BE1022453B1 - Method for predicting a placebo effect in a subject - Google Patents

Abstract

The present invention relates to a method for predicting a placebo response in a subject, comprising collecting data by - questioning said subject about personality and health traits; and / or - performing one or more social and / or (bio) physical learning tests on said subject; characterized in that said data is used in a mathematical model stored on a computer to calculate a correlation between the input data, thereby assigning a Rating Factor to said subject, whereby said Rating Factor is a measure of propensity to issue a placebo response and / or a measure of the intensity of said response.

Description

METHOD FOR PREDICTING A PLACEBO EFFECT IN A SUBJECT

TECHNICAL FIELD The invention relates to the technical field of methods for obtaining improved therapeutic treatments and improved clinical trials for therapeutic treatments. More particularly, this relates to methods for predicting a response or placebo effect and systems for obtaining such predictions and for utilizing the data produced from predictions.

BACKGROUND

The clinical development of new drugs or treatments for major therapeutic indications such as chronic pain (including neuropathic pain, migraines, ...), mental disorders, depression, epilepsy, Parkinson's disease, asthma is complex and not effective. This is mainly due to the fact that many Phase 2 and 3 clinical trials are abandoned or fail due to safety or inability to demonstrate a clear superiority of the drug tested versus placebo despite promising results seen in vitro and / or preclinical studies. The reason for this is that, in therapeutic fields such as for example pain or depression, the placebo response itself has a pronounced effect on the primary results of clinical studies. More specifically, it is now recognized that the behavior of the investigator vis-à-vis his patient as well as the expectations of patients (in terms of drug efficacy and overall well-being) have an impact. strong on the patient's assessment of the effectiveness of the medication.

Thus, the sharp increase in the drop-out rate in drug development is a major concern for both clinicians and pharmaceutical industries who face major difficulties in obtaining marketing authorization for new drugs in the marketplace. current therapeutic fields such as for example pain and depression. From a caregiver's point of view, properly managing the placebo response / effect can contribute positively to the better well-being of its patients. From the point of view of the pharmaceutical industry, controlling the placebo effect is essential to properly designing a clinical trial that allows a clear differentiation between the physiological effect of the drug under study and the other side. effects considered collectively as the placebo effect.

Furthermore, (i) the high impact of the placebo response on the evaluation of the efficacy of the drug and (ii) the lack of common traits among patients that can measure, at the population level, how far the placebo response interferes with the physiological evaluation of a new drug candidate makes it very difficult to demonstrate its superiority. As a result, scientists in clinical research as pharmaceutical industries need to design improved clinical trials and better characterize the patient's ability to distinguish the placebo response from the physiological effect of the drug being tested.

It has been recognized that the placebo effect is by nature multifactorial. On one side the effect is a learning phenomenon, which is influenced by the manipulation of different variables including patient waiting, previous (bio) physical experiences, observation and social learning as well as personal traits. Therefore, the placebo effect is mainly dependent on the patient. Each subject may report a different response based on his or her therapeutic history and personality aspects.

In addition it has been recognized that the placebo effect depends on the disease, whereby a subject will show an effect that varies from disease to illness.

In addition, it has been recognized that the placebo effect depends on time, whereby a subject will show a placebo response that changes with time or time of treatment. Therefore, patients may respond to a placebo effect differently at the beginning of a treatment compared to the response level during or at the end of a treatment. Subjects who respond to placebo or who report a propensity for "change of response" or "response drift" may be more likely to decrease dosages, improved outcomes, self-reported improvements perceived as superior, quality of life or similar.

In the same way, a subject can have a nocebo effect that changes with the time or the moment of the treatment. Therefore, patients may respond to a nocebo effect differently at the beginning of a treatment compared to the response level during or at the end of a treatment. Subjects who have a nocebo response or who report a propensity for 'change of response' or drift of response may be more likely to increase dosages, decreased therapeutic outcomes, self-reported improvements perceived as inferior, quality of life or similar.

Several questionnaires, biophysical tests or virtual reality tools have already been developed and used to evaluate some aspects of the placebo effect in a subject. However, because of their unconnected and very restrictive nature, these questionnaires and biophysical tests do not allow to give an accurate estimate of a placebo effect present in the subject. WO 2005 027 719 describes a method for predicting placebo effect predisposition based on biological markers. The procedure is very one-sided, and does not take into account the multifactorial nature of the placebo effect. WO 2013 039 574 discloses a method for selecting participants for a clinical trial whereby participants are filtered on the basis of their reactivity to the placebo treatment. The method in WO 2013 039 574 uses in this respect an evaluation of the body image of self or of self-actualization, for example a perception of the subject of his own self as a function of or in relation to his body. The method described in WO 2013 039 574 is one of the methods available in the prior art for classifying subjects in placebo responders and placebo nonresponders but is based solely on the evaluation of the adaptability to the perception of a subject of his image of body self. The evaluation according to WO 2013 039 574 fails to provide a method based on its own understanding of interrelationships between various factors of both psychological and physiological nature that contribute to a placebo effect. Therefore, WO 2013 039 574 fails to describe an authentication or a global and unbiased placebo response model of a subject. US 2014 0 006 042 describes a methodology for conducting studies, thereby producing a responder placebo index. The index is obtained by comparing data obtained in a patient with previously obtained data. The use of a comparative approach to determine a putative placebo response is not desired because such a comparison must be based on previously obtained data. If such an earlier data is erroneous or if there is even the slightest difference in the circumstances of the test, then the comparison may be unreliable. In addition, a deviation in the result may occur if the compared data does not come from the same subject. This may result in distortion in the result obtained. At present, both to reduce the level of drop-out rates in clinical trials and to improve the accuracy of the contribution of the physiological effect of (medical) treatment on a patient's overall response to disease treatment where the placebo effect or, more generally, to improve a treatment of diseases where the placebo effect occurs, prior art inappropriately solves the problem of precisely defining the propensity of a subject to respond placebo or to reveal a placebo effect. Secondly, the existing methods, especially the questionnaires, are time-consuming and impose a heavy burden on the patient to be tested.

The present invention aims to solve at least some of the problems mentioned above.

SUMMARY OF THE INVENTION

The present invention aims to provide a method and a tool for predicting the propensity of a placebo effect in a subject, said prediction is constructed on a multifactorial approach of features that are related to the placebo effect. The methodology and tool start from a predefined amount of data, derived from the subject, that is used in a mathematical model to define a correlation between the input data, whereby the correlation provides a measure of the response. placebo. The invention provides a way to produce a precise placebo score using a limited number of input variables. It has been unexpectedly observed that the relationship between the input variables (correlations or other forms of mathematical relationships between one or more random variables or data points) can be used to have a "direct" prediction of the placebo response (without "excessive" questioning of patients).

Due to the multifaceted approach of the present invention, said prediction is more reliable than other methods currently known in the art. Since it is based on a correlation between intrinsic characteristics and data obtained from a subject, often at a specific time point, neglecting the need to compare the latter with previously obtained data (for example from other topics), it is more reliable. Thus, the results of the present method can be deployed at various stages of patient treatment and / or clinical trials, including the distribution of placebo responders into various groups (classes) of a clinical study, all of which are known to be sensitive to placebo effect. The present invention relates to a method for predicting a placebo response in a subject according to claim 1. In other aspects, the present invention also relates to a method and a computer processing product and a complementary tool. diagnosis. The present invention also relates to methodologies that can be used in clinical trials or to improve the quality of the results thereof.

DESCRIPTION OF THE FIGURES

Figure 1 shows a schematic overview of an embodiment of the methodology according to the present invention.

Figure 2 shows a screenshot of a computer interface according to an embodiment of the present invention, whereby the intensity of a placebo response is predicted based on the input characteristics.

Figure 3 shows a decision tree following example 2.4.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides methodologies for determining a placebo effect in a subject, or for determining the propensity of a subject to respond to a placebo effect. The importance of the placebo effect in clinical trials and for treating patients began to be recognized only in the last decade. Some of the neuroanatomical and neurophysiological supports of the placebo effect have been elucidated in recent years, but the development of placebo prediction tools has so far been largely underexposed. It is the object of the present invention to develop a methodology and system for predicting a placebo response in a subject and for implementing it in drug and clinical trial design.

Unless otherwise defined, all the terms used in the disclosure of the invention, including technical and scientific terms, have the meaning commonly understood by the professional of the field to which this invention belongs. With additional guidance, definitions of terms are introduced to better appreciate the teaching of the present invention.

