ASSAY

Abstract

The present invention relates to methods of determining the clinical significance of an HPV infection in a subject and associated methods, systems and kits. The method involves detection of biomarkers associated with clinically significant HPV infections and associated dysplasia.

Claims

1. A method of determining the clinical significance of an HPV infection in a subject, the method comprising: providing a biological sample, preferably a tissue sample, from said subject; determining the abundance in at least a portion of said sample of at least one biomarker, or the abundance of mRNA encoding said at least one biomarker, the at least one biomarker selected from the group consisting of CCL2, CCL3, CCL4, CXCL1 and CXCL12; determining therefrom the clinical significance of the HPV infection.

2. The method of claim 1 wherein subject is a human.

3. The method of claim 1 wherein the biological sample is a cervical tissue sample that comprises cervical cells.

4. The method of claim 1 wherein the HPV infection is an hrHPV infection.

5. The method of claim 1 which is a method of detecting or predicting the occurrence of HPV-induced high-grade precancerous lesions and/or HPV-induced invasive cancers.

6. The method of claim 1 which is a diagnostic method for differentiating between subjects who have an HPV infection which is likely to resolve without treatment or which does not currently merit further investigation and/or treatment from those in which a more active management is required.

7. The method of claim 1 which is for differentiating between patients likely to exhibit normal cervical tissue or CIN1 histology/low grade squamous intraepithelial lesion (LSIL) cytology, and those likely to have CIN2 or CIN3 histology/high grade squamous intraepithelial lesion (HSIL) cytology.

8. The method of claim 1 which is part of a diagnostic method performed by a clinician to allow them to decide upon a suitable course of treatment.

9. The method of claim 1 which comprises identifying an alteration in expression levels of at least one biomarker selected from the group consisting of CCL2, CCL3. CCL4, CXCL1 and CXCL12.

10. The method of claim 9 wherein the alteration is relative to: a pre-determined reference value; a value derived from contemporaneously obtained tissue form the subject which is not infected with HPV; or a value derived from previously obtained sample.

11. The method of claim 9 wherein the alteration is an increase in abundance of the biomarker protein itself, or in mRNA encoding the biomarker protein.

12. The method of claim 1 comprising the step of obtaining a sample of cervical tissue from a subject.

13. The method of claim 1 which is a method of screening a subject for risk of developing cervical cancer, the method comprising: obtaining a cervical tissue sample containing a plurality of cervical cells from the subject; detecting the presence of HPV infected cells in the sample; determining the abundance, in at least a portion of said cervical sample, of at least one biomarker, or the abundance of mRNA encoding for said at least one biomarker, the at least one biomarker selected from the group consisting of CCL2, CCL3, CCL4, CXCL1 and CXCL12; and determining therefrom an indication of risk of the subject developing cervical cancer.

14. The method of claim 1 comprising the step of isolating protein from cells in the sample.

15. The method of claim 1 comprising the step of isolating nucleic acids, especially mRNA, from cells in the sample.

16. The method of claim 1 comprising the step of selecting subjects to undergo further investigation and/or selecting subjects for treatment.

17. The method of claim 1 comprising determining the abundance of at least two of CCL2, CCL3, CCL4, CXCL1 and CXCL12.

18. (canceled)

19. The method of claim 1 comprising determining the abundance of at least three of CCL2, CCL3, CCL4, CXCL1 and CXCL12.

20. (canceled)

21. The method of claim 1 comprising determining the abundance of at least four of CCL2, CCL3, CCL4, CXCL1 and CXCL12.

22. The method of claim 1 comprising determining the abundance of at least CCL2, CXCL1 and CXCL12 and/or the abundance of mRNA encoding CCL2, CXCL1 and CXCL12.

23. The method of claim 1 comprising determining the abundance of CCL2, CCL3, CXCL1 and CXCL12, and/or the abundance of mRNA encoding CCL2, CCL3, CXCL1 and CXCL12.

24. The method of claim 1 comprising determining the abundance of at least CCL3, CCL4 and CXCL1, and/or the abundance of mRNA encoding CCL3, CCL4 and CXCL1.

25. (canceled)

26. The method of claim 1 further comprising determining of the abundance of CCL28, Chemerin, CXCL5, IL-16, CXCL11, CXCL9, CCL5, CCL17, GP130 and/or Transferrin Receptor, and/or the abundance of mRNA encoding CCL28, Chemerin, CXCL5, IL-16, CXCL11, CXCL9, CCL5, CCL17, GP130 and/or Transferrin Receptor.

27. The method of claim 1 comprising determining the abundance of one or more reference biomarkers, such as IL-8, or total protein level or mRNA level.

28. The method of claim 1 comprising testing for the presence of non-HPV infections.

29. (canceled)

30. A method of treatment of a subject, the method comprising the steps of: providing a biological sample from said subject; determining the abundance in at least a portion of said sample of at least one biomarker, or the abundance of mRNA encoding said at least one biomarker, the at least one biomarker selected from the group consisting of CCL2, CCL3, CCL4, CXCL1 and CXCL12; determining therefrom the clinical significance of the HPV infection; deciding whether treatment is required; and if treatment is required, deciding upon an appropriate treatment and carrying out said treatment.

