bioRxiv preprint doi: https://doi.org/10.1101/2020.05.22.110080; this version posted May 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. A clinically validated human capillary blood transcriptome test for global systems biology studies Ryan Toma1, Ben Pelle2, Nathan Duval1, Matthew M Parks1, Vishakh Gopu1, Hal Tily1, Andrew Hatch1, Ally Perlina1, Guruduth Banavar1, and Momchilo Vuyisich1* 1 Viome, Inc., Bellevue, WA, USA, viome.com, 2 The University of Washington, Seattle, USA *Corresponding author: Momchilo Vuyisich, [email protected] Abstract should be integrated into longitudinal, population- scale, systems biology studies. Chronic diseases are the leading cause of morbidity and mortality globally. Yet, the majority of Introduction them have unknown etiologies, and genetic Quantitative gene expression analysis contribution is weak. In addition, many of the chronic provides a global snapshot of tissue function. The diseases go through the cycles of relapse and human transcriptome varies with tissue type, remission, during which the genomic DNA does not developmental stage, environmental stimuli, and change. This strongly suggests that human gene health/disease state (Frith et al., 2005; Lin et al., 2019; expression is the main driver of chronic disease onset Marioni et al., 2008; Mortazavi et al., 2008; Velculescu and relapses. To identify the etiology of chronic et al., 1999). Changes in the expression patterns can diseases and develop more effective preventative provide insights into the molecular mechanisms of measures, a comprehensive gene expression analysis disease onset and progression. In the era of precision of the human body is needed. Blood tissue is easy to medicine, methods to non-invasively assess an access and contains a large number of expressed individual’s transcriptional changes can reveal genes involved in many fundamental aspects of our predictive markers of disease, inform the best choice physiology. of therapy, and enable the development of novel We report here the development of a whole therapies. For example, using systems biology blood transcriptome clinical test that is high approaches to interrogate the transcriptome are now throughput, automated, inexpensive, and clinically a standard strategy to stratify cancer patients and validated. The test requires only 50 microliters of blood select the best available therapies (Cabanero and from a finger prick, enabling access by diverse Tsao, 2018; Chauhan et al., 2019; Rosell et al., 2012). populations that have been traditionally under- Methods to analyze the transcriptome have become a represented in clinical research. The transcripts in the vital part of trying to better understand chronic samples are preserved at the time of collection and diseases with unknown etiologies. can be stored and/or transported at ambient In the United States, chronic, non- temperatures for up to 28 days. The sample communicable diseases have become an epidemic preservative protects integrity, while also inactivating with an ever-growing share of health care all pathogens (bacteria, fungi, and viruses), enabling expenditures (CDC, 2020). Many chronic diseases go safe transportation globally. Given its unique set of through cycles of elevated disease activity (disease usability features and clinical performance, this test relapse) and periods of low disease activity (disease remission) (Carlberg et al., 2019; Haberman et al., bioRxiv preprint doi: https://doi.org/10.1101/2020.05.22.110080; this version posted May 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. A clinically validated human capillary blood transcriptome test for global systems biology studies 2019; Ostrowski et al., 2019; Stephenson et al., 2018; sequencing technologies has enabled the rapid Zhao et al., 2013). The human genome has provided analysis of the transcriptome from small sample valuable information on disease predisposition and volumes. To our knowledge, there is currently no susceptibility, but the overall genetic contribution to easily scalable blood test for population-based studies many chronic diseases is weak (Rappaport, 2016). using whole-blood transcriptome analysis. Here we Furthermore, many environmental or lifestyle factors describe, for the first time, a high-throughput method increase the risk of developing a chronic disease, for interrogating the human transcriptome from a small implicating a change in gene expression. This has finger prick blood sample. The method is automated, been shown in large twin and epidemiological studies inexpensive, and clinically validated. An RNA (Hand et al., 