metformin (SPIOMET), normalizes the PCOS phenotype more than does a treatment with an oral contraceptive (OC) [Ibanez et a/. (2017) J Ado/esc Health 61, 446-453].

The sequence of events leading to PCOS may be subject to genetic and epigenetic modulations influencing both the phenotype and possibly the outcome during and after intervention. Indeed, different loci related to PCOS have been identified by genome-wide association studies (GWAS) but together only explain < 10% of heritability. For this reason, there is mounting interest on the effects of epigenetic regulation - and in particular on the effects of microRNAS (miRNAs) - in the phenotypic variation of PCOS. miRNAs are single-stranded, small non-coding RNAs aproximately 19-25 nucleotides in length that bind to the 3'UTR region of mRNAs inhibiting their translation, and thus working as negative posttranscriptional regulators of gene expression. miRNAs play pivotal roles in numerous biological processes, and have emerged as diagnostic biomarkers in different diseases. In women with PCOS, miRNA expression has been reported to be altered in serum, adipose tissue, follicular fluid, and in granulosa and theca cells [Long et al. (2014) Cell Physiol Biochem 33, 1304-1315; Song et al. (2016) PLoS One 11, e0163756; Chen et al. (2013) Diabetes 62, 2278-2286; Naji et a/. (2017) Sci. Rep. 7, 14671; Xue et at. (2018) J Cell Biochem 119, 3913-3921]. However, those studies have been performed long beyond adolescence, in heterogeneous populations (due to the use of different criteria for PCOS diagnosis), and only in cross-sectional designs. Several miRNAs have been found in the prior art to be involved in the pathophysiology of diabetes, and specifically in the impaired glycemic homeostasis. Patients with type 2 diabetes display decreased levels of circulating miR-206 and miR-451a [Ding et a/. (2016) Biosci Biotechnol Biochem 80, 461-465], whereas miR- 106, miR-652 and miR-451a have been reported to increase insulin sensitivity via different mechanisms; for example, miR-106 ameliorates hyperglycemia and vascular endothelial cell dysfunction and promotes adipogenesis; miR-652 purportedly increases insulin-stimulated lipogenesis in adipocytes and miR-451a inhibits hepatic gluconeogenesis by down-regulating glycerol-kinase. A reduced presence of miR-451a has been documented in the fatty liver of ob/ob mice [a type 2 diabetes animal model with spontaneous non-alcoholic fatty liver disease (NAFDL) development], in mice with high fat diet (HFD)-induced non-alcoholic steatohepatitis (NASH), in rats receiving a HFD or high-fructose diet or a combination of both, and also in humans with NASH. Conversely, overexpression of miR-451a is able to decrease triglyceride accumulation in the liver of mice, and also in cultured hepatic cells [Zeng et at. (2018) Mol Cell Endocrinol 474, 260-271].

SUMMARY OF THE INVENTION

The present invention discloses a miRNA profile of adolescent girls with PCOS, distinct from that of healthy control girls, and reflecting the [perpetuating vs normalizing] effects of interventions [respectively with an oral contraceptive (OC) vs with a low- dose combination of spironolactone-pioglitazone-metformin (SPIOMET) for 1 year] on the PCOS-underpinning pathophysiology.

The miRNA profiling was performed by RNA sequencing, differentially expressed miRNAs being validated by qRT-PCR in 13 control and 31 PCOS girls.

Girls with PCOS had markedly reduced concentrations of circulating miR-451a, miR- 652-3p, miR-106b-5p and miR-206; pathway enrichment analysis showed that these miRNAs target genes involved in energy homeostasis and cell-cycle control. In the present study, miR-451a could diagnose PCOS (versus a healthy condition) with 100% sensitivity and 100% specificity. SPIOMET (but not OC) was accompanied by on-treatment normalization of the miRNA profile in PCOS girls; miR-451a concentrations after 1 year on OC or SPIOMET treatment associated closely (r=0.66; P<0.0001) with post-treatment ovulation rates (assessed by salivary progesterone). Circulating miR-451a is a biomarker suitable to guide diagnosis and treatment of PCOS in adolescent girls.

The invention relates to the use miRNA expression as a marker for diagnosing PCOS and monitoring the efficacy of a PCOS treatment.

The invention is further summarised in the following statements:

I.In vitro use of reagents for measuring the expression level of at least miR-451a in a body sample, in the diagnosis of PCOS in an adolescent girl.

The use implies the use of reagents such as probes and PCR primers for detecting the expression level of miRNA in a biological sample such as blood.

2. The use according to statement 1, wherein the adolescent girl is non-obese.

3. The use according to statement 1 or 2, wherein the reagents are oligonucleotides for performing RT-PCR.

4. The use according to any one of statements 1 to 3, further comprising reagents for measuring the expression level of one more of miR-106b-5p, miR-206 and miR-652- 3p.

5. In vitro use of reagents for measuring the expression level of of at least miR-451a in a body sample, in monitoring the efficacy of a PCOS treatment in an adolescent girl.