31. A method for monitoring the efficacy of a treatment for HPV infection or a lesion associated with an HPV infection in a subject, the method comprising: providing a suitable biological sample, preferably a tissue sample, from said subject which is undergoing or which has undergone treatment; determining the abundance in at least a portion of said sample of at least one biomarker, or the abundance of mRNA encoding said at least one biomarker, the at least one biomarker selected from the group consisting of CCL2, CCL3, CCL4, CXCL1 and CXCL12; determining therefrom the clinical significance of the HPV infection; and based upon the result, assessing the efficacy of the treatment.

32. The method of claim 31 which comprises comparing the results obtained with the results obtained in an equivalent procedure carried out previously.

33. (canceled)

34. (canceled)

Description

BRIEF DESCRIPTION OF THE FIGURES

[0101] FIG. 1Proteomic microarray of chemokine expression in pooled LBC samples from women with A) normal cervical morphology and no infection; B) hrHPV infection and mild disease (CIN1); C) hrHPV infection and moderate disease (CIN2); and D) hrHPV infection and severe disease (CIN3). Each chemokine that is present is represented by a pair of white dots. The dots in the top left and right and the bottom left corners are the internal reference positive controls for the assay.

[0102] FIG. 2Multiplex data showing four chemokines which show significant difference between hrHPV+ve samples from women with no (normal=51) or low grade (CIN1, n=30) disease and those from women with moderate (CIN2, n=63) or severe (CIN3, n=62) disease; *p<0.05; **p<0.01; ***p<0.001.

[0103] FIG. 3Fold increase in mRNA expression for CCL2 in LBC samples with no infection and no disease compared to samples with hrHPV infection and CIN3. **=p<0.01.

[0104] FIG. 4Multiplex data showing four chemokines for HPV+ve samples from women with normal cytology compared to all other groups analysed by Kruskall Wallis test with Dunn's post-test. Data shown as mean+1-95% CI; NS=not significant; *=p<0.05; **=p<0.01; ***=p<0.001

[0105] FIG. 5Multiplex data showing four chemokines for HPV-ve samples from women with normal cytology compared to all other groups analysed by Kruskall Wallis test with Dunn's post-test. Data shown as mean+1-95% CI; NS=not significant; *=p<0.05; **=p<0.01; ***=p<0.001

[0106] FIG. 6Graph of quantitative PCR results for CCL2 mRNA levels. One dot=p<0.05; two dots=p<0.01.

[0107] FIG. 7Graph of quantitative PCR results for CCL5 mRNA levels. One dot=p<0.05; two dots=p<0.01.

SPECIFIC DESCRIPTION OF EMBODIMENTS OF THE INVENTION

[0108] While the making and using of various embodiments of the present invention are discussed in detail below, it should be appreciated that the present invention provides many applicable inventive concepts that can be embodied in a wide variety of specific contexts. The specific embodiments discussed herein are merely illustrative of specific ways to make and use the invention and do not delimit the scope of the invention.

[0109] Terms defined herein have meanings as commonly understood by a person of ordinary skill in the areas relevant to the present invention. Terms such as a, an and the are not intended to refer to only a singular entity, but include the general class of which a specific example may be used for illustration. The terminology herein is used to describe specific embodiments of the invention, but their usage does not delimit the invention, except as outlined in the claims.

[0110] Cervical cancer remains one of the most common cancer types affecting women worldwide. The biological pathway to cervical carcinoma begins with normal intraepithelial cells, and develops through low and then high grade dysplasia before malignancy occurs. Cytologists mark the passage to malignancy as progression from normal epithelial cells to atypical squamous cells of undetermined significance (ASCUS) to low grade squamous intraepithelial lesions (LSIL) and then high grade squamous intraepithelial lesions (HSIL) before carcinoma in situ and finally malignancy result. Histologists mark the progression from normal cells to various grades of cervical intraepithelial neoplasia (CIN1, 2 and 3), then to carcinoma in situ and finally malignancy. CIN1 is considered low grade dysplasia comparable to LSIL. CIN 2 and 3 are considered high grade dysplasia comparable to HSIL.

[0111] The current standard of care includes regular cytologic testing with a Papanicolau (Pap) smear to identify abnormalities as indicating dysplasia or carcinoma in patient cells. When high grade dysplasia is detected and confirmed by histological examination, the transformation zone of the patient's cervix is removed immediately by loop excision or cone biopsy. More radical procedures are required when carcinoma is detected. At the same time, however, the progression from normal to malignancy is not strict and the presence of low grade dysplasia does not necessarily indicate that the patient will progress to high grade dysplasia or malignancy. Significantly, the negative predictive value of cytologic methods (e.g., Pap smears) for detecting high grade dysplasia is poor. Thus, low grade dysplasia may be misdiagnosed as high grade, thereby subjecting the patient to unwarranted treatment and high grade dysplasia may be misdiagnosed as low grade dysplasia, thereby delaying appropriate treatment. Accordingly, there is a need for a diagnostic method that will accurately distinguish between low and high grade dysplasia.

