Adaptive Bayesian analysis of serum creatinine as a marker for drug-induced renal impairment in an early-phase clinical trial

Evaluation of Serum Creatinine Levels with Reference Change Value in Patients Receiving Colistin Treatment

OBJECTIVE

In this study, we aimed to evaluate the serum creatinine (SCr) levels with the reference change value (RCV) in patients receiving colistin treatment.

METHODS

We retrospectively recorded the SCr levels of 47 patients receiving colistin treatment before treatment and on days 3 and 7 after treatment. RCV was calculated with the asymmetrical RCV formula (Z = 1.64, P < .05). Percent (%) increase in the SCr results of the patients was compared with RCV and values exceeding RCV were regarded as statistically significant.

RESULTS

The RCV was calculated as 15.6% for SCr. Compared with pretreatment values, SCr value on day 3 was 32/47 and on day 7 it was 36/47; as these results exceeded RCV, they were considered statistically significant.

CONCLUSION

Use of RCV in the interpretation of results between serial measurements will provide a more rapid and sensitive method when making decisions.

Serum cystatin C and βeta-2 microglobulin as potential biomarkers in children with lupus nephritis

Objectives: In this study, we aimed to assess serum levels of Cystatin C (Cys C) and beta-2 microglobulin (β2M) in juvenile systemic lupus erythematosus (JSLE) patients and to investigate their role as potential biomarkers of lupus nephritis (LN) and overall disease activity.

Patients and methods: Between December 2018 and November 2019, a total of 40 patients with JSLE (11 males, 29 females; mean age: 12.6±2.5 years; range, 7.5 to 16 years) and 40 age- and sex-matched controls (10 males, 30 females; mean age: 12.3±2.4 years; range, 7 to 16 years) were included in this study. Serum (s) Cys C and β2M levels were compared between the groups. The SLE Disease Activity Index (SLEDAI-2K), the renal SLEDAI (rSLEDAI), and the Renal Damage Index were used.

Results: JSLE patients had significantly elevated mean sCyc C and sβ2M levels (1.4±0.8 mg/mL and 2.8±0.9 mg/mL, respectively) compared to the controls (0.6±0.1 mg/mL and 2.0±0.2 mg/mL, respectively; p<0.00). The mean sCys C and sβ2M levels were significantly higher in the LN group, compared to non-LN patients (1.8±0.7 mg/mL and 3.1±1.0 mg/mL, respectively vs. 0.8±0.3 mg/mL and 2.4±0.6 mg/mL, respectively; p=0.002 and p=0.02, respectively). The sCys C levels had significant positive correlations with erythrocyte sedimentation rate (r=0.3, p=0.05), serum creatinine (r=0.41, p= 0.007), 24-h urinary protein (r=0.58, p<0.001), anti-double stranded deoxyribonucleic acid antibodies titers (r=0.55, p=0.002), extra-renal SLEDAI scores (r=0.36, p=0.04), rSLEDAI (r=0.46, p=0.002), and renal class (r=0.7, p=0.0001). Serum β2M levels were significantly negatively correlated with complement 4 levels (r=-0.31, p=0.04) and significantly positively correlated with extra-renal SLEDAI scores (r=0.3, p=0.05).

Conclusion: These findings confirm that sCys C and sβ2M levels are increased in JSLE patients in association with the overall active disease. However, sCys C level may act as a promising non-invasive biomarker for predicting kidney disease activity and biopsy classes in children with JSLE.

A comparison of methods to generate adaptive reference ranges in longitudinal monitoring

In a clinical setting, biomarkers are typically measured and evaluated as biological indicators of a physiological state. Population based reference ranges, known as 'static' or 'normal' reference ranges, are often used as a tool to classify a biomarker value for an individual as typical or atypical. However, these ranges may not be informative to a particular individual when considering changes in a biomarker over time since each observation is assessed in isolation and against the same reference limits. To allow early detection of unusual physiological changes, adaptation of static reference ranges is required that incorporates within-individual variability of biomarkers arising from longitudinal monitoring in addition to between-individual variability. To overcome this issue, methods for generating individualised reference ranges are proposed within a Bayesian framework which adapts successively whenever a new measurement is recorded for the individual. This new Bayesian approach also allows the within-individual variability to differ for each individual, compared to other less flexible approaches. However, the Bayesian approach usually comes with a high computational cost, especially for individuals with a large number of observations, that diminishes its applicability. This difficulty suggests that a computational approximation may be required. Thus, methods for generating individualised adaptive ranges by the use of a time-efficient approximate Expectation-Maximisation (EM) algorithm will be presented which relies only on a few sufficient statistics at the individual level.

