key: cord-0869628-cad09wvn authors: Almarzooq, A. A. title: Exploration of interethnic variation and repurposed drug efficacy in the treatment of SARS-CoV-2 Infection (COVID-19) date: 2021-03-08 journal: nan DOI: 10.1101/2021.03.07.21253095 sha: 9960240e346f706df13169b6c8c7950daa1bcd1e doc_id: 869628 cord_uid: cad09wvn The COVID-19 global pandemic has led to repurposing of drugs, with little underlying evidence for treatment safety and efficacy. This may increase complications for patients with acute viral respiratory infections. UGT1A1 and CYP2D6 enzymes are involved in the metabolism of atazanavir and fluvoxamine repurposed for COVID-19. This study aimed to elucidate the role of interethnic variation in these enzymes in the efficacy of repurposed drug therapies. A retrospective cohort of 101 Jordanian Arab samples were genotyped using Affymetrix DMET Plus Premier Package. Comprehensive global population genetic structure analyses were performed for CYP2D6 and UGT1A1 allele frequencies across multi-ethnic populations of over 131,000 global subjects from 417 published reports, revealing that Jordanian Arabs share the closest sequence homology to European and Near East populations. The East Asian populations have significantly over-representaiton of individuals with diplotype pairs for reduced atazanavir metabolism compared to the African populations and are more likely to show impaired UGT1A1 metabolism. East Asian populations are also 4.4x more likely to show impaired fluvoxamine metabolism than South Central Asian and Oceanian populations, and 8x more likely than other ancestry populations. The results here support previous findings that interethnic variation should be used for developing proper population-specific dosage guidelines. The global coronavirus disease 2019 pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has led to more than 105 million infected patients and more than 2,600,000 deaths worldwide 1 . With the rise of multiple and potentially more virulent variant strains, managing the treatment and spread of the disease is paramount. The management of the disease includes preventive and therapeutic strategies, treatment of the acute respiratory distress syndrome and the cytokine storm 2 . Patients who are hospitalized with COVID-19 may be treated with multiple therapeutics, including repurposed treatments and adjuvant therapies 3, 4 , in addition to any concomitant medications already prescribed to the patients 5 . As such, it is vital for clinicians to be able to quickly and easily integrate patient-specific factors, such as genetic variation to optimize treatment 6, 7 . (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2021. ; https://doi.org/10.1101/2021.03.07.21253095 doi: medRxiv preprint 4 metabolizers (IMs) and poor metabolizers (PMs). The PMs (i.e. carriers of two decreased function alleles) are more likely to develop jaundice, which may cause non-adherence to medications, and alternative drugs need to be considered. The risk of discontinuation is low or very low for individuals carrying one or zero decreased-function alleles (IMs and NMs) 15 . Meanwhile, the recent report of Lenze and colleagues indicated that the selective serotonin reuptake inhibitor (SSRI) fluvoxamine might reduce the rate of clinical deterioration in outpatients with symptomatic COVID-19 16 . Fluvoxamine, an antidepressant, is primarily used for the treatment of obsessive-compulsive disorder, but it is also used for depression and anxiety disorders, such as panic disorder, social anxiety disorder and post-traumatic stress disorder 17 . Fluvoxamine has been considered for use in the early symptom treatment of COVID-19. Although there are no recommendations for the treatment of COVID-19 with fluvoxamine, a Phase I clinical trial is currently being conducted on outpatients to assess if serious complications, such as shortness of breath, can be prevented 18 . It was previously shown that fluvoxamine reduced the inflammatory response "cytokine storm" during sepsis 19 . The second phase of COVID-19 can involve a serious inflammatory reaction 20 , which may be prevented if fluvoxamine is used for treatment. However, SSRIs including fluvoxamine increase the risk of developing gastrointestinal side effects when a non-efficient cytochrome P450 (CYP2D6) enzyme variant is present 21 . When administered in similar doses, CYP2D6 poor metabolizers have significantly greater exposure to fluvoxamine when compared to NMs 21, 22 . This increase in drug exposure may be a risk factor for drug-induced side effects or toxicity. The US Food and Drug Administration (FDA) states that fluvoxamine should be used cautiously in patients known to have reduced levels of CYP2D6 activity 23 . The Clinical Pharmacogenetics Implementation Consortium (CPIC) guideline recommends a 25-50% dose reduction for the starting dose of fluvoxamine, or the use of an alternative drug that is not metabolized by enzyme CYP2D6 for CYP2D6 PMs (*3/*4, *4/*4, *5/*5, and All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2021. ; https://doi.org/10.1101/2021.03.07.21253095 doi: medRxiv preprint 5 *5/*6) 24 . Therefore, host genetics and demography associated with COVID-19 are crucial aspects of infection and prognosis. Hence, integral medication dosing might need adjustments based on a patient's genetic information. The CPIC has