[0112] The development and application of suitable triage tests which can risk-stratify primary screen Human Papilloma Virus (HPV) infected women is arguably the key priority for cervical screening.

[0113] The present inventors have determined that looking at host proteins involved in innate immunity at epithelial surfaces, such as the cervix, can form the basis of such a test. It appears that persistent infection alters the local environment and sets up a different tissue response from that seen where there is no disease or resolving infection. This type of response, that is by definition ineffective at clearing virus or virally infected cells, seems likely to involve production of relatively large amounts of reactive host proteins both in infected cells and in the immediately surrounding uninfected cells. Testing for these proteins in this context has the advantages of: not being hrHPV type specific; not looking for the very small quantities of viral protein expressed in infected epithelial cells; and amenable to being developed as a plate based high-throughput assay does not require preparation, staining and examination of cells under a microscope.

Example 1Analysis of Biomarkers

[0114] Methodology

[0115] Ethics Statement

[0116] Ethical approval was obtained from Scotland A Research Ethics Committee (REF 12/SS/0034). All cervical samples were collected into ThinPrep-preservcyt liquid based cytology transport medium (Hologic, Crawley, UK). For this study, anonymised, curated, cervical smear samples were obtained from the Scottish National HPV archive, which holds Generic Scotland A Research Ethics Committee approval for Research Tissue banks (REC Ref 11/AL/0174) for provision of samples for HPV related research after approval from an Independent Scottish HPV Archive Steering Committee.

[0117] Samples

[0118] Aliquots (1.5 ml) from 350 LBC samples were obtained from the Scottish HPV archive (http://www.shine.mvm.ed.ac.uk/archive.shtml). Disease status defined by either cytology and/or cervical biopsy was obtained for each sample.

[0119] hrHPV Typing

[0120] All samples were analysed by the archive staff for hrHPV infection using a clinically validated PCR test from Abbot Molecular, RealTime High Risk HPV (http://www.abbottmolecular.com/products/infectious-diseases/realtime-pcr/hepatitis-high-risk-hpv-assay.html), following the manufacturer's instructions. This assay is a qualitative in vitro PCR assay that utilizes homogeneous target amplification and detection technology for the detection of hrHPV DNA in cervical cells collected in liquid cytology media. It detects 14 high risk HPV genotypes (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68) and partially genotypes 16, 18 from the other 12 high risk genotypes.

Extraction of Protein from LBC Samples

[0121] This method has been previously published by the present inventors [1]. Briefly, LBC samples are collected into 20 ml ThinPrep-preservcyt medium (Hologic, Crawley, UK) which contains 50% methanol. Briefly, 100 l ice cold chloroform is added to 900l of sample in an Eppendorf tube on ice, vortexed, incubated on ice for 5 min and spun for 5 min at 10,000 rpm at 4 C. in a bench top microfuge. The aqueous top layer is discarded, 300 l methanol is added and the tube vortexed, and spun for 10 min at 10,000 rpm at 4 C. in a bench top microfuge. The supernatant is removed and the pellet re-suspended in 100 l Tris Buffered Saline. Protein concentration is determined by Pierce BCA assay (Thermo Scientific, UK).

Proteomic Array

[0122] Antibody-based, human chemokine proteomic arrays (Cat No. ARY017, R&D Systems, UK) were carried out as per manufacturer's instructions. To identify proteins of interest, for the arrays, pools of samples from 9 women with different grades of disease were compared. Each pool contained 10 g of protein from each individual sample (see Results). Developed arrays were scanned and the images reversed for use with Image-J analysis software. Each chemokine present in the pooled samples is represented by a pair of dots (see FIG. 1). Image-J software was used to measure the pixel density of each dot and the average pixel density of the two dots present for each chemokine present is given in Table 1.

Procarta Multiplex Assay

[0123] Protein extracted as above from individual samples with different grades of disease were tested for abundance of CCL11, CCL2, CCL3, CCL4, CCL5, CXCL1, CXCL12, CXCL10 and CXCL8 by ProcartaPlex Human Chemokine Panel 1 (9-plex) supplied by Affymetrix e-Bioscience, UK, (Cat. No. EPX090-12187-901). Ten g protein from each sample was added to a microplate well and the assay developed and read on a BioRad Bio-Plex 200 HTF multiplex assay system following the manufacturer's instructions. Data were analysed using SPSS software. Univariate analysis was done by Kruskall-Wallis rank sum test and by ANOVA; multivariate analysis was done by logistic regression.

