A young man sifting through complex data on a large virtual screen, the data projected in front of him

What is CEME?

CEME, or Cutting-Edge Medical Education, is a proprietary process, developed by McCormack that is entirely unique within the pharmaceutical industry.

The primary function of CEME is to enhance patient management. In this way, CEME generates new sales by robustly differentiating the client’s brand from the competition.

CEME is invaluable because the problem that exists for pharma companies is the sheer volume of available experimental and clinical data.

Importantly, individual pieces of data are often of little value in isolation. It is only when numerous diverse facts are seen in relation to each other that new teachings evolve.

CEME has the power to generate new teachings by searching for these previously-unknown relationships.

Click for a more detailed explanation

A young man sifting through complex data on a large virtual screen, the data projected in front of him

What is CEME?

CEME is an acronym for Cutting-Edge Medical Education. However, by contrast with CME (Continuing Medical Education), CEME outcomes are always unique and exciting.

The systems of correlation and association that formed the basis of CEME were first created in 1983 by Keith McCormack. Since that time he has accumulated an unprecedented knowledge base of the existing pharmacopoeia. Today, CEME is a modified version of an association rules learning algorithm with the power to locate previously-unknown relationships between product data and data for a licensed indication. These novel outputs constitute new teachings that subsequently provide the means for enhanced brand awareness and increased sales.

New teachings mean robust differentiation

Because of their inseparable association with the client’s brand it is these new teachings that enable robust differentiation by significantly enhancing patient management, often by fulfilling unmet clinical needs, together with the prospect for creating novel IP. The CEME process is complete when each outcome represents a significant contribution to medicine and science that enhances understanding. The prescriber/healthcare professional and other stakeholders learn breaking news about the brand’s actions and effects that they could not have learned anywhere else.

Additionally, in today’s value culture, CEME has also evolved as a tool for searching real-world data for observational clinical correlates, and in the design of longitudinal studies of patients at the point-of-care.

Primary objectives of CEME

CEME provides the healthcare professional and researcher with two things, both of which are inseparably associated.

The first is novel teachings that take the form of a new understanding within a selected area of medicine. Within this selected area of medicine, the second is differentiation of the candidate brand.

Origin and basis of CEME

It is an inescapable fact that despite screening billions and billions of drug candidates, in the US pharmacopoeia for example, there are only 1200 regularly-prescribed small molecules.

It is also a fact that these 1200 small molecules converge upon about 500 human genome-derived druggable targets; and approximately 30,000 diseases and disorders converge upon this small number of targets.

A diagram showing how thirty thousand diseases and 1200 regularly-prescribed small molecules converge upon just 500 human genome-derived druggable targets.

From this core concept of convergence we can conclude that a universe of information is contained within this small population of drugs that may be described as a collective pharmacophore for most known diseases and disorders. CEME has the power to explore, probe and ultimately exploit this untapped universe of hidden and previously-unknown relationships.

​Each drug’s chemical structure codes for clinical outcomes. This code is modified by degrees of freedom in context, formulation and route of administration.

CEME’s relational databases probe this code in search of previously-unknown relationships between data, differentiating the drug and/or brand through new teachings.

The fundamental mechanics of CEME

Data mining is a powerful tool that can be employed in order to identify relationships within large volumes of data.

Such relationships are not discernible through simple observation, and they are not obvious and were not previously known. Generally, the principles of data mining are not evident within pharmaceutical marketing strategies that traditionally are derived using individual pieces of raw data that characterize a brand. These data points are then compared and contrasted with similar data for competing brands. However, with data mining there is the exciting prospect for discovering new teachings that tell us more about the brand and the patient groups for which it is licensed, and in this way, the brand is differentiated from the competition. Data mining is knowledge discovery whereby what is retrieved is not explicit in the databases being interrogated. Importantly, outcomes from data mining must be of value.

There are three main reasons for using data mining. First, there is too much data (facts) and too little information. Second, there is a need to extract useful information which is then interpreted in the light of our best understanding. Third, the extracted information should advance our best understanding; that is, the extracted information represents new teachings.

The customized relational databases of CEME employ data mining algorithms that search for previously-unknown relationships between data in moderate to large size datasets, and no prior assumptions are made about what those relationships might be. CEME outcomes are always exciting and always represent new teachings.

Chemical structure codes for clinical outcomes

All experimental outcomes (animal, pharmacological, physicochemical and interventional clinical studies) are determined by a drug’s chemical structure.

Differences in structure will code for differences in outcomes between related structures. Likewise, for the same molecule, differences in formulation and administration will also code for differences in outcomes. Thus, for every approved drug, the information contained within its chemical structure, formulation and route of administration, collectively codes for all clinical outcomes. Importantly, as discussed below this code is context-dependent whereby each context represents a variable that defines patient heterogeneity in response.

CEME identifies patient subgroups

By identifying previously-unknown relationships between data, McCormack’s technology decodes the uniqueness that is inherent within every approved drug. That is, CEME-derived, previously-unknown relationships represent an expression of this inherent uniqueness. As pharmacodynamic and clinical correlates, these relationships invariably reveal new insights into potential patient heterogeneity in both response and unwanted effects.

In this way, CEME identifies maximally-responding patient subgroups, non-responders and “reversible” non-responders (= potential responders). CEME may also facilitate the identification of markers that can be used to identify these patient categories, or provide new evidence for a previously-unknown therapeutic benefit in the patients for whom the drug is indicated, that for example may include disease-modifying effects in some groups.

Diagram showing a homogenous array of pale purple butterflies with a few picked out in bright orange
CEME identifies patient subgroups / © Shutterstock

CEME is also a probe for searching real-world data

Searching real-world data can be compared to looking for a white softball in snow.

Since the full spectrum of clinical outcomes for a drug is ultimately apparent within real-world (observational) data then additionally, CEME clearly fulfills a new role as a sensitive probe of data gathered at the point-of-care and elsewhere, or in the design of longitudinal studies of patients treated at the point-of-care. That is, by decoding the available experimental data, CEME identifies differences in outcomes and response within different patient groups that are likely to be observed in real-world situations.