Therapeutic Drug Monitoring (TDM) in its initial years, the 1960/1970s, was mainly equated with the measurement of drug concentrations in serum, plasma, or whole blood to individualize dosage with the goal of maintaining drug concentrations within a therapeutic target range pharmacokinetic (PK) TDM. Nevertheless, from the beginning, the vision of TDM was much more ambitious. As Charles Pippenger, the first Editor-in-Chief of Therapeutic Drug Monitoring, mentioned in an interview some years ago: even in these early years, TDM was seen as a global concept for individualized drug therapy to result in better patient care that would expand to include a wide variety of drugs and to be adapted to a variety of disciplines. “In 1979, the world of TDM recognized the importance of ‘personalized drug therapy’; only we called it ‘individualized drug therapy’” said Pippenger.1
Indeed, according to the current definition proposed by the International Association of Therapeutic Drug Monitoring and Clinical Toxicology, “TDM is a multi-disciplinary clinical specialty aimed at improving patient care by individually adjusting the dose of drugs for which clinical experience or clinical trials have shown it improved outcome in the general or special populations. TDM can be based on a a priori pharmacogenetic, demographic and clinical information, and/or on the a posteriori measurement of blood concentrations of drugs (pharmacokinetic monitoring) and/or biomarkers [pharmacodynamic (PD) monitoring].”2
Although PK monitoring is certainly an important instrument to adjust doses to compensate for interpatient and intrapatient PK variability, to monitor patient adherence to therapy and particularly to avoid side effects related to overdosing and drug–drug interactions, it is limited by the fact that it does not reflect PD and toxicodynamic interactions such as those caused by individual and environment-related factors, which are important for the efficacy and safety of drug therapy (Fig. 1). These include but are not limited to an individual's (eg, genetically determined) sensitivity to a particular medication and/or its adverse effects, additive (synergistic or antagonistic) pharmacological effects of coadministered drugs, the development of tolerance, the potential influences of coexisting morbidity, the level of immune responsiveness, age, patient stature, dietary, and overall life habits. On a molecular level, in vivo factors (such as the density of receptors on the cell surface, the functionality of second messengers in signal transmission or of regulatory factors that control gene translation and protein expression, translational modifications, and stability) also affect drug response and influence the association between drug kinetics and the corresponding PD or toxicodynamic effects.3,4
Owing to the said limitations of PK TDM and the increased interest in personalized drug therapy, PD TDM based on molecular “biomarkers” has become increasingly important. Moreover, this development is driven by new developments in instrumentation, such as mass spectrometry and array technologies, and in computational biology/pharmacology, databases, and bioinformatics. Most importantly, these are technologies that allow for the measurement of tens and sometimes thousands of parameters in a single analytical run such as the so-called “omics” technologies.5,6
PD monitoring is not a new approach in the management of drug therapy per se. Clinical PD markers, typically representing physical variables or symptoms (fever, pain, rush, bleeding, diarrhea, abnormal behavior or thinking, etc.), are often monitored and used either directly or as a part of disease activity scores to guide therapy. Although such physician-reported outcome parameters are very well established and in general reflect changes in molecular PD biomarkers, they are often subjective and have very limited specificity for the action of a particular drug. Moreover, they often suffer from a considerable lag time between the development of events on a molecular and a clinical level, and sometimes, careful data collection (particularly for clinical scores) is time-consuming. Instrumental tools such as electroencephalography, structural magnetic resonance imaging, electrocardiography, endoscopic evaluations, and positron emission tomography integrated with computed tomography provide a more objective measure of pathological processes and their response to drug effects. However, the read out is not quantitative, and the changes observed are often not drug specific. Histological examination of tissue biopsies, the gold standard to assess graft injury after solid organ transplantation as a consequence of underimmunosuppression or drug toxicity, is an invasive procedure, and its informative value is depended on the quality of the obtained tissue sample as well as the expertise of the pathologist evaluating it.4
By contrast, molecular PD biomarkers, an emerging field of PD monitoring, are a more promising alternative because molecular biomarkers are objective, often highly sensitive, and provide quantitative measurements of pathobiochemical processes occurring in vivo. For several of said biomarkers, there are already analytically well-validated commercial tests available that can be run on consolidated automated analytical systems without the need for very specific technical skills and deliver results within a very short turnaround time. Analysis is often established with different sample matrices, and sequential monitoring is possible.3,7
Molecular PD biomarkers can be divided into drug-specific biomarkers that reflect changes in the biochemical and physiological functions in the body produced directly by the drugs (eg, interaction with specific receptors or alteration of target enzyme activity or target gene expression) and general (non–drug-specific) biomarkers that describe nonspecific physical or chemical interactions.4
