Therapeutic drug monitoring (TDM) is useful for planning and individualizing drug therapy. Proper interpretation of plasma levels requires both certain prerequisites and physician knowledge. This review outlines the rational use of TDM, key indications, timing of blood samples, and the evidence for therapeutic ranges. It also highlights common pitfalls and provides guidance to support clinicians in the effective use of TDM when prescribing medications to patients.

Medication reviews in Denmark are best conducted as a cross-sectoral, multidisciplinary effort, primarily anchored in the responsibility of general practice. This literature review documents that several Danish national guidelines recommend medication reviews, which underscores the need for a shared understanding of their scope and content. The Danish Society for Clinical Pharmacology has defined the framework, scope and content for a »clinical pharmacological medication review«, which are offered as a part of the advisory services offered in all regions of Denmark.

Background: Fluoropyrimidines (FP) are widely used chemotherapy agents with substantial interindividual variability and risk of severe toxicity. Following the European Medicines Agency (EMA) recommendation in 2020, pre-emptive dihydropyrimidine dehydrogenase (DPD) testing with dose individualisation have been advocated to improve patient safety. However, robust real-world evidence on clinical outcomes after implementation remains limited. We evaluated the uptake and clinical impact of routine pre-emptive DPD testing. Methods: We conducted a retrospective, multicentre, interrupted time-series analysis using electronic health record data from patients receiving FP-containing chemotherapy at four specialised oncology departments in two Danish regions. Outcomes were compared over two years before and three years after implementation of routine DPD testing. Associations between DPD-guided dosing and hospitalisation, neutropenia, and 30-day mortality were assessed using multivariable logistic regression. Findings: Overall, 9,981 patients received FP, 7,939 of whom were included in the analyses. Within six months of implementation, over 95% of patients receiving FP were DPD tested before treatment. In the post-implementation phase, FP dosing was individualised according to guidelines and was associated with lower risk of hospitalisation (adjusted odds ratio [OR] 0.85 [95% CI 0.76-0.94]), neutropenia (OR 0.35; 0.16-0.79) and severe neutropenia (OR 0.22; 0.10-0.46) compared to the pre-implementation phase. The post-implementation thirty-day mortality after the first treatment cycle was lower (OR 0.63; 0.40-0.99) compared to the pre-implementation phase. No difference in 30-day mortality was observed in subsequent cycles. In the post-implementation phase, ddespite dose adjustment, patients with DPD deficiency had higher 30-day mortality after cycle 1 than those with normal DPD activity (OR 3.72; 1.36-10.18). Interpretation: In this real-world, population-based cohort study, pre-emptive DPD-guided FP dosing was rapidly adopted and was associated with lower risk of hospitalisation, neutropenia, and 30-day mortality, indicating notable safety benefits. Bias due to residual confounding and temporal changes in supportive care cannot be excluded.

AIMS:To analyse geographical variation in use of glucose-lowering drugs (GLDs) for type-2 diabetes (T2DM) in Denmark. METHODS:We included all adults who filled a prescription for a non-insulin GLD indicated for T2DM from 2010 to 2023 in Denmark. Stratified by each of the five regions, we calculated the annual incidence rate, the annual prevalence proportion and the total volume of dispensed GLDs from 2010 to 2023. We calculated the proportion of incident prescriptions for GLDs that were issued by different prescriber types and characterized non-insulin GLD users in each region in 2023. Finally, we analysed variation at the municipal level. RESULTS:Totally, data from 478 118 adults were included. The user characteristics were comparable across the regions with no differences in the median age (range 66-67 years) or in the sex distribution (range 56-58% males). In all regions, general practitioners were the main prescribers of GLDs. There were minor regional differences in the use of GLDs, with a predominant and increasing use of metformin, accelerating use of SGLT-2is and GLP-1RAs, and declining use of SUs over time. In Q4 2023 the ranges of prevalence per 1000 inhabitants were 31 to 44 for metformin, 16 to 23 for SGLT2is, 14 to 23 for GLP-1RAs and 1.37 to 1.58 for SUs. There were some differences in characteristics of GLD users across regions, with a slightly higher prevalence of diabetes-related complications and larger involvement of hospital physicians in the Capital Region compared to the other regions. We identified four municipality clusters that differed marginally in the prescribing pattern of non-insulin GLD. CONCLUSION:The minor differences in the use of GLDs indicate equal access to GLDs across Denmark. Differences at the municipal level call for future studies to investigate if these reflect differences in clinical practice or differences in T2DM populations.

