Time- and Race-Specific Haematological Reference Intervals for Healthy Volunteer Trials: A Retrospective Analysis of Pooled Data From Multiple Phase I Trials.

Simon Coates,Ulrike Lorch,Tomasz Pierscionek,Dilshat Djumanov,Jörg Täubel,Duolao Wang,Sara Fernandes

doi:10.3389/fphar.2020.00314

Abstract

Most UK hospitals, laboratories, and research institutions use uniform reference intervals (RI) that do not take into account known diurnal and racial variation in total white blood cells (WBC) count and its constituent parameters. These risks of excluding potentially suitable ethnic minority volunteers from participating in phase I clinical trials could call into question the validity of a trial’s findings or limit its scientific applications and ability to accurately observe drug effects upon WBC parameters. This study pools data from multiple phase I trials, assesses the effects of race and time of day on WBC count, and compares it to the existing literature to establish race and time-specific RIs. A total 13,332 venous blood samples obtained from 7,157 healthy male and female volunteers at the time of screening or admission (predosing) who took part in 35 phase I trials over a period of seven years were pooled and the data were analyzed using generalised estimating equation models. Adjusted RI of total WBC count and its individual parameters were then calculated according to time of day (morning vs. evening) for both black and nonblack populations. This study indicates that black individuals on average had lower total WBC, neutrophil, monocyte, eosinophil, and basophil counts than individuals from nonblack racial groups. Black volunteers had higher mean lymphocyte counts relative to their nonblack counterparts. These differences were deemed statistically significant. Statistically significant increases in total WBC, neutrophil, lymphocyte, and monocyte counts were also observed over the course of daily sampling. Eosinophil counts decreased during this time period, but this finding was only statistically significant in the nonblack population. Despite an observed mild diurnal increase in basophil count in both populations, this was not considered statistically significant. This high-powered study adds significant weight to the known evidence for diurnal and racial variation in WBC parameters. Importantly, it proposes specific RIs that more precisely reflect race and time of day. These could ensure increased participation of black volunteers in clinical trials for improved population representation. Furthermore, the proposed RIs allow for more accurate postdose safety monitoring and reporting, and ensure improved monitoring of postdose WBC count changes.

Highlights

Most UK hospitals, laboratories, CROs, and other research institutions use uniform reference intervals (RI) for white blood cell (WBC) parameters, irrespective of an individual’s race, gender, age, or the time of day
All subjects provided written informed consent for that specific clinical trial and all 35 trials were approved by the Medicines and Healthcare Products Regulatory Agency (MHRA) and a Research Ethics Committee
Total WBC count rose by an average of 20% from the morning until the end of daily sampling, while the increase for nonblack volunteers was 29% (Figure 1A)

Summary

Introduction

Most UK hospitals, laboratories, CROs, and other research institutions use uniform reference intervals (RI) for white blood cell (WBC) parameters, irrespective of an individual’s race, gender, age, or the time of day. Adopting a single reference range is potentially unsuitable and could result in the underrepresentation of volunteers from ethnic minority groups in early phase clinical trials (Haley et al, 2017; Sheffet et al, 2018). A poor representation of these groups in phase I studies may call into question the validity of a clinical trial’s findings (Hussain-Gambles et al, 2004), limit its scientific applications, and amplify health disparities (Chen et al, 2014). Low participation rates in clinical trials by historically underrepresented groups has previously contributed to failures in detecting potential harm as population representation proved insufficient to spot negative effects (Roberts et al, 2012). A single reference range may limit the ability to accurately observe drug effects on WBC parameters

Methods

Results

Conclusion