Extraction of RNA from LBC Samples

[0124] As published[5] RNA was extracted from 1.5 ml aliquots of LBC samples using miRNeasy mini kit (Qiagen, UK) following the manufacturer's instructions. RNA was quantified on a Nanodrop spectrophotometer and stored at 80 C. until use. cDNA was made from 500 ng RNA using the Quantitect Reverse Transcription Kit (Qiagen, UK) as per manufacturer's instructions and stored at 20 C. until use. Duplex qPCR was carried out in Lightcycler Nano (Roche) real-time PCR system in duplicate in 8 well PCR strips using plates, primers, probes and reagents from Life Technologies Applied Biosystems, UK. Wells contained 2 l cDNA, 1 l gene of interest primer mix, 0.12 l each of 18S forward and reverse primers, 1.6 l 18S probe, and 13 l mastermix. The PCR programme was 2 min at 50 C., 10 min at 95 C., followed by 40 cycles of 15 sec at 95 C. and 1 min at 60 C.

qRT-PCR for CCL2

[0125] RNA was extracted from 1.5 ml aliquots of LBC samples using miRNeasy mini kit (Cat. No. 217004, Qiagen, UK) following the manufacturer's instructions. RNA was quantified on a Nanodrop spectrophotometer and stored at 80 C. until use. cDNA was made from 500 ng RNA using the Quantitect Reverse Transcription Kit (Cat. No. 205313, Qiagen, UK) as per manufacturer's instructions and stored at 20 C. until use. Duplex qPCR was carried out in Lightcycler Nano (Roche) real-time PCR system in duplicate in 8 well PCR strips using plates, primers and probe (Cat. No. 4331182), and reagents from Life Technologies Applied Biosystems, UK. Wells contained 2 l cDNA, 1 l gene of interest primer mix, 0.12 l each of 18S forward and reverse primers, 1.6 l 18S probe, and 13 l mastermix. The PCR programme was 2 min at 50 C., 10 min at 95 C., followed by 40 cycles of 15 sec at 95 C. and 1 min at 60 C.

Results

[0126] Proteomic Microarray Reveals Several Chemokines are Upregulated in High Grade (CIN3) hrHPV Induced Cervical Disease

[0127] Pooled protein derived from cervical samples from women with 1) no HPV infection and no disease 2) hrHPV infection and CIN1 3) hrHPV infection and CIN2, 3) hrHPV infection and CIN3 were tested and compared for chemokine protein expression as described in Materials and Methods. FIG. 1 shows that several chemokines were up-regulated in hrHPV+ve samples from women with CIN3. The results of the analysis of pixel density shown in Table 1 demonstrate that a number of chemokine proteins are up-regulated in hrHPV+ve samples from women with CIN3 compared to uninfected controls and to hrHPV+ve infected samples from women with CIN1.

TABLE-US-00001 TABLE 1 Pixel density analysis of chemokine proteomic array from pooled LBC samples. Image-J calculated average pixel density HR-HPV HR-HPV HR-HPV NORMAL POS POS POS ANALYTE HPV-ve CIN1 CIN2 CIN3 REF 17856 20326.5 20302 18544 REF 18592.5 20346 20312.5 19143.5 CCL21 0 0 0 938 CCL28 0 0 0 17818 CXCL16 0 0 11 0 CHEMERIN 0 0 0 18752.5 CXCL5 0 0 0 18946.5 CCL26 0 0 0 1371 CX3CL1 0 1 0 65 CXCL1 2039.5 18736.5 14601 3358 CCL14 0 0 0 86.5 CCL1 0 0 0 589.5 CXCL8 11177.5 20351 20229.5 13834 IL-16 996 1581 3614.5 4361 CXCL10 33.5 0 0 18962 CXCL11 78 0 0 17183 LYMPHOTACTIN 4 0 0 695 CCL2 33.5 0 0 18878 CCL7 0 0 0 420 CCL22 223.5 0 0 3188.5 MIDKINE 338.5 12900 2990 2510 CXCL9 100 4 0 19133 CCL3/CCL4 0 0 0 431 CCL15 35.5 0 0 167 CCL20 0 0 0 0 CCL19 48 0 0 1773.5 CXCL7 1653.5 19882 1851 5553.5 CCL18 0 9.5 3 1044.5 CXCL4 371 17923 223 4083 CCL5 82.5 0 0 19038.5 CXCL12 15 0 0 18759 CCL17 45 0 0 1007 CXCL17 0 0 0 0 NEG CONTROL 0 0 0 0 REF 17774.5 20305 20365.5 18795

Multiplex Chemokine Assay of Individual Samples

[0128] The threshold for treatment of women with hrHPV cervical infection is CIN2. Univariate analysis of the results showed that this platform indicated significant differences between hrHPV+ve samples from women with high grade disease (CIN2 and CIN3) and women with no or low grade (CIN1) disease for four chemokines, CCL2, CCL3, CXCL1 and CXCL12 (FIG. 1). Multivariate analysis showed that combining the data obtained on this platform for three of the chemokines CCL3, CCL4 and CXCL1 improved the sensitivity and specificity compare to any of the chemokines alone (Table 2). The most relevant data obtained in the Procarta multiplex assay is set out below in Tables 3 to 12.