AIM:Gestational diabetes mellitus (GDM) has been associated with reduced postprandial glucagon-like peptide 1 (GLP-1) responses. As pregnancy induces changes in gallbladder motility and bile acids stimulate GLP-1 secretion, we investigated postprandial gallbladder emptying and GLP-1 responses in women with GDM. METHODS:Women with and without GDM underwent two 240-min mixed meal tests; one during third trimester of pregnancy and one 3-6 months postpartum. We evaluated ultrasonography-assessed gallbladder emptying, plasma concentrations of glucometabolic hormones including GLP-1, paracetamol absorption (proxy for gastric emptying) and circulating factors known to affect gallbladder dynamics. RESULTS:Fifteen women with GDM and 15 pregnant women with normal glucose tolerance (NGT) (baseline median age 31 (interquartile range 29;33) versus 32 (28;33) years, body mass index (BMI) 27.2 (24.7;30.7) versus 28.4 (26.2;31.0) kg/m2, HbA1c 30 (29;32) versus 30 (28;31) mmol/mol) were included. No differences in postprandial gallbladder emptying or GLP-1 responses were observed between women with and without GDM, neither during pregnancy nor postpartum. Pregnancy increased fasting gallbladder volumes by 69 (30;122)% and 103 (59;156)% and postprandial gallbladder emptying by 77 (28;236)% and 99 (37;190)% compared with postpartum in women with and without GDM, respectively. Postprandial GLP-1 responses were reduced by 60 (3;82)% and 81 (11;90)% during pregnancy compared with postpartum in women with and without GDM, respectively. CONCLUSION:Pregnancy-induced changes in gallbladder motility seem to play no or a limited role in previously reported GDM-associated reduced postprandial GLP-1 responses as gallbladder emptying was greater and postprandial GLP-1 response was lower in pregnancy than postpartum regardless of GDM status.

Background: QT prolongation is a potential serious adverse drug reaction, and assessing the risk of QT-prolonging drugs is routinely included in psychotropic medication reviews. However, the actual clinical benefits of such assessments are unknown. We investigate whether QT prolongation (QTc value > 480 ms) manifests in psychiatric inpatients at risk of QT prolongation as identified by assessing drug regimens. Secondly, we test the predictive value of well-known risk factors for QT prolongation. Results: The median patient age was 49 years (IQR 34–64) for patients treated with a median of nine drugs (IQR 6–12) and a median QT-prolonging drug sum of three daily defined dosages (IQR 1.88–4.76). We extracted 290 ECGs for patients where pharmacist-led-medication reviews (PMRs) identified an increased risk of QT prolongation and 190 ECGs for patients with no such risk, identifying 33 cases of verified QT prolongation equally distributed between groups. Unadjusted regression analysis revealed that advanced age (OR 3.27 CI 95% 1.60–6.84) and cardiovascular comorbidity (OR 3.53 CI 95% 1.71–7.29) were associated with manifest QT prolongation, while the QT-prolonging drug load was not. Methods: We reviewed electronic health records (EHRs) of 799 psychiatric inpatients exposed to PMRs made from 1 September 2016 to 31 December 2018 in Region Zealand Denmark. Conclusions: Patients at risk of QT prolongation as identified by drug reviews rarely manifests with actual QT prolongation. Non-pharmacological risk factors seem to be better predictors for identifying patients with QT prolongation.

