Method of Assessing Risk

Abstract

The application relates to a method of assessing the risk of a patient for cervical pre-cancer or cancer, especially where there is no sign of a malignancy.

Claims

1. A method of identifying a patient at risk of developing cervical pre-cancer or cervical cancer comprising: (a)Measuring the level of methylation of a cervical cancer marker in a sample obtained from the patient; and (b) Determining the level of risk based on the level of methylation and the patient's age.

2. The method of claim 1 wherein the cervical cancer marker is selected from POU4F3 or HS3ST2.

3. The method of claim 1 wherein the patient is over 30 years of age.

4. The method of claim 1 wherein the level of risk is determined by correlating the level of marker methylation with the patients' age using the isorisk curve shown in FIG. 1 or FIG. 2.

5. The method of claim 1 comprising the steps of (a) obtaining a sample of nucleic acid from a cervical cell from a subject; (b) treating the nucleic acid obtained with sodium bisulfite; and (c) measuring the level of methylation of a cervical cancer marker within the treated nucleic acid; and (d) determining the level of risk of developing cervical pre-cancer or cancer based on the level of methylation in combination with the age of the subject.

6. The use of the level of methylation of a cervical cancer marker, in combination with a patients age, to determine the level of risk of developing cervical cancer or pre-cancer.

Description

[0039] The invention will be described in the examples below with reference to the following figures:

[0040] FIG. 1 shows an isorisk curve map for POU4F3 showing the ratio of confirmed pre-cancer or cancer cases to presumed histology negatives plotted according to epigenetic result and age of the patient. 1,595 data points representing the epigenetic test result (vertical axis, log of index value) and the age (horizontal axis) of HPV-positive women tested in the clinical trial were plotted in 2 groups: red dots are cases with a confirmed dysplasia or cancer diagnosis; blue dots are presumed histology negative samples. The isorisk curves show the shifting ratio of the 2 groups. Background colour indicates the standard error of the risk level estimate.

[0041] FIG. 2 shows an expanded view of FIG. 1 at lower levels of methylation.

[0042] FIG. 3 shows a comparison of readings from three groups of patients generated with POU4F3 marker.

[0043] Group 1 (N=1855): HPV & CITO & CIN2 (i.e. do not have cervical cancer; healthy controls)

[0044] Group 2 (N=712): HPV+ & (CITOASCH|HSIL|CIS|CC) & CIN2 (Give a negative result on traditional smear test, but may be at risk of developing cervical cancer as have HPV infection)

[0045] Group 3 (N=59): CIN2+ (i.e., Histologically confirmed to have cervical cancer)

[0046] FIG. 4 shows a ROC curve generated with POU4F3 marker indicating that the test of the invention is highly specific and sensitive. The cut-off points at 2%, 10%, and 40% level of risk for CIN2+ are shown. These correlate with the threshold values provided by the National Cancer Institute. Patients falling below the 2% threshold do not require follow up treatment; Patients falling between 2% and 10% require follow up treatment; Patients falling between 10% and 40% require colposcopy; Patients with a result over 40% threshold require treatment even if there is no sign of a lesion.

EXAMPLE

[0047] 1,595 clinical samples (Rovers cervical brush), all HPV positives, aged between 20 and 60, collected in a multicentre clinical trial conducted in Hungary were tested. Including HPV negatives over 6,000 women were enrolled.

[0048] Samples were bisulfate converted, and tested with Taqman reaction using the primers and probes described in Chen et al (2014) Methylomics analysis identifies epigenetically silenced genes and implies an activation of -catenin signaling in cervical cancer International Journal of Cancer 135(1):117-27 to determine the level of methylation.

[0049] The generalized additive model (GAM) model and logistic regression was used to create the isorisk map shown in FIGS. 1 and 2.

[0050] The results were further verified by comparing three groups of patients.

[0051] Group 1 (N=1855): HPV & CITO & CIN2 (i.e. do not have cervical cancer; healthy controls)

[0052] Group 2 (N=712): HPV+ & (CITO ASCH|HSIL|CIS|CC) & CIN2 (Give a negative result on traditional smear test, but may be at risk of developing cervical cancer as have HPV infection)

[0053] Group 3 (N=59): CIN2+ (i.e. Histologically confirmed to have cervical cancer)

[0054] These results show that the level of general methylation increases with age, but that the increase with age is far greater in patients with cervical cancer.

[0055] The sensitivity and specificity of the method was assessed and a ROC curve generated. In HPV-positive women aged 25 years or older sensitivity for severe cervical dysplasia or worse condition at baseline was 91.94% (95% CI 82.17-97.33) meanwhile specificity in the same group was 74.35% (95% CI 71.78-76.81). (See FIG. 4)

[0056] The level of methylation alone is insufficient to determine the level of risk. This must be used in combination with the patient age. For example, the methylation level could be 50 for in two patient samples. However, the ages of the patients are 32 and 52. Accordingly their calculated risk is 13% (95% CI 0.091 to 0.19) and 2.9% (95% CI 0.012 to 0.067) respectively. These results are significantly different.