Models to estimate biological variation components and interpretation of serial results: strengths and limitations

Biological variation (BV) has multiple applications in a variety of fields of clinical laboratory. The use of BV in statistical modeling is twofold. On the one hand, some models are used for the generation of BV estimates (within- and between-subject variability). Other models are built based on BV in combination with other factors to establish ranges of normality that will help the clinician interpret serial results for the same subject. There are two types of statistical models for the calculation of BV estimates: A. Direct methods, prospective studies designed to calculate BV estimates; i. Classic model: developed by Harris and Fraser, revised by the Working Group on Biological Variation of the European Federation of Laboratory Medicine. ii. Mixed-effect models. iii. Bayesian model. B. Indirect methods, retrospective studies to derive BV estimates from large databases of results. Big data. Understanding the characteristics of these models is crucial as they determine their applicability in different settings and populations. Models for defining ranges that help in the interpretation of individual serial results include: A. Reference change value and B. Bayesian data network. In summary, this review provides an overview of the models used to define BV components and others for the follow-up of patients. These models should be exploited in the future to personalize and improve the information provided by the clinical laboratory and get the best of the resources available.

Modelos para estimar la variación biológica y la interpretación de resultados seriados: bondades y limitaciones

La variación biológica (VB) tiene múltiples aplicaciones en diversos campos del laboratorio clínico. Hay dos formas de relacionar el concepto de VB y los modelos estadísticos. Por un lado existen modelos para el cálculo de estimados de VB (intra e inter individual) y por otro, existen modelos que tienen en cuenta la VB y otros factores para la definición de rangos que ayudan a la interpretación de resultados seriados en un mismo individuo. Dentro de los modelos estadísticos dirigidos al cálculo de los estimados de VB existen dos tipos: A. Métodos directos. Estudios prospectivos, diseñados exclusivamente para el cálculo de estimados de VB: i. Modelo clásico: desarrollado por Harris y Fraser, revisado por EFLM-BVWG. ii. Modelos de efectos mixtos iii. Modelo bayesiano. B. Métodos indirectos. Estudios retrospectivos basados en extraer estimados de VB a partir de resultados que provienen de grandes bases de datos. Big-data. Ambos tipos presentan una serie de características que es importante conocer porque pueden condicionar su aplicabilidad en diferentes situaciones o poblaciones. Entre los modelos para definir rangos que ayudan a la interpretación de resultados seriados en un individuo encontramos: A. Valor de referencia del cambio (VRC). B. Red de datos bayesiana. En resumen, esta revisión pretende dar un enfoque general sobre los modelos para definir los componentes de VB así como otros para aplicarlos en el seguimiento de pacientes, que deberían ser explorados en el futuro para personalizar y mejorar la información aportada por el laboratorio clínico, aprovechando al máximo los recursos disponibles.

A bedside clinical tool using creatinine kinetics to predict worsening renal injury and early recovery

Background

Changing creatinine concentrations during acute renal failure are often confusing to clinicians to interpret and can cloud the patient's true current state of renal injury. By modifying the formula for kinetic estimate of glomerular filtration rate (KeGFR), a simple bedside clinical tool can be used to identify subtle changes in renal function.

Methods

The KeGFR was rewritten to instead calculate a predicted peak creatinine after renal injury. By comparing the changes in predicted peak creatinine at two or more subsequent time intervals, the patient's current state of renal injury can be determined: whether improving, worsening or unchanged from prior.

Results

Three case examples are provided using the equation for predicted peak creatinine. In each case, the creatinine concentration has continued to rise at three sequentially measured times. The change in predicted peak creatinine is analyzed for each case, demonstrating scenarios involving (i) improving renal injury, (ii) unchanged renal injury continued by unfavorable hemodynamics and (iii) worsening renal injury despite interventions.

Conclusions

The use of this model may provide clinicians with an easy bedside tool to assess a patient's state of acute kidney injury. Reassessment of how the creatinine is changing is already a nonquantitative part of a nephrologist's approach to acute kidney injury. Providing an assessment of the patient's changing renal function would be a useful addition to potentially detect early renal recovery or worsening renal injury and appropriately adjust treatment strategies.

The pressing need for real-time risk assessment of hospital-acquired acute kidney injury

Acute Kidney Injury (AKI) is associated with short- and long-term outcomes that reflect the severity of the injury. Recent studies have suggested that 'early' initiation of renal replacement therapy may alter the course of AKI and improve short-term outcomes like inpatient mortality. The current Kidney Disease Improving Global Outcomes (KDIGO) consensus definition of AKI has been criticized for misclassification bias, lack of sensitivity and the static manner in which AKI stages are defined. This editorial reviews various approaches to improving the specificity and sensitivity of the KDIGO AKI criteria, and also concludes that a staging system based on creatinine trajectories would be better suited for developing a prognostic index for real-time, dynamic risk assessment that the current KDIGO staging criteria.

