Assessing accuracy of ChatGPT in response to questions from day to day pharmaceutical care in hospitals.

Background: The advent of Large Language Models (LLMs) such as ChatGPT introduces opportunities within the medical field. Nonetheless, use of LLM poses a risk when healthcare practitioners and patients present clinical questions to these programs without a comprehensive understanding of its suitability for clinical contexts. Objective: The objective of this study was to assess ChatGPT's ability to generate appropriate responses to clinical questions that hospital pharmacists could encounter during routine patient care. Methods: Thirty questions from 10 different domains within clinical pharmacy were collected during routine care. Questions were presented to ChatGPT in a standardized format, including patients' age, sex, drug name, dose, and indication. Subsequently, relevant information regarding specific cases were provided, and the prompt was concluded with the query "what would a hospital pharmacist do?". The impact on accuracy was assessed for each domain by modifying personification to "what would you do?", presenting the question in Dutch, and regenerating the primary question. All responses were independently evaluated by two senior hospital pharmacists, focusing on the availability of an advice, accuracy and concordance. Results: In 77% of questions, ChatGPT provided an advice in response to the question. For these responses, accuracy and concordance were determined. Accuracy was correct and complete for 26% of responses, correct but incomplete for 22% of responses, partially correct and partially incorrect for 30% of responses and completely incorrect for 22% of responses. The reproducibility was poor, with merely 10% of responses remaining consistent upon regeneration of the primary question. Conclusions: While concordance of responses was excellent, the accuracy and reproducibility were poor. With the described method, ChatGPT should not be used to address questions encountered by hospital pharmacists during their shifts. However, it is important to acknowledge the limitations of our methodology, including potential biases, which may have influenced the findings.

Association of Busulfan Exposure and Outcomes after HCT for Patients with an Inborn Error of Immunity.

Allogeneic HCT is a potentially curative treatment strategy for patients with inborn errors of immunities (IEIs). Objective of this study was to assess the optimal busulfan exposure prior to allogeneic HCT for patients with an IEI who received an intravenous busulfan-based conditioning regimen between 2000 and 2023. Patients from 17 international centers were included. Main outcome of interest was event-free survival (EFS). Patients were categorized into 4 IEI subgroups: combined-immunodeficiency (CID), severe combined immunodeficiency (SCID), neutrophil disorders and hemophagocytic lymphohistiocytosis (HLH)-related disorders. Busulfan exposure was calculated by individual centers (AUCCENTER) and was re-estimated using a validated model (AUCNONMEM). Overall, 562 patients were included: 173 (30.8%) CID, 154 (27.4%) SCID, 101 (18.0%) HLH-related disorders, and 134 (23.8%) neutrophil disorders. Median busulfan AUCNONMEM was 69.0 mg×h/L and correlated poorly with AUCCENTER (r2=0.54). Patients with SCID, HLH-related, and neutrophil disorders were analyzed together (n=389), because CID disease subtype was an effect modifier (p=0.03). Estimated 2-year EFS was 78.5%. In patients with the found optimal busulfan AUCNONMEM of 70-90 mg×h/L, 2-year EFS was superior to <70 mg×h/L (adj-HR 1.97, 95% CI 1.11-3.49, p=0.02), and >90 mg×h/L (adj-HR 5.05, 95% CI 2.43-10.49, p<0.0001). Full donor chimerism increased with higher busulfan AUCNONMEM, plateauing at 90 mg×h/L. For CID patients, optimal AUCNONMEM for donor chimerism was found to be >70 mg×h/L. Improved EFS and higher donor chimerism may be achieved by targeting a cumulative busulfan AUCNONMEM of 80 mg×h/L (range 70-90). Our study stresses the importance to uniformly using a validated population PK-model to estimate the AUCNONMEM.

Association of Lamotrigine Plasma Concentrations with Efficacy and Toxicity in Patients with Epilepsy: A Retrospective Study.

BACKGROUND: There is limited evidence to support the currently suggested lamotrigine (LTG) therapeutic reference range of 2.5-15 mg/L for the treatment of seizures. The objective of this study was to evaluate the association of LTG plasma concentrations with the efficacy and toxicity of the treatment in patients with epilepsy. METHODS: Patients whose LTG plasma concentration was measured between January 2013 and February 2022 were included. Efficacy was defined as seizure freedom for at least 6 months around the time of measured LTG concentration. Toxicity was defined as any LTG-related adverse drug effect documented in each patient's health record or when the reason for measuring the LTG concentration was toxicity. In addition, the dose-concentration relationship of LTG was assessed. RESULTS: In total, 549 concentrations from 259 patients with epilepsy were included. The most common reasons for therapeutic drug monitoring were suspected inefficacy (39%) and pregnancy (21%). The LTG plasma concentration was not associated with efficacy (adjusted odds ratio = 0.94; 95% confidence interval, 0.85-1.04). The LTG plasma concentration was positively associated with the incidence of toxicity after adjusting for age, sex, and number of antiepileptic drugs (odds ratio = 1.11; 95% confidence interval, 1.04-1.19). The daily dose had a significant linear correlation with the LTG plasma concentration ( P < 0.001). CONCLUSIONS: The LTG plasma concentration was associated with toxicity, whereas no association with efficacy was found. A reference range of 2.5-10 mg/L may be considered to decrease the risk of toxicity while maintaining similar efficacy. Therapeutic drug monitoring may be useful when LTG-related toxicity is suspected and in cases of pharmacokinetic changes (eg, pregnancy and concomitant use of interacting drugs) that can influence the LTG plasma concentration.