6. The use according to statement 5, wherein the PCOS treatment is a treatment with with oestro-progestagen contraceptive such spironolactone, pioglitazone and metformin (SPIOMET).

The present invention allows thus to monitor the efficacy whereby a treatment normalizes the underpinning PCOS pathophysiology in an adolescent girl with PCOS.

7. The use according to statement 5 or 6, wherein the adolescent girl is non-obese.

8. The use according to any one of statement 5 to 7, further comprising the use of reagents for measuring the expression level of miR-106b-5p and/or miR-652-3p.

9. The use according to any one of statements 5 to 8, wherein the monitoring identifies responders or non responders to said treatment.

10. The use according to anyone of statements 5 to 9, wherein an increase of the miRNA expression level of said at least one miRNA, compared to the level prior to the treatment is indicative of the patient being a responder.

I I. The use according to statement 10, wherein the increase is a rise of the circulating concentrations of said at least one miRNA, above the -2 SD limit of a reference range. 12. The use according to any one of statements 5 to 11, wherein the monitoring is performed prior to the occurrence of physical changes due to the spiomet treating.

13. The use according to statement 12, wherein the physical change is a normalisation of the ovulation rate.

The methods of the invention thus allow the miRNA based monitoring independently of other assessments such as determining the normalisation of the ovulatory functione.

14. In vitro method of diagnosing PCOS in an adolescent girl, comprising the steps of:

determining in a body sample of said adolescent girl, the level of at least miR-451a, comparing the determined level with a reference value of a healthy adolescent girl or a reference value obtained by combining reference values from a groups of, healthy adolescent girl,

determining from said comparison, whether said adolescent girl is a PCOS patient. 15. The method according to statement 14, wherein a low expression level of said at least one miRNA in the sample compared to said reference value is indicative for said adolescent girl being a PCOS patient.

16. In vitro method for monitoring the efficacy of a PCOS treatment in an adolescent girl, comprising the steps of:

determining in a body sample of said adolescent girl prior to or at the onset of said treatment, a first expression level of at least miR-451a,

determining during the treatment in a body sample of said adolescent girl, a second expression level of at least miR-451a,

comparing the first and the second expression level of said miRNA, wherein an increase of the second expression level, compared to the first expression level is indicative of the adolescent girl being a reponder to said treatment.

The method can be equally performed when the first measurement is made when the treatment is already ongoing, for example after 1, 2 or 3 weeks after the start of the treatment, or for example after 1, 2, or 3 months after the start of the treatment. A change in miRNA compared with a later measurement caan then be equally determined.

17. The method according to statement 16, wherein the treatment is a treatment with oestro-progestagen contraceptive such as spironolactone, pioglitazone and metformin (SPIOMET). 18. The method according to statement 16 or 17, wherein the step of determining the second level is performed between 1 and 12 months ((eg 1, 2, 4, 6 9 or 12 months), after the onset of said treatment or after the first measurement.

DETAILED DESCRIPTION

Description of the figures

Figure 1

miRNAs differentially expressed in adolescent girls with PCOS as compared to control girls. The results are expressed as the ratio of the average expression fold change (FC) in girls with PCOS versus control girls (Z-score). miRNAs depicted in dark bars are those confirmed in the entire study population (n= 13 control girls and 31 girls with PCOS). P-values are adjusted for multiple testing.

Figure 2

Panel A: Relative expression of selected candidate miRNAs in adolescent girls with PCOS (n=31) at baseline. Results are expressed as Z-scores derived from the relative expression in healthy control girls (n= 13). *** P<0.0001; ** P=0.007

Panel B: Correlations between the circulating concentrations of differentially expressed miRNAs and a selection of clinical, biochemical and imaging markers across the study population of 13 healthy controls and 31 girls with PCOS. Correlations with P<0.001 are shown in bold.

Figure 3

Panel A: Longitudinal changes in the relative expression of circulating miR-451a, as expressed in Z-score, in girls with PCOS, who were randomized to receive an oral contraceptive (OC, n= 16, lower line), or low-dose spironolactone, pioglitazone and metformin (SPIOMET, n= 15, upper line) for 1 year, and who received no treatment over the subsequent year. miRNA Z-score increases towards normal in the SPIOMET subgroup, whereas no changes are observed in the OC subgroup.

The vertical box depicts the active treatment phase; the upper and lower limits of the horizontal box correspond, respectively, to a Z-score of + 1 and -1 in control girls (n= 13).

* Between group differences at time 1 yr and 2 yr (p<0.0001; by two-sided t-test) *Within group differences for changes between consecutive timepoints (p<0.0001; by one-way-ANOVA)

Panel B: Post-treatment number of ovulations over 6 months (mo 15-18 + mo 21- 24) vs expression levels of circulating miR-451a after intervention in girls with PCOS who were randomized to receive an OC (n= 16, lower box), or low-dose SPIOMET (n= 15, upper box) for 1 yr.

Figure 4

Girls with PCOS with spare samples in whom miRNAs assessment was performed.