BACKGROUND:Machine learning can operationalize the rich and complex data in electronic patient records for exploratory pharmacovigilance endeavours. OBJECTIVE:The objective of this review is to identify applications of machine learning and big patient data in exploratory pharmacovigilance. METHODS:We searched PubMed and Embase and included original articles with an exploratory pharmacovigilance purpose, focusing on medicinal interventions and reporting the use of machine learning in electronic patient records with ≥1000 patients collected after market entry. FINDINGS:Of 2557 studies screened, seven were included. Those covered six countries and were published between 2015 and 2021. The most prominent machine learning methods were random forests, logistic regressions, and support vector machines. Two studies used artificial neural networks or naive Bayes classifiers. One study used formal concept analysis for association mining, and another used temporal difference learning. Five studies compared several methods against each other. The numbers of patients in most data sets were in the order of thousands; two studies used what can more reasonably be considered big data with >1 000 000 patients records. CONCLUSION:Despite years of great aspirations for combining machine learning and clinical data for exploratory pharmacovigilance, only few studies still seem to deliver somewhat on these expectations.

ImportanceThis is the first network meta-analysis to assess outcomes associated with multiple conventional synthetic disease-modifying antirheumatic drugs and glucocorticoid.ObjectiveTo analyze clinical outcomes after treatment with conventional synthetic disease-modifying antirheumatic drugs and glucocorticoid among patients with rheumatoid arthritis.Data SourcesWith no time restraint, English language articles were searched in MEDLINE, Embase, Cochrane Central, ClinicalTrials.gov, and reference lists of relevant meta-analyses until September 15, 2022.Study SelectionFour reviewers in pairs of 2 independently included controlled studies randomizing patients with rheumatoid arthritis to mono–conventional synthetic disease-modifying antirheumatic drugs, glucocorticoid, placebo, or nonactive treatment that recorded at least 1 outcome of tender joint count, swollen joint count, erythrocyte sedimentation rate, and C-reactive protein level. Of 1098 assessed articles, 130 articles (132 interventions) were included.Data Extraction and SynthesisThe review followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses reporting guideline, and data quality was assessed by the Cochrane risk of bias tool RoB 2. Data were extracted by a single author and checked independently by 2 authors. Data were analyzed using a random effect model, and data analysis was conducted from June 2021 to February 2023.Main Outcomes and MeasuresA protocol with hypothesis and study plan was registered before data recording. The most complete of recorded outcomes (tender joint count) was used as primary outcome, with imputations based on other outcomes to obtain a full analysis of all studies. Absolute change adjusted for baseline disease activity was assessed.ResultsA total of 29 interventions in 275 treatment groups among 132 randomized clinical trials (mean [range], 71.0% [27.0% to 100%] females in studies; mean [range] of ages in studies, 53 [36 to 70] years) were identified, which included 13 260 patients with rheumatoid arthritis. The mean (range) duration of RA was 79 (2 to 243) months, and the mean (range) disease activity score was 6.3 (4.0 to 8.8). Compared with placebo, oral methotrexate was associated with a reduced tender joint count by 5.18 joints (95% credible interval [CrI], 4.07 to 6.28 joints). Compared with methotrexate, glucocorticoid (−2.54 joints; 95% CrI, −5.16 to 0.08 joints) and remaining drugs except cyclophosphamide (6.08 joints; 95% CrI, 0.44 to 11.66 joints) were associated with similar or lower tender joint counts.Conclusions and RelevanceThis study’s results support the present role of methotrexate as the primary reference conventional synthetic disease-modifying antirheumatic drug.

This review offers a summary of the current knowledge of pshychotropic drugs and glaucoma. If exposed to psychotropic drugs, some patients may develop angle-closure glaucoma. Although rarely contraindicated, exposed predisposed and diagnosed patients should be followed-up by an ophthalmologist. It is still unclear if serotonin reuptake inhibitors increase the risk of angle-closure glaucoma. Tricyclic antidepressants and benzodiazepines should be used with caution in predisposed patients. The same applies to antipsychotic drugs, where first-generation antipsychotic drugs might have a smaller impact on the intraocular pressure than second-generation antipsychotic drugs.