As used herein, the following terms have the following meanings: "From", "a" and "the" as used herein refer to both the singular and the plural unless the context clearly justifies it. Using an example, "a bucket" refers to one or more compartments. "Approximate" as used herein with reference to a measurable value such as a parameter, a quantity, a duration, and the like means to include variations of +/- 20% or less, preferably +/- 10% or less, more preferably of +/- 5% or less, more preferably of +/- 1% or less, and more preferably of +/- 0.1% or less of and from the value of specified, to the extent that such variations are appropriate to carry out the disclosed invention. However, it must be understood that the value to which the moderator "about" refers is itself also revealed in a specific way. "Understand", "understand", and "understand" and "understand" as used herein are synonymous with "include", "include", "include" or "contain", "contains", "contains" and are Global or specific terms specific to the presence of the following eg a component without excluding or excluding the presence of components, features, elements, members, additional steps, not listed, known to the field or disclosed herein. The enumeration of numeric ranges by threshold values includes all numbers and sub-totalized fractions in this range, as well as the threshold values enumerated. The term "% by weight" (percent by weight), herein and throughout the specification unless otherwise defined, refers to the relative weight of the component under consideration in relation to the overall weight of the formulation.

The present invention provides a method for predicting a placebo response in a subject. The method may include collecting data via the steps of: - querying said subject on personality traits and health status; and / or - to practice one or more social and / or (bio) physical learning tests in said subject.

In a preferred embodiment, a Notation Factor will be assigned to said subject, whereby said Notation Factor is a measure of propensity to issue a placebo response and a measure of the intensity of the response. For this purpose, the obtained data is used in a mathematical model, the result of said model being the Notation Factor.

This is different from what is currently known in the field. So far, no mathematical model or tool for qualifying, quantifying and / or predicting the placebo effect in a subject exists that takes into account a subset of the aspects that contribute to the placebo effect such as traits. personality, health characteristics, (bio) physical measures of a subject, etc. Single questionnaires or only (bio) physical tests currently used never give a value of a placebo effect, being stand-alone approaches. Not only do they fail to take into account the multifactorial nature of the placebo effect, but if the domain professional decides to use all of them (together or separately), it will fail to provide a measure of the placebo effect because corresponding investigations and tests is not reliable.

In the context of the present invention, the term "predictor" and any of its derivatives (predictive, prediction, ...) must be understood as providing a probabilistic image of an analyzed characteristic, said image is preferably calculated according to a computer model. If not or in addition, predicting must be understood as anticipating the evolution of said characteristic in time or during a predefined period of time.

With respect to the present invention, the term "pain disorder" should be understood as an acute or chronic pain experienced by a patient. The said pain disorders can be subdivided into three groups: pain associated with psychological factors pain associated with a general psychological and pathological state pain disorder associated with a general pathological condition.

Thus, said pain: - may be due to lesions or diseases that affect the somatosensory system (neuropathic pain); - be due to the activation of nociceptors (nociceptive pain); - to be provoked or augmented by mental, emotional or behavioral factors (psychogenic pain); - transient painful access, for example caused by cancer; or - inherent in sudden activity (painful episode).

With respect to the present invention, said "correlation" or "correlating" must be understood as a mathematical relationship between two or more random variables or data points. Preferably, said correlation is predictive or makes it possible to identify a predictive relationship between the variables analyzed.

In the context of the present invention, the term "Placebo" may be any of the usually inert or active substances, formulations, drug treatments or non-medicinal treatments administered, given or used in a patient, for example, tablets, suspensions or injections of inert principles, for example, sugar pills or starch pills, or other false treatments, for example, false surgeries, false psychiatric care, or others that have been used, usually as witnesses, of a " true "putative treatment (in order to achieve a claimed, deemed, or estimated therapeutic effect on a symptom, disorder, condition, or disease, or prescribed, recommended, endorsed, or promoted, knowingly or unknowingly, to another patient, despite the fact that the treatment is currently ineffective, has no known physiological effect, or is not specifically effective on the symptom, disorder, condition, or disease through iter).

In the context of the present invention, the term "placebo effect" means any of the specific or non-specific psychobiological phenomena attributable to placebo and / or treatment context regardless of whether the placebo is usually or not administered. The placebo effect as meant in the context of the present invention highlights the central role between expectations and suggestions, in the placebo-related phenomenon, and diseases.

In the context of the present invention, the term "placebo response" means the result of the placebo effect as expressed, perceived or measured by one or more subjects to qualify or quantify both improvement and degradation (nocebo response). a symptom or physiological condition in the context of placebo and / or treatment.

The placebo response not only includes the presence or absence of the response itself but also refers to the intensity of the response given or expressed by the subject.

Said placebo response may be dependent on the disease and / or time.

In the context of the present invention, the term "response change" or "response drift" means a change in the placebo response during a treatment, a clinical trial, or any intervention related to the treatment. health.

In the context of the present invention, "line" or "lines" should be understood as all kinds of variables, directly or indirectly related to a subject, which can be introduced into the model according to the present invention, and which are used to enter in the Notation Factor. In more detail, said traits are identified by a professional on the basis of the present understanding of the different aspects potentially related to a placebo aspect, and routinely collected using questionnaires and / or existing tests.

In the context of the present invention, "personality traits" should be understood as the subject's subject-related characteristics, the physical characteristics of the subject, and / or the personal history information of that subject. Said characteristics of the psyche may include, but not be limited to, emotional characteristics, behavioral characteristics, general beliefs of the subject and / or emotional traits.

The said health traits may include any information relating to the subject's health as well as that of the subject's family. Said health traits may for example include, but not be limited to, old and present diseases, treatments received, current and past use of drugs, potential health risks, genetic predisposition to develop a disease, etc.

In the context of the present invention, said social learning can be understood as a process where subjects observe the behavior of others and its consequences, or specific situations and patterns to modify their own behavior accordingly. Said social learning test includes providing to a subject information or behavioral, environmental and / or exemplary stimuli, thereby eliciting (or not) a response from said subject, based on the information received.

In the context of the present invention, said (bio) physical test must be understood as any test, in connection with the measurement or detection of a biophysical parameter. For example, said (bio) physical test may include but is not limited to measuring or analyzing a biological compound of said subject; measuring or detecting a biological reaction of said subject; performing a neurological test on said subject; measuring or sensing a sensory reaction; to perform a tactile test on said subject.

For example, the Somedic Thermotest device (Somedic AS, Stockholm, Sweden) can be used to deliver a quantized and reproducible thermal pulse via a Peltier thermometer - 2.5 □ 5 cm (12 cm2) - applied to the thenar eminence of the non-dominant hand.

Preferably, said (bio) physical test involves a neurological, somatosensory, tactile or analytical test, or virtual reality tools or any combination thereof.

Examples of such objective tests may include controlling heart rate, controlling blood pressure, controlling respiration, measuring one or more blood components or metabolites (eg, blood chemistry) or other biological fluid, measuring parameters. skin conditions such as blood flow, temperature, or skin conduction; or other physiological measures including measuring any brain or neurological activity, resonance cutaneous conduction (SCR), electroencephalography (EEG), quantitative EEG (QEEG), magnetic resonance imaging (MRI), functional MRI, (fMRI), CT-assisted tomography computer (CT), positron emission tomography (PET), electronystagmography (ENG), computer-assisted mono-photonic emission tomography (SPECT), magnetoencephalography (MEG), superconducting quantum interference device (SQUIDS), electromyography, eye tracking , and / or change in pupillary diameter, pain tests such as thermal pain procedure.

In the context of the present invention, said Notation Factor should be understood as a measure of a certain characteristic analyzed (in this case the propensity to exhibit a placebo effect or response). The Notation Factor may be a numerical factor or parameter, as an indication of the analyzed characteristic based on a specific scale, whereby the higher the numerical factor is up (or down) on the scale, the more likely the analyzed characteristic is to to be present. For example, in the context of the present invention, said Notation Factor may provide a scale with respect to the propensity of a subject to be eligible for a placebo effect. In another embodiment, said Notation Factor may be a classification of an analyzed subject. For example, in the context of the present invention, said Notation Factor may determine whether a subject is responder or nonresponsive to a placebo effect ("yes" or "no"). In yet another embodiment, said Notation Factor is a profile or overview of the Placebo response. In general, said Notation Factor is a (predictive) value (for example a color code, a definition, a term, a numerical factor, etc.) of the placebo response or the placebo effect of a subject.

In one embodiment, said Notation Factor will be compared to one or more threshold or threshold values to determine the presence of a placebo response in a subject. If said Notation Factor is higher than a predefined limit value, this is an indication of the presence of a placebo response, or a high propensity to develop it.

