WO2023067186A1

WO2023067186A1 - Methods of predicting stable remission in ulcerative colitis

Info

Publication number: WO2023067186A1
Application number: PCT/EP2022/079466
Authority: WO
Inventors: Christian Boerde ARKTEG; Rasmus GOLL
Original assignee: Universitetet I Tromsø - Norges Arktiske Universitet
Priority date: 2021-10-22
Filing date: 2022-10-21
Publication date: 2023-04-27
Also published as: GB202115242D0

Abstract

The present invention relates to a method of determining whether a patient in remission from ulcerative colitis (UC) is in stable remission, said method comprising quantifying the level of gene expression of IFNG and IL33 in a gastrointestinal (Gl) mucosal sample from said patient and comparing said expression levels; in particular wherein a high level of expression of IL33 relative to IFNG is prognostic of stable remission and a level of expression of IL33 relative to IFNG which is not high, is prognostic of relapse.

Description

Methods of Predicting Stable Remission in Ulcerative Colitis

Ulcerative Colitis (UC) is a chronic inflammatory disease of the colon, and it is typically characterized by a relapsing-remitting course. Previously, the main costs associated with UC were linked to hospitalization. However, thanks to the arrival of expensive biological agents in the last two decades, the main costs of UC are becoming increasingly associated with medication. Although high in cost, biological agents have proved beneficial in lowering colectomy rates and maintaining remission. Recent studies report that 43% of UC patients have a relapsing and remitting course during the first 12 months following diagnosis, and 16% relapse within 1 year of de- escalation of biological treatment. Over a 10-year time-frame, 67-83% of patients relapse, dependent on the clinical situation. Therefore, it is of interest to identify UC patients at risk of relapse so that treatment can be optimized to ensure longer periods in remission and fewer relapses using minimal, but adequate, therapy.

There is little knowledge about what initiates a relapse. Most research is focused on targets for treatment and, to a lesser degree, on surveillance markers. The problem with this approach is that it does not differentiate between the mechanisms of resolution and initiation of inflammation. Factors that can indicate when inflammation is resolved are not necessarily the same as the ones that can give an early warning of an impending relapse.

An initiative by experts under the auspices of the International Organization for the Study of Inflammatory Bowel Diseases (IOIBD) recommends a set of goals across different clinical factors in order to achieve optimal treatment. According to this expert panel, only symptoms, endoscopy and faecal calprotectin are recommended for surveillance of UC. These three markers can detect manifesting inflammation, but they have not been found to be capable of detecting imminent replace.

Biomarkers are used in many conditions for diagnosis and monitoring of disease states, but there are, however, still limitations in the utility and clinical relevance of mucosal markers. Despite research in this area, to date the data has been insufficient to make recommendations for patients on a variety of different treatment regimens, on when a patient is likely to relapse from a state of remission from UC.

WO2014/096873 reported that immunological markers can aid in the prediction of relapse in Inflammatory Bowel Disease (IBD) patients, within a limited patient group. The results of this study were applicable only to IBD patients that had finished a course of treatment with biological agents, such as anti-TNFa therapy. It was found that, within this limited patient group, those with increased levels of TNFa, IL-17 and IFN-y cytokines within their mucosa were more likely to relapse than those with normalised levels, when compared to healthy controls. While it was shown that specific cytokine profiles can reflect the disease state of an IBD patient, the methods are not applicable in a clinical setting, where patients are treated with a broad spectrum of medications and not just anti-TNFa therapies. Patients may also be in different stages of disease, i.e. patients may be in the middle of treatment rather than at the end.

In Gundersen MD et al., (Scl. Rep. 6, 35403 (2016)) it was found that a loss of IL33 expression could be a marker for disease remission in UC patients that had finished anti-TNFa therapy. This finding suggested that levels of IL33 are induced and maintained by acute disease inflammation, and that a loss of IL33 in patients - post anti-TNFa therapy - showed that therapy had been successful and had achieved a state of remission. Again, this finding was limited in that it applied to patients that had completed a course of anti-TNFa therapy, meaning that the finding is not applicable for the heterogeneous patient groups that are relevant for clinical applications.

There therefore remains a need for a method of predicting whether a UC patient in remission is likely to relapse that can be applied widely, irrespective of the treatment regimen previously applied to each patient. A method of this type would prevent unnecessary escalation of the treatment of patients that are unlikely to relapse, preventing redundant costs and unpleasant side effects.

The present inventors have surprisingly shown that patients in remission from UC can be monitored for relative expression levels of IFNG and IL33 in their gastrointestinal mucosa, and can be stratified into two groups. A relatively higher level of IL33, when compared to the level of IFNG, is predictive of stable remission in UC patients. Patients with relative levels of IFNG and IL33 expression that are similar, or wherein the patient has relatively higher IFNG expression, are more likely to experience imminent relapse. This is useful to know in terms of patient management and choosing the most beneficial methods of treatment in order to achieve or maintain stable remission. This finding applies to patients in clinical and endoscopic remission, regardless of their disease duration or treatment regimen.

Thus, the invention provides a method of determining whether a patient in remission from ulcerative colitis (UC) is in stable remission, said method comprising quantifying the level of gene expression of IFNG and IL33 in a gastrointestinal (Gl) mucosal sample from said patient and comparing said expression levels.

Preferably, the level of gene expression of IFNG and IL33 is determined by analysing a test sample, which is obtained from or removed from said patient by an appropriate means. The determination is thus preferably carried out in vitro.

