Coming full circle: The potential utility of real-world evidence to discern predictions from a physiologically based pharmacokinetic model.

Today real word data (RWD) are playing a greater role in informing health care decisions. A physiologically based pharmacokinetic model (PBPK) and observed exposure-risk relationship predicted an increased bleeding risk induced by rivaroxaban (RXB) in patients with mild to moderate chronic kidney disease (CKD) taking concomitant medications that are combined Pgp-CYP3A inhibitors. In this commentary, we explore the potential use of RWD to assess the clinical consequence of this complex drug-drug interaction predicted from PBPK. This is a retrospective, case control, pilot study using a RWD dataset of 896,728 patients with mild to moderate chronic kidney disease and rivaroxaban use that was refined based upon combined Pgp-CYP3A inhibitor exposure and report of drug-induced bleeding (DIB). The odds ratio of patients with mild to moderate chronic kidney disease taking rivaroxaban with or without concurrent Pgp-CYP3A inhibitor use having a DIB was calculated. The odds ratio for DIB was 2.04 (CI95 1.82, 2.3; p < 0.001) suggesting an approximate doubling of bleeding risk which is consistent with the rivaroxaban exposure changes predicted by the published PBPK model and observed exposure-risk relationship. This exploratory analysis demonstrated the potential utility of RWD to assess model-based predictions as part of a drugs life cycle management.

Artificial Intelligence and Machine Learning for Lead-to-Candidate Decision-Making and Beyond.

The use of artificial intelligence (AI) and machine learning (ML) in pharmaceutical research and development has to date focused on research: target identification; docking-, fragment-, and motif-based generation of compound libraries; modeling of synthesis feasibility; rank-ordering likely hits according to structural and chemometric similarity to compounds having known activity and affinity to the target(s); optimizing a smaller library for synthesis and high-throughput screening; and combining evidence from screening to support hit-to-lead decisions. Applying AI/ML methods to lead optimization and lead-to-candidate (L2C) decision-making has shown slower progress, especially regarding predicting absorption, distribution, metabolism, excretion, and toxicology properties. The present review surveys reasons why this is so, reports progress that has occurred in recent years, and summarizes some of the issues that remain. Effective AI/ML tools to derisk L2C and later phases of development are important to accelerate the pharmaceutical development process, ameliorate escalating development costs, and achieve greater success rates.

Implementing Predictive Models Within an Electronic Health Record System: Lessons from an External Validation of a Suicide Risk Model.

Over the past 5 years, there has been an increase in the development of EHR-based models for predicting suicidal behaviour. Using the McGinn (2000) framework for creating clinical prediction rules, this study discusses the broad validation of one such predictive model in a context external to its derivation. Along with reporting performance metrics, our paper high-lights five practical challenges that arise when trying to undertake such a project including (i) validation sample sizes, (ii) availability and timeliness of data, (iii) limited or incomplete documentation for predictor variables, (iv) reliance on structured data and (v) differences in the source context of algorithms. We also discuss our study in the context of the current literature.

Application of exposure bracketing to streamline the development of contraceptive products.

Developing new long-acting products of well-characterized contraceptive drugs is one way to address some of the reasons for unmet need for modern methods of family planning among women in low- and middle-income countries. Development and approval of such products traditionally follow a conventional paradigm that includes large Phase 3 clinical trials to evaluate efficacy (pregnancy prevention) and safety of the investigational product. Exposure-bracketing is a concept that applies known pharmacokinetics and pharmacodynamics of a drug substance to inform its safe and efficacious use in humans. Several therapeutic areas have applied this concept by leveraging established drug concentration-response relationships for approved products to expedite development and shorten the timeline for the approval of an investigational product containing the same drug substance. Based on discussions at a workshop hosted by the Bill & Melinda Gates Foundation in December 2020, it appears feasible to apply exposure-bracketing to develop novel contraceptive products using well-characterized drugs.

Systemic quinolones and risk of retinal detachment III: a nested case-control study using a US electronic health records database.