published a pharmacogenetic guideline on SSRI and atazanavir, with specific therapeutic recommendations based on CYP2D6 and UGT1A1 phenotypes, respectively 15,24 . The phenotype was derived from an activity score, obtained by the sum of two individuals' allele scores. However, recent meta-analysis of concentrations using the Pharmacogenomics Knowledge Base (PharmGKB), showed strong correlations between the UGT1A1*80 allele with higher plasma levels of atazanavir, and the CYP2D6*3, *4, *5, *6 and *10 alleles with higher plasma levels of fluvoxamine 25 . Based on these results, PharmGKB assigned the highest level of evidence (level 1A) to these associations, indicating that these biomarkers were extensively studied for years and are already validated in clinical practice. The goal of this study was to demonstrate that current pharmacogenomics databases can be leveraged to enhance the identification of gene alleles, and to determine population differences in associated drug response/toxicity events. Results from this work have wide ranging impacts on the targeted treatment of COVID-19 patients across broad geographic ranges and ethnic backgrounds, thereby facilitating the drug development processes and guide towards safer therapeutic prescriptions. Sixty-two variants across 101 Jordanian individuals of Arab descent associated with reduced UGT1A1 and CYP2D6 enzymes function were selected (Table 1 ). Allele and genotype frequencies of the selected variants are listed in Table 2 . SNPs were tested for Hardy-Weinberg Equilibrium (HWE), and all the studied polymorphisms did not deviate from the All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2021. ; https://doi.org/10.1101/2021.03.07.21253095 doi: medRxiv preprint 6 HWE (p-value > 0.05). For CYP2D6, the defective allele *2 was the most abundant variant (0.243), followed by allele *4 (0.163), while the defective allele UGT1A1*80 was the most abundant variant (0.381), followed by allele *28 (0.371). In addition, three less common alleles, CYP2D6 *10, *17 and *41 were also detected. UGT1A1 and CYP2D6 common alleles together accounted for 0.752 and 0.619, respectively. Genetic structure analysis across populations validated our findings that Jordanian Arabs share the closest variants sequence homology to Near East and European populations ( Asian populations (Fig. 1b) . In addition, a defined cluster of European and Near Eastern populations were found, which was further validated using pairwise Fst analyses (Table S1 ). The lowest level of differentiation was observed between the Jordanian Arab population and a North American population which was self-defined as 'Caucasian' and living in the United States (Fst = 1.48x10 -5 ), followed by Portuguese (Fst = 2.88x10 -5 ) and Iberian from Portugese All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2021. ; https://doi.org/10.1101/2021.03.07.21253095 doi: medRxiv preprint 7 and Spanish populations (Fst = 1.34x10 -4 ). The greatest affinity with Near Eastern was observed with the Ashkenazi Jews in Argentina (Fst = 6.75x10 -4 ). Across the nine biogeographical groups, about 50% of subjects were of Europeans origin, followed by East Asian (17%), African Americans/Afro-Caribbean's (10%), Latinos (8%), Sub-Saharan Africans (4%), South Central Asians (4%), Americans (4%), Near Easterns (3%), and Oceanians (1%; Table 3 ). Distinct differences were found among these populations, with direct impacts on atazanavir and fluvoxamine clinical outcomes (Fig. 2 ). The UGT1A1*80 allele frequency was significantly higher in the Sub-Saharan African origin Table S2 ). These significant variant alleles and genotypes were classified as PharmGKB Level 1A evidence with reduced enzyme function, and are therefore associated with recommended changes to atazanavir dosing 15 . Inferring from the specific dosing guidelines for individual with diplotype pairs for reduced fluvoxamine metabolism 24 , the results here indicates that the East Asian populations are about 8x more likely to show impaired CYP2D6 metabolism than African, European, Near Eastern, American and Latino populations, and 4.4x more likely than South Central Asian and Oceanian populations (Table S2) . Interestingly, a large number of generally less common alleles were also identified. Allele CYP2D6*2 was significantly over-represented in South Central Asians (0.295) and Europeans (0.277), while CYP2D6*35 was significantly over-represented in Europeans (0.055, Table S2 ). Alleles CYP2D6*17, *29, and *45 were significantly over-represented in All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Of the 101 individuals of Jordanian Arab descent, ~15% were poor metabolizers of atazanavir based as the first sequence variant analysis of UGT1A1 for a Near Eastern population. However, mapped frequencies of CYP2D6 and UGT1A1 genotypes based on both the PharmGKB meta-analysis 25 , and the CPIC updated reports 15,24 , showed that East Asian populations are approximately 6.7x more likely to show impaired fluvoxamine metabolism than other groups and that African populations are approximately 7x more likely to show impaired atazanavir metabolism than East Asian populations (Fig. 3, Table S2 ). Therefore, a higher proportion of European, South Central Asian, Near Eastern, American, Latino, Oceanian ancestry populations have normal fluvoxamine or atazanavir metabolism, and