As the method involves the determination of the level of more than one cytokine (i.e. IFNG and IL33, sometimes referred to for convenience herein as cytokine(s)), the method may be performed on a single sample in parallel (e.g. by multiplex analysis) or on a single sample in sequence (e.g. where the single sample is assayed multiple times, once for each cytokine (i.e. IFNG and IL33) that is to be detected or the level thereof determined). Alternatively, the analysis may be performed on multiple (e.g. 2, 3, 4, 5, 6, 7, 8 or at least 2, 3, 4, 5, 6, 7, or 8) samples obtained from the same patient.

The term stable remission (SR), as adopted herein, is used to describe those patients who are more likely to continue in long-term (12 months or more) remission (and thus have a reduced likelihood of relapse) on their current treatment regimen. For the purposes of the present invention and in view of the data presented herein, a patient is classified as being in SR if the patient does not require escalation in their treatment within a 12-month follow-up period from when the analysis of IFNG and IL33 levels was performed. In preferred embodiments, a patient whose cytokine expression levels indicate SR in accordance with the methods of the present invention will remain in SR for at least 12, or at least 18, or at least 24 months. In other words, SR means that the patient is unlikely to relapse in the 12 months after performance of an analytical method of the invention, relapse being defined as an escalation in their treatment regimen. As discussed elsewhere herein, in the field of UC there is a widely accepted pyramid structure of escalating therapeutic options.

Thus patients who are in SR have a reduced likelihood of relapse compared to those who are not in SR. Relapse in this context may mean a recurrence of the symptoms of UC. The symptoms experienced by the individual patient are diverse but the common feature is the presence of lesions in the Gl tract and, in view of this, relapse is characterised by a return of such lesions and/or an assessment based on one of the clinical scoring systems discussed herein.

Alternatively viewed, the present invention provides a method of determining whether a patient in remission from UC is likely to relapse (e.g. in the next 12 months), said method comprising quantifying the level of gene expression of IFNG and IL33 in a gastrointestinal (Gl) mucosal sample from said patient and comparing said expression levels.

Alternatively viewed, the present invention provides a method of quantifying the level of gene expression of IFNG and IL33 in a gastrointestinal (Gl) mucosal sample from a patient in remission from UC and comparing said expression levels.

The methods of the invention as described herein can be carried out on any subject which may suffer from UC. The methods are generally carried out on mammals, for example humans, other primates (e.g. monkeys), laboratory mammals (e.g. mice, rats, rabbits, guinea pigs), livestock mammals (e.g. horses, cattle, sheep, pigs) or domestic pets (e.g. cats, dogs). In preferred embodiments, the mammals are humans. However, in other embodiments, IFNG and IL33 and can be used as SR markers in any appropriate animal model.

Scoring systems based on a visual examination of the Gl tract exist to determine the status and severity of UC, and these scoring systems are intended to ensure that uniform assessment of different patients occurs, despite the fact that patients may be assessed by different medical professionals, in diagnosis and monitoring of these diseases as well as in clinical research evaluations.

Clinical scoring systems also exist, with the same purpose. The findings on endoscopy or other examination of the mucosa can be incorporated into these clinical scoring systems, but these scoring systems also incorporate data based on symptoms such as stool frequency, rectal bleeding and physician's global assessment. UC has a variety of symptoms that affect quality of life, so certain of these scoring systems also take into account a quantitative assessment of the effect on quality of life as well as the quantification of symptoms.

One example of a scoring system for UC is the Mayo scoring system (Schroeder KW et al., N Eng J Med 1987, 317(26): 1625-1629). In the context of the present invention, ‘a patient in remission from UC’ will preferably have a Mayo score of 0 or 1 , with a Mayo endoscopic subscore of 0 or 1, wherein no points are tolerated for rectal bleeding and the Mayo score is no larger than 1. The methods of the present invention are suitable for assessing patients in remission. The Mayo scoring system is shown below.

Table 1 : shows the components of the Mayo scoring system

Patients will have typically been subjected to some form of UC treatment in order to achieve a status of remission as defined herein. The patient in remission may be a patient that has a complex and long disease history with UC, optionally where previous treatments have not worked successfully. Said patient may have had at least two rounds of remission and relapse over the duration of their disease.

In the methods of the present invention, “quantifying the level of gene expression” is to determine the level of gene expression of a target gene and ascribe to it a quantitative value. Levels of gene expression can be measured by measuring the levels of encoding nucleic acid molecules in said sample. mRNA is preferably measured, from which gene expression levels can be inferred. In the methods of the present invention, the expression levels may be quantified using qPCR (quantitative polymerase chain reaction). Additionally, the expression levels of IFNG and IL33 may each be determined relative to one or more reference genes. The expression levels of IFNG and IL33 may be quantified in terms of CT (cycle threshold) values in an amplification reaction. The CT values can be used to calculate the amount of target sequence and therefore the amounts of cytokine target nucleic acid in the sample. The expression levels of IFNG and IL33 may be quantified in terms of CT values in relation to one or more reference genes. dCT values can be calculated by subtracting the CT value of the reference gene (or the mean CT value if there are multiple reference genes) from the CT value of the gene of interest (i.e. IFNG and IL33).

In the methods of the present invention, the comparing of expression levels of IFNG and IL33 may comprise calculating the ratio between said expression levels. The ratio may be calculated, in the form of a ddCT value, by subtracting the dCT value of IL33 from the dCT value of IFNG. In the methods of the present invention, the ratio value may be compared to a threshold value. A threshold value is a (ratio) value that has been found to provide good discrimination between patients in SR and patients who are not. A ratio value above the threshold value may be prognostic for stable remission and a ratio below the threshold value may be prognostics for relapse.