TABLE-US-00002 TABLE 2 Multivariate logistic regression analysis of multiplex chemokine data from samples analysed with Procarta assay as detailed in Materials and Methods. Analyte AUC Accuracy Accuracy 95% CI Kappa Sensitivity Specificity PPV NPV CCL5 0.533 0.578 (0.477, 0.676) 0.138 0.627 0.512 0.638 0.5 CXCL1 0.629 0.657 (0.556, 0.748) 0.276 0.78 0.488 0.676 0.618 CXCL8 0.59 0.608 (0.506, 0.703) 0.196 0.661 0.535 0.661 0.535 CXCL10 0.515 0.431 (0.334, 0.533) 0.123 0.407 0.465 0.511 0.364 CCL2 0.638 0.608 (0.506, 0.703) 0.261 0.424 0.86 0.806 0.521 CCL3 0.677 0.637 (0.536, 0.730) 0.302 0.508 0.814 0.789 0.547 CCL4 0.632 0.618 (0.516, 0.712) 0.294 0.373 0.953 0.917 0.526 CXCL12 0.594 0.608 (0.506, 0.703) 0.244 0.492 0.767 0.744 0.524 All 0.693 0.676 (0.577, 0.766) 0.347 0.678 0.674 0.741 0.604 CCL3, 0.639 0.608 (0.506, 0.703) 0.261 0.424 0.86 0.806 0.521 CCL4, CXCL1 COMBINED PPV = Positive Predictive Value; NPV = Negative Predictive Value.
mRNA Analysis by Real-Time rtPCR for CCL2 in LBC Samples Confirms Protein Data.

[0129] To reassure that the protein data was reflected at a different biological level from protein presence in the LBC samples, we extracted mRNA from aliquots of LBC samples from women with no or high grade (CIN3) disease and carried out real-time rtPCR analysis for one of the chemokines (CCL2) shown to be different at the protein level. FIG. 3 shows a significant 34-fold increase in expression of CCL2 mRNA in hrHPV+ve samples with CIN3 compared to samples from uninfected women with no disease. This indicates that mRNA can be used in place of (or in addition to) protein as a biomarker.

Further DataMarker Abundance as Determined by Procarta Multiplex Protein Assay

[0130] The abundance values of the chemokines CCL2, CCL3, CCL4, CXCL1 and CXCL12 correlated to disease state, as determined in the Procarta experiments, are set out below in Tables 3 to 12.

CXCL1 (GRO-Alpha, pg Protein Per 200 g Protein)

[0131]

TABLE-US-00003 TABLE 3 Grouped by normal + CIN1 versus CIN2 + CIN3 Min. 1st Qu. Median Mean 3rd Qu. Max. Normal_CIN1 6.05 71.68 136.8 225.8 247.5 3057 CIN2_CIN3 2.25 137 246.5 488 546.3 3525 Kruskal-Wallis rank sum test p = 7.941e06 ANOVA p = 0.00103

TABLE-US-00004 TABLE 4 Grouped by normal, CIN1, CIN2, and CIN3 Min. 1st Qu. Median Mean 3rd Qu. Max. Normal 6.05 62.79 99 215.9 179.4 3057 CIN1 38.83 93.09 182.8 243.7 299.4 829.4 CIN2 36.05 172 248.4 502.1 546.3 2523 CIN3 2.25 93.31 230 473.7 524.4 3525 Kruskal-Wallis rank sum test p = 8.675e06 ANOVA p = 0.0126

CCL2 (MCP1, pg Protein Per 200 g Protein)

[0132]

TABLE-US-00005 TABLE 5 Grouped by normal + CIN1 versus CIN2 + CIN3 Min. 1st Qu. Median Mean 3rd Qu. Max. Normal_CIN1 1.2 4.672 7.015 14.51 12.25 184.6 CIN2_CIN3 0 7.2 12.93 30.15 36.42 346.8 Kruskal-Wallis rank sum test p = 1.385e05 ANOVA p = 0.00469

TABLE-US-00006 TABLE 6 Grouped by normal, CIN1, CIN2, and CIN3 Min. 1st Qu. Median Mean 3rd Qu. Max. Normal 1.2 4.48 5.68 11.52 9.34 121.5 CIN1 1.96 5.61 10.45 19.99 18.32 184.6 CIN2 2.95 5.97 11.36 32.14 41.15 346.8 CIN3 0 7.532 16.7 28.13 32 127.8 Kruskal-Wallis rank sum test p = 1.113e05 ANOVA p = 0.0267

CCL3 (MIP-1Alpha, pg Protein Per 200 g Protein)

[0133]