Figure 5

Venn diagram depicting the miRNAs expressed in the serum of untreated girls with PCOS and healthy control girls.

Figure 6 Longitudinal changes in serum relative expression levels of miR-652-3p, miR-106-5p and miR-206 in girls with PCOS, who were randomized to receive an oral contraceptive (OC, n= 16, lower line), or low-dose spironolactone, pioglitazone and metformin (SPIOMET, n= 15, upper line) for 1 year, and who received no treatment over the subsequent year.

The vertical dotted box represents the active treatment phase; the upper and lower limits of the box correspond, respectively, to a Z-score of + 1 and -1 in healthy control girls (n = 13).

Panel A: miR-652-3p Z-score increased significantly in the SPIOMET subgroup after

1 year. *p=0.0004 by one-way ANOVA; #P<0.0001 between subgroups after 1 and

2 years, by two-sided t-test.

Panel B: miR-106-5p Z-score increased significantly in the SPIOMET subgroup after

1 year. *p<0.0001 by one-way ANOVA; #P<0.0003 between subgroups after 1 and

2 years, by two-sided t-test.

Panel C: miR-206 Z-score decreased significantly in the OC subgroup over 2 years. *p<0.0001 by one-way ANOVA; #P<0.0001 between subgroups after 1 and 2 years, by two-sided t-test.

Figure 7 Receiver operating characteristics (ROC) curves (panels A to D) and summary (Panel E) of validated miRNAs for discriminating girls with PCOS from healthy controls.

"Polycystic ovary syndrome (PCOS)" is a prevalent disorder in adolescent girls, commonly driven by hepato-visceral fat excess, usually presenting with hirsutism and menstrual irregularity, and often followed by subfertility and type 2 diabetes. Criteria for diagnosis of PCOS in adolescent girls are described in detail in [Ibanez et at. (2017) Horm Res Pediatr 88, 371-395] and include irregular menses or oligomenorrhea, biochemical and-clinical evidence of hyperandrogenism.

"Adolescence" refers to the age range between 10 and 24 years more typically between 10 and 19 years. "Non-obese" refers to a body mass index below 30, typically below 27,5 and more typically below 25.

"Responder" refers to a PCOS patient wherein a treatment is accompanied by a rise of the circulating concentrations of a particular miR, or of a particular combination of miRs, typically above the -2 SD limit of a reference range.

"expression level" may be determined by measuring the amount of microRNA in the sample fluid using a variety of suitable reagents. The expression level of the microRNA can be determined, for example, with an assay for global gene expression in a biological fluid (e.g. using a microarray assay for microRNA expression profiling analysis, or a ready-to-use microRNA qPCR plate), or by specific detection assays, such as quantitative PCR, quantitative reverse-transcription (real-time) PCR (qRT- PCR), locked nucleic acid (LNA) real-time PCR, or northern blotting. The measurement of the expression level of a microRNA in a biological fluid may be carried out with an oligonucleotide probe specific for the detection of said microRNA. Said oligonucleotide probe may bind directly and specifically to the microRNA, or may specifically reverse transcribe said microRNA. Alternatively, said oligonucleotide probe may bind a cDNA obtained from said microRNA. Said oligonucleotide probe may also amplify a cDNA obtained from said microRNA.

"kit" refers to any combination of reagents or apparatus that can be used to perform a method of the invention. Kits for use in the present invention comprise probes for detection of a limited amount of dedicated miRNA (up to 10, 15 or 20 miRNA) to distinguish over arrays containing probes for more than 100 or 1000 miRNA. Apart from the miRNA as recited in the claims the dedicated miRNA may comprise housekeeping miRNA or exogenous miRNA from other species that serve as control. The kit may also comprise instructions for use to diagnose whether a subject classifies as a responder to the IL10, PINS, anti-CD3 diabetes therapy.

"oligonucleotide probe" refers to a short sequence of nucleotides that match a specific region of a microRNA or a cDNA obtained from said microRNA, or fragments thereof, and then used as a molecular probe to detect said microRNA or cDNA sequence.

An oligonucleotide probe "specific for the detection of a microRNA" for example refers to an oligonucleotide probe that bind directly and specifically to a microRNA or a fragment thereof, or specifically reverse transcribe said microRNA. Alternatively, said oligonucleotide probe may bind specifically to a cDNA obtained from said microRNA. Said oligonucleotide probe may also specifically amplify a cDNA obtained from said microRNA. "control miRNA" refers to one or more miRNA which expression is determined using a similar, preferably identical, methodology, as the experimental miRNA. As illustrated in the examples, the expression of numerous miRNA does not differ between responders and non-responders. One or more of these miRNA can be used as internal control. Alternatively, an miRNA (e.g. a non-human) is added to a sample and acts as an external control.

Alternatively a reference is obtained by using a sample or a pool of samples from a population of validated responders or by using a sample or a pool of samples from a population of validated non-responders. Experimental data can be compared with control data and classified as belonging to the responder or non-responder group. Compared with a control group of non-responders a non-responder will have about equal expression levels, and a responder will have lower expression levels. Equally, compared with a control group of responders a responder will have about equal expression levels, and a non-responder will have higher expression levels.