According to the present invention the ratio value is preferably in the form of a ddCT value (dCT IFNG - dCT/L33) and preferably said threshold value is between 3.6 and 5.4, preferably between 4 and 5, e.g. around 4.2 or around 4.8. A threshold value between 4 and 5 corresponds to a level of expression of IL33 that is 16 to 32 times higher than the level of expression of IFNG, according to the equation; fold change = 2ddCT

In preferred methods of the present invention, a level of expression of IL33 that is 16 to 32 times higher (or at least 16 times higher) than the level of expression of IFNG is prognostic of stable remission. Alternatively viewed, in preferred methods of the present invention, a level of expression of IL33 that is not 16 to 32 times higher than the level of expression of IFNG is prognostic of relapse. Thus the ratio may be expressed as the number of times higher or lower one of the cytokines is than the other.

The methods of the invention optionally comprise the additional step of carrying out an endoscopic evaluation and/or a clinical evaluation of the patient.

The invention also further provides a method of obtaining information relevant to the prognosis of a patient in remission from ulcerative colitis (UC), said method comprising quantifying the level of gene expression of IFNG and IL33 in a gastrointestinal (Gl) mucosal sample from said patient and comparing said expression levels. Preferred aspects of this method are as disclosed elsewhere herein.

In such a method, information about the level of gene expression of IFNG and IL33 may be used together with one or more other clinical or laboratory investigations in order to provide information about the SR status of a patient. Examples of such other clinical or laboratory investigations include visual examination of the Gl tract and clinical assessment using the UC scoring systems discussed elsewhere herein.

The invention also further provides a method for the prognosis of a patient in remission from ulcerative colitis (UC), said method comprising quantifying the level of gene expression of IFNG and IL33 in a gastrointestinal (Gl) mucosal sample from said patient and comparing said expression levels, wherein a high level of expression of IL33 relative to IFNG is prognostic of stable remission and a level of expression of IL33 relative to IFNG which is not high, is prognostic of relapse. Preferred aspects of this method are as disclosed elsewhere herein.

In general the higher the level of IL33, relative to IFNG, the worse the prognosis.

The term "prognosis" as used herein includes a risk prediction of the probable course and clinical outcome associated with a disease. Associated with this is also the ability to classify or discriminate patients according to the probability of whether various treatment options may be of benefit to an individual. In the present case, said prediction of clinical outcome includes a prediction of any relapse with the patient’s current treatment and hence the need or benefit to the patient of an escalation of treatment.

The methods of the invention can also be used to identify subjects where their current treatment regimen is appropriate and sufficient for managing their disease, or subjects which might benefit from an escalation of therapeutic treatment.

Optimal treatment of UC depends on the patient, the nature of the specific disease and the severity of disease. Various therapeutic strategies are available, including the use of aminosalicylates (5-ASAs such as mesalazine), steroids (such as prednisolone), and biological therapies (such as infliximab) and the immunosuppressive agents azathioprine, methotrexate, 6-mercaptopurine. A combination of steroids and immunosuppresives is often used in more severe cases and an individual's treatment program will vary over time. Particularly severe cases may require surgery, such as subtotal colectomies, ileo pouch-anal anastomosis, proctocolectomy or a temporary or permanent colostomy or ileostomy.

A relatively new treatment for UC is anti-TNFa therapy. By anti-TNFa therapy, it is meant any therapy that inhibits or antagonises TNFa. This could include inhibiting the production of TNFa or its receptor, e.g. by inhibiting its transcription or translation, or inhibiting its activity, directly or indirectly. Various methods for achieving this are known in the art. Inhibitors and antagonists of TNFa thus include antisense molecules, RNAi molecules, ribozymes, antibodies (e.g. a monoclonal antibody) or other binding proteins and small molecules. Any of these may be directed against TN Fa or its receptor. Infliximab (sold as Remicade) and adalimumab (sold as HIIMIRA) have been used recently for the treatment of UC. This treatment is in general used for patients who have not responded to other drugs, where these drugs have caused side effects, or when surgery is not considered the right treatment. T reatment with infliximab is via infusion of the drug, over the course of several hours, requiring the patient to visit a hospital or medical centre. Current treatment protocols recommend infusion approximately every 6 to 8 weeks until visual healing of the mucosa is obtained. Treatment with adalimumab is by subcutaneous injection, once every 1 to 2 weeks.

Escalation of the patient’s treatment regimen according to the present invention includes the introduction of stronger or alternative medications or an increase in the dosage of a currently used medication.

In the field of UC there is a widely accepted pyramid structure of escalating therapeutic options. Baseline treatment is with a 5-ASA (aminosalicylate). The level above that is steroid treatment. The level above that is populated by immunosuppressive agents and the level above that employs so called biological agents, i.e. antibodies, in particular antibodies which target TNF-a. The final treatment option is surgery. Thus escalation can involve commencement of therapy to manage the UC, but more usually the patient will already be on medication and escalation will involve increased dosing within each level (as discussed above) or movement up the pyramid to a “higher” level of therapy.

Viewed alternatively, the methods of the present invention provide a method for the prognosis of a patient in remission from ulcerative colitis (UC), said method comprising quantifying the level of gene expression of IFNG and IL33 in a gastrointestinal (Gl) mucosal sample from said patient and comparing said expression levels, wherein a similar level of expression of IL33 relative to IFNG, or a lower level of expression of IL33 relative to IFNG is prognostic of relapse. Thus, the methods of the present invention provide a method of determining whether a patient is likely to relapse.