TABLE-US-00007 TABLE 7 Grouped by normal + CIN1 versus CIN2 + CIN3 Min. 1st Qu. Median Mean 3rd Qu. Max. Normal_CIN1 0 3.155 6.96 10.85 13.06 77.8 CIN2_CIN3 0 6.02 12.07 22.33 23.58 231.9 Kruskal-Wallis rank sum test p = 0.0001189 ANOVA p = 0.0029

TABLE-US-00008 TABLE 8 Grouped by normal, CIN1, CIN2, and CIN3 Min. 1st Qu. Median Mean 3rd Qu. Max. Normal 0 3.12 6.21 9.956 12.95 67.24 CIN1 0.23 4.22 7.94 12.49 15.26 77.8 CIN2 0 5.1 13.42 19.7 23.77 116.7 CIN3 0 6.08 11.68 25 23.12 231.9 Kruskal-Wallis rank sum test p = 0.001437 ANOVA p = 0.017

CCL4 (MIP-1Beta, pg Protein Per 200 g Protein)

[0134]

TABLE-US-00009 TABLE 9 Grouped by normal + CIN1 versus CIN2 + CIN3 Min. 1st Qu. Median Mean 3rd Qu. Max. Normal_CIN1 0.37 15.02 23.64 43.8 41.99 415.5 CIN2_CIN3 0 18.59 42.32 74.93 74.33 1209 Kruskal-Wallis rank sum test p = 0.0023 ANOVA p = 0.062

TABLE-US-00010 TABLE 10 Grouped by normal, CIN1, CIN2, and CIN3 Min. 1st Qu. Median Mean 3rd Qu. Max. Normal 0.37 11.94 22.24 37.56 33.67 310.1 CIN1 4.58 17.3 38.71 55.21 54.24 415.5 CIN2 0 15.03 45.62 55.43 79.2 284 CIN3 0 20.52 40.8 94.76 70.38 1209 Kruskal-Wallis rank sum test p = 0.00532 ANOVA p = 0.0605

IL-8 (pg Protein Per 200 g Protein)

[0135]

TABLE-US-00011 TABLE 11 Grouped by normal + CIN1 versus CIN2 + CIN3 Min. 1st Qu. Median Mean 3rd Qu. Max. Normal_CIN1 200.8 1059 1911 2865 3573 12000 CIN2_CIN3 5.69 1519 2854 4001 5417 16210 Kruskal-Wallis rank sum test p = 0.003507 ANOVA p = 0.00988

TABLE-US-00012 TABLE 12 Grouped by normal, CIN1, CIN2, and CIN3 Min. 1st Qu. Median Mean 3rd Qu. Max. Normal 200.8 1069 1961 2922 4066 12000 CIN1 372.3 1056 1886 2761 3354 11690 CIN2 358.7 1519 2813 4052 6460 16210 CIN3 5.69 1554 3139 3949 5388 14740 Kruskal-Wallis rank sum test p = 0.03402 ANOVA p = 0.0818

CONCLUSIONS

[0136] Our data indicate that using the Procarta panel MIP-1alpha, MIP-1beta and GRO-A as a combined biomarker panel achieves a specificity and a PPV which out-performs the all of the current market competitors (86% and 80% respectively). MIP1 beta alone delivers a specificity and PPV of 95% and 92% respectively. While the sensitivityincluding for the combined panelis currently suboptimal, we have a clear pathway to boosting this via (1) the use of alternative platforms which have been shown to detect lower levels of protein, and (2) the interrogation of real-time reverse transcription PCR strategies for the targets described.

Example 2Increased Sample Numbers

[0137] As a continuation of Example 1, further samples were examined to seek to further validate the earlier findings.

[0138] There is shown in FIGS. 4 and 5 the combined multiplex results for 482 samples. This adds in >100 samples to the earlier dataset. The results indicate that the previous conclusions are supported in these additional samples. It is observed that there is somewhat better discrimination and less noise with the HPV+ve samples, but this may just be down to the difference in numbers between the groups (326 HPV+ve and 156 HPV-ve).

[0139] The results are presented slightly differently to previously, and thus, as well as analysing samples from women with normal or borderline cytology who were not sent for colposcopy versus CIN1 2 and 3, a sent to colposcopy group (mild or above cytology results) and a needed treatment at colposcopy group (CIN2 or above) has been included.

[0140] FIG. 4 shows the results for the HPV+ve group, and additional details for this group are:

326 ABBOTT Test HR-HPV+Ve LBC Samples.

[0141] Multiplex Procarta kit results on extracted protein.

TABLE-US-00013 Normal/borderline n = 73 (women not sent for colposcopy after cytology) Mild or above cytology (women sent for colposcopy) n = 256 CIN2 or above n = 211 (women who required treatment at colposcopy) CIN1 n = 45 CIN2 n = 91 CIN3 n = 115

[0142] SDF1 shows smaller differences compared with the other markers. MIP1beta very clearly allows differentiation of CIN1 and above; Gro-alpha and MCP1 clearly allow differentiation of CIN2 and above.