If the Notation Factor is below the limit value, but above a second limit value, a placebo response could then be present. Below the second limit value, the placebo response is not present.

In another embodiment, said Notation Factor will be located on or compared to a predefined scale, whereby the height of the Notation Factor is then directly proportional to the propensity to develop a placebo response or to the presence of a placebo response. in a subject.

The present method has the advantage of providing a model of a placebo effect or response, thus adopting a multifactorial and multi-integrated approach. Models of the placebo response have so far focused on a very limited amount of information, and studies have failed to provide a consistent link between the data collected and the placebo response. The present methodology and tools derived from it endeavor to take into account multiple facets of the placebo effect, thus providing a reliable tool for predicting a placebo response in a large amount of medical indications. For this purpose, the present invention describes a methodology and tools that make use of objectified data (for example obtained by testing and / or questioning a subject), and which must be considered as the "contribution" to the final prediction.

In a preferred embodiment, said method will include data from: - one or more personality surveys; - one or more health surveys; - one or more social learning tests; and - one or more (bio) physical tests that relate to a subject or to a subject.

In another embodiment, said method comprises any combination of 2 or 3 of the above investigations and / or tests.

Figure 1 shows a schematic overview of a possible methodology according to the present invention.

In one embodiment, said personality survey includes one or more selected questions from sets of questions for characterizing traits or personality characteristics of a subject that are stable over time and attributable to the person and are not the effect of his environment. The said personality-related set of questions includes one or more questions for measuring the five major (easily known domain) components of the personality, namely, individual openness to experience, awareness, extroversion, amiability, and neuroticism (or emotionality), all well known to the professional of the field.

In another embodiment, the survey includes one or more questions selected from question sets for measuring or evaluating the impact of a subject's environment on his or her perception of health issues.

The set of questions related to the impact of the environment includes: - one or more questions to measure the impact of the behavior (pleasant, open, severe ...) or the intervention (oral, acts ...) of the caregiver, - one or more questions related to the sensation of contagion, suggestibility or any other factor capable of influencing the balance between reflection and automatic processing of information on the appearance, evaluation, relief, evolution of a health-related symptom ... - one or more questions to assess the level of anxiety, fear, discouragement, despair, depression related to the environment of a clinical setting or a caregiver.

In another embodiment, said survey includes one or more selected questions from question sets for assessing the impact of a subject's environment on his belief in a just world, his psychological well-being, his psychological quality of life, satisfaction with life, resistance to stress and depression.

In another embodiment, said survey includes one or more questions selected from sets of questions for measuring subject's expectations of an external stimulus, positive and negative results of an intervention or treatment. , and to assess their propensity to have a positive or negative attitude toward external factors or health-related symptoms, specific treatments for relieving health-related symptoms.

In another embodiment, said survey includes one or more questions that are asked after exposure of said subject to information influenced by a wait or neutral information. For the purpose of the present invention, said information includes any information, directly or indirectly related to the test performed and / or placebo given and the mode of action of said placebo.

In another embodiment, the survey includes one or more questions selected from sets of questions to evaluate the response to a subject's attitudes or emotions to external stimuli. The set of questions includes questions to measure the level of control the subject believes he has over his life, the level of mastery of external factors or health-related symptoms of his life such as luck, fate, life events, or other strong influences (such as parents, health professionals, work colleagues, etc.) and to measure the level of mastery of other strong influences such as parents or social learning, ... on his attitude to resist, fight or overcome aggressive external factors or health-related symptoms.

In another embodiment, the survey includes one or more selected questions from question sets for assessing the level (severity) of health related symptoms.

The set of questions may include one or more questions to measure to what extent the subject considers that health-related symptoms influence his overall physical and psychological condition including the functioning of his body, his activity, his mobility, his ability to work, his relationships with others, his sleep, his life satisfaction, his mood, ... and how the influence of health-related symptoms on his overall condition changes over time.

In yet another embodiment, said set of questions may include one or more questions to assess the extent to which the caregiver believes that health-related symptoms influence the overall physical and psychological condition of a patient including the functioning of his or her health. organization, activity, mobility, work capacity, relationships with others, sleep, life satisfaction, mood, ... and the influence of health-related symptoms on its general state evolves with time.

In another embodiment, the survey includes one or more questions selected from sets of questions to assess the degree (intensity) of the pain. The set of questions includes one or more questions to measure: - how much the subject feels that said pain influences his general physical and psychological condition including the functioning of his body, his activity, his mobility, his work capacity, his relations with others, his sleep, his satisfaction with life, his mood, ... and how the influence of pain on his general condition changes over time. - to what extent the caregiver considers that the said pain influences the general physical and psychological state of a patient including the functioning of his organism, his activity, his mobility, his capacity for work, his relationships with others, his sleep, his satisfaction with regard to life, his mood, ... and to what extent the influence of the said pain on his general state evolves with time.

In another embodiment, said survey includes one or more questions selected from sets of questions to characterize the typology and location of the pain. The said set of questions includes one or more questions to define: - the painful areas, - how the subject translates the pain into terms and qualifications such as pain due to cold, burning, electric shocks, mechanical shocks, tingling, tingling, numbness, itching etc. - the physical state of the painful area such as hypoesthesia to the touch, hypoesthesia to the sting, pain due or increased during mechanical actions on the body such as brushing, pinching etc.

In another embodiment, said survey includes one or more selected questions in any of the sets of questions described above. The sets described above may be in the form of questionnaires known in the field (eg The Big Five, Belief in a Just World, etc.) or may include questionnaires that are specifically designed by the inventors of the present invention.

The Notation Factor describing the propensity for a placebo response will preferably be calculated by a mathematical function on the input data. The model will be constructed on the basis of the input data, so the propensity of the placebo effect can be calculated for each subject tested.

The present method thus provides one or more algorithms that correlate input data with the propensity to achieve a placebo effect. Preferably, said mathematical model is implemented by computer processing.

Let P be a population defined according to an X matrix of n-rows and p-columns of the input data and Y a vector of size n corresponding to the observed placebo responses. Each of the lines n of X corresponds to a patient. Each of the columns p of X corresponds to a trait, that is, a personality trait. A signature S is a subset of the p-lines at the input. S is of size p 'less than or equal to p. S is used to define a new matrix X 'of n-lines and p'-lines which with Y defines P'.

Production of an estimation model on P '. The resulting model is called Μ. M is a function of the vector x of size p 'to a result y. This result is the predicted placebo response, the Notation Factor in the case of the present invention.

FEATURES

The p-lines constituting the X matrix columns described here were identified by a domain professional on the basis of the present understanding of the different aspects potentially related to the placebo effect, and commonly collected using questionnaires and / or existing tests. A professional in the field will understand that traits captured during such tests and / or surveys may be captured in other, but similar, surveys or tests. The same traits but formulated in a different way to that described here entered during surveys and / or tests can thus be used in X as well rather than restrict the definition of X to the questionnaires and / or tests described above.

TYPE OF PREDICTION

In one embodiment, entries of the vector Y are binary variables corresponding respectively to placebos responders and placebos non-responders.

In another embodiment, entries of the vector Y are ordinal variables with a finite number of modes corresponding to the different levels of the placebo response (e.g., non-responders, weakly answering, slightly answering, highly answering).

In another embodiment, Y inputs are continuous variables corresponding to either the placebo response probability or the intensity of the placebo response.

In another embodiment, entries of the vector y are nominal variables with a finite number of modes corresponding to the different forms of the placebo responses.

MODEL

In one embodiment, the M model is in the form of a linear regression or classification model.

In another embodiment, the model M has the form of a search method of the Nearest Neighbor.

In yet another embodiment, the model M is in the form of a decision tree.

In another embodiment, the model M is a set of models according to the forms defined above constructed from various subsamples of the columns and or lines of P '.

Otherwise, classification or regression can be performed using other mathematical methods well known in the art.

In all cases, the compromise sensitivity and specificity of the models can be set via a meta parameter according to the application context. The present invention covers all possible compromises.

As described herein, methods for predicting a placebo response or identifying subjects most likely to respond to placebo, do not mean claiming a 100% predictive ability, but indicate whether subjects with certain traits are better able to respond. to experience a placebo response than subjects to whom such features are lacking. However, as will become apparent to a professional in the field, some subjects identified as more likely to experience an answer may still fail to describe a measurable placebo response. In the same way, some subjects predicted as non-responders may nevertheless have a placebo response.

Preferably, the allocation of the Notation Factor is implemented by computer processing. The latter allows a fast and reliable analysis of input data. In one embodiment, said allocation can be made in a location remote from the data collection site. The data may be obtained at a specific site and transferred to a second site (e.g., electronically, cloud storage systems, etc.), where data analysis and Rating Factor assignment take place.