The methods of the invention can also be used to monitor the progress of disease in a subject. The methods of the present invention can thus be used to monitor the progress of UC, to assess the effectiveness of therapy (e.g. anti-TNFa therapy) or to monitor the progress of therapy, i.e. can be used for active monitoring of therapy. An increase in the level of IL33 relative to IFNG or maintenance of relatively higher levels of IL33 compared to IFNG levels is indicative of improvement or effective therapy. A decrease in the level of IL33 relative to IFNG or maintenance of similar levels of IL33 and IFNG is indicative of deterioration.

Preferred embodiments, features and additional steps described above in the context of other methods of the invention apply, mutatis mutandis, to this method and, indeed, to all methods or kits of the invention.

If the levels of IL33 are determined to be similar to, or relatively lower than levels of IFNG, then the clinical decision would usually be to escalate treatment (i.e. introduce a stronger therapeutic or increase the dosage of the current therapeutic). If the level of IL33 is determined to be relatively higher than IFNG levels then the clinical decision would usually be to maintain the current treatment regimen.

As noted above, the invention has particular utility in terms of determining whether a patient requires the escalation of treatment to avoid relapse. A patient who is in SR is less likely to require escalated treatment to remain relapse free, or to remain in continued remission, than a patient who is not in SR. As such, if a determination is made that a patient is not in SR, it may be beneficial to escalate treatment. If a determination is made that a patient is in SR, current treatment will usually be continued.

Thus in a further aspect, the method of determining whether a patient in remission from ulcerative colitis (UC) is in stable remission, as defined herein, further includes the step of maintaining the patient’s current treatment regimen if said patient is in SR or escalating the patient’s current treatment regimen if said patient is not in SR.

In a further aspect the present invention provides a method of determining whether a patient in remission from ulcerative colitis (UC) is in stable remission, said method comprising quantifying the level of gene expression of IFNG and IL33 in a gastrointestinal (Gl) mucosal sample from said patient and comparing said expression levels, wherein a high level of expression of IL33 relative to IFNG is prognostic of stable remission and the patient’s current treatment regimen is maintained, and wherein a level of expression of IL33 relative to IFNG which is not high is prognostic of relapse and the patient’s current treatment regimen is escalated.

If treatment is continued then the method of the invention may be repeated at regular intervals (e.g. every 3, 6, 9 or 12 months), such that at each time point a decision is made (and optionally acted upon) about whether to continue or escalate treatment. If the first time that the method is carried out it is found that the patient is not in SR, treatment can then be escalated and continued until such a time as it is found that the patient is in SR. Preferably the assessment of IFNG and IL33 levels is only carried out on a patient in endoscopic remission (e.g. having visually healed mucosa) and all methods referred to above thus further optionally comprises the step of carrying out an endoscopic assessment of said patient before assessing the levels of gene expression of IFNG and IL33.

The invention also provides an assay method comprising:

(i) obtaining a Gl mucosal sample from a patient with UC;

(ii) optionally processing said sample;

(iii) measuring the level of IFNG and IL33 gene expression in said sample;

(iv) comparing the levels measured in step (iii) (wherein a high level of expression of IL33 relative to IFNG is prognostic of stable remission, whereas a level of expression of IL33 relative to IFNG which is not high, is prognostic of relapse).

This method may conveniently be used in order to decide whether or not to escalate treatment of the patient.

Said assay method may be used to determine whether said patient is in SR, in accordance with the methods described elsewhere herein. Preferably step (iii) is only carried out on a patient in remission as defined herein and said method further optionally additionally comprises the step of carrying out an endoscopic assessment of said patient before step (iii). Methods of endoscopic scoring (e.g. Mayo) are described herein.

The Gl tract contains mucous membranes (mucosa). These are the linings of the Gl tract and they are of mostly endodermal origin, and covered in epithelium. The mucosa thus contains epithelium, the lamina propria and the muscularis mucosa. The samples of use in the invention are mucosal samples. Such tissue samples may be obtained by biopsy e.g. (and preferably) during an endoscopic procedure, or may be obtained from part of a Gl tract tissue which has been removed surgically and may be a portion of the excised tissue. Samples may be collected via endoscopy at routine follow-up appointments. Preferably the sample is obtained from the lower Gl tract, i.e. from the jejunum, the ileum, the cecum, the (sigmoid) colon, the rectum or the anus; the sigmoid colon and rectum are particularly preferred.

The methods of the invention may thus optionally further comprise the step of obtaining the sample from the patient, e.g. the step of obtaining a sample of Gl mucosa from the patient. The sample may be used in the methods of the invention in the form in which it was initially retrieved. The sample may also have undergone some degree of manipulation, refinement or purification before being used in the methods of the invention. Thus the term "sample" also includes preparations thereof, e.g. relatively pure or partially purified starting materials, such as semi-pure preparations of the above mentioned samples. The term "sample" also includes preparations of the above mentioned samples in which the RNA of the sample, has undergone reverse transcription.

The way in which the sample is manipulated, refined or purified will depend on the form in which the levels of gene expression of IFNG and IL33 is to be detected. The purification may be slight, for instance amounting to no more than the concentration of the tissue, solids, or cells of the sample into a smaller volume or the separation of cells from some or all of the remainder of the sample. Representative cell isolation techniques and tissue manipulation, refinement or purification are described in the art.