Background Antipsychotic-induced weight gain is a contributing factor in the reduced life expectancy reported amongst people with psychotic disorders. CYP2D6 is a liver enzyme involved in the metabolism of many commonly used antipsychotic medications. We investigated if CYP2D6 genetic variation influenced weight or BMI among people taking antipsychotic treatment. Methods We conducted a systematic review and a random effects meta-analysis of publications in Pubmed, Embase, PsychInfo, and CENTRAAL that had BMI and/or weight measurements of patients on long-term antipsychotics by their CYP2D6-defined metabolic groups (poor, intermediate, normal/extensive, and ultra-rapid metabolizers, UMs). Results Twelve studies were included in the systematic review. All cohort studies suggested that the presence of reduced-function or non-functional alleles for CYP2D6 was associated with greater antipsychotic-induced weight gain, whereas most cross-sectional studies did not find any significant associations. Seventeen studies were included in the meta-analysis with clinical data of 2,041 patients, including 93 poor metabolizers (PMs), 633 intermediate metabolizers (IMs), 1,272 normal metabolizers (NMs), and 30 UMs. Overall, we did not find associations in any of the comparisons made. The estimated pooled standardized differences for the following comparisons were (i) PM versus NM; weight = –0.07 (95%CI: –0.49 to 0.35, p = 0.74), BMI = 0.40 (95%CI: –0.19 to 0.99, p = 0.19). (ii) IM versus NM; weight = 0.09 (95% CI: –0.04 to 0.22, p = 0.16) and BMI = 0.09 (95% CI: –0.24 to 0.41, p = 0.60). (iii) UM versus EM; weight = 0.01 (95% CI: –0.37 to 0.40, p = 0.94) and BMI = –0.08 (95%CI: –0.57 to 0.42, p = 0.77). Conclusion Our systematic review of cohort studies suggested that CYP2D6 poor metabolizers have higher BMI than normal metabolizers, but the data of cross-sectional studies and the meta-analysis did not show this association. Although our review and meta-analysis constitutes one of the largest studies with comprehensively genotyped samples, the literature is still limited by small numbers of participants with genetic variants resulting in poor or UMs status. We need further studies with larger numbers of extreme metabolizers to establish its clinical utility in antipsychotic treatment. CYP2D6 is a key gene for personalized prescribing in mental health.

BACKGROUND:In clinical oncology, systemic 5-fluorouracil (5-FU) and its oral pro-drugs are used to treat a broad group of solid tumours. Patients with dihydropyrimidine dehydrogenase (DPD) enzyme deficiency are at elevated risk of toxicity if treated with standard doses of 5-FU. DPYD genotyping and measurements of plasma uracil concentration (DPD phenotyping) can be applied as tests for DPD deficiency. In April 2020, the European Medicines Agency recommended pre-treatment DPD testing to reduce the risk of 5-FU-related toxicity.OBJECTIVES:The objective of this study is to present the current evidence for DPD testing in routine oncological practice.METHODS:Two systematic literature searches were performed following the PRISMA guidelines. We identified studies examining the possible benefit of DPYD genotyping or DPD phenotyping on the toxicity risk.FINDINGS:Nine and 12 studies met the criteria for using DPYD genotyping and DPD phenotyping, respectively.CONCLUSIONS:The evidence supporting either DPYD genotyping or DPD phenotyping as pre-treatment tests to reduce 5-FU toxicity is poor. Further evidence is still needed to fully understand and guide clinicians to dose by DPD activity.