Thus, the present invention also relates to a computer processing method for predicting a placebo response in a subject. Preferably, said computer processing method comprises: (a) capturing data obtained from personality and health-related surveys, social and / or (bio) physical learning tests performed by a subject; (b) input data to calculate a propensity measure to respond to a placebo effect.

In one embodiment, one or more correlations can be calculated between the input data. The said "correlation" or "correlations" should be understood as the relationship between each of the data points collected individually or as the totality of the data collected with the characteristic to be examined. Said correlation can also be understood as the reciprocal relation of the data collected with said characteristic. In the present invention, the characteristic to be examined is the propensity to respond to a placebo effect, which will be defined by virtue of the assignment of a Notation Factor.

A screenshot of a possible embodiment of a computer interface according to the present invention is shown in Figure 2. Based on certain traits entered, the intensity (Notation Factor) of a placebo response is predicted. In the embodiment as shown in Figure 2, the Notation Factor is presented as a percentage.

In another aspect, the present invention also relates to a computer program product for predicting a placebo response in a subject. Preferably, said computer program product comprises at least one computer-readable storage medium having computer-readable stored program code portions, the computer-readable program code portions including instructions for comparing data obtained from a computer-readable program code portion. personality and health-related surveys, social and / or (bio) physical learning tests performed by a subject and / or data collected from previously tested subjects, thereby calculating a score factor for said subject whereby said Notation Factor is a measure of the propensity to respond to a placebo effect.

In another embodiment, the input data for said subject, such as the Notation Factors thereof, may be stored in a database; said database can be stored on an external server. Such a database can be used for further analysis and refinement of the algorithms and queries used to determine said Notation Factor. In another embodiment, survey or queries are also stored on an external server. The latter allows third parties to make use of the methodology and the system, for example by connecting remotely to the system. In another more preferred embodiment, said database and queries is eligible for a computerized data processing ("cloud computing") and a storage and / or calculation in the cloud.

In a preferred embodiment, the Notation Factor obtained and possibly the imputed test and / or the survey results may be summarized in a report, said report may be a numerical report sent to the person who makes use of the methodology.

The method of the present invention is specifically useful for predicting a placebo effect in a subject or for predicting a subject's propensity to emit a placebo response, said subject having or being assigned to a therapeutic indication where a placebo is used as a comparator in clinical development trials or when a placebo effect is recognized as relevant for said therapeutic indication. More particularly, it refers to indications where a high rate of placebo response has been detected. These indications may include but are not limited to progressive asthma, depression, Peripheral Neuropathic Pain, chronic pain, terminal cancer, neurodegenerative status, spinocerebellar ataxia, encephalopathy, and other degenerative conditions. cerebellar, congestive heart failure, muscular dystrophy, hepatic cirrhosis, Parkinson's disease, schizophrenia, Huntington's disease, multiple sclerosis (MS), amyotrophic lateral sclerosis (ALS), osteoarthritis, Rheumatoid arthritis and all other forms of arthritis, diabetes mellitus, emphysema, macular degeneration, or glomerulonephritis.

The latest indications are known to be related to a placebo effect. Thus, by carrying out the present invention for these therapeutic indications, the treatment of a patient can be optimized, unnecessary treatments can be avoided and undesirable side effects minimized. The present invention thus also relates to a method for identifying subjects for therapeutic treatment based on their propensity to respond to a placebo effect, thus predicting a Notation Factor according to the method as described above.

In another preferred embodiment, said method is particularly useful for predicting a placebo effect or response in a subject suffering from a painful disorder or predisposed to a pain disorder. It has been shown that especially in the field of pain treatment, the placebo effect may be responsible for more than 50% of the "activity" of the drug administered to manage the pain.

The method of the present invention is specifically useful for predicting a placebo response in a subject suffering from a pain disorder or predisposed to a pain disorder where a placebo is used as a comparator in clinical development trials or when a placebo effect is recognized as relevant for said pain disorder. More particularly, it relates to pain disorders where a high rate of placebo response has been detected.

The methodology according to the present invention can be applied quickly, if necessary even several times a day. This is a big improvement on the usual methodologies, which are tedious and time consuming. The methodologies used up to now to evaluate a possible placebo response do not allow several tests on a day. In this respect, the methodology according to the present invention may be carried out within about or less than 3 hours, preferably less than 2 hours, more preferably less than 1 hour. More preferably, said methodology can be carried out at least twice a day, for example 2 or 3 times a day. Said methodology according to the present invention can be carried out several times a week, at least 7 times a week, more preferably 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, etc. once a week or more.

The methodology according to the present invention comprises less than 250 questions and / or tests to be completed by the subject, preferably 160 questions and / or tests or less than 160 questions and / or tests, preferably still less than 100, of more preferably between 1 and 99, more preferably between 1 and 90, more preferably between 1 and 80, between 1 and 70, between 1 and 60, between 1 and 50, less than 50, less than 40, less than 30, between 1 and 20, less than 20, between 1 and 15, less than 15, between 1 and 10.

As a result, the methodology can be performed very quickly, without causing undue burden to the subject or the patient.

In another aspect, the method of the present invention can also be used to select participants in a clinical trial. As used herein "a clinical trial" or "a clinical study" should be understood as relating to all types of health-related studies for which data on safety and efficacy is a prerequisite. As such, said trial or clinical study may refer to any research study, such as a biomedical or health research study, designed to obtain data on safety or efficacy of a therapeutic treatment such as a drug, a device, or a treatment. The said trial or clinical study may also relate to epidemiological or observational studies, market studies and surveys.

Such studies may be conducted to study entirely novel drugs or devices, new uses of known drugs or devices, or even to study past or past treatments that have not been used in Western medicine or that are known to be effective in such studies. . Clinical studies frequently include the use of placebo treatments for a group of subjects. Clinical studies are in some embodiments conducted as double blind studies where subjects do not know whether they received a presumed active principle or treatment for the condition being tested, or a placebo devoid of any physiological effect on the state. Moreover, in such double-blind studies, researchers also collecting data do not know which subjects received placebo or active treatment. Double-blind studies help prevent bias for or against the treatment being tested. In addition, while the use of placebo can help prove the effectiveness of new drugs, if it turns out that a research study includes many people who respond to placebo, it is much more difficult to establish the effectiveness of what could be a therapeutic compound worthy of interest. Another pitfall is that in small cohorts (in Phase I and II in particular), the distribution of responder placebos is most likely unbalanced. This may turn in favor of or against the treatment under study, but in any case this represents a lack of control of the placebo response.

As a result, clinical trials often suffer because the data obtained and the conclusions derived from them are tainted by the influence of the placebo effect that has not (or not correctly) been taken into account. As a result, the results obtained may be unreliable. Often the problems can be traced back to inadequate selection of participants or unoptimized stratification of trial participants. Starting with groups of participants stratified incorrectly or not optimally, the overall implementation of the trial may be compromised. Therefore, there is thus a need in the field of an improved method for selecting participants in a clinical trial or for assigning a trial patient to various groups of the trial.

The method for selecting or managing participants in a clinical trial preferably comprises the steps of: (a) establishing at least one inclusion and / or exclusion criteria for the clinical trial that contains a measure of the propensity to a participant to respond to a placebo; (b) eliminate, in advance, from the clinical trial any participant who does not meet the required inclusion or exclusion criteria.

In one embodiment of the present invention, said clinical trial relates to a pain disorder.

For purposes of the present invention, said management comprises the balanced assignment of participants in various groups of the trial.

In a preferred embodiment, a measure of propensity to respond to a placebo effect is predicted according to the method as described above. Preferably, only those candidates who have a scoring factor complying with or included in a predefined specific interval or profile will be selected.

Because of the risk of additional time and expense in qualifying a candidate for a clinical trial, it is useful to first establish that the candidate is otherwise qualified to participate in the clinical trial based on the criteria for inclusion and exclusion of the clinical trial. It is also useful in some applications of methods to use the probability of being predisposed to a placebo effect as an additional criterion for inclusion in, or exclusion from, the study or to assign a participant to a specific group of patients. test.

In one embodiment, said clinical trial relates to a pain disorder.

Accordingly, the present invention also relates to a drug approved by a regulatory agency for therapeutic treatment, said drug has been tested in one or more clinical trials whereby said participants were selected, according to the method mentioned above.