The sample may be manipulated to enable the components of the detection method to access the material to be detected. For example the contents of individual cells in the sample may be released from the cells, e.g. by homogenisation or by lysing the cells of the sample, or by making tissue sections. The material to be detected e.g. mRNA may furthermore be fully or partially purified therefrom using standard methods which are known in the art.

For example, the invention may use a preparation of the nucleic acid from the above mentioned samples. Such preparations include reverse transcription products and/or amplification products of such samples or nucleic acid preparations thereof. The predominant nucleic acid of the nucleic acid preparation is mRNA or cDNA obtained by reverse transcription therefrom.

Techniques for the isolation of nucleic acid from samples, including complex samples, are numerous and well known in the art and are described at length in the literature.

The methods disclosed herein thus optionally further include the steps of manipulation, refinement or purification that are required to prepare the mucosal sample for analysis using any one or more of the above described processes, alone or in combination. In a particularly preferred embodiment the steps of obtaining RNA from said sample and/or reverse transcription of RNA from the sample are carried out. The level of gene expression of IFNG and IL33 in the sample may be determined by any appropriate means, a number of which are well known and documented in the art and some of which are commercially available. The level of gene expression of IFNG and IL33, or fragments thereof, in a mucosal sample can be measured by measuring the levels of mRNA encoding IL33 and IFN-y.

The level of gene expression of IFNG and IL33 may be expressed as an absolute amount (e.g. mRNA), an amount per unit of tissue or unit of other material (e.g. total RNA) or may be expressed as a relative measure (e.g. compared to the level of a housekeeping gene or protein or other control gene or protein).

A preferred method for measuring the level of mRNA encoding IL33 and IFN-y is qRT-PCR.

The amount of gene expression of IFNG and IL33 in the sample from the patient under investigation may be determined by measuring the levels of encoding nucleic acid molecules (mRNA) in said sample.

Levels of mRNA can be quantitatively measured by hybridisation techniques in which the binding of a probe to the nucleic acid molecule (referred to herein as the target nucleic acid) is measured. In such cases the probe may be labelled in such a way that it can be detected directly or indirectly. After contact of such a probe with the sample under conditions which allow hybridisation, and typically following a step (or steps) to remove unbound labelled oligonucleotide and/or non-specifically bound oligonucleotide, the strength of the signal from the label of the probe emanating from the sample under investigation (i.e. the amount of label bound to the sample) will be proportional to the amount of hybridised oligonucleotide and thus to the amount of target nucleic acid. In preferred embodiments the label is selected such that it is detectable only when the probe is hybridised to its target.

A preferred approach for measuring target nucleic acid abundance is based on amplification techniques. If the appropriate conditions are selected, such a reaction can be performed such that the amount of amplification product obtained will be proportional to the amount of target nucleic acid in the sample. Thus, for example, the amount of product the amplification reaction provides is proportional to the amount of the IFNG and IL33 mRNA in the sample.

Amplification can be achieved by any convenient primer-dependent nucleic acid amplification reaction. Most conveniently the polymerase chain reaction (PCR) will be used, although the skilled man would be aware of other techniques. For instance LAR/LCR, SDA, Loop-mediated isothermal amplification and nucleic acid sequence based amplification (NASBA)/3SR (Self-Sustaining Sequence Replication) may be used.

Many variations of PCR have been developed, for instance real time PCR (also known as quantitative PCR, qPCR), hot-start PCR, competitive PCR, and so on, and these may all be employed where appropriate.

In one basic embodiment of the invention using a PCR based amplification an appropriate primer pair is contacted with a reaction mixture containing the sample and free nucleotides in a suitable buffer under conditions which allow hybridisation. Thermal cycling of the resulting mixture in the presence of a DNA polymerase results in amplification of the sequence characteristic of the cytokine (i.e. IL33 and IFN-y). Optimal performance of the PCR process is influenced by choice of temperature, time at temperature, and length of time between temperatures for each step in the cycle.

Modifications of the basic PCR method such as qPCR (real time PCR) have been developed that can provide quantitative information on the template being amplified. Numerous approaches have been taken although the two most common techniques use double-stranded DNA binding fluorescent dyes or selective fluorescent reporter probes, e.g. SYBR Green qPCR assays or hydrolysis probe assays.

Double-stranded DNA binding fluorescent dyes, for instance SYBR Green, associate with the amplification product as it is produced and when associated the dye fluoresces. Accordingly, by measuring fluorescence after every PCR cycle, the relative amount of amplification product can be monitored in real time. Through the use of internal standards and controls, this information can be translated into quantitative data on the amount of template at the start of the reaction.

To simplify the detection of IFNG and IL33 expression levels in patient samples in a clinical setting, a one-step qPCR protocol in which the reverse transcription and qPCR reactions are combined within the one reaction step is preferable to traditional protocols which require a separate reverse transcription step. Methods for optimising a one-step qPCR protocol are known in the art and kits for performing one-step reactions are available commercially. To further simplify diagnosis in a clinical setting, reagents and primers for the one step qPCR procedure could be premixed within the wells of a suitable multi-well plate such as a 96 or a 384 well plate. Enzymes and primers premixed in wells could be stabilised for storage by maintaining at a low temperature such as at -20 or -80°C and/or lyophilising the mixture in the wells. An alternative method of simplifying the one-step qPCR procedure for use in a clinical setting is to optimise the reagent as beads, whereby one bead can be added for one reaction. To simplify the diagnostic procedure for use in a clinical setting, the qPCR protocol can be a duplex or a multiplex assay containing primers complementary to at least one of IFNG and IL33 and at least one housekeeping gene. Primers and fluorescent reporter probes complementary to at least one of IFNG and IL33 are contained within a reaction mixture comprising a reporter probe complementary to one or more housekeeping genes. Housekeeping genes are selected based on their stability in UC vs normal intestinal mucosa. Preferably, the expression levels of the housekeeping genes are similar to the expression levels of at least one of IFNG and IL33, whichever it is intended to measure in the assay in question, in intestinal mucosa.