BACKGROUND: Quinolones are popular antibiotics that are known for their potency, broad coverage, and reasonable safety. Concerns have been raised about a possible association between quinolones and retinal detachment (RD). METHODS: We conducted a nested case-control study using electronic health records (EHR) from the Health Facts® Database. The initial cohort included all patients who were admitted between 2000 and 2016, with no history of eye disease, and had a minimum medical history of one year. Eligible cases comprised inpatients who were first admitted with a primary diagnosis of RD between 2010 and 2015. Each eligible case was matched without replacement to five unique controls by sex, race, age, and period-at-risk. We used conditional logistic regression to calculate RD risk, adjusting for exposure to other medications, and major risk factors. RESULTS: We identified 772 cases and 3860 controls. Whereas our primary analysis of all subjects revealed no quinolone-associated RD risk, elevated but non-significant risks were noted in African Americans (ciprofloxacin and levofloxacin), those aged 56-70 years old (moxifloxacin), and women (ciprofloxacin). CONCLUSION: Our study did not identify an elevated RD risk within 30 days following systemic administration of quinolone antibiotics. Suggestions of increased risk observed in some population subgroups warrant further investigation.

Use of RWE to Inform Regulatory, Public Health Policy, and Intervention Priorities for the Developing World.

For low- and middle-income countries (LMICs) to benefit from real-world evidence (RWE)/real-world data (RWD) in both product registration ("regulatory") decision making and in product utilization policy ("policy") decision making, they need to overcome several challenges. They need to deploy more electronic health records systems (EHRs), adjust for confounder variables, build trust between stakeholders, and create laws and regulations for local generation of data that are assented for secondary use. The role of procurers and their use of RWE/RWD in the LMIC context likewise is in a state of ongoing development. Procurers of health products are strong players currently in the "access" chain as LMICs continue to work on strengthening governmental health technology assessment (HTA) bodies. Procurers' use of RWE is presently at an early stage and is mostly indirect, leveraging RWE results that are produced by researchers in high-income countries (HICs), often under considerably different regulatory and policy objectives and constraints compared to LMICs' epidemiology and priorities. Pending wider deployment of EHRs and other RWE sources, stakeholders must realize that populations from HIC RWE (i) can be devised to closely resemble phenotypic patterns in LMIC populations and (ii) can be analyzed to align with LMICs' unmet needs.

Characterizing Long COVID: Deep Phenotype of a Complex Condition.

BACKGROUND: Numerous publications describe the clinical manifestations of post-acute sequelae of SARS-CoV-2 (PASC or "long COVID"), but they are difficult to integrate because of heterogeneous methods and the lack of a standard for denoting the many phenotypic manifestations. Patient-led studies are of particular importance for understanding the natural history of COVID-19, but integration is hampered because they often use different terms to describe the same symptom or condition. This significant disparity in patient versus clinical characterization motivated the proposed ontological approach to specifying manifestations, which will improve capture and integration of future long COVID studies. METHODS: The Human Phenotype Ontology (HPO) is a widely used standard for exchange and analysis of phenotypic abnormalities in human disease but has not yet been applied to the analysis of COVID-19. FUNDING: We identified 303 articles published before April 29, 2021, curated 59 relevant manuscripts that described clinical manifestations in 81 cohorts three weeks or more following acute COVID-19, and mapped 287 unique clinical findings to HPO terms. We present layperson synonyms and definitions that can be used to link patient self-report questionnaires to standard medical terminology. Long COVID clinical manifestations are not assessed consistently across studies, and most manifestations have been reported with a wide range of synonyms by different authors. Across at least 10 cohorts, authors reported 31 unique clinical features corresponding to HPO terms; the most commonly reported feature was Fatigue (median 45.1%) and the least commonly reported was Nausea (median 3.9%), but the reported percentages varied widely between studies. INTERPRETATION: Translating long COVID manifestations into computable HPO terms will improve analysis, data capture, and classification of long COVID patients. If researchers, clinicians, and patients share a common language, then studies can be compared/pooled more effectively. Furthermore, mapping lay terminology to HPO will help patients assist clinicians and researchers in creating phenotypic characterizations that are computationally accessible, thereby improving the stratification, diagnosis, and treatment of long COVID. FUNDING: U24TR002306; UL1TR001439; P30AG024832; GBMF4552; R01HG010067; UL1TR002535; K23HL128909; UL1TR002389; K99GM145411.