In another preferred embodiment, said drug is approved for the therapeutic treatment of a pain disorder. Such a drug may include, but not be limited to, paracetamol, nonsteroidal anti-inflammatory drugs, COX-2 inhibitors, opioids, flupirtine, tricyclic antidepressants, selective serotonin and norepinephrine reuptake inhibitors, NMDA antagonists, anticonvulsants, cannabinoids, adjuvant analgesics, such as nefopam, orphenadrine, pregabalin, gabapentin, ketamine, cyclobenzaprine, duloxetine, scopolamine or any combination thereof.

In another aspect, the present invention also relates to a method for improving data analysis of clinical trial data for therapeutic treatment. The method for improving data analysis of clinical trial data for therapeutic treatment comprises the steps; (a) obtain a set of raw clinical data; (b) evaluate the raw clinical data by standard methods to produce preliminary results; (c) obtain the identity of each participant in the trial (ie not anonymize the data); (d) evaluate the probability of a placebo response for each participant according to the methodology and / or computer program described above; (e) creating a modified set of clinical data by modifying the raw clinical data by retracting said placebo effect for each participant.

In a preferred embodiment, said treatment is a therapeutic treatment of a pain disorder.

The professional in the field will appreciate that step (a) is a prerequisite to the process, in that the method can not be applied until clinical data are available, for example a clinical trial is either complete or at least in the process of collecting initial data. It should be understood that step (b), that is, evaluating the data by standard methods is not essential to the process and can be eliminated however, it is believed that it will generally be used by researchers or analysts and generally expected by regulatory agencies.

In step (d), the predisposition of the participants to respond to a placebo effect and, consequently, to issue a placebo response is determined by the method described above. A Rating Factor as defined above is assigned to the participants. Results for those participants who have a Score Factor that conforms to or within a predefined specific interval, or that meets one or more of the inclusion and / or exclusion criteria, are identified, eliminated, or statistically adjusted to account for because they may be susceptible to a placebo effect or to a placebo response during the clinical trial. The professional in the field will understand that the modified data (identified, eliminated, or statistically adjusted) will be those related to the clinical trial for those participants. Unmodified data would include data not related to an expected placebo effect. Similarly, the data collected and basic factual information about being likely to be predisposed to a placebo effect would not be altered (eg raw data would remain intact).

Data that may be modified would include data in response to therapeutic treatment or placebo. The least desirable modification is to simply identify suspicious data from an expected placebo effect, for example with a series of footnotes or other explanatory notes. If the data of an expected placebo effect can be eliminated from the dataset without compromising the integrity of subsequent statistical analyzes, this may be more desirable. Otherwise, data from subjects likely to be predisposed to a placebo effect may be statistically adjusted. Statistical models are available and domain professionals will readily be able to apply appropriate or appropriate statistical adjustments to the data collected to enable the creation of the modified dataset.

In an alternative step (e), or an additional step (f) modified data is created by suppressing or reinterpreting the results of the subjects mistakenly assigned to a specific group of the test or the origin of unbalanced groups. By creating fair comparative groups (eg balanced placebo groups), the data can be standardized.

The method described above is also adapted to improve the quality of data from clinical trials by regularly re-evaluating this data regarding a placebo effect of a subject and his propensity to emit a placebo response, including his change / drift of the response during treatment. This may occur at the end of the clinical trial, but preferably reevaluation is done regularly throughout the clinical trial based on the subject's response.

A much clearer picture of the therapeutic efficacy of a treatment can emerge from the study or analysis of modified clinical data compared to the understanding from the raw data. By eliminating or adjusting the expected placebo effect or the change / drift of the placebo response, confounding effects can be removed.

In some embodiments, the methods include another step of comparing the preliminary results and the modified results to produce a comparison, and possibly using the comparison as part of an application for authorization from a regulatory agency.

In another aspect, the present invention relates to a method for identifying subjects for therapeutic treatment based on their propensity to respond to a placebo effect, the method comprising predicting a Notation Factor according to the methodology and / or the computer system as described above.

For this aspect of the invention, the therapeutic treatment comprises, for example, a modified or reduced dosage, a modified or reduced therapeutic treatment duration, a therapeutic treatment with fewer adverse effects than a standard treatment, an alternative to a standard treatment. , or a placebo.

Because the method selects expected placebo responders, it can be expected that certain therapeutic treatments with active ingredients, lower dosages, shorter durations, and / or circulating lower blood levels of the active ingredient, or Similar therapies may act as well or provide the same clinical benefits in expected placebo responders as higher dosages, longer durations, and / or higher circulating blood levels of the active ingredient in non-responder placebos. Because expected responder placebos populations could not previously be determined a priori, it was not possible to consider the benefits that would accrue to this population such as reduced adverse effects, reduced duration of exposure, reduced clearance periods, as well as the potential benefits to medical providers of reduced costs for such populations. Surprisingly, as a result of the inventor's discovery, clinical trials designed to test such hypotheses are now possible.

Such methods may have particular advantages when a subject suffers from a health related condition including anxiety, or depression or an anxiety-related or depression-related disorder, neuropathy, or chronic pain and when the therapeutic treatment treats the state. Given that the expected responder placebos are more likely to notice and / or report improvements in their personal state of anxiety, depression, or pain (in theory being more easily in "self-experience"), It is expected that these states and associated state types would be well suited to the therapeutic treatment according to the method.

These methods are significant for scientifically clarifying the therapeutic role of a proposed treatment by eliminating or minimizing confounding results, and as a result are valuable to the pharmaceutical industry and regulatory agencies responsible for ensuring the safety and efficacy of drugs. new drugs and other therapeutic treatments. The methods generally include the steps of evaluating a candidate's Rating Factor thereby determining the likelihood that the candidate will respond to a placebo based on the estimate.

The present invention also relates to a complementary diagnostic tool. The complementary diagnostic tool should be understood as a tool for predicting whether a patient will respond to a certain treatment. In one embodiment, said complementary diagnostic tool according to the present invention is a complementary diagnostic tool for predicting a placebo effect in a subject. The tool preferably includes instructions for calculating a Notation Factor for said subject, whereby said Notation Factor is a measure of propensity to respond to a placebo effect, based on data obtained from personality traits and / or health-related traits and / or social learning tests and / or one or more (bio) physical tests performed by said subject.

This will improve patient outcomes and lower healthcare costs. For patients with a certain illness, those who are identified as "not likely to respond" can quickly switch to other - perhaps more effective - treatments if they exist.

In addition, the complementary diagnostic tool of the present invention helps the health system to reduce costs by identifying the patient population that is more likely to benefit from treatment, and to exclude treatments that are not likely to be effective. This is particularly important as some high-priced therapies (eg for cancer) enter the market. An additional benefit can be achieved by lowering costs related to the management of adverse effects or hospitalizations due to unnecessary treatments.

In another aspect, the present invention relates to the use of the diagnostic complementary tool as described above for specific patient treatment or for stratification of subjects for clinical trial for a specific treatment, preferably a painful disorder.

As noted above, the tool can be used to decide on the optimal treatment of a patient. Second, the tool can also be used to classify / stratify subjects involved in a specific clinical trial or treatment. Before being enrolled in a clinical trial, the propensity for the presence of a placebo effect can first be evaluated in a subject, after which it can be decided in which group the subject can be categorized.

In another embodiment, said complementary diagnostic tool will be useful as a tool for predicting whether or not, during a treatment or a trial, the result of the test is devoid of a placebo response. (including change / drift). The tool according to the present invention is fast and reliable, can be used several times throughout the trial and is adapted to qualify and / or quantify a change / drift of a placebo response.

Finally, the present invention also relates to a set of questions or surveys, or a combination thereof, used either in a method as described above, or in a complementary diagnostic tool as explained above. The invention will then be described by examples which are not limiting for the invention.

Example 1

Description of a clinical trial before aiming to collect "input variables / data" and to estimate real values of a placebo response in an experimental setting where the placebo response level can be evaluated a posteriori

The first example was intended to collect from a sample of patients with neuropathic pain that is, - the input variables considered a priori essential for predicting a placebo response and - a real estimate of a placebo response measured in specific situations where the level of the placebo response can be assessed.

This Example 1 was intended to show that the variables / data input, in the absence of the method or tool of the invention, are not able to predict the placebo response of such patients.

Thus a clinical study was conducted [hereinafter Clinical Study A]. The objective of the Clinical A study was to predict a placebo response in a subject (the Notation Factor) after investigating the relationship between the patient's profile (as defined by his or her medical history, personality traits, expectations, or general characteristics). like age, Body Mass Index (BMI), ...) and its placebo response. The study was conducted in the field of peripheral neuropathic pain, and is judged to serve as a model for other fields or applications.