To the general biomarkers belong a large list of well-established biochemical clinical markers such as creatinine, C-reactive protein, enzyme activities, bilirubin, and troponins that primarily reflect the severity of disease or tissue injury but can also be used for a crude estimation of PD effects triggered by the prescribed drugs. Moreover, they are often slow indicators of PD effects and an individual's response to therapy. Therefore, they are usually inappropriate to precisely monitor therapy, with drug side effects and clinical complications remaining common. Other biological/biochemical markers such as the international normalized ratio (vitamin K antagonists), glucose/HbA1c (antihyperglycemic drugs), low-density lipoprotein–cholesterol (statins), leukocytes (immunosuppressants), CK-18 and CK-18 fragments (antitumor agents), or viral load (antiviral treatment) are more specific for the therapeutic effects of a particular group of drugs, but still not for the PD effects of a single drug as a part of a combination therapy, because they do not reflect the specific molecular mechanism(s) of drug action. However, this becomes an advantage if the analysis aims at the evaluation of additive drug effects.4
Molecular biomarkers that specifically reflect the interaction between a drug and its pharmacological target have the advantage of providing a tool for precise dose adjustment based on an individual's molecular responsiveness to the therapy. Moreover, such biomarkers have the important potential to support the selection of the most appropriate therapeutic approach for an individual. Typical drug-specific PD biomarkers monitor the activity or gene expression of the inosine monophosphate dehydrogenase, the therapeutic target of mycophenolate,8 the activity of calcineurin phosphatase under therapy with calcineurin inhibitors,8 the activity of factor VIII under desmopressin,9 the inhibition of thrombin formation under dabigatran,10 or B-lymphocyte suppression for the therapy with anti-CD20 monoclonal antibodies.11 The analysis of circulating tumor DNA as an instrument to select patients with epidermal growth factor receptor–mutant non–small cell lung cancer for specific targeted therapy, if a tumor sample is unavailable, has been approved recently by the European Medical Agency and FDA.12 According to the updated guidelines published by the Clinical Pharmacogenetics Implementation Consortium, consideration of polymorphisms related to the warfarin PK and PD (CYP2C9, VKORC1, and optionally CYP4F2 genes) is strongly recommended to predict the appropriate warfarin dose in addition to well-established clinical and international normalized ratio monitoring.13 This example illustrates the potential added value of integrating various biomarker tools in drug dosing algorithms for a timely and precise prediction and adjustment of dose requirements.
The present Focus Issue of Therapeutic Drug Monitoring appears in the year that we celebrate the 40th anniversary of the journal; it focuses on the state-of-the-art implementation of PD monitoring as a tool to complement PK TDM (Fig. 1). It includes 11 review articles that aim to highlight the introduction of modern TDM approaches to diverse clinical situations to better predict, understand, and individualize PD effects (both wanted and unwanted) of prescribed drug therapies. While recognizing the currently well-established PK/PD-guided approach to manage the therapy with antimicrobials, the present Focus Issue concentrates on topics for which the interplay between PK and PD monitoring is just emerging.14
The contributions in the present Focus Issue provide a critical analysis of current practices of TDM (both PK and PD) with their limitations, introduce newer promising biomarkers that are on the path between preclinical development and clinical validation, discuss the challenges faced to date in translating preclinical tools into clinical settings, and point out recent advances in the establishment of modeling approaches that apply PK/PD as well as pharmacogenetic information. The opportunities for a widespread implementation of PD monitoring to deliver a more effective treatment and the actions needed on the way to realization are highlighted. Potential benefits that go beyond the primary goal of PD monitoring—the optimized response to therapy—such as impact on costs of therapy and patient adherence to prescribed medication(s) are also addressed. The list of topics approached is broad and includes therapy with biologics,15 anticancer,16 immunosuppressive,17 antiepileptic,18 and psychotropic,19 as well as anticoagulant20 and procoagulant21 drugs. Two articles pursue specifically the PD assessment of drug-induced kidney22 and liver23 toxicity and provide an update on recent developments including critical analysis. Concentration–effect relationships of psychoactive drugs and the problem to calculate therapeutic reference ranges in psychiatry are further critical appraisals addressing the question why simple correlation analyses of psychoactive drug concentrations in blood and clinical effects are obsolete for flexible dose studies or TDM databases.19 Another contribution approaches the emerging field of pharmacogenetic biomarkers predictive of drug PD.13 Last but not least, a critical review stresses the rising role of “liquid biopsy” techniques to monitor diseases and to guide therapies using diagnostic and prognostic applications of circulating cell-free DNA in personalized cancer therapy as an example.12
Although PD monitoring, particularly based on molecular biomarkers, is already recognized as an essential element of the diagnostic armamentarium striving to a new level of personalized, precisely guided drug therapy, many important questions still remain. For example: Which are the most appropriate biomarkers for a particular clinical situation? Are said biomarkers ready for clinical implementation? How to best combine them with other diagnostic tools? Are there specific recommendations of target levels and dose adjustment based on such biomarkers developed? Are supportive tools, for example, software, mathematical models, and algorithms needed and, if yes, are they already available to complement clinical implementation? Are robust, reliable, easy to perform, and affordable analytical assays available? Are measures to assure the quality of analysis and results interpretation provided? Moreover, vice versa, the establishment of reliable biomarkers for PD monitoring by itself is likely to further foster the successful development and use of new molecularly targeted pharmaceuticals.