Association Between the Magnitude of Intravenous Busulfan Exposure and Development of Hepatic Veno-Occlusive Disease in Children and Young Adults Undergoing Myeloablative Allogeneic Hematopoietic Cell Transplantation.

Intravenous busulfan is widely used as part of myeloablative conditioning regimens in children and young adults undergoing allogeneic hematopoietic cell transplantation (HCT). Hepatic veno-occlusive disease/sinusoidal obstruction syndrome (VOD/SOS) is a serious clinical problem observed with busulfan-based conditioning HCT. The development of VOD/SOS may be associated with busulfan exposure. Getting more insight into the association between busulfan exposure and the development of VOD/SOS enables further optimization of dosing and treatment strategies. The objective of this study was to assess the association between the magnitude of busulfan exposure and the occurrence of VOD/SOS in children and young adults undergoing myeloablative conditioning with a busulfan-containing regimen before allogeneic HCT. In this observational study we included all patients who underwent allogeneic HCT with intravenous busulfan as part of the conditioning regimen at 15 pediatric transplantation centers between 2000 and 2015. The endpoint was the development of VOD/SOS. The magnitude of busulfan exposure was estimated using nonlinear mixed effect modeling and expressed as the maximal concentration (Cmax; day 1 and day 1 to 4 Cmax), cumulative area under the curve (AUC; day 1, highest 1-day AUC in 4 days, and 4-day cumulative AUC), cumulative time above a concentration of 300 µg/L, and clearance on day 1. A total of 88 out of 697 patients (12.6%) developed VOD/SOS. The number of alkylators in the conditioning regimen was a strong effect modifier; therefore we stratified the regression analysis for the number of alkylators. For patients receiving only busulfan as one alkylator (36.3%, n = 253), cumulative busulfan exposure (>78 mg × h/L) was associated with increased VOD/SOS risk (12.6% versus 4.7%; odds ratio [OR] = 2.95, 95% confidence interval [CI] 1.13 to 7.66). For individuals receiving busulfan with one or two additional alkylators (63.7%, n = 444), cumulative busulfan exposure (≤78 and >78 mg × h/L) did not further increase the risk of VOD/SOS (15.4% versus 15.2%; OR = 1.03, 95% CI 0.61 to 1.75). The effect of the magnitude of busulfan exposure on VOD/SOS risk in children and young adults undergoing HCT is dependent on the number of alkylators. In patients receiving busulfan as the only alkylator, higher cumulative busulfan exposure increased the risk of VOD/SOS, whereas in those receiving multiple alkylators, the magnitude of busulfan exposure did not further increase this risk.

Unbound Fraction of Clozapine Significantly Decreases with Elevated Plasma Concentrations of the Inflammatory Acute-Phase Protein Alpha-1-Acid Glycoprotein.

BACKGROUND: During inflammation, elevated total (unbound plus protein-bound) clozapine plasma concentrations have been observed. Elevated alpha-1-acid glycoprotein concentrations during inflammation are suggested to cause increased plasma clozapine-alpha-1-acid glycoprotein binding, resulting in elevated total clozapine plasma concentrations without significant changes in unbound concentrations. Here, we investigated the association between alpha-1-acid glycoprotein plasma concentrations and clozapine unbound fraction. METHODS: First, 25 and 60 µL of alpha-1-acid glycoprotein solution (20 mg/mL) were added to plasma samples (n = 3) of clozapine users (spiking experiment). Second, the association between alpha-1-acid glycoprotein plasma concentration and clozapine unbound fraction was assessed in patient samples (patient study). Samples were determined by liquid chromatography-tandem mass spectrometry. Data were analyzed with a paired t test (spiking experiment) and an unpaired t test (patient study). RESULTS: The spiking experiment showed significantly lower mean unbound fractions following 25- and 60-µL alpha-1-acid glycoprotein spikes (relative reductions of 28.3%, p = 0.032 and 43.4%, p = 0.048). In the patient study, total clozapine plasma concentrations were 10% higher in elevated (n = 6) compared with normal alpha-1-acid glycoprotein (n = 20) samples [525 µg/L vs. 479 µg/L, mean difference = 47 µg/L (95% confidence interval -217 to 310), p = 0.72]. Elevated alpha-1-acid glycoprotein samples had a 26% lower mean unbound fraction compared with normal samples [1.22% vs. 1.65%, mean difference = -0.43% (95% confidence interval -0.816 to -0.0443), p = 0.03]. CONCLUSIONS: Both the spiking experiment and patient study showed a significant association between elevated alpha-1-acid glycoprotein plasma concentrations and a lower clozapine unbound fraction. Future studies should include clinical data to examine whether this association is clinically relevant, suggesting any clozapine dose adjustments.