Detection of oligo nucleotides by way of a duplex or multiplex assay is accomplished using fluorescent reporter probes, which can be detected at different wavelengths for each gene to be examined. Examples of technologies suited to duplex and multiplex experiments are Scorpion® (DxS ltd.) and TaqMan® (Roche Molecular Systems, Inc.) nprobe and primer sets.

In the methods of the invention the step of determining (quantifying) the level of gene expression of IFNG and IL33 preferably comprises:

(i) contacting the sample with oligonucleotides specific for the target sequence;

(ii) performing a primer-dependent nucleic acid amplification reaction; and

(iii) determining the amount of amplification product produced from the oligonucleotide in said primer-dependent nucleic acid amplification reaction.

If the amplification method used is itself quantitative, e.g. amplification methods in which internal standards and controls are incorporated (for instance qPCR) the method can provide quantitative data.

The cycle threshold (CT) can be used to calculate the amount of target sequence and therefore the amounts of IL33 or IFNG target nucleic acid in the sample. In all qPCRs there is a threshold at which the fluorescence of the amplification product is detected above background. The cycle at which this threshold is crossed is the CT. In the exponential phase of the reaction, the quantity of DNA theoretically doubles every cycle and so relative amounts of DNA can be calculated between samples by comparing CT values falling in the exponential phase. If the comparison is made with samples with a known quantity of template, the quantity of template in the test sample can be calculated and the amount of target nucleic acid present in the sample can be determined.

For any of the above methods for detecting nucleic acid molecules, appropriate primers or probes can be designed based on the publicly available sequences for IFNG and IL33. The skilled person is thus able to design probes or primers for any particular application or detection method and suitable examples are set out in the Examples hereto.

In a yet further aspect of the present invention is provided a kit for use in the methods described herein, said kit comprising agents suitable for determining the level of gene expression of IL33 or IFNG in a Gl mucosal sample. Such agents will typically have binding affinity for the nucleic acids encoding IL33 and IFN-y. Other preferred agents are labelled or unlabelled oligonucleotide probes or primers suitable for detecting IFNG and IL33 nucleic acid molecules. Kits may also comprise labelled or unlabelled oligonucleotide probes or primers suitable for detecting reference genes, e.g. HPRT1, RPLPO and/or ACTB. Kits may also comprise detection means and sampling means. Kits may comprise a receptacle adapted to receive a sample of Gl mucosa. The kit may also comprise reaction mixtures and buffers etc. The kit may be provided with instructions, in printed or other form, which may include control values for comparison with test values obtained using the kit to perform the methods of the invention.

The invention will be further described with reference to the following nonlimiting Examples with reference to the following figure in which:

Figure 1 is a survival plot of the IFNG:IL33 ratio. The Kaplan-Meier plot illustrates the difference in survival probability between a high and low ratio of IFNG:IL33, tested with a log-rank test. Dotted lines represent 12 months survival.

Examples

Example 1 - Predicting Relapse in Patients in Remission from Ulcerative Colitis (UC)

Methods

Patients

This study was a part of the Advanced Study of Inflammatory Bowel (ASIB) prospective study at the University Hospital of Northern Norway, Tromso. All study participants gave written, informed consent. The study, including storage of biological material, was approved of by the Regional Committees for Medical and Health Research Ethics, division North (REK Nord ID:2012/1349).

41 Ulcerative Colitis (UC) patients in clinical remission were recruited from the Advanced Study of Inflammatory Bowel Disease (ASIB) at three different time points: at a routine follow-up, at a 6-month control after initiation of biological treatment, or at an evaluation for terminating biological treatment. UC was diagnosed according to established diagnostic definitions (Magro F et al., Journal of Crohn's and Colitis. 2017; 11(6):64970). Remission was defined as a UCCS/Mayo clinical score of 0 or 1, with a Mayo endoscopic subscore of 0 or 1 (Magro F et al., Journal of Crohn's and Colitis. 2017;11(6):64970). No points were allowed for rectal bleeding, and patients with a total Mayo score larger than 1 were not included. Independent of their medication, only patients with remission as confirmed by this criteria were included in the study. A patient’s current treatment regimen was recorded at time of inclusion. The different steroid and anti-TNFa treatments were grouped together into categorical yes/no variables. Relapse was defined as contact with a clinician, where the clinician determined the need for escalation of treatment, regardless whether the escalation was an increase in dosage of mesalazin or addition of biological treatment.

Histopathology

All mucosal biopsies were immediately fixed in 10% formalin after sampling and embedded in paraffin. Multiple 3pm sections were cut with a Micron microtome (HM355S, Thermo Fisher, Tudor Rd, Runcorn WA7 1TA, United Kingdom) and stained with haematoxylin and eosin. Two general pathologists, blinded to the endoscopic score and biopsy location, investigated the slides. If there were disagreements, a third gastrointestinal (Gl)-specialized pathologist gave the final score. Each slide was evaluated with the three most applied histologic UC indices; Geboes Score (GS), Robarts Histopathological Index (RHI) and Nancy Index (Nl). The histological evaluation method is further described in Arkteg CB et al., (PloS one.