Purpose: While the beneficial effects of medications are numerous, drug-drug interactions may lead to adverse drug reactions that are preventable causes of morbidity and mortality. Our goal was to quantify the prevalence of potential drug-drug interactions in drug prescriptions at Danish hospitals, estimate the risk of adverse outcomes associated with discouraged drug combinations, and highlight the patient types (defined by the primary diagnosis of the admission) that appear to be more affected. Methods: This cross-sectional (descriptive part) and cohort study (adverse outcomes part) used hospital electronic health records from two Danish regions (similar to 2.5 million people) from January 2008 through June 2016. We included all inpatients receiving two or more medications during their admission and considered concomitant prescriptions of potentially interacting drugs as per the Danish Drug Interaction Database. We measured the prevalence of potential drug-drug interactions in general and discouraged drug pairs in particular during admissions and associations with adverse outcomes: post-discharge all-cause mortality rate, readmission rate and length-of-stay. Results: Among 2 886 227 hospital admissions (945 475 patients; median age 62 years [IQR: 41-74]; 54% female; median number of drugs 7 [IQR: 4-11]), patients in 1 836 170 admissions were exposed to at least one potential drug-drug interaction (659 525 patients; median age 65 years [IQR: 49-77]; 54% female; median number of drugs 9 [IQR: 6-13]) and in 27 605 admissions to a discouraged drug pair (18 192 patients; median age 68 years [IQR: 58-77]; female 46%; median number of drugs 16 [IQR: 11-22]). Meropenem-valproic acid (HR: 1.5, 95% CI: 1.1-1.9), domperidone-fluconazole (HR: 2.5, 95% CI: 2.1-3.1), imipramine-terbinafine (HR: 3.8, 95% CI: 1.2-12), agomelatine-ciprofloxacin (HR: 2.6, 95% CI: 1.3-5.5), clarithromycin-quetiapine (HR: 1.7, 95% CI: 1.1-2.7) and piroxicam-warfarin (HR: 3.4, 95% CI: 1-11.4) were associated with elevated mortality. Confidence interval bounds of pairs associated with readmission were close to 1; length-of-stay results were inconclusive. Conclusions: Well-described potential drug-drug interactions are still missed and alerts at point of prescription may reduce the risk of harming patients; prescribing clinicians should be alert when using strong inhibitor/inducer drugs (i.e. clarithromycin, valproic acid, terbinafine) and prevalent anticoagulants (i.e. warfarin and non-steroidal anti-inflammatory drugs - NSAIDs) due to their great potential for dangerous interactions. The most prominent CYP isoenzyme involved in mortality and readmission rates was 3A4.

Antipsychotic-induced weight gain (AIWG) is a serious adverse effect. Studies have linked genetically-predicted CYP2D6 metabolic capacity to AIWG. The evidence, however, is ambiguous. We performed multiple regression analyses examining the association between genetic-predicted CYP2D6 metabolic capacity and AIWG. Analyses were based on previously unpublished data from an RCT investigating the clinical utility of routine genotyping of CYP2D6 and CYP2C19 in patients with schizophrenia. A total of 211 patients, corresponding to 71% of the original study population, were included. Our analyses indicated an effect of genetically predicted CYP2D6 metabolic capacity on AIWG with significant weight gain in both CYP2D6 poor metabolizers (PMs) (4.00 kg (95% CI: 0.80; 7.21)) and ultrarapid metabolizers (UMs) (6.50 kg (95% CI: 1.03; 12.0)). This finding remained stable after adjustment for covariates (PMs: 4.26 kg (0.88; 7.64), UMs: 7.26 kg (1.24; 13.3)). In addition to the CYP2D6 metabolic capacity, both baseline body mass index (−0.24 (95% CI: −0.44; −0.03)) and chlorpromazine equivalents per day (0.0041 (95% CI: 0.0005; 0.0077)) were statistically significantly associated with weight change in the adjusted analysis. Our results support that the genetically predicted CYP2D6 metabolic capacity matters for AIWG.