General techniques useful in microRNA detection are disclosed in "MicroRNA Expression Detection Methods", Wang Zhiguo, Yang Baofeng, 2010, XX; "Circulating MicroRNAs Methods and Protocols", series: Methods in Molecular Biology, Vol. 1024 Ochiya, Takahiro (Ed.) 2013.

Determination of expression values obtained by qPCR can be expressed as ^-dct values as explained for example in Livak 8i Schmittgen (2001) Methods 25, 402- 408. This value represents the fold change in expression of the target miRNA relative to a control miRNA.

The term "statistically significant" differences between the groups studied, relates to condition when using the appropriate statistical analysis (e.g. Chi-square test, t- test) the probability of the groups being the same is less than 5%, e.g. p<0,05. In other words, the probability of obtaining the same results on a completely random basis is less than 5 out of 100 attempts.

Any technique known to one of skill in the art for detecting and measuring RNA expression levels can be used in accordance with the methods described herein. Non limiting examples of such techniques include microarray analysis, Northern blotting, nuclease protection assays, RNA fingerprinting, polymerase chain reaction, ligase chain reaction, Qbeta replicase, isothermal amplification method, strand displacement amplification, transcription based amplification systems, quantitative nucleic acid amplification assays (e.g., polymerase chain reaction assays), combined reverse transcription/nucleic acid amplification, nuclease protection (SI nuclease or RNAse protection assays), Serial Analysis Gene Expression (SAGE), next generation sequencing, gene expression microarray, as well as other methods.

The probe can be labelled by any of the many different methods known to those skilled in this art. The labels most commonly employed for these studies are radioactive elements, enzymes, chemicals that fluoresce when exposed to ultraviolet light, and others. A number of fluorescent materials are known and can be utilized as labels. These include, but are not limited to, fluorescein, rhodamine, auramine, Texas Red, AMCA blue and Lucifer Yellow. The radioactive label can be detected by any of the currently available counting procedures. Non-limiting examples of isotopes include ³H, ¹⁴C, ³²P, ³⁵S, ⁵¹Cr, ⁵⁷Co, ⁵⁸Co, 59Fe, ⁹⁰Y, ¹²⁵I, ¹³¹I, and ¹⁸⁶Re.

The present invention discloses a serum miRNA profile of girls with PCOS versus healthy controls, the longitudinal changes of the differentially expressed miRNAs in PCOS girls during randomized interventions, and their association with endocrine- metabolic and imaging markers. The miRNA signature of non-obese adolescents with PCOS was reveals a down-regulation of miR-106b-5p, miR-206, miR-652-3p, and particularly of miR-451a, which distinguished most between healthy controls and girls with PCOS. Randomized interventions over one year disclosed that treatment with SPIOMET normalized those strikingly low miRNA concentrations, but that treatment with an OC failed to do so. More normal on-treatment miR-451a levels were followed by more normal post-treatment ovulation rates. SPIOMET-related increments in miR- 106b-5p, miR-652-3p and miR-451a concentrations were partly maintained after treatment discontinuation.

The PCOS-discerning miRNAs identified herein in adolescent girls with PCOS were not observed in prior art consecutive studies of women with PCOS [Song et al. (2016) PLoS One 11, e0163756; Naji et al. (2017) Sci. Rep. 7, 14671; Jiang et al. (2016) Endocrine 53, 280-290; Sorensen et al. (2016) Chem Biol Interact 259 (Pt A), 8- 16; Eisenberg et al. (2017) Fertil Steril 107, 269-275; Li et al. (2019) Mol Hum Reprod. 25, 638-646 ; Kong et al. (2019) Mol Cell Endocrinol 486, 18-24].

Differences in diagnostic criteria and the heterogeneity of the populations assessed in the prior art (ethnic group; age; previous and/or ongoing treatments; presence of obesity and/or diabetes) apparently result in different phenotypes adding further variety to the syndrome. In the present invention, the differentially expressed miRNAs, in particular miR-451a and miR-652-3p, were closely and inversely related to androgen excess, and to the degree of mismatch between (reduced) prenatal and (augmented) postnatal weight gain. Given that these miRNAs did not differ between healthy women and PCOS women, the present findings corroborate the concept that adolescent PCOS and adult PCOS may have different features [Ibanez et a/. (2017) Horm Res Paediatr 88, 371-395], possibly because adolescent PCOS relates more closely (than adult PCOS) to central obesity, and to a mismatch between prenatal and postnatal weight gain [de Zegher et al. (2018) Trends Endocrinol Metab 29, 815- 818; de Zegher et al. (2017) Obesity (Silver Spring) 25, 1486-1489].