2021 ;16(3) :e0248224). The strict histological definition of remission was GS < 2A.1, RHI < 4 and Nl = 0 and a more relaxed definition was defined as GS < 3.1, RHI < 5 and Nl < 2.

Quantitative Polymerase Chain Reaction (qPCR)

Quantification of gene expression was performed as close to the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) guidelines as possible (Bustin SA, et al., Clinical chemistry. 2009;55(4):61122). The geometric mean between the expression of HPRT1 and RPLPO was used as reference gene for the SYBR-Green assays, and ACTB was used for the hydrolysis probe assays. A more detailed description of the qPCR method can be found in Arkteg CB et al., Scandinavian Journal of Gastroenterology. 2020:19. The genes tested were IL1B, IL6, IL8, IL10, IL18, IL21, IL23, IL33, TNF, TGFb, IFNG, TLR4, ST2, SPI1, TBX21, F0XP3, GATA3, RORC, ACTB, IL17, and IL4. The corresponding primers and probes are listed in the table below.

Table 2: Genes with the corresponding primers, probe and tested efficiency

Immunostaining

Endoscopic biopsies from the sigmoid colon of 9 UC patients in remission and 10 non-IBD controls were analysed. Formalin-fixed, paraffin embedded, 4pm sections were deparaffinised and rehydrated through graded steps of xylene and alcohol.

Antigen retrieval solution (DAKO, Glostrup, Denmark) was used and the sections were boiled in a water bath for 20 minutes, followed by 20 minutes cooling at room temperature. Goat serum 10% was used for blocking (20 minutes) prior to primary antibody incubation. Monoclonal antibodies for IL33 [1ug/ml] (anti-mouse, Nessyl, Enzo Life Sciences) and vWBF [1/100] (anti-rabbit, Abeam) were incubated overnight at 4°C. Secondary goat antibodies conjugated with alexa555 or alexa647 for rabbit or mouse (Life technologies) were used as appropriate at [1/1000] and incubated for 90 minutes at room temperature. Hoechst 33258 (Life technologies) was used for nuclear staining. Sections were mounted with Fluoromount aqueous mounting medium (Sigma Aldrich/Merck, St. Louis, USA). Isotype and concentration matched antibodies were used (IgG mouse, rabbit, Cell signal Technology, Danvars, MA). Tonsillar tissue served as positive controls.

A Zeiss LSM780 CLSM microscope (Carl Zeiss Microscopy, Zena, Germany) was used with the Zen 2012 software (black edition) for taking images. Three representative images at x20 magnification were taken of each section. Nuclear IL33 signal was analysed using the Velocity® 6.3 software using positive fluorescent signal of total nuclear area/positive nuclear area. Image processing was performed with Adobe Photoshop CC (Adobe System Software, Ireland Ltd, Dublin) with histogram adjustments only applied for whole images.

Statistics

Missing data was imputed using a nearest neighbour averaging method. Variables with more than 20% missing values were excluded from the analysis.

The data set had more variables than cases, which challenges the ordinary approach of backward/forward selection for identifying potential predictors, due to the risk of producing an overfitted model (all variables are listed in Table 5). To avoid this we applied a strategy often used for high dimensional data such as microarrays and sequencing data (Hesterberg T, et al. Statist Surv. 2008; 2:6193, and Bovelstad HM et al. Predicting survival from microarray data a comparative study. Bioinformatics. 2007;23(16):20807).

To narrow down the number of potential covariates for the model building, a LASSO regression was run 10,000 times with different seed each time. This resulted in a range of different penalizing factors, which again produces a range of covariates with nonzero coefficients, i.e. covariates with potential significance. In this study the lowest penalty i.e. the lowest lambda (A) value of the range, was chosen to avoid excluding potential covariates. The LASSO regression was done with glmnet package for R.

The non-zero covariates then went through a forward stepwise model selection by AIC and univariate cox regression to identify covariates for a relapse-predicting model. So as to not violate the “10 events per covariate” rule of thumb, two of the most significant variables were transformed to a ratio. The optimal cut-off for relapse prediction was determined with a ROC analysis and Youden J statistics. This cut-off was used to dichotomize the ratio into a high and a low category, which then was analysed with the Kaplan-Meier method to evaluate and illustrate the difference in relapse free survival between the two categories.

The calculations were done with Rstudio Version 1.3.1056. The packages used are listed in Table 3.

Table 3: R packages

Results

Of the 41 patients in UC remission, 14 experienced a relapse during the follow-up period. The median remission time was 6.5 months (Interquartile Range (IQR) 6.6), whilst those who did not experience relapse had a median follow-up period of 12 months (IQR 11).

A total of 42 variables, across clinical, medication, histopathological and gene expression categories, were evaluated (table 5). Among the initial clinical characteristics there were no statistical differences between those who relapsed and those who did not (p<0.05) (table 4).

Table 4: Baseline data.

Abbreviations: Ulcerative Colitis Clinical Score (UCCS), Mayo Endoscopic Score (MES), Geboes Score (GS), Nancy Index (Nl), Robarts Histopathological Index (RHI)

Variables Tested Table 5: All covariates tested with the LASSO regression

Abbreviations: Ulcerative Colitis Clinical Score (UCCS), Mayo Endoscopic Score (MES)

After 10,000 iterations, 8 variables had been identified as non-zero by LASSO regression. IL33, IFNG and MES were the variables with the lowest AIC in a forward stepwise regression. Although MES had a non-zero coefficient, it failed to reach significance.

Table 6: Result from forward selection.