[0143] FIG. 5 shows the results for the HPV-ve group, and additional details for this group are:

156 ABBOTT Test HR-HPV-Ve LBC Samples.

[0144] Multiplex Procarta kit results on extracted protein.

TABLE-US-00014 Normal/borderline n = 126 (women not sent for colposcopy after cytology) Mild or above cytology (women sent for colposcopy) n = 30 CIN2 or above n = 7 (women who required treatment at colposcopy) CIN1 n = 23

[0145] SDF1 shows no significant difference. MIP1beta allows differentiation of CIN1; Gro-alpha and MCP1 clearly allow differentiation of CIN1 and above.

Example 3mRNA Analysis

[0146] To further validate the approach of using mRNA expression levels as an alternative to protein levels, measurements of mRNA levels were undertaken for CCL2 and CCL5 to confirm whether differences were differences in protein levels were mirrored in differences in mRNA levels. The methodology is described in Example 1 for CCL2, and an identical method with primers and probes for CCL5 was used (Cat. No. 4331182).

[0147] The results for CCL2 are shown in FIG. 6. Significant differences were observed between all three cytology+ groups and HPV cytology.

[0148] Sample numbers were as follows: [0149] HPV cytology n=22, [0150] HPV+ cytology n=20, [0151] CIN1 n=15, [0152] CIN2 n=13, [0153] CIN3 n=15

[0154] The results for CCL5 are shown in FIG. 7. Significant differences between CIN1 and CIN 2 and HPV cytology.

[0155] Sample numbers were as follows: [0156] HPV cytology n=20, [0157] HPV+ cytology n=19, [0158] CIN1 n=15, [0159] CIN2 n=13, [0160] CIN3 n=15

[0161] As expected, these results further support supposition that both protein levels and mRNA levels can be successfully used in the present invention.

Example 4Illustrative Screening Method and Preliminary Thresholding

[0162] Under the current system of cytology screening, approximately 3% of women are offered a colposcopy appointment for detected abnormality. 75% of women sent for colposcopy do not require treatment (i.e. are judged to have normal cervical morphology or CIN1). HPV testing is estimated to show overall positivity of around 10% of women. Approximately 70% of CIN1 lesions will spontaneously regress but 30% will progress. At present there is no way of distinguishing these and women with CIN1 are offered rescreening and colposcopy if necessary after 6-12 months.

[0163] Preliminary thresholding of the current dataset has been carried out using an arbitrary threshold as the median value of all HPV positive samples (n=320) for each of five chemokines (CXCI1, CXCL12, CCL2, CCL3, & CCL4). This threshold was then applied to the data from those samples which were below the treatment threshold. The data was analysed to discriminate between samples which should indicate women be sent for checking (i.e. which showed an increase in 3 or more chemokines at this threshold, indicating a higher level of tissue involvement), and those which should not need colposcopic examination (i.e. which showed an increase in 0, 1 or 2 chemokines, indicating low tissue involvement). The analysis indicated that an assay using this thresholding approach, using multiplex protein detection, would have sent only 38% (16/47) of those women found to have normal morphology or CIN1 on examination for colposcopy, i.e. representing a significant reduction in the number of women undergoing unnecessary colposcopy.

[0164] It should be noted that this screening method, and the thresholding system applied, is merely a preliminary approach, and is not intended to represent an final, optimal screen for any particular patient group or situation. However, it does serve to illustrate the utility of the present method. Optimisation of a screening method of the present invention for use in particular situations, or with particular patient groups, can be readily carried out by the skilled person.

Discussion of Health Economics

[0165] From the 2011 census, there are 28.5 m women in the UK. Using the anticipated value of 10-15% women being HPV positive, and assuming all women are screened once every three years, approximately 950,000 to 1.4 m women would screen positive for HPV annually. This is potentially a massive burden on the NHS in terms of further investigations.

[0166] Adding an additional triage test for positive HPV screens in this patient cohort, would add an additional cost initially, but result in large savings further down the line. The benefits of this triage step are multiple. As is often the case in healthcare, the costs and savings generated by this test occur across different organisations and budget lines, so a holistic overview of the impact is required.

[0167] The main area of budget saving associated with the introduction of the triage test is reduced numbers of referrals to colposcopy. It is estimated that unnecessary colposcopy referrals may be reduced by up to 70% from their current rate. Further investigation of the numbers of women currently referred for colposcopy who have an outcome of no biopsy and no treatment is also relevant hereif the adoption of a triage test leads to a decrease in this outcome, not only is the test reducing the total numbers overall, it is also helping to improve the intrinsic performance of colposcopy as a test, which would be good for patients. Further detailed modelling work is required to pick up all the nuances of cost and clinical benefit associated with reduced demand for colposcopy, including the possible impact on waiting times etc., which is not possible within the scope of this early assessment. However a simplistic estimate of the impact can be made, as follows.