The present invention shows that miR-451a is the most down-regulated miR in PCOS girls, and is the most differentiating between healthy controls and PCOS girls, and is a biomarker contributing to guide PCOS diagnosis. In addition, circulating miR-451a associated closely with hepato-visceral fat, which is thought to be a key PCOS driver, and is shown in the present invention to be the most up-regulated miRNA after SPIOMET treatment, increasing almost 4-fold. Relatively elevated levels of miR-451a were observed after 1 year on SPIOMET treatment (and even longer), and these levels associated with higher post-treatment ovulation rates. These observations indicate that circulating miR-451a is a biomarker integrating the broad spectrum of SPIOMET-associated normalizations.

The present invention discloses that adolescent girls with PCOS display a distinct miRNA profile as compared to healthy control girls; SPIOMET treatment over one year normalizes the expression of several down-regulated miRNAs related to glucose homeostasis, energy metabolism, and the control of the cell cycle; miR-451a was the most up-regulated miR after SPIOMET treatment and showed the highest sensitivity and specificity to differentiate girls with PCOS from healthy controls, indicating its use as a biomarker for PCOS diagnosis and treatment outcome.

EXAMPLES

Example 1. Research Design & Methods

Study Population

The original study population consisted of 36 non-obese adolescent girls with PCOS who participated in a randomized, open-label, controlled trial (ISRCTN29234515), whose primary endpoint was ovulation rate after OC or SPIOMET intervention [Ibanez et al. (2017) J Adolesc Health 61, 446-453] at the Adolescent Endocrinology Unit of Sant Joan de Deu University Hospital, Barcelona, Spain. Longitudinal measurement of circulating miRNAs was a secondary endpoint, and was performed in spare serum samples available from a total of 31 girls (mean age, 15.7 yr; BMI, 23.1 Kg/m²; flow chart, Figure 4).

As described in Ibanez et al. (2017) J Adolesc Health 61, 446-453, the inclusion criteria were: hirsutism (score of 8 or more on modified Ferriman 8i Gallwey scale); oligomenorrhea (menstrual intervals spanning more than 45 days); menarche at least 2 years before study onset; no need for contraception (no sexual activity). Exclusion criteria were congenital adrenal hyperplasia due to 21 -hydroxylase deficiency; glucose intolerance or diabetes mellitus; thyroid, liver, or kidney dysfunction; hyperprolactinemia; use of medications affecting gonadal or adrenal function, or carbohydrate or lipid metabolism. The girls were randomized to receive for 1 year an OC containing 20 pg of ethinylestradiol plus 100 mg of levonorgestrel for 21/28 days (and placebo for 7/28 days) or SPIOMET, a low-dose combination of spironolactone 50 mg/d, pioglitazone 7.5 mg/d, and metformin 850 mg/d, and were followed in the subsequent year off treatment. All girls were asked to adhere to a Mediterranean diet, and to perform exercise regularly.

Thirteen age- and BMI-matched, healthy girls (mean age, 15.7 yr) who were recruited in nearby schools served as controls. None was hirsute or taking medications, and all had regular cycles.

Auxology, endocrine-metabolic variables, ovulation, body composition and hepato- visceral fat assessment

Birth weight, birth length, and BMI (and their Z-scores) were retrieved from medical records; endocrine-metabolic variables and carotid intima-media thickness (cIMT) were assessed as described in Ibanez et al. (2017) J Adolesc Health 61, 446-453. HOMA-insulin resistance (HOMA-IR) was calculated as [fasting insulin in mU/L] x [fasting glucose in mg/dl_]/405. Ovulations were derived from salivary progesterone measurements, combined with the information contained in the menstrual diaries kept by the girls. Salivas were obtained weekly during the second and fourth trimesters of the post-treatment year over 12 consecutive weeks in each trimester, as reported in Ibanez et al. (2017). Progesterone was measured by ELISA (Novatec, Inmundiagnostica, Dietzenbach, Germany); intra- and inter-assay coefficients of variation were 5.5% and 6.0%.

Body composition was assessed by dual X-ray absorptiometry with a Lunar Prodigy and Lunar software (version 3.4/3.5, Lunar Corp, WI); abdominal fat (subcutaneous and visceral) and hepatic fat were assessed by magnetic resonance imaging (MRI) with a multiple- slice MRI 1.5 Tesla scan (Signa LX Echo Speed Plus Excite, General Electric, Milwaukee, WI). Central fat was arbitrarily defined as the sum of visceral fat (in cm²) and hepatic fat (in %).