Systemic quinolones and risk of acute liver failure III: A nested case-control study using a US electronic health records database.

BACKGROUND AND AIM: Quinolones are globally popular antibiotics with proven potency, broad coverage, and reasonable safety. However, some concerns were raised as to their possible association with acute liver failure (ALF). The aim of this study is to assess ALF risk within 30 days of receiving a systemically administered quinolone antibiotic, in individuals with no history of liver/diseases. METHODS: We conducted a nested case-control study using electronic health records from the Cerner Health Facts. The initial cohort (n = 35 349 943) included all patients who were admitted between 2000 and 2016, with no history of liver diseases, and had a minimum medical history of one year. Eligible cases were inpatients who were first diagnosed with ALF between 2010 and 2015. Using incidence density sampling, each case was matched with up to five unique controls by sex, race, age at index encounter, and period-at-risk. We used conditional logistic regression to calculate the odds ratio and 95% confidence interval for ALF risk, upon adjusting for exposure to other medications, and major confounders (diabetes mellitus and alcohol abuse). We used the STROBE Statement for reporting on our study. RESULTS: We identified 3151 cases and 15 657 controls. Our primary analysis did not reveal an association between quinolones and ALF risk. However, some risk was identified among those with no or few comorbidities, those ≤ 60 years of age, women, men, African Americans, and Caucasians. CONCLUSION: Although our study does not suggest an overall association between quinolones and ALF, elevated risks seen in some subgroups warrant further investigation.

Pressure Ulcer Risk Factors in Persons with Mobility-Related Disabilities.

OBJECTIVE: To assess pressure ulcer (PU) risk in persons with mobility impairments using a large data set to identify demographic, laboratory, hemodynamic, and pharmacologic risk factors. METHODS: The cohort of interest was persons with disabilities who have mobility impairments and are diagnostically at risk of PUs. To define this cohort, diagnoses that qualify patients for skin protection wheelchair cushions were used. Data were obtained from the Cerner Health Facts data warehouse. Two cohorts were defined: persons with and without a history of PUs. Analysis included descriptive statistics and multivariate logistic regression modeling. Variables retained in the model were identified using LASSO, gradient boosting, and Bayesian model averaging. MAIN RESULTS: The resulting cohorts included more than 87,000 persons with a history of PUs and more than 1.1 million persons who did not have a PU. The data revealed seven disability groups with the greatest prevalence of PUs: those with Alzheimer disease, cerebral palsy, hemiplegia, multiple sclerosis, paraplegia/quadriplegia, Parkinson disease, and spina bifida. Ulcers in the pelvic region accounted for 82% of PUs. Persons with disabilities who were male or black had a greater prevalence of PUs. Physiologic risk factors included the presence of kidney or renal disease, decreased serum albumin, and increased serum C-reactive protein. CONCLUSIONS: The results indicate that, although persons with disabilities can exhibit a wide functional range, they remain at risk of PUs and should be evaluated for proper preventive measures, including support surfaces and wheelchair cushions.

External validation and comparison of two variants of the Elixhauser comorbidity measures for all-cause mortality.