Based on the ever-increasing levels of knowledge in the field of drug PD as well as the availability of advanced and accessible technical and computational approaches to strengthen research in the field of molecular biomarkers and PD/toxicodynamic TDM, we can look forward with eager anticipation to new developments being made in this area in the near future.
1. Pippenger CE. Therapeutic drug monitoring is turning 30—Happy Birthday. The IATDMCT Newsletter Compass, December 2009.
3. Brunet M, Shipkova M, van Gelder T, et al. Barcelona consensus on biomarker-based immunosuppressive drugs management in solid organ transplantation. Ther Drug Monit. 2016;38(suppl 1):S1–S20.
4. Shipkova M. Biomarker monitoring in immunosuppressant therapy: an overview. In: Oellerich M, Dasgupta A eds. Personalized Immunosuppression in Transplantation: Role of Biomarker Monitoring and Therapeutic Drug Monitoring. 1st ed: Elsevier; 2015:125–152.
5. Aronson JK, Ferner RE. Biomarkers-a general review. Curr Protoc Pharmacol. 2017;76:9.23.1–9.23.17.
6. Yehudit Hasin Y, Seldin M, Lusis A. Multi-omics approaches to disease. Genome Biol. 2017;18:83.
7. Shipkova M, López OM, Picard N, et al. Analytical aspects of the implementation of biomarkers in clinical transplantation. Ther Drug Monit. 2016;38(suppl 1):S80–S92.
8. Bergan S, Bremer S, Vethe NT. Drug target molecules to guide immunosuppression. Clin Biochem. 2016;49:411–418.
9. Schütte LM, van Hest RM, Stoof SCM, et al. Pharmacokinetic modelling to predict FVIII:C response to desmopressin and its reproducibility in nonsevere haemophilia a patients. Thromb Haemost. 2018;118:621–629.
10. Rigano J, Ng C, Nandurkar H, et al. Thrombin generation estimates the anticoagulation effect of direct oral anticoagulants with significant interindividual variability observed. Blood Coagul Fibrinolysis. 2018;29:148–154.
11. Kaplan B, Kopyltsova Y, Khokhar A, et al. Rituximab and immune deficiency: case series and review of the literature. J Allergy Clin Immunol Pract. 2014;2:594–600.
12. Oellerich M, Schütz E, Beck J, et al. Circulating cell-free DNA- diagnostic and prognostic applications in personalized cancer therapy. Ther Drug Monit. 2019;41:115–120.
13. Haufroid V, Picard N. Pharmacogenetics biomarkers predictive of drug pharmacodynamics as an additional tool to therapeutic drug monitoring. Ther Drug Monit. 2019;41:121–130.
14. de Velde F, Mouton JW, de Winter BCM, et al. Clinical applications of population pharmacokinetic models of antibiotics: challenges and perspectives. Pharmacol Res. 2018;134:280–288.
15. Dreesen E, Gils A. Pharmacodynamic monitoring of biological therapies in chronic inflammatory diseases. Ther Drug Monit. 2019;41:131–141.
16. Veal GJ, Amankwatia EB, Paludetto MN, et al. Pharmacodynamic therapeutic drug monitoring for cancer: challenges, advances and future opportunities. Ther Drug Monit. 2019;41:142–159.
17. Millán O, Wieland E, Marquet P, et al. Pharmacodynamic monitoring of mTOR inhibitors. Ther Drug Monit. 2019;41:160–167.
18. Brandt C. Pharmacodynamic monitoring of antiepileptic drug therapy. Ther Drug Monit. 2019;41:168–173.
19. Hiemke C. Concentration-effect relationships of psychoactive drugs and the problem to calculate therapeutic reference ranges. Ther Drug Monit. 2019;41:174–179.
20. Wieland E, Shipkova M. Pharmacokinetic and pharmacodynamic drug monitoring of direct acting oral anticoagulants: where do we stand? Ther Drug Monit. 2019;41:180–191.
21. Preijers T, Schütte LM, Kruip MJHA, et al. Strategies for individualized dosing of clotting factor concentrates and desmopressin in hemophilia A and B. Ther Drug Monit. 2019;41:192–212.
22. Griffin BR, Faubel S, Edelstein CL. Biomarkers of drug-induced kidney toxicity. Ther Drug Monit. 2019;41:213–226.
23. Neuman MG, Biomarkers of drug-induced liver toxicity. Ther Drug Monit. 2019;41:227–234.