Small RNA sequencing

The PCOS-discerning miRNAs [n=6 from control girls and 12 from girls with PCOS (n=6 randomized to OC, and n=6 randomized to SPIOMET)] were selected for miRNA deep sequencing, performed at Making Genetics (Villava, Navarra, Spain; www.makinQ-Qenetics.eu). Small cDNA libraries were generated from 200 ng of total RNA using a TruSeq Small RNA Library Prep Kit (Illumina, San Diego, USA), according to the manufacturer's instructions. Briefly, small RNA fractions consisting of 16-26 nucleotides were purified from total RNA and enriched using denaturing polyacrylamide gel electrophoresis (PAGE). Adapters were ligated at 3' and 5' ends of the enriched fragments using T4 ligase and then amplified by RT-PCR. cDNA libraries were purified with an automated agarose gel separation system (Labchip XT, PerkinElmer, Waltham, MA, USA) and sequenced on a Hiseq 2500 system (Illumina, San Diego, CA, USA), using default parameters (single read lx50bp). The expression of miRNAs in each library was extracted from the RNA-seq data using featureCounts function available in Bioconductor R package Rsubread [Liao et at. (2013) Nucleic Acids Res 41, el08; Liao et a/. (2014) Bioinformatics 30, 923-930], after alignment on GRCh38 with Hisat2 of clipped/trimmed reads, as provided by the manufacturer. The data were filtered, and only those genes with more than 1 reads per million mapped reads in at least two libraries were kept in the analysis; intra- and inter sample normalization was done using the trimmed mean of M-values (TMM) method implemented in the edgeR Bioconductor package [Robinson et al. (2010) Bioinformatics 26, 139-140; McCarthy et al. (2012) Nucleic Acids Res 40, 4288- 4297]. After pre-processing, linear model fitting and differential miRNA expression analysis were performed using the eBayes moderated t-statistic by limma package and individual miRNAs P values were derived [Ritchie et a/. (2015) Nucleic Acids Res 43, e47]. For each miRNA the log2 fold-change (FC) was calculated from readcount. The results are expressed as the ratio of the average expression in girls with PCOS versus the average expression in control girls (Z-score). The resulting P values were adjusted using the Benjamini-Hochberg procedure and a False Discovery Rate (FDR) cut-off of 0.05 was used as statistical significant threshold. A miRNA was considered to be differentially expressed if its adjusted P value was <0.01. In order to confirm the RNA sequencing results, the most differentially expressed miRNAs were validated in the entire study population (13 controls, 31 girls with PCOS; Figure 4) by RT-qPCR (see below for details). RNA extraction

Total RNA was isolated from 300 mI of serum using the mirVana PARIS kit (Life technologies, Ambion®, Waltham, MA, USA) according to the manufacturer's instructions. The quantity of the isolated RNA was determined with a Nanodrop ND- 100 spectrophotometer (nanodrop Technologies, Wilmington, DE), and the purity and integrity of each RNA was checked with an Agilent Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA). Isolated RNA samples were frozen and kept at -80°C.

Multiple miRNA Reverse Transcription

miRNAs were reverse-transcribed using the TaqMan MicroRNA Reverse Transcription Kit (Thermo Fisher Scientific, Waltham, MA, USA), and the associated miRNAs- specific stem-loop primers (TaqMan microRNA assay kit, #4427975), with minor modifications. A customized RT primer pool was prepared by pooling all miRNAs- specific primers of interest. In brief, miRNA-specific primers were pooled and diluted in lX Tris-EDTA (TE) buffer to obtain a final dilution of 0.05X each; 6 mI of this mixture was added to the reaction mix containing 0.3 mI 100 mM dNTP, 3 mI enzyme (50 U/mI), 1.5 mI 10X RT buffer, 0.19 mI RNase inhibitor (20u/mI) and 3 mI of serum RNA. A final volume of 15 mI was reverse-transcribed under the following conditions: 30 min at 16°C to anneal primers, 30 min at 42°C for the extension phase, and 5 min at 85°C to stop the reaction. cDNA was then stored at -20°C.

Pre-amplification and multiple cDNAs pre-amplification mix

In order to increase the amount of cDNA and to improve the sensitivity of the TaqMan qPCR reaction, a pre-amp step was performed on serum cDNAs. Pre-amp PCR conditions consisted in 10 min at 95°C, 2 min at 55°C, and 2 min at 72°C, followed by 12 cycles of 15 s at 95°C, 4 min at 60°C and a final step of 10 min at 99.9°C. At the end of the run, the pre-amp products were diluted 4X in 0.1X TE buffer pH 8.0, and stored at -20°C.

To simultaneously pre-amplify multiple cDNAs, a pre-amp primer pool was built targeting the same miRNAs that were reverse-transcribed, and containing 10 mI of each individual 20X TaqMan small RNA assay (part of #4427975), diluted in 1000 mI IX TE. The reaction mix was prepared by combining 3.75 mI of pre-amp primer pool with 2.5 mI of RT product, 12.5 mI of TaqMan PreAmp Master Mix (2X) (#4391128) and 6.25 mI Nuclease-free water.