Because of the opposite effect of the IFNG and IL33 values, we subtracted the IL33 dCT value from the IFNG dCT, corresponding to a ratio (JFNG:IL33) in linear values. The ROC analysis of this ratio resulted in an optimal cut-off of 4.2, which had a specificity of 92.6% and a sensitivity of 57.1%, with an AUC of 76.5. With this cut-off, the positive predictive value (PPV) and negative predictive value (NPV) were 80% and 80.6%.

Survival Analysis

The Kaplan-Meier analysis on the dichotomized IFNG:IL33 ratio (low ratio < 4.2 and high ratio > 4.2) revealed a statistically significant better outcome for participants with high ratio compared to the low ratio group (log rank test, p-value of p < 0.001) (Figure 1). The one-year cumulative relapse was 26.7% in the low ratio group and 74.8% in the high ratio group.

The IFNG:IL33 ratio was a better predictor of patient relapse than both endoscopic score and clinical score. The univariate cox analysis showed 5.3 times (95%CI = 1.815.4) higher risk of relapse in the low ratio group than in the high ratio group.

As IL33 showed an association to relapse, its relationship with its receptor IL1RL1 was investigated. IL33 and IL1RL1 showed a positive correlation, but there was no difference between those who relapsed and those who did not. Immunostaining confirmed the presence of nuclear IL33 in both UC and normal controls. IL33 was present in endothelial cells and mononuclear cells in the lamina propria. No intestinal epithelial cells were positive for IL33, neither in UC nor in the control group. Statistical analysis did not reveal a significant difference between the two groups. Furthermore, no clear difference in IL33 pattern was observed between relapse and non-relapse patients.

Discussion

In this study, we found a 5.3 times increased risk of relapse in UC patients with a low ratio between IFNG and IL33. There was no apparent difference in IL33 distribution in the mucosa between relapsers and non-relapsers. Histology could not predict relapse in our data.

The IFNG:IL33 ratio has not been previously described as a predictor of relapse in UC. The beneficial ratio consists of relatively higher dCT of IFNG than IL33.

This IFNG:IL33 ratio may serve as an early detection warning for patients in remission, as the median time for relapse for the low ratio group was 7.5 months while the high ratio group never reached median time during the follow up. The ratio achieved an adequate sensitivity of 57.1%, but with high specificity of 92.6%. Despite the only adequate sensitivity, the positive predictive value was 80.0% with a negative predictive value of 80.1%, indicating that a patient with a low ratio is likely to experience a relapse. In a clinical setting, this ratio could aid the clinical decision making associated with patient management; the high specificity would prevent unnecessary escalation of treatments, which in turn results in better patient care.

In this study several clinical factors where investigated. None of the UCCS, MES or histology factors showed an association with relapse. UCCS and MES showed potential, as they had nonzero coefficients, but ultimately failed to reach significance.

In conclusion, the ratio between IFNG and IL33 gene expression is a new and promising biomarker for predicting disease relapse in UC patients in remission, independent of medication. The mechanisms behind this ratio may have interesting implication to the pathophysiology of UC relapse.

Claims

- 25 - Claims

1. A method of determining whether a patient in remission from ulcerative colitis (UC) is in stable remission, said method comprising quantifying the level of gene expression of IFNG and IL33 in a gastrointestinal (Gl) mucosal sample from said patient and comparing said expression levels.

2. The method of claim 1 wherein comparing said expression levels of IFNG and IL33 comprises calculating the ratio between said expression levels.

3. The method of claim 1, wherein a high level of expression of IL33 relative to IFNG is prognostic of stable remission and a level of expression of IL33 relative to IFNG which is not high, is prognostic of relapse.

4. The method of any preceding claim wherein, when the level of expression of IL33 is 16 to 32 times higher than that of IFNG, that is prognostic of stable remission.

5. The method of any preceding claim wherein the patient is not receiving biological treatment for their UC.

6. The method of any one of claims 1 to 4 wherein the patient is receiving biological treatment for their UC.

7. The method of claim 6 wherein the biological treatment is an anti-TNFalpha therapy.

8. The method of any one of claims 1 to 7 wherein said patient is in clinical and endoscopic remission.

9. The method of any preceding claim wherein a decision about the patient’s future treatment regimen is made following said comparison of expression levels.

10. The method of any one of claims 1 to 9 wherein the expression levels are quantified using qPCR (quantitative Polymerase Chain Reaction).

11. The method of any one of claims 1 to 10 wherein expression levels of IFNG and IL33 are each determined relative to one or more reference genes.

12. The method of claim 10 or 11 wherein expression levels are quantified in terms of CT (cycle threshold) values.

13. The method of claim 12 wherein expression levels are quantified in terms of dCT values in relation to one or more reference genes.

14. The method of claim 13 wherein the ratio of dCT IL33 to dCT IFNG is calculated.

15. The method of claim 14 wherein the ratio value is compared to a threshold value.

16. The method of claim 15 wherein a ratio value above the threshold value is prognostic for stable remission and a ratio below the threshold value is prognostic for relapse.

17. The method of claim 15 or 16 wherein the threshold value is between 4 and 5.

18. A method of determining whether a patient in remission from ulcerative colitis (UC) is in stable remission, said method comprising quantifying the level of gene expression of IFNG and IL33 in a gastrointestinal (Gl) mucosal sample from said patient and comparing said expression levels, wherein a high level of expression of IL33 relative to IFNG is prognostic of stable remission and the patient’s current treatment regimen is maintained, and wherein a level of expression of IL33 relative to IFNG which is not high is prognostic of relapse and the patient’s current treatment regimen is escalated.