The diagnostic value of rescaled renal biomarkers serum creatinine and serum cystatin C and their relation with measured glomerular filtration rate

BACKGROUND

Serum creatinine (Scr) is the major contributing variable in glomerular filtration rate (GFR) estimating equations. Serum cystatin C (ScysC) based GFR estimating (eGFR)-equations have also been developed. The present study investigates the relation between 'rescaled' levels of these renal biomarkers (with reference interval of [0.67-1.33]) and measured GFR (mGFR).

METHODS

We evaluated the diagnostic ability to detect impaired kidney function of the rescaled renal biomarkers in 8584 subjects from 12 cohorts with measured GFR, standardized Scr and ScysC. We calculated sensitivity and specificity of the rescaled biomarkers to identify kidney disease, with reference to a fixed (60mL/min/1.73m²) as well as an age-dependent threshold for mGFR.

RESULTS

The upper reference limit of 1.33 for rescaled renal biomarkers is closely related to the age-dependent threshold for defining kidney status by mGFR with sensitivity and specificity for the rescaled biomarkers close to 90% for all ages. If the fixed threshold of 60mL/min/1.73m² for mGFR is used, then lower specificity in children and sensitivity in older adults are observed.

CONCLUSIONS

Impaired kidney function can be diagnosed by rescaled renal biomarkers instead of eGFR-equations using the fixed threshold of 1.33 for all ages, consistent with an age-dependent threshold of mGFR.

The Effects of CYP2C19 genotype on the susceptibility for nephrosis in cardio-cerebral vascular disease treated by anticoagulation

In recent years, the genetic factor has become one of the important predisposing factors of nephropathy susceptibility. There is a high incidence of nephropathy in CCVd. The CYP2C19 enzyme metabolizes most the drugs, including proton pump inhibitors commonly used medicines to treat CCVd, CYP2C19 genetic polymorphisms is association with multi-pathogenesis factors of nephropathy. The purpose of the study is to reveal the association between CYP2C19 genotype and the susceptibility of nephropathy in the CCVd patients. The study is composed of 623 samples from CCVd treated by anticoagulation. The patients were studied, including CCVd with hyperuricemia, coronary heart disease, diabetes, and other complication. Biochemical tests and CYP2C19 variants measurements were performed by the gene chip method. The association among CYP2C19 variants, complications, and nephropathy was analyzed in the CCVd. There is no correlation between nephropathy and complications in CCVd. In hyperuricemia, coronary heart disease and diabetes groups, the differences of renal function tests were significant between CYP2C19 mutant (P < 0.05). The nephropathy risk of wild genotype is 3.288 times higher than of mutation genotype in hyperuricemic group, 1.928 times higher than mutation genotype in coronary heart disease group, and 5.248 times higher than CYP2C19 mutation genotype in the diabetic group. There was significant correlation between the CYP2C19 wild type and the nephropathy susceptibility in CCVd patients. The CYP2C19 gene plays a potential maker to evaluate nephropathy in CCVd patients. We deduced that identification of CYP2C19 gene type may benefit for reducing and avoiding nephropathy caused by abnormal metabolism function in CCVd patients.

Adaptive Bayesian approach to clinical trial renal impairment biomarker signal from urea and creatinine

A major concern with the identification of renal toxicity using the traditional biomarkers, urea and creatinine, is that toxicity signal definitions are not sensitive to medically important changes in these biomarkers. Traditional renal signal definitions for urea and creatinine have not adequately identified drugs that have generated important medical issues later in development. Here, two clinical trial databases with a posteriori known drug induced renal impairment were analyzed for the presence of a renal impairment biomarker signal from urea (590 patients; age 26-92, median 65) and creatinine (532 patients; age 26-97, median 65). Data was analyzed retrospectively using multiple definitions for the biomarker signal to include values outside stratified reference intervals, values exceeding twofold increases from baseline, values classified by the 2009 NIAID renal toxicity table, change from baseline represented as a Z-score based on intra-individual biological variations, and an adaptive Bayesian methodology that generalizes population- with individual-based methods for evaluating a biomarker signal. The data demonstrated that the adaptive Bayesian methodology generated a prominent drug induced signal for renal impairment at the first visit after drug administration. The signal was directly related to dose and time of drug administration. All other data analysis methods produced none or significantly weaker signals than the adaptive Bayesian approach. Interestingly, serum creatinine and urea are able to detect early kidney dysfunction when the biomarker signal is personalized.