If you’re even the tiniest bit familiar with the field of precision dosing, you must have come across multiple terms for — basically — the same concept: terms like therapeutic drug monitoring, or target concentration intervention and various others all refer to the concept of using drug concentrations (or other biomarkers) to evaluate and potentially adapt the dosing regimen to reach a certain exposure target.
Below is an overview of the terms that I’ve come across in the last decade or so. The table also includes the number of “hits” on Google, to give an indication of its popularity.
|Therapeutic drug monitoring||TDM||486,000|
|Adaptive Feedback Control||AFC||38,600|
|Target Concentration Intervention||TCI||5,150|
|Model-based Precision Dosing||MBPD||331|
|Model-informed Precision Dosing||MIPD||169|
* I couldn’t get a good Google estimate for Concentration Control, since this term is used a lot in chemical engineering.
The question now is: Are all these terms referring to the same exact concept, or are there important differences between them?
Let’s start with TDM. Looking on PubMed, first mentions of the term Therapeutic Drug Monitoring are found in the 1970s. While it is hard to pinpoint the actual first use of this term, the concept of measuring a drug concentration to evaluate drug therapy is certainly even older than that, going back to the early days of the bioanalysis field. TDM is unequivocally the most well known term of those listed in the table, also indicated by the largest number of google “hits”. There is even an organization of TDM professionals (the International Association of Therapeutic Drug Monitoring and Clinical Toxicology, IATDMCT). This association defines TDM as:
Therapeutic drug monitoring is a multi-disciplinary clinical specialty aimed at improving patient care by individually adjusting the dose of drugs […]”
While TDM is quite narrowly related to the pharmacology field of clinical pharmacology, the term Adaptive Feedback Control is a term borrowed from broader fields such as electrical and mechanical engineering, where it is very common parlance. Used in the context of clinical pharmacology and precision dosing, I agree that this term indeed covers the concept of precision dosing based on a model of pharmacokinetics (and or pharmacodynamics) quite nicely. Usage of this term also makes it convenient to explain the concept to engineers outside the pharmacology field, and allows adoption of algorithms and quantitative methods from the broader AFC field.
I do believe, however, that the difference with TDM is mainly in its scope: it seems to refer only to the control of drug levels, i.e. to calculations to adjust the dose based on some biomarker. However, the bioanalysis (quantification of drugs in the body) of the drugs is also an integral of TDM, and this is not inherently covered in the term AFC (or more specifically Concentration Control, CC). Additionally, TDM is a broader term since it can be performed both with and without modeling & simulation, while AFC is inherently model-based. The popularity of this term is hard to gauge from the number of google hits, since this includes other engineering fields as well, but there seem to be few articles on PubMed that refer to this methodology in this particular way. However, one semantic drawback of the term AFC is that it doesn’t automatically indicate that it refers to drug dosing or how it is to be used in a clinical context.
Since perhaps a decade or so, the term Target concentration intervention (TCI) has gained some traction in this field as well. TCI has been proposed as a conceptually better approach to precision dosing than TDM, as TDM narrowly focuses on a specific target range (with the lower end defined by drug efficacy and the higher end by drug toxicity). The argument then follows that dosing using a range ultimately leads to imprecise “lazy” dosing, since clinicians that perform TDM will be happy whenever the concentration falls inside the range, but will not try to calculate and prescribe the optimal dose to hit a specific target concentration.
While there are some merits to this argumentation, several objections against it should be pointed out too:
- TDM doesn’t necessarily use a target range. While in practice this is indeed often the case, to me — and IATDMCT — TDM really is an umbrella term covering many different ways of precision dosing. Therefore, rather than a replacement of TDM, TCI is just a more specific approach to TDM.
- Many drugs should not be dosed targeted to a single specific concentration, but rather on averaged or integrated exposure measure (area-under the concentration-time curve, AUC, or sometimes cumulative AUC). Examples include many antibiotics such as vancomycin, or cytotoxic drugs like busulfan or methotrexate. Additionally, some drugs might be better dosed based on monitoring of specific pharmacodynamic biomarkers such as neutrophil counts. TCI inherently excludes those cases, and is therefore too narrow a definition.
- In many cases, there is no specific target concentration available that is proven optimal. Or, reported targets from various clinical studies differ considerably, such as for imatinib.
- TCI also neglects the fact that frequent dose or interval changes can more easily lead to dosing errors, due to human mistake. Although I do not know of a specific study to back up this claim, many clinicians I have spoken to agree with this assertion. Especially for drugs in which the concentration-effect or concentration-toxicity relationship is not very strong, it will then actually be safer to use a target range than to aim for a specific concentration and make frequent dose adjustments.
- Related, TCI seems to leave little room for clinician input: e.g. in old and fragile patients a clinician might prefer to be less aggressive than in young and ‘healthy’ patients. (I will get back to this point in a future blog post to share some thoughts on how this could be done more appropriately but still in a quantitive and scientific way).
Finally, in the last few years the term of Model-based Precision Dosing (MBPD) has taken root, probably in the wake of the succesful introduction and maturation of Model-based Drug Development in drug industry, which happened primarily in the last decade. In my mind, MBPD is indeed a more narrow term than TDM but it does more accurately describes the concept of what we (at InsightRX) are doing, and the concept that underlies the platform that we are building.
More recently, I came across the perhaps even slightly more appropriate term Model-informed Precision Dosing. In my mind, the term MIPD — compared to MBPD — additionally acknowledges the fact that dosing decisions are commonly not solely based on mathematical models & simulations, but are also ruled by factors and decision that cannot be captured easily by equations.
At InsightRX we generally prefer to refer to this methodology only as precision dosing, as we feel it is self-evident that to achieve good precision and predictive ability a model-based approach should be applied.