[0168] Colposcopy in England is remunerated at the outpatient appointment rate for gynaecology, 131 in 2014/15. In England, 3.57 million screening samples were examined by pathology laboratories in 2012-13. Data from the English cervical screening service suggests that 167,394 referrals to colposcopy were made from screening in 2012-13, so 4.7% of the screening samples resulted in a referral to colposcopy. So, using current prices and last year's activity (the most recent available) we can estimate the annual cost of screening-related colposcopy activity to be approximately 19.8 m for England, or 25 m for the whole of the UK (recognising that there is no comparable tariff system in some of the other countries so this is harder to assess).

[0169] Reducing these referrals by 70% would therefore save 105,000 colposcopies and 13.8 m from the current annual spend. These savings would occur at hospital level, in the form of reduced income. Hospitals may disinvest to cope with smaller volumes, or may reallocate their resources to the higher number of patients actually requiring treatment, rather than just colposcopy, which will result from the higher number of patients found through screening who require treatment for pre-cancerous cell changes. Hospitals in England are paid separately for this activity, so that income stream would be expected to increase, compensating in part for the loss of pure colposcopy income.

[0170] This analysis does not take into account harder to measure costs, such as patient's time, travel, childcare and intangible but significant factors such as anxiety. Patients may also consult their GP more frequently if they are called for colposcopy, which would be reduced by reducing the numbers required to take the test.

[0171] Pure savings are not however the whole story. It is also important to compare the costs of using a triage test according to the present invention (referred to as CINck) to the costs of alternative triage. If CINck is not used as a triage test, the recommended pathway would be to subject HPV positive screens to subsequent cytology screening as a triage step. Using a quoted cost of 21.36 as the lab-based cost of a liquid based cytology screen, and applying the same assumptions as above, it can be seen that the costs of this alternative triage will be 30.4 m to 20.2 m per year. While this is potentially cheaper than triage with CINck, it will not deliver the same sensitivity, meaning that some screens are interpreted as false negatives at a potentially high cost to the patients concerned. The element of subjectivity inherent in cytology based screening also remains.

[0172] On the basis of the initial estimates above, using CINck to triage HPV positive samples appears to be more expensive than cytology-based triage, but confers greater clinical advantage. The possible savings resulting from fewer colonoscopies appear to be adequate to fund this additional cost differential between the two triage approaches.

[0173] Overall, a triage test according to the present invention has the power to prevent a clinically unnecessary invasive diagnostic procedure for many women, which can be the cause of significant anxiety as well as cost to the NHS.

REFERENCES

[0174] 1. Szarewski A, Mesher D, Cadman L, Austin J, Ashdown-Barr L, Ho L, Terry G, Liddle S, Young M, Stoler M, McCarthy J, Wright C, Bergeron C, Soutter W P, Lyons D, Cuzick J (2012) Comparison of seven tests for high-grade cervical intraepithelial neoplasia in women with abnormal smears: the Predictors 2 study. J Clin Microbiol 50: 1867-1873. JCM.00181-12 [pii]; 10.1128/JCM.00181-12 [doi]. [0175] 2. Dijkstra M G, van N D, Rijkaart D C, van Kemenade F J, Heideman D A, Snijders P J, Meijer C J, Berkhof J (2014) Primary hrHPV DNA testing in cervical cancer screening: how to manage screen-positive women? A POBASCAM trial substudy. Cancer Epidemiol Biomarkers Prev 23: 55-63. 1055-9965.EPI-13-0173 [pii]; 10.1158/1055-9965.EPI-13-0173 [doi]. [0176] 3. Kavanagh K, Pollock K G, Potts A, Love J, Cuschieri K, Cubie H, Robertson C, Donaghy M (2014) Introduction and sustained high coverage of the HPV bivalent vaccine leads to a reduction in prevalence of HPV 16/18 and closely related HPV types. Br J Cancer 110: 2804-2811. bjc2014198 [pii]; 10.1038/bjc.2014.198 [doi]. [0177] 4. De Strooper L M, van Z M, Steenbergen R D, Bleeker M C, Hesselink A T, Wisman G B, Snijders P J, Heideman D A, Meijer C J (2014) CADM1, MAL and miR124-2 methylation analysis in cervical scrapes to detect cervical and endometrial cancer. J Clin Pathol 67: 1067-1071. jclinpath-2014-202616 [pii]; 10.1136/jclinpath-2014-202616 [doi]. [0178] 5. Canham M, Charsou C, Stewart J, Moncur S, Hoodless L, Bhatia R, Cong D, Cubie H, Busby-Earle C, Williams A, McLoughlin V, Campbell J D, Cuschieri K, Howie S (2014) Increased Cycling Cell Numbers and Stem Cell Associated Proteins as Potential Biomarkers for High Grade Human Papillomavirus+ve Pre-Neoplastic Cervical Disease. PLoS One 9: e115379. 10.1371/journal.pone.0115379 [doi]; PONE-D-14-30656 [pii].