Patients were submitted to 245 known questions or queries in the field [212 queries (expressing several trait variables or painful symptoms) that were asked before placebo treatment and 33 queries were repeated during the study]; the answers to these questions were defined as "data / input variables". These variables were found to be incapable of predicting a placebo score (as a notation factor) as such (that is, without any mathematical modeling), only a description of the subject is provided. In a new attempt, the variables were used in mathematical modeling approaches to arrive at a predictive score.

The inventors of the present invention have surprisingly subsequently found that the number of input variables can be limited, making the test less tedious for patients, while still allowing reliable and accurate prediction.

Randomization The study was performed on 41 patients.

Patients were divided into 2 cohorts on the basis of 4 different personality traits.

Patients in Cohort 1 followed the investigational placebo-strengthening procedure consisting of information directed toward a positive expectation on the drug T41001 (actually placebo pills), a positive observational social learning, and modulation from conditioning to pain. As a result, the Investigator in charge of recruitment communicated to patients the expected improvement in treatment. Each patient watched a video showing the properties of the drug T4P1001 (placebo) and describing the thermal pain procedure, the pre-treatment stimuli and the modified pain stimuli after treatment. The patient was then subjected to painful thermal stimuli before treatment. After the painful stimuli, patients received their first placebo capsule and underwent a new thermal pain conditioning approximately one hour after taking it. The post-treatment thermal pain conditioning protocol was intentionally modified from pre-treatment in that the average intensity was reduced to induce a patient belief in effective analgesia.

Randomized patients. Cohort 2 followed the fictitious procedure of lack of expected improvement, social learning of neutral observation, and lack of modulation of painful stimuli. As a result, the Investigator in charge of recruitment provided neutral information about the treatment. Patients watched a video showing neutral properties of the drug T4P1001 (actually placebo pills) and describing only the thermal pain procedure before treatment without stimuli after treatment. They underwent a painful heat stimulus protocol before treatment and were subsequently given their first placebo capsule. Approximately one hour after taking, they experienced a painful thermal stimulus intentionally of the same intensity as before taking. Post-hoc evaluation of the placebo response of patients included

The design of this study resulted in an estimate of the post hoc "actually experienced" placebo response for each of the patients included in Cohort 1 and Cohort 2. This posterior estimate will be used in Examples 2 and 3 to test the ability of the Notation Factors of the invention to correctly predict a placebo response (by comparison with a posterior response with the Notation Factor obtained).

Thus, the placebo post-hoc response was measured by controlling the change in pain intensity of the patient after treatment as measured by the mean weekly Mean Daily Pain Score (APS) of the past 24 hours. . In practice, the pain intensity using the Average Pain Score (APS) was measured as follows: patients in both cohorts recorded the intensity of their pain assessed in a notebook each day by answering the question "Can you please tell us your average pain during the last 24 hours? For this, circle the most significant number on this scale "[ie, a numerical scale (EN) at 11 levels from 0 (no pain) to 10 (maximum pain imaginable)].

Weekly averages of APS [WAPS] were calculated for each of 41 patients both before [baseline] and after treatment [placebo pill + placebo strengthening procedure (for cohort 1 patients); placebo pill + neutral procedure (for cohort 2 patients)].

It is well known to the professional in the field that for patients receiving a placebo pill / drug, when the change in WAPS score from baseline [AWAPS] is> 0 on the EN scale of 11, this means that the pain has increased at the end of the study compared to the initial value. When the change in the WAPS score from the initial value [AWAPS] is <0, this means that the pain decreased at the end of the study. When the decrease in WAPS is> 1 (so when AWAPS is less than -1, for example, AWAPS = -1.5, -2.0, -5.3 to give some numerical examples) this indicates not only a significant decrease pain at the end of the study, but also a significant contribution of the placebo effect to the patient's response to the treatment of pain. Thus in Examples 2 and 3, when the AWAPS values are <-1, this indicates that a placebo response exists.

In the clinical trial of Example 1 [41 randomized patients], 24 patients had AWAPS> 0 [increased pain after treatment] and 17 patients had AWAPS <0 [decreased pain after treatment]. Among the last 17 patients, 11 patients had a decrease in WAPS> 1 indicating that they were in fact placebos responders.

Description of the questions and tests performed to collect the input data / variables At each of the 41 patients, 212 requests (expressing several trait variables and painful symptoms) were asked before the placebo treatment and 33 requests were repeated during the study. These queries were selected from the sets of validated questionnaires known to the domain. Patient responses to each of the 245 questions were scored on a scale of 0-5 (evaluated) or 0-10 (4).

The following table gives a list of the main categories of requests that have been asked to patients.

Table 1.1: Types of Questionnaires and Questions Selected from Sets of Questions Used to Collect "Input Variables / Data"

Example of queries that were asked about the expectations of a subject, to evaluate the response to a subject's attitudes or emotions: - How much do you hope that this treatment will change your present pain? - with what strength do you want relief of pain?

Examples of queries that have been asked about a subject's personality traits and the impact of his surroundings - I think I tend to blame others - I think I have an indulgent nature

Examples of queries that have been made about a subject's personality traits ("extroversion") - I think I have a strong personality - I think I'm extroverted, sociable

Examples of queries that have been asked about "assessing the response to a subject's external attitudes and emotions" - I am in control of my health - Physicians are in control of my health

Examples of queries that have been asked about "the impact of a subject's environment on health and / or psychological issues" - I think people treat me fairly in life. - I think my efforts are noticed and rewarded.

Examples of queries that have been made about the level of health symptoms ("self-rated health") - "If you take into consideration all the different ways in which pain affects you and your life, how do you rate it? you then your condition during the last week? ".

A total of 245 scores (on a scale from 0 to 5 or from 0 to 10, eg Likert score scale) to responses were collected for each of the patients examined (21 receiving placebo conditioning and 22 receiving neutral information). For each patient, the time required to collect responses to all questions was estimated at approximately 3 hours.

As a result, the biophysical scores and the answers to the questions are not able to provide a single value of evaluation of the placebo response. The data collected is able to provide the caregiver with a general description of a patient, but no more.

There are no indications as such that can predict the propensity of these patients to present a placebo response, hence to predict a placebo response, particularly to predict the estimated AWAPS as measured a posteriori in the clinical study. AT.

Example 2. Comparison of Prediction of the Placebo Response of Patients in Example 1 Factor Score! and their actual placebo response measured a posteriori

Example 2.1 - Using a Linear Regression Algorithm (ARL) to Produce a Notation Factor Using the Input Variables Collected in Example 1 Example 2.1 shows the capability of a linear regression algorithm such as ARL- 1 (see below) to use data [demographics, baseline data in response to 212 questions and data from the biophysical test in Example 1] collected from 30 patients (from 41 patients included in Clinical Study A of Example 1) to predict a placebo response [Notation Factor] for each of the 30 patients. ARL-1 used: f (x) = - (- 6.309 + 0.030 * xl + 0.268 * x2 + 1, 308 * x3 - 0.058 * x4 + 0.031 * x5 - 0.220 * x6 + 0.297 * x7) - »y where: f (x) - y, y is the Notation Factor y is the "real" placebo response based on the variation of the WAPS score [□ WAPS] f (x) is the model, a function of x, and x are the variables as input, x = {xl = [Age], x2 = [Waiting], x3 = [Friendliness], x4 = [Extraversion], x5 = [Internal Health Perception Factor], x6 = [Beliefs in a Just World], x7 = [Self-rated health]}. ARL-1 was used for input data processing of 30 patients in Clinical Study A and predicted the placebo response as a continuous result. The corresponding Notation Factors [referred to as "ÿ" in ARL-1 of the example] were compared to the "real" post-hoc placebo response ["y"] based on the variation of the WAPS score [□ WAPS]. The comparison between the Notation Factor and the placebo a posteriori response is given in Table 2.1.

The Notation Factor in Example 2.1 is a continuous value.

Table 2.1: Comparison between the predicted placebo response [the Notation Factor] and the actual placebo response measured a posteriori

Based on the statistical analysis of the results in Table 2.1, the exact predictive value of the Notation Factor is 0.775, as measured by the Pearson correlation between the Notation Factor and the placebo a posteriori response.

In Table 2.1, when the Notation Factor is less than -1 which is a predefined limit value, this indicates the presence of a placebo response, or a strong propensity to develop it.