Assessing prevalent comorbidities is a common approach in health research for identifying clinical differences between individuals. The objective of this study was to validate and compare the predictive performance of two variants of the Elixhauser comorbidity measures (ECM) for inhospital mortality at index and at 1-year in the Cerner Health Facts® (HF) U.S. DATABASE: We estimated the prevalence of select comorbidities for individuals 18 to 89 years of age who received care at Cerner contributing health facilities between 2002 and 2011 using the AHRQ (version 3.7) and the Quan Enhanced ICD-9-CM ECMs. External validation of the ECMs was assessed with measures of discrimination [c-statistics], calibration [Hosmer-Lemeshow goodness-of-fit test, Brier Score, calibration curves], added predictive ability [Net Reclassification Improvement], and overall model performance [R2]. Of 3,273,298 patients with a mean age of 43.9 years and a female composition of 53.8%, 1.0% died during their index encounter and 1.5% were deceased at 1-year. Calibration measures were equivalent between the two ECMs. Calibration performance was acceptable when predicting inhospital mortality at index, although recalibration is recommended for predicting inhospital mortality at 1 year. Discrimination was marginally better with the Quan ECM compared the AHRQ ECM when predicting inhospital mortality at index (cQuan = 0.887, 95% CI: 0.885-0.889 vs. cAHRQ = 0.880, 95% CI: 0.878-0.882; p < .0001) and at 1-year (cQuan = 0.884, 95% CI: 0.883-0.886 vs. cAHRQ = 0.880, 95% CI: 0.878-0.881, p < .0001). Both the Quan and the AHRQ ECMs demonstrated excellent discrimination for inhospital mortality of all-causes in Cerner Health Facts®, a HIPAA compliant observational research and privacy-protected data warehouse. While differences in discrimination performance between the ECMs were statistically significant, they are not likely clinically meaningful.

Associations between Anticholinergic Burden and Adverse Health Outcomes in Parkinson Disease.

BACKGROUND: Elderly adults should avoid medications with anticholinergic effects since they may increase the risk of adverse events, including falls, delirium, and cognitive impairment. However, data on anticholinergic burden are limited in subpopulations, such as individuals with Parkinson disease (PD). The objective of this study was to determine whether anticholinergic burden was associated with adverse outcomes in a PD inpatient population. METHODS: Using the Cerner Health Facts® database, we retrospectively examined anticholinergic medication use, diagnoses, and hospital revisits within a cohort of 16,302 PD inpatients admitted to a Cerner hospital between 2000 and 2011. Anticholinergic burden was computed using the Anticholinergic Risk Scale (ARS). Primary outcomes were associations between ARS score and diagnosis of fracture and delirium. Secondary outcomes included associations between ARS score and 30-day hospital revisits. RESULTS: Many individuals (57.8%) were prescribed non-PD medications with moderate to very strong anticholinergic potential. Individuals with the greatest ARS score (≥ 4) were more likely to be diagnosed with fractures (adjusted odds ratio (AOR): 1.56, 95% CI: 1.29-1.88) and delirium (AOR: 1.61, 95% CI: 1.08-2.40) relative to those with no anticholinergic burden. Similarly, inpatients with the greatest ARS score were more likely to visit the emergency department (adjusted hazard ratio (AHR): 1.32, 95% CI: 1.10-1.58) and be readmitted (AHR: 1.16, 95% CI: 1.01-1.33) within 30-days of discharge. CONCLUSIONS: We found a positive association between increased anticholinergic burden and adverse outcomes among individuals with PD. Additional pharmacovigilance studies are needed to better understand risks associated with anticholinergic medication use in PD.

Enhancing Nursing Staffing Forecasting With Safety Stock Over Lead Time Modeling.

In balancing competing priorities, it is essential that nursing staffing provide enough nurses to safely and effectively care for the patients. Mathematical models to predict optimal "safety stocks" have been routine in supply chain management for many years but have up to now not been applied in nursing workforce management. There are various aspects that exhibit similarities between the 2 disciplines, such as an evolving demand forecast according to acuity and the fact that provisioning "stock" to meet demand in a future period has nonzero variable lead time. Under assumptions about the forecasts (eg, the demand process is well fit as an autoregressive process) and about the labor supply process (≥1 shifts' lead time), we show that safety stock over lead time for such systems is effectively equivalent to the corresponding well-studied problem for systems with stationary demand bounds and base stock policies. Hence, we can apply existing models from supply chain analytics to find the optimal safety levels of nurse staffing. We use a case study with real data to demonstrate that there are significant benefits from the inclusion of the forecast process when determining the optimal safety stocks.

Off-label use of cancer therapies in women diagnosed with breast cancer in the United States.