Real-Time quantitative PCR

All RT-qPCR were performed on an ABI PRISM 7500 thermal cycler from Applied Biosystems with the following conditions: one denaturing step at 95°C for 10 min, followed by 45 cycles consisting of denaturing at 95°C for 15 s, and annealing and elongation at 60°C for 60 s, followed by an inactivation step of 10 min at 99.9°C. Real-time PCR reactions with multiple cDNAs were performed in a 20 pi final volume. A reaction mix containing 5 mI of diluted pre-Amp cDNAs, 1 mI of 20X TaqMan Small RNA assay (#4427975) and 10 mI 2X Taq-Man Universal PCR Master Mix, no UNG (#4304437) was loaded in each well. Samples were normalized to the housekeeping gene U6 snRNA. Relative expression was calculated according to the 2 ^ACT method. Statistical Analysis & Ethics

Statistical analysis were performed with the GraphPad Software (La Jolla, California, USA). Results are expressed as mean ± SEM. For comparisons within and between groups at each time point, two-tailed Student's t-test was used. A one-way-ANOVA with post-hoc Bonferroni's test was performed to assess longitudinal changes in the levels of expression of miRNAs within groups. A receiver operating characteristic (ROC) curve was plotted for each validated miRNA to determine their discriminating effect. The sensitivity and specificity of detecting cases and controls were assessed by the area under curve (AUC) and 95% confidence interval. The cut-off value for each miRNA was obtained using Youden's index. P<0.05 was considered statistically significant.

The study was approved by the Institutional Review Board of Hospital Sant Joan de Deu, Barcelona, after written consent by the parents and after assent by each of the study girls.

EXAMPLE 2. Effects of OC versus SPIOMET on clinical , endocrine-metabolic and imaging parameters

The characteristics of the study population are summarized in Table 1.

OC and SPIOMET treatment had no detectable effects on BMI, bone mineral density, lean mass or total fat, and both treatments attenuated the markers of androgen excess comparably. However, they had divergent effects on cardio-metabolic risk markers, as judged by the reduction in HOMA-IR, C-reactive protein, cIMT, hepato- visceral fat, and by the increase in HDL-cholesterol, and high-molecular-weight- adiponectin levels with SPIOMET but not with OC intervention, as reported in Ibanez et al. (2017) J Adolesc Health 61, 446-453.

Table 1. Clinical, endocrine-metabolic, and imaging data data from healthy control girls (n= 13) and from girls with PCOS who were randomized to receive an oral contraceptive (OC, n= 16) or low-dose Spironolactone, Pioglitazone and Metformin (SPIOMET, n= 15), for 1 year and who remained untreated in the subsequent year.

Table legend:

& By LC-MS; BMI, body mass index; SHBG, sex hormone-binding globulin; FAI, free androgen index; HOMA-IR: homeostasis model assessment insulin resistance; OGTT, oral glucose tolerance test; cIMT, carotid intima-media thickness; DXA, dual X-ray absorptiometry; MRI, magnetic resonance imaging. Glycemia and insulinemia Z- scores were derived as described (8). Values are mean + SEM.

* p< 0.05, *p< 0.01, and ^†p<0.001 between controls and PCOS girls at baseline.

^§ No differences between PCOS subgroups at start.

^**p <0.05, ¹¹p<0.01 and *p <0.001 within PCOS subgroups for changes from 0-1 yr. ^§§p <0.05,^††p<0.01 and **p <0.001 within PCOS subgroups for changes from 1-2 yr. ^M p<0.005, ^***p< 0.01 and ^{11 11}p<0.001 between subgroups for 0 to 1 and for 1 to 2 yr changes (D)

EXAMPLE 3 Identification and validation of differentially expressed circulating miRNAs in girls with PCOS

miRNA deep sequencing detected a total of 320 miRNAs in serum samples from healthy controls and from untreated PCOS girls (Figure 5). Among those, 196 miRNAs were shared by the two subgroups. Sixteen miRNAs were differentially expressed in PCOS girls (FDR <5%), as compared to control girls (miR-451a, miR- 652-3p, miR-106b-5p, miR-206, miR-22-3p, miR-92a-3p, miR-4732, miR-25-3p, miR-16-5p, miR-16-2-3p, miR-194-5p, miR-130b-5p, miR-15a-5p, miR-10b-5p, miR-584-5p, and miR-151a-3p) (Figure 1). Among those, n = 13 were down- regulated, and n=3 were up-regulated. The differentially expressed miRNAs were ranked according to their expression fold change, and miR-451a, miR-652-3p, miR- 106b-5p and miR-206, which were within the first quartile of the lowest adjusted P- values (< 10 ⁴) (Figure 1, dark bars) were selected for validation by qRT-PCR in the entire study population (n= 13 control girls and n=31 girls with PCOS). The relative expression levels of those miRNAs, as expressed in Z-score derived from the expression levels obtained in control girls, confirmed that miR-451a, miR-652-3p, miR-106b-5p and miR-206 were down-regulated in girls with PCOS (Figure 2, panel A).

EXAMPLE 4. Correlation between circulating miRNAs expression and clinical endocrine-metabolic and imaging variables in girls with PCOS and in control girls

The results of the correlation analysis are summarized in Figure 2, right panel. The expression of miR-451a, miR-652-3p, miR-106b-5p and miR-206 was negatively correlated with the change in Z-score from weight at birth to BMI at PCOS diagnosis, and with circulating testosterone, free androgen index, and HOMA-IR, whereas a positive association was found for circulating sex hormone-binding globulin (SHBG) levels (all P<0.01 to <0.0006). Hepato-visceral (central) fat was negatively correlated with the expression of miR-451a and miR-652-3p (all P=0.03 to 0.004).