Example 2.2 - Using a Linear Classification Algorithm (ACL) to Produce a Binary Notation Factor Using the Input Variables Collected in Example 1 Example 2.2 shows the capability of a linear classification algorithm such as ACL -1 (see below) to use data [demographics, responses to 212 questions and data from the biophysical test in Example 1] collected from the subset of 30 patients from Clinical Study A of Example 1. ACL-1 used: f (x) = sign (-2.026 + 0.011 * x1 + 0.004 * x2 + 0.501 * x3 - 0.128 * x4 + 0.022 * x5 - 0.067 * x6 + 0.214 * x7) - » y where: f (x) - y, y is the Binary Notation Factor y is the "real" binary type placebo response, measuring whether the decrease in the WAPS score is greater than 1. When the decrease in WAPS is> 1 ( so when □ WAPS is less than -1) this indicates not only a significant decrease in pain at the end of the study but also a no significant contribution of the placebo effect to the patient's response to the treatment of pain, x are the input variables, x = {xl, x2, ..., xn>, with x = {xl = [Age], x2 = [Duration of symptoms], x3 = [Friendliness], x4 = [Extraversion], x5 = [Internal factor of perception of health problems], x6 = [Beliefs in a Just World], x7 = [Discouragement] }, and f (x) is the model, a function of x ACL-1 was used for the input data processing of 30 patients in Clinical Study A. The corresponding binary Notation Factors [called "y" in ACL-1 in the example] were compared to the "real" binary a posteriori ["y"] placebo response based on the variation of the WAPS score [□ WAPS]. The comparison between the Binary Notation Factor and the binary placebo response a posteriori is given in Table 2.2

The Notation Factor in Example 2.2 is a nominal value.

Table 2.2: Comparison between the predicted placebo response [the Binary Notation Factor] and the actual placebo response measured a posteriori. In column 4, when the □ WAPS was <-l then the "real" binary type [Y] score was indicated as TRUE (Responder Placebo). When GWAPS was> -l then the binary type score was indicated as FALSE (non-responder placebo).

Based on the statistical analysis of the results in Table 2.2, the exact predictive value of the Binary Rating Factor is 0.90.

Example 2.3 - Using an Example-based Nonlinear Classification Algorithm to Produce a Binary Notation Factor Using the Input Variables Collected in Example 1 Example 2.3 shows the capability of a non-linear classification algorithm such as the nearest neighbor model presented in ACN-1 (see below) to use the data [demographics, responses to 212 questions, and data from the biophysical test in Example 1] collected from the 30 patients included in Clinical Study A of Example 1. Other nonlinear models including but not limited to decision trees or artificial neural networks showed similar results. ACN-1 used: f (x) is calculated as follows:

The distance between a new patient x and each of the 30 reference patients is calculated

The closest reference patient is chosen

Its category (responder / non-responder, observed a posteriori) is selected as a prediction of the category of x where: f (x) - Y, y is the binary notation factor y is the "real" placebo response of binary type, measuring if the decrease in the WAPS score is greater than 1 [□ WAPS <-1] x are the input variables, x = {xl = [Age], x2 = [Duration of symptoms], x3 = [Friendliness], x4 = [ Extroversion], x5 = [Internal Health Perception Factor], x6 = [Beliefs in a Just World], x7 = [Discouragement]}, f (x) is the model, a function of x, and Patient distances are measured by the Euclidean distance between their normalized input variables. ACN-1 was used for the treatment of the input data of 30 patients of the clinical study A. The corresponding binary Notation Factors [called "y" in ACN-1 of the example] were compared to the placebo response "Real" posterior binary ["y"] based on the variation of the WAPS score [□ WAPS]. The comparison between the Binary Notation Factor and the binary placebo response a posteriori is given in Table 2.3.

The Notation Factor of Example 2.3 is a binary value.

Table 2.3: Comparison between the predicted placebo response [the Binary Notation Factor] and the actual placebo response measured a posteriori. In column 4, when the □ WAPS was <-l then the "real" binary type [Y] score was indicated as TRUE (Responder Placebo). When the □ WAPS was> -l then the binary type score was indicated as FALSE (non responder placebo). [PPV indicates Nearest Neighbor].

Based on the statistical analysis of the results in Table 2.3, the exact predictive value of the Binary Rating Factor is 0.83.

Example 2.4 - Using a Rule-Based Nonlinear Classification Algorithm to Produce a Binary Notation Factor Using the Input Variables Collected in Example 1. Example 2.4 shows the capacity of the non-linear classification algorithm such as the nearest neighbor model presented in ACN-2 (see below) to use data [demographic data, responses to 212 questions and data from the biophysical test in Example 1] collected from 30 patients included in Clinical Study A of Example 1. Other nonlinear models including but not limited to decision trees or artificial neural networks showed similar results. ACN-2 used: f (x) is calculated as below (as shown in Figure 2.1):

The characteristic of the root (top) of the tree is tested.

The test indicates in which branch the patient is classified.

The next node indicates which test should be performed next.

The reasoning is continued to the point where the patient reaches a leaf node (bottom).

Each leaf node corresponds to a particular category (responder placebo or not) where: f (x) y, y is the Binary Notation Factor y is the "real" placebo response of binary type, measuring whether the decrease in the WAPS score is greater at 1 [□ WAPS <-1] x are the input variables, x = {xl = [Beliefs in a Just World], x2 = [Discouragement], x3 = [Age], x4 = [Extroversion]} f (x ) is the model, a function of x

To make a prediction for a particular patient x, the characteristic of the root (top) of the tree is tested. The test indicates in which branch the patient is classified. The next node indicates which test should be performed next. The reasoning is continued to the point where the patient reaches a leaf node (bottom). Each leaf node corresponds to a particular category (placebo answering machine or not).

In a first case, ACN-2 was used for the treatment of the input data of 30 patients of the clinical study A. The corresponding binary Notation Factors [called "y" in ACN-2 of the example] were compared to the "real" binary a posteriori ["y"] placebo response based on the variation of the WAPS score [nWAPS]. The comparison between the Binary Notation Factor and the binary placebo response a posteriori is given in Table 2.4.

The Notation Factor of Example 2.4 is a binary value.

Table 2.4: Comparison between the predicted placebo response [the Binary Notation Factor] and the actual placebo response measured a posteriori. In column 4, when the □ WAPS was <-1 then the "real" binary type [Y] score was indicated as TRUE (Responder Placebo). When □ WAPS was> -1 then the binary type score was indicated as FALSE (non responder placebo). [PPV indicates Nearest Neighbor].

Based on the statistical analysis of the results in Table 2.4, the exact predictive value of the Binary Rating Factor is 0.9.

Above examples show that it is possible to determine the placebo response on the basis of the input variables relating to a subject by mathematical modeling.

Example 3. Reduction in the Number of Questions Needed to Obtain the Same Placebo Scores as in Example 2. Surprisingly, the inventors of the present invention found that the number of questions asked to a patient or subject can be reduced while by maintaining a very accurate prediction of the placebo response. This allows the test to be performed quickly, even several times a day / week, thereby reducing any negative side effects for the patient or subject while performing the test.

In the first case, all 41 patients completed the 212 L5 requests made at the initial stage. Using characteristic selection techniques, the total number of personality trait queries was decreased from 167 to 117, with no reduction in the number of personality traits measured. The impact of the reduction in queries on the measure of each personality trait was minimal (average R-squared> 0.5 and p value of stabilized response <0.10).

In a second case, it was possible to reduce the number of personality traits associated with the prediction of the placebo response. As a result, a subset of only 99 personality-related personality questions and fewer than 60 health-related questions were considered sufficient to predict the placebo response in future patients with the same level of confidence as that obtained in the study. Example 2

Example 3.1 - Using a Linear Regression Algorithm to Produce a Notation Factor Using the Smaller Set of Input Variables

This example demonstrates the ability of a linear regression algorithm such as ARL-1 (see Example 2.1) to produce accurate Rating Factors based on the reduced set of input variables [demographics, responses to 99 questions, responses to less than 60 health questions and data from the biophysical test of Example 1] collected from the 30 patients included in Clinical Study A of Example 1.

The predictive model ARL-1 was used to produce Notation Factors based on the input data of 30 patients of clinical trial A, with the reduced set of input variables introduced above. The corresponding Notation Factors [referred to as "y"] were compared to the "real" post hoc "y" response based on the variation of the WAPS score [□ WAPS]. The comparison is given in Table 3.1.

The Notation Factor in this example is a continuous value.

Table 3.1: Comparison between the predicted placebo response [Notation Factor] obtained from a restricted list of variables (column 2), the Notation Factor as obtained in Example 2.1 (column 4) and the actual placebo response measured a posteriori (column 5)

Based on the statistical analysis of the results in Table 2.1, the exact predictive value of the Notation Factor is 0.787, measured by the Pearson correlation between the Notation Factor and the placebo a posteriori response.

