Perspectives on Model‐Informed Precision Dosing in the Digital Health Era: Challenges, Opportunities, and Recommendations

• DOI: 10.1002/cpt.2049
• Publication Date: 2020-10-17T13:07:57Z
• Authors: Franziska Kluwe; Robin Michelet; Anna Mueller‐Schoell; Corinna Maier; Lena Klopp‐Schulze; Madelé van Dyk; Gerd Mikus; Wilhelm Huisinga; Charlotte Kloft
• Keywords: 0301 basic medicine; 0303 health sciences; Dose-Response Relationship, Drug; 540; MIPD; 3. Good health; Model‐informed precision dosing; 03 medical and health sciences; Pharmaceutical Preparations; Drug approval; Humans; Precision Medicine; 500 Naturwissenschaften und Mathematik::540 Chemie::540 Chemie und zugeordnete Wissenschaften

Drug approval is based on exposure, response, and variability of studied populations, typically excluding comorbidities/medications very ill patients, thus not representing real-world populations. This results in wide therapeutic outcome for individual patients. Model-informed precision dosing (MIPD) can characterize/quantify this variability, support optimal dose selection, enable individualized therapy. The aim perspective to raise awareness MIPD, identify challenges hindering its implementation clinical practice, provide recommendations, highlight opportunities. MIPD aims at tailoring doses patients' needs, therefore presents a promising tool increase treatment success.1 Within Bayesian framework, prior knowledge about drug pharmacokinetics (PK) exposure-response relationships are patient characteristics ("covariates," e.g., age, weight, sex, disease characteristics, or comedication) PK biomarker data obtain model parameters (maximum a-posteriori estimates). Recently, assimilation methods have come into focus, overcoming major limitations maximum posteriori-based approaches by enabling accurate uncertainty quantification propagation.2 In contrast traditional well-established drug/biomarker monitoring (TDM), provides quantitative decision healthcare professionals populations integrating multi-level data. With the available computing power,3 further methodological advances comprehensive quantification,2 numerous publications demonstrating benefits MIPD,4 also user-friendliness few already existing tools,5 question arises as why practice—with exception local initiatives academic hospital centers6—still largely fails. following, we summarize selected key that need be addressed, perspectives beyond (see Figure 1, Table 1). We propose an alignment terminology across scientific disciplines, collaborative work. Furthermore, literature overview current applications, review articles, innovative methodology, initiatives, software tools (Table S1). rapidly evolving research area which multiple disciplines/communities other stakeholders meet. Due diverse origins projects tools, various terms exist these methodologies As result, labels used interchangeably, although there many common features, parallel without touch points. Harmonization definitions different areas disciplines crucial.7 Following successful example "model-informed discovery development" joint effort unified appearance under consensus term "MIPD" will ensure greater visibility extension target audience facilitate accelerate development practice. Complementary this, must recognized between ought leveraged investigating developing approaches/tools. Only through efforts, realized become reality. Moreover, "precision medicine" comprises (e.g., pharmacogenomics), often seen unconnected with times even competing concepts discourse whereas they should rather understood complementary take full advantage them. offers potential serve platform simultaneously therapy individualization optimization.4 A obstacle still constituted fact curricula physicians, pharmacists, lacking in-depth training pharmacology, needed understanding, application, evaluation tools. needs addressed via increasing publishing good examples "best practices" right journals,7 presenting conferences workshops) education (offer more opportunities acquire expertise model-informed/quantitative approaches). User-friendly integration workflow ultimately remove remaining barriers unleash power Healthcare involved understand "one-dose-fits-all" replaced improve outcome. An offer support, but rests treating physician pharmacist, who integrates overall status Additionally, regulatory registration medical devices) system reimbursement framework) level discourage application routine. Joint efforts among community promoting professionals, pharmaceutical companies, regulatory/healthcare environment required. approved investigational drugs, however, all therapies might benefit. Drugs high associated costs, severe adverse reactions narrow index, and/or interindividual particularly qualified dosing. However, reliable correlation surrogate concentration (preferably readily matrix) effect. addition, robust PK, pharmacodynamic, PK/pharmacodynamic targets, utility functions, ranges established thoroughly evaluated, warrant large-scale trials observational studies "real-world" if yet during development. challenging drugs delayed (un-)desired effect chemotherapy), tolerance developed depression) active metabolites. Next, identified suitable candidate than one methodology applicable. Pooling (or models) same drug-disease continuous updating new incoming allows capture most realistic population scenarios single appropriate use MIPD. future, frequent sampling (biosensors, wearables, point-of-care, home-sampling advanced data-analysis machine learning), better informed automated could achieved. Nevertheless, future using big learning, quality mechanistic understanding underlying processes crucial. Despite emerging from academic/clinical collaborations, routine care Translation findings easy-to-use crucial part implementation, pharmacology/pharmacometrics needed. Implementation required: strategies (training, education, adaptability, flexibility, changed clinician behavior) explored evaluated uptake tools5 (and S1), establishment guidelines fostered. collaboration societies, how implement successfully realize specific systems established, currently lacking. Next TDM infectious diseases, immunology, transplantation medicine), fields Besides questions, practical First, infrastructure adapted ready bedside, providing ambulant home-based digital devices wearable biosensors point-of-care allow patients measure report online concentrations. hospital, difficult time-consuming paper-based analysis, reporting, communication samples. electronic health record widely available, interfaces expanded trend. today's technical progress, soon no longer pose challenge least high-income countries systems. model-informed design contribute overcome "classical" problems, such inappropriate timing, quality, quantity Often highlighted concerns comprise long bioanalytical turnaround samples (several hours weeks), lack standardization workflows, instrumentation costs complex sample preparation.8 Raising (often earlier less) timepoints, streamlining internal shorten times, introducing point-of-care/bedside analytics, biosensors/wearables, home-monitoring systems9), only plasma, also, example, saliva, interstitial fluid, capillary blood, solutions listed above. Nowadays, pharmacometric within (MID3) been well accepted agencies. Therefore, exploiting their postapproval phases, investigate well-studied pregnant women obese patients) encouraged. For development, it acknowledged accepting trigger higher response rates fewer reactions, facilitating reimbursement.10 agencies increasingly value open input discourse. frameworks adapt to, time changing trial designs, flexible recommendations. Of course, does apply candidates early projected low indices costs), incorporation consideration approval, reimbursement, use.10 First regard were kicked-off agencies11 so-called "companion diagnostics." Global instead historic "one-dose-fits-all-approach" newly rising. New mobile gathering sources complexity making increases. Keeping pace era possible field includes user-friendly, scalable decision-support integrated workflow, improved physicians "MIPD practices," eventually over practices. envision integral intrinsically motivated requirement agencies, training, widespread availability, goes centers becomes optimized Multistakeholder collaborations ranging bedside validate, implement, demonstrate MIPD.11 Open access funding enabled organized Projekt DEAL. No was received C.K. W.H. grants industry consortium (AbbVie Deutschland GmbH & Co. K.G., AstraZeneca, Boehringer Ingelheim Pharma Grünenthal GmbH, F. Hoffmann-La Roche Ltd., Merck KGaA, SANOFI) PharMetrX program. reports Innovative Medicines Initiative-Joint Undertaking ("DDMoRe") Diurnal Ltd. Federal Ministry Education Research Programming Initiative Antimicrobial Resistance (JPIAMR). All outside submitted F.K. L.K.-S. employees K.G. respectively. authors declared interests Please note: publisher responsible content functionality any supporting information supplied authors. Any queries (other missing content) directed corresponding author article.