EXAMPLE 5. Effect of SPIOMET or OC intake on expression levels of validated miRNAs

In a next step the circulating concentrations of the four validated miRNAs after 1 year of treatment with OC or SPIOMET, and again at the end of the subsequent year without intervention. Expression levels of miR-652-3p and miR-106-5p (Figure 6), and particularly those of miR-451a (Figure 3, panel A ), normalized with SPIOMET and remained above baseline levels after treatment discontinuation. Circulating miR- 451a concentrations after treatment with OC or SPIOMET for 1 year associated with post-treatment ovulation rate (Figure 3, panel A).

EXAMPLE 6. Predictive value of candidate circulating miRNAs for PCOS diagnosis

To evaluate the predictive value of the candidate miRNAs for differentiating girls with PCOS from healthy control girls, a ROC curve was performed for each of the four validated miRNAs and calculated the AUC in each case (Figure 7). According to the maximal Youden's index (sensitivity + specificity -1) a cut-off value was identified of 0.06, 0.31, 0.08 and 0.78 respectively for miR-451a, miR-652-3p, miR-106b-5p and miR-206. The capacity of circulating miR-451a, miR-106b-5p and miR-652-3p to differentiate was remarkable, with a sensitivity and specificity of 100% and 100%, 100% and 85%, and 81% and 100%, respectively (Figure 7).

EXAMPLE 7. Prediction of miRNA target genes and pathway enrichment analysis

The miRSystem web tool was used to identify the gene targets and the signaling pathways regulated by miR-451a, miR-652-3p, miR-106b-5p and miR-206. KEGG and Reactome pathway enrichment analysis showed that genes targeted by those miRNAs, were all part of a complex regulatory network involved in glucose and lipid metabolism as well as in inflammation (Table 2).

Table 2. Pathway analysis of predicted miRNA gene targets

Claims

1. In vitro use of reagents for measuring the expression level of at least miR- 451a in a body sample, in the diagnosis of PCOS (PolyCystic Ovary Syndrome) in an adolescent girl between 10 and 24 years.

2. The use according to claim 1, wherein the adolescent girl is between 10 and 19 years.

3. The use according to claim 1 or 2, wherein the adolescent girl is non-obese.

4. The use according to claim 1, 2 or 3, wherein the reagents are oligonucleotides for performing RT-PCR.

5. The use according to any one of claims 1 to 4, further comprising reagents for measuring the expression level of one more of miR-106b-5p, miR-206 and miR-652-3p.

6. In vitro use of reagents for measuring the expression level of of at least miR- 451a in a body sample, in monitoring the efficacy of a PCOS treatment in an adolescent girl between 10 and 24 years.

7. The use according to claim 6, wherein the adolescent girl is between 10 and 19 years.

8. The use according to claim 6 or 7, wherein the PCOS treatment is a treatment with with spironolactone, pioglitazone and metformin (SPIOMET).

9. The use according to claim 6, 7 or 8, wherein the adolescent girl is non-obese.

10. The use according to any one of claim 6 to 9, further comprising the use of reagents for measuring the expression level of miR-106b-5p and/or miR-652- 3p.

11. The use according to any one of claims 6 to 10, wherein the monitoring identifies responders or non responders to said treatment.

12. The use according to anyone of claims 6 to 11, wherein an increase of the miRNA expression level of said at least one miRNA, compared to the level prior to the treatment is indicative of the patient being a responder.

13. In vitro method of diagnosing PCOS in an adolescent girl between 10 and 24 years, comprising the steps of:

determining in a body sample of said adolescent girl, the level of at least miR- 451a,

comparing the determined level with a reference value of a healthy adolescent girl

determining from said comparison, whether said adolescent girl is a PCOS patient.

14. The method according to claim 13, wherein the adolescent girl is between 10 and 19 years.

15. The method according to claim 13 or 14, wherein a low expression level of said at least one miRNA in the sample compared to said reference value is indicative for said adolescent girl being a PCOS patient.

16. In vitro method for monitoring the efficacy of a PCOS treatment in an adolescent girl between 10 and 24 years, comprising the steps of:

- determining in a body sample of said adolescent girl prior to or at the onset of said treatment, a first expression level of at least miR-451a,

- determining during the treatment in a body sample of said adolescent girl, a second expression level of at least miR-451a,

- comparing the first and the second expression level of said miRNA, wherein an increase of the second expression level, compared to the first expression level is indicative of the adolescent girl being a reponder to said treatment.

17. The method according to claim 16, wherein the adolescent girl is between 10 and 19 years.

18. The method according to claim 16 or 17, wherein the treatment is a treatment with spironolactone, pioglitazone and metformin (SPIOMET).

19. The method according to claim 16, 17 or 18, wherein the step of determining the second level is performed between 1 and 12 months after the onset of said treatment.