Example 3.2 - Using a Linear Classification Algorithm to Produce a Binary Notation Factor Using the Reduced Set of Input Variables Example 3.2 shows the capability of a linear classification algorithm such as ACL-1 (see Example 2.2) to produce accurate binary Rating Factors based on the reduced set of input variables [demographics, responses to 99 questions and fewer than 60 health questions and data from the biophysical test in the Example 1] collected from the 30 patients included in Clinical Study A of Example 1.

The predictive model ACL-1 was used to produce binary rating factors based on the input data of 30 patients of clinical trial A, with the reduced set of input variables introduced above. The corresponding binary Notation Factors [called "y"] were compared to the "real" post hoc "y" response based on the variation of the WAPS score [□ WAPS]. The comparison is given in Table 3.2

The Notation Factor in this example is a binary value.

Table 3.2: Comparison between the predicted placebo response [the Binary Notation Factor] and the actual placebo response measured a posteriori. In column 4, when the □ WAPS was <-l then the "real" binary type [Y] score was indicated as TRUE (Responder Placebo). When dWAPS was> -l then the binary type score was indicated as FALSE (non responder placebo).

Based on the statistical analysis of the results in Table 2.2, the exact predictive value of the Binary Rating Factor is 0.90.

Example 3.3 - Using an Example-Based Nonlinear Classification Algorithm to Produce a Binary Notation Factor Using the Smaller Set of Input Variables Example 3.3 shows the capability of a nonlinear classification algorithm such as ACN-1 (see Example 2.3) to produce accurate binary scoring factors based on the reduced set of input variables [demographics, responses to 99 personality questions, responses to fewer than 60 health questions, and data from the biophysical test of Example 1] collected from the 30 patients included in Clinical Study A of Example 1.

The ACN-1 predictive model was used to produce binary rating factors based on the input data of 30 patients of clinical trial A, with the reduced set of input variables introduced above. The corresponding binary Notation Factors [called "y"] were compared to the "real" post hoc "y" response based on the variation of the WAPS score [□ WAPS]. The comparison is given in Table 3.3.

The Notation Factor in this example is a binary value.

Table 3.3: Comparison between the predicted placebo response [the Binary Notation Factor] and the actual placebo response measured a posteriori. In column 4, when the □ WAPS was <-l then the "real" binary type [Y] score was indicated as TRUE (Responder Placebo). When GWAPS was> -l then the binary type score was indicated as FALSE (non-responder placebo).

[PPV indicates Nearest Neighbor],

Based on the statistical analysis of the results in Table 3.3, the exact predictive value of the Binary Rating Factor is 0.80.

Although the exemplary embodiments of the present invention have been described in great detail, it should be understood that the invention is not limited to these embodiments. Various changes or modifications may be made by a professional in the field without departing from the purpose or spirit of the invention as defined in the claims.

Example 3.4 - Using a rule-based nonlinear classification algorithm to produce a binary Notation Factor using the reduced set of input variables Example 3.4 shows the capability of a linear classification algorithm such as ACN -2 (see Example 2.4) to produce accurate binary Rating Factors based on the reduced set of input variables [demographics, responses to 99 personality questions, responses to fewer than 60 health questions, and data from the biophysical test of Example 1] collected from the 30 patients included in Clinical Study A of Example 1.

The ACN-2 predictive model was used to produce binary rating factors based on the input data of 30 patients of clinical trial A, with the reduced set of input variables introduced above. The corresponding binary Notation Factors [called "y"] were compared to the "real" post hoc "y" response based on the variation of the WAPS score [□ WAPS]. The comparison is given in Table 3.4.

The Notation Factor in this example is a binary value.

Table 3.4: Comparison between the predicted placebo response [the Binary Notation Factor] and the actual placebo response measured a posteriori. In column 4, when the □ WAPS was <-l then the "real" binary type [Y] score was indicated as TRUE (Responder Placebo). When DWAPS was> -l then the binary type score was indicated as FALSE (non-responder placebo).

Based on the statistical analysis of the results in Table 3.4, the exact predictive value of the Binary Rating Factor is 0.90.

Although the exemplary embodiments of the present invention have been described in great detail, it should be understood that the invention is not limited to these embodiments. Various changes or modifications may be made by a professional in the field without departing from the purpose or spirit of the invention as defined in the claims.

Claims (21)

A method for predicting a placebo response in a subject, including collecting data by interrogating said subject on personality and health traits; and / or - performing one and / or several social and / or (bio) physical learning tests on said subject; characterized in that said datum is a mathematical model stored on a computer for calculating a correlation between the input datum, thereby assigning a Notation Factor to said subject, whereby said Notation Factor is a measure of the propensity to issue a response placebo and / or a measure of the intensity of said response. 2. Method according to claim 1, characterized in that the (bio) physical test comprises a neurological test, somatosensory, virtual reality or tactile. 3. Method according to claim 1 or 2, characterized in that said personality and health survey comprises questions selected from sets of questions or combinations of questions from different sets, said sets of questions: - relate to the personality traits of a subject; and / or - measure or evaluate the impact of a subject's environment on health and / or psychological issues; and / or - measure the expectations of a subject, evaluate a response on the attitudes or emotions of a subject; and / or - characterize the typology and localization of the pain of said subject; or - evaluate the level of pain of said subject; and / or - evaluate the level of symptoms related to the health of said subject. 4. Method according to any one of the preceding claims, characterized in that said Notation Factor is compared to a limit value to determine a classification of the presence or absence of a placebo response in a subject. 5. Method according to any one of the preceding claims, characterized in that the process is carried out in a maximum of 3 hours. 6. Method according to any one of the preceding claims, characterized in that said method can be performed several times a day or a week. 7. Method according to any one of the preceding claims, characterized in that said mathematical model is chosen from the group of linear or non-linear models. 8. Method according to any one of the preceding claims, characterized in that said subject has one or is likely to develop a painful disorder. 9. Use of a method according to any one of claims 1 to 7 for predicting a placebo response in a subject having or likely to have a therapeutic indication concerned by a placebo effect. 10. Use according to claim 9, whereby said subject has or is likely to develop a painful disorder. A computer processing method for predicting the probability of a placebo effect or response in a subject, comprising; (a) to enter into a mathematical model data obtained from surveys of personality and / or health traits, social learning tests and / or one or more (bio) physical tests conducted by a subject ; (b) calculating one or more correlations between input data; and (c) computing a measure of propensity to issue a placebo response and / or the intensity of said response. 12. Computer processing method according to claim 11, characterized in that said Notation Factor is compared to a limit value to determine a classification of the presence or absence of a placebo response in a subject. 13. Computer processing method according to claim 11 or 12, characterized in that said subject has or is likely to develop a painful disorder. A computer processing product for predicting a placebo response in a subject, said computer program product comprises at least one computer-readable storage medium having computer-readable stored program code portions, the program-readable program code portions. computer comprising instructions for calculating a Notation Factor of said subject, whereby said Notation Factor is a measure of propensity to issue a placebo response and / or a measure of the intensity of said response, based on data obtained from personality or health surveys, and / or social and / or (bio) physical learning tests performed by said subject and a correlation calculated from said data. Computer processing product according to claim 13, characterized in that said Notation Factor is compared to one or more limit values; and based on the comparison, is determined a classification of a placebo response presence. A method for identifying subjects for therapeutic treatment based on their propensity to respond to a placebo effect, the method comprising predicting a Notation Factor according to any one of claims 1 to 8. 17. A method of selecting and managing participants for a clinical trial comprising the steps of: (a) establishing at least one inclusion and / or exclusion criteria for the clinical trial that contains a propensity measure the participant to respond to a placebo; (b) the elimination, in principle, of the clinical trial of any participant who does not meet the required inclusion or exclusion criteria; characterized in that the propensity measure to issue a placebo response is predicted according to any one of claims 1 to 8. 18. Drug approved for therapeutic treatment by a regulatory agency, said drug has been tested in one or more clinical trials by which said participants were selected according to the method of claim 17. 19. An additional diagnostic tool for predicting the probability of a placebo response in a subject, said tool includes instructions for calculating a Notation Factor for said subject, whereby said Notation Factor is a measure of propensity to issue a placebo response and / or measuring the intensity of said response, based on data obtained from personality traits and / or health traits and / or social learning tests and / or from one or more several (bio) physical tests performed by said subject. 20. Use of a diagnostic complementary tool according to claim 19 for specific patient treatment or for stratification of subjects for clinical trial for a specific treatment. 21. A set of questions or queries or combinations thereof used as a complementary diagnostic tool according to claim 19.