2016; Wilkins et al., 2019). Changes in preservation buffer mixed with the blood sample at the gene expression have been shown in type I diabetes point of collection inactivates all pathogens (bacteria, (Reynier et al., 2010), type II diabetes (Berisha et al., fungi, and viruses), enabling safe transportation 2011; Homuth et al., 2015), and cardiovascular globally at ambient temperatures for up to 28 days, disease (Pedrotty et al., 2012; Wirka et al., 2018), while preserving RNA integrity for gene expression often associated with immune regulation (Chaussabel analysis. et al., 2010). Unfortunately, little progress has been made in understanding the role of the transcriptome in Methods chronic disease due to the lack of a low-cost and Ethics statement scalable method for which samples can be collected All procedures involving human participants by anyone and anywhere, instead relying on complex were performed in accordance with the ethical logistics of collecting peripheral blood at a clinic and standards and approved by a federally accredited transporting it frozen to a laboratory. In addition, there Institutional Review Board (IRB) committee. Informed are very few clinically validated blood transcriptome consent was obtained from all participants. tests available, so analyses of data from disparate methods can lead to erroneous conclusions due to Sample collection biases introduced in each method. Capillary blood was collected by study Circulating blood is easily accessible and participants using an at-home kit, following the provides a non-invasive alternative to tissue biopsies included instructions. The kit contains a 1.5mL screw for molecular profiling of human disease and disease cap tube (Sarstedt) containing 200uL of RNA risk (Liew et al., 2006). Peripheral blood mononuclear preservative buffer (RPB), a 50uL EDTA-coated cells (PBMC) have been the primary choice for minivette (Sarstedt), a 70% isopropyl alcohol wipe transcriptomic investigations from whole blood samples (Sen et al., 2018). Unfortunately, methods available for scrutinizing the transcriptome from Figure 1. Primary components of the at- PBMCs are useful for research methods, but not ideal home capillary blood for broader applications at scale. Furthermore, collection kit and outline of the sample analysis methods using PBMCs fail to measure the cell-free method. The lancet is transcripts, which have been associated with certain used to prick the finger; disease conditions, particularly for cancer patients and the minivette is a capillary device with a piston; the pregnant women (Ng et al., 2003; Pinzani et al., 2010; sample collection tube is Pös et al., 2018; Souza et al., 2017). a microcentrifuge tube with 200 microliters of The current state of the art focuses on the RPB. whole blood transcriptome, typically requiring a venous blood draw. The availability of high-throughput 2 bioRxiv preprint doi: https://doi.org/10.1101/2020.05.22.110080; this version posted May 23, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. A clinically validated human capillary blood transcriptome test for global systems biology studies (Dynarex), a gauze pad, a push-button 17-gauge blood collection method described above. To validate disposable lancet (Acti-Lance), and instruction sheet. the storage stability, samples were stored for 7, 14, Participants were instructed to fast for 8 hours prior to and 28 days at room temperature (RT) or -80oC and collection and collect the samples within one hour of compared to the technical replicates analyzed the day waking up, without exercising or drinking (except of collection (day 0). A subset of replicate samples was water). The finger was wiped with a 70% isopropyl shipped twice during storage times. alcohol wipe and allowed to dry. Using the disposable RNA extraction, library preparation, and lancet, the finger was punctured, and the first drop sequencing wiped away using the gauze pad. The blood was drawn by capillary action into a 50uL EDTA-coated RNA was extracted using silica beads and a minivette and dispensed into the sample collection series of washes, followed by elution in molecular tube that contained 200uL of RPB. The tube was biology grade water. DNA was degraded using vigorously shaken for 15 seconds to thoroughly mix RNase-free DNase. Polyadenylated messenger RNAs RPB with blood. The puncture was wiped clean and a (mRNAs) were isolated from the total RNA using an band-aid was applied. oligo dT-based magnetic mRNA isolation kit. mRNAs were converted to directional sequencing libraries with Peripheral (venous) blood was collected by a