PURPOSE: To determine the level of off-label cancer therapy use in a population of female breast cancer patients and to establish whether this use was evidence-based. METHODS: A study was conducted by sampling Cerner's data warehouse for all women diagnosed with breast cancer between January 2000 and June 2009 who received at least one cancer therapy approved by the US-FDA during the study period. Drug encounters were considered off-label if the circumstances of use did not match the age or medical diagnoses specified on the product label at the time of study. The level of evidence for the use of these drugs in a breast cancer setting was evaluated from randomized phase III trials using a tiered approach. RESULTS: The study included 2,663 women with a median age of 59 years. A total of 1,636 off-label encounters were recorded, representing 13.0% of all encounters. Of the 65 cancer therapies investigated, 55.4% were prescribed off-label. The drugs with the highest off-label use were, in a descending order, vinorelbine, carboplatin, bevacizumab, leuprolide, liposomal doxorubicin and cisplatin. Most off-label encounters were evidence-based and more likely to be associated with private insurance coverage, younger age, ethnicities other than Caucasian, smaller treatment centres and drugs with limited labeled indications that have a longer market history. CONCLUSIONS: Off-label prescribing is common practice in oncology and is an integral component of breast cancer treatment strategies. While this practice tends to be associated with specific socio-demographic factors and disease characteristics, the majority of off-label encounters appear to be evidence-based.

Detection of prion protein in urine-derived injectable fertility products by a targeted proteomic approach.

BACKGROUND: Iatrogenic transmission of human prion disease can occur through medical or surgical procedures, including injection of hormones such as gonadotropins extracted from cadaver pituitaries. Annually, more than 300,000 women in the United States and Canada are prescribed urine-derived gonadotropins for infertility. Although menopausal urine donors are screened for symptomatic neurological disease, incubation of Creutzfeldt-Jakob disease (CJD) is impossible to exclude by non-invasive testing. Risk of carrier status of variant CJD (vCJD), a disease associated with decades-long peripheral incubation, is estimated to be on the order of 100 per million population in the United Kingdom. Studies showing infectious prions in the urine of experimental animals with and without renal disease suggest that prions could be present in asymptomatic urine donors. Several human fertility products are derived from donated urine; recently prion protein has been detected in preparations of human menopausal gonadotropin (hMG). METHODOLOGY/PRINCIPAL FINDINGS: Using a classical proteomic approach, 33 and 34 non-gonadotropin proteins were identified in urinary human chorionic gonadotropin (u-hCG) and highly-purified urinary human menopausal gonadotropin (hMG-HP) products, respectively. Prion protein was identified as a major contaminant in u-hCG preparations for the first time. An advanced prion protein targeted proteomic approach was subsequently used to conduct a survey of gonadotropin products; this approach detected human prion protein peptides in urine-derived injectable fertility products containing hCG, hMG and hMG-HP, but not in recombinant products. CONCLUSIONS/SIGNIFICANCE: The presence of protease-sensitive prion protein in urinary-derived injectable fertility products containing hCG, hMG, and hMG-HP suggests that prions may co-purify in these products. Intramuscular injection is a relatively efficient route of transmission of human prion disease, and young women exposed to prions can be expected to survive an incubation period associated with a minimal inoculum. The risks of urine-derived fertility products could now outweigh their benefits, particularly considering the availability of recombinant products.

Intensive care unit length of stay: Benchmarking based on Acute Physiology and Chronic Health Evaluation (APACHE) IV.