Benjamin Skov Kaas-Hansen,1–3 Cristina Leal Rodríguez,2 Davide Placido,2 Hans-Christian Thorsen-Meyer,2,4 Anna Pors Nielsen,2 Nicolas Dérian,5 Søren Brunak,2 Stig Ejdrup Andersen1 1Clinical Pharmacology Unit, Zealand University Hospital, Roskilde, Denmark; 2NNF Center for Protein Research, University of Copenhagen, Copenhagen, Denmark; 3Section for Biostatistics, Department of Public Health, University of Copenhagen, Copenhagen, Denmark; 4Department of Intensive Care Medicine, Copenhagen University Hospital (Rigshospitalet), Copenhagen, Denmark; 5Data and Development Support, Region Zealand, DenmarkCorrespondence: Benjamin Skov Kaas-Hansen, Department of Intensive Care, Copenhagen University Hospital — Rigshospitalet, Blegdamsvej 9, Copenhagen, 2100, Denmark, Tel +45 60 19 68 01, Email epiben@hey.com View the original paper by Dr Kaas-Hansen and colleagues This is in response to the Letter to the Editor

Introduction Dosing of renally cleared drugs in patients with kidney failure often deviates from clinical guidelines but little is known about what is predictive of receiving inappropriate doses. Methods and materials We combined data from the Danish National Patient Register and in-hospital data on drug administrations and estimated glomerular filtration rates for admissions between 1 October 2009 and 1 June 2016, from a pool of about 2.9 million persons. We trained artificial neural network and linear logistic ridge regression models to predict the risk of five outcomes (>0, ≥1, ≥2, ≥3 and ≥5 inappropriate doses daily) with index set 24 hours after admission. We used time-series validation for evaluating discrimination, calibration, clinical utility and explanations. Results Of 52,451 admissions included, 42,250 (81%) were used for model development. The median age was 77 years; 50% of admissions were of women. ≥5 drugs were used between admission start and index in 23,124 admissions (44%); the most common drug classes were analgesics, systemic antibacterials, diuretics, antithrombotics, and antacids. The neural network models had better discriminative power (all AUROCs between 0.77 and 0.81) and were better calibrated than their linear counterparts. The main prediction drivers were use of anti-inflammatory, antidiabetic and anti-Parkison’s drugs as well as having a diagnosis of chronic kidney failure. Sex and age affected predictions but slightly. Conclusion Our models can flag patients at high risk of receiving at least one inappropriate dose daily in a controlled in-silico setting. A prospective clinical study may confirm this holds in real-life settings and translates into benefits in hard endpoints.

Dihydropyrimidine dehydrogenase (DPD) genotype and phenotype among Danish cancer patients: prevalence and correlation between DPYD-genotype variants and P-uracil concentrations

We sought to craft a drug safety signalling pipeline associating latent information in clinical free text with exposures to single drugs and drug pairs. Data arose from 12 secondary and tertiary public hospitals in two Danish regions, comprising approximately half the Danish population. Notes were operationalised with a fastText embedding, based on which we trained 10 270 neural‐network models (one for each distinct single‐drug/drug‐pair exposure) predicting the risk of exposure given an embedding vector. We included 2 905 251 admissions between May 2008 and June 2016, with 13 740 564 distinct drug prescriptions; the median number of prescriptions was 5 (IQR: 3–9) and in 1 184 340 (41%) admissions patients used ≥5 drugs concomitantly. A total of 10 788 259 clinical notes were included, with 179 441 739 tokens retained after pruning. Of 345 single‐drug signals reviewed, 28 (8.1%) represented possibly undescribed relationships; 186 (54%) signals were clinically meaningful. Sixteen (14%) of the 115 drug‐pair signals were possible interactions, and two (1.7%) were known. In conclusion, we built a language‐agnostic pipeline for mining associations between free‐text information and medication exposure without manual curation, predicting not the likely outcome of a range of exposures but also the likely exposures for outcomes of interest. Our approach may help overcome limitations of text mining methods relying on curated data in English and can help leverage non‐English free text for pharmacovigilance.