OBJECTIVE: To revise and update the Acute Physiology and Chronic Health Evaluation (APACHE) model for predicting intensive care unit (ICU) length of stay. DESIGN: Observational cohort study. SETTING: One hundred and four ICUs in 45 U.S. hospitals. PATIENTS: Patients included 131,618 consecutive ICU admissions during 2002 and 2003, of which 116,209 met inclusion criteria. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The APACHE IV model for predicting ICU length of stay was developed using ICU day 1 patient data and a multivariate linear regression procedure to estimate the precise ICU stay for randomly selected patients who comprised 60% of the database. New variables were added to the previous APACHE III model, and advanced statistical modeling techniques were used. Accuracy was assessed by comparing mean observed and mean predicted ICU stay for the excluded 40% of patients. Aggregate mean observed ICU stay was 3.86 days and mean predicted 3.78 days (p < .001), a difference of 1.9 hrs. For 108 (93%) of 116 diagnoses, there was no significant difference between mean observed and mean predicted ICU stay. The model accounted for 21.5% of the variation in ICU stay across individual patients and 62% across ICUs. Correspondence between mean observed and mean predicted length of stay was reduced for patients with a short (< or =1.7 days) or long (> or =9.4 days) ICU stay and a low (<20%) or high (>80%) risk of death on ICU day 1. CONCLUSIONS: The APACHE IV model provides clinically useful ICU length of stay predictions for critically ill patient groups, but its accuracy and utility are limited for individual patients. APACHE IV benchmarks for ICU stay are useful for assessing the efficiency of unit throughput and support examination of structural, managerial, and patient factors that affect ICU stay.

Acute Physiology and Chronic Health Evaluation (APACHE) IV: hospital mortality assessment for today's critically ill patients.

OBJECTIVE: To improve the accuracy of the Acute Physiology and Chronic Health Evaluation (APACHE) method for predicting hospital mortality among critically ill adults and to evaluate changes in the accuracy of earlier APACHE models. DESIGN: : Observational cohort study. SETTING: A total of 104 intensive care units (ICUs) in 45 U.S. hospitals. PATIENTS: A total of 131,618 consecutive ICU admissions during 2002 and 2003, of which 110,558 met inclusion criteria and had complete data. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We developed APACHE IV using ICU day 1 information and a multivariate logistic regression procedure to estimate the probability of hospital death for randomly selected patients who comprised 60% of the database. Predictor variables were similar to those in APACHE III, but new variables were added and different statistical modeling used. We assessed the accuracy of APACHE IV predictions by comparing observed and predicted hospital mortality for the excluded patients (validation set). We tested discrimination and used multiple tests of calibration in aggregate and for patient subgroups. APACHE IV had good discrimination (area under the receiver operating characteristic curve = 0.88) and calibration (Hosmer-Lemeshow C statistic = 16.9, p = .08). For 90% of 116 ICU admission diagnoses, the ratio of observed to predicted mortality was not significantly different from 1.0. We also used the validation data set to compare the accuracy of APACHE IV predictions to those using APACHE III versions developed 7 and 14 yrs previously. There was little change in discrimination, but aggregate mortality was systematically overestimated as model age increased. When examined across disease, predictive accuracy was maintained for some diagnoses but for others seemed to reflect changes in practice or therapy. CONCLUSIONS: APACHE IV predictions of hospital mortality have good discrimination and calibration and should be useful for benchmarking performance in U.S. ICUs. The accuracy of predictive models is dynamic and should be periodically retested. When accuracy deteriorates they should be revised and updated.

Therapeutic potential of implantable replacement hearts.

In patients hospitalized with decompensated life-threatening heart failure, the impact of newer pharmacologic therapies and mechanical circulatory support has not yet been realized, except for those who are bridged to cardiac transplantation. For long-term support of transplant-ineligible patients who have severe biventricular failure that is refractory to optimized pharmacologic therapy, replacement of the natural heart with a totally implantable mechanical replacement heart, capable of producing blood flow of up to 8 to 10 L/min, may become the most well tolerated and effective treatment. This article summarizes the current status of the first generation implantable replacement heart (AbioCor trade mark, ABIOMED. Inc., Danvers, MA). With regard to optimizing the further enhancement of treatment options for end-stage heart failure and other life-threatening illnesses, the pharmacodynamics-like principle of therapeutic efficiency should play a role in the development of both drugs and devices. In keeping with that principle, we recommend that adjusting a product's design input requirements to maximize the therapeutic effect per exposure and;to separate the cumulative therapeutic effects of the product from the cumulative adverse effects (of the product, and of the comorbid disease processes in the patients treated) should be part of the good product development process for any therapeutic product.