therefore are less susceptible to complications for these two drugs. Although our simulation is based upon phenotype allele frequencies instead of directly assaying patient biospecimens, these allele frequencies were previously well-described, and were already validated in clinical practice 29 Existing CYP2D6 and UGT1A1 genotype results may provide the potential benefit of identifying populations who are at an increased risk of experiencing adverse drug reactions or therapeutic failure. Collectively, This work demonstrates the capability and application of All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. large-scale pharmacogenomics studies to elucidate genetic variation effects on treatment efficacy in COVID-19 patients. Ultimately, the implementation of pharmacogenetics in clinical settings is increasing as it leads to more efficient and cost-effective treatments. In times of a global healthcare crisis, these therapeutic improvements become crucial. This study retrospectively included 101 unrelated Jordanian participants, of which 56 were male and 45 were female. After a signed informed consent, a 3 ml venous blood sample was Genomic DNA was extracted from each blood sample using the QIAamp DNA Micro Kit (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. SNPs with a call rate of less than 99% were excluded from subsequent analyses. Statistical and genetic analyses were performed for selection and validation using Microsoft Excel and SPSS v16 (Fig. 4) . The genotype and allele frequencies were calculated and tested for deviations from Hardy-Weinberg equilibrium using the chi-square (χ 2 ) test (p > 0.05). Fisher's exact test was used to correct for small population size. These were implemented with the HWChisq and HWextact functions in HardyWeinberg R package 34 . A Holm's sequential correction for multiple comparisons was applied 35 . Population structure was examined using Wright's Fixation Index (Fst) for pairwise distances between populations based on the UGT1A1 (12 SNPs) and CYP2D6 (18 SNPs) (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (Table S4) 36 . Fixation indices were estimated with the calcFst function in the polysat R package 37 . The frequencies of the 27 actionable pharmacogenetics biomarkers were assessed cumulatively for the Jordanian Arabs against nine biogeographical groups, consisting of 130,856 individuals from 417 global populations 15,24 . These nine groups were defined by global autosomal genetic structure and based on data from large-scale sequencing initiatives and are used to illustrate the broad diversity of global allele frequencies in this study. Furthermore, this biogeographic grouping system meets a key need All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2021. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The total frequency of six SNPs with a Level 1A for the Jordanian Arab population within nine geographically-defined groups were mapped for global impact visualization of allele frequency on atazanavir and fluvoxamine response (Table S2 , Fig. 4 ). Inferred frequencies for UGT1A1*1 and CYP2D6*1 were excluded from our biogeographical analyses as no population studies have tested for all known variant alleles, and *1 was not genotyped directly in many studies 15,24 . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Consent for publication of the data obtained in this study was retrospectively approved, following de-identifying information of individuals. The datasets generated during the current study are included in the supplementary files. Publicly available repositories used in the analyses were previously published under PMID: . The author declares that there is no conflict of interest. No funding declared. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted March 8, 2021. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Pharmacogenetics analyses with nine geographically-defined groups compiled with CPIC guidelines (>100,000 subjects, 417 reports) estimating genotype-predicted phenotype status across world populations. WHO. Coronavirus disease (COVID-19) Situation Report -124 The current understanding and potential therapeutic options to combat COVID-19 Pharmacologic Treatments for Coronavirus Disease 2019 (COVID-19): A Review Off-label use of medicines for COVID-19 Precision Dosing: The Clinical Pharmacology of Goldilocks The importance of race and ethnic background in biomedical research and clinical practice Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for UGT1A1 and Atazanavir Prescribing Fluvoxamine vs Placebo and Clinical Deterioration in Outpatients With Symptomatic COVID-19: A Randomized Clinical Trial An updated review of its use in the management of adults with anxiety disorders Placebo-controlled Clinical Trial of Fluvoxamine for Symptomatic Individuals With COVID-19 Infection (STOP COVID) Modulation of the sigma-1 receptor-IRE1 pathway is beneficial in preclinical models of inflammation and sepsis The trinity of COVID-19: immunity, inflammation and intervention A comparison of the risk of QT prolongation among SSRIs Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for CYP2D6 and CYP2C19 Genotypes and Dosing of Selective Serotonin Reuptake Inhibitors Pharmacogenomics knowledge for personalized medicine Remdesivir for the Treatment of Covid-19 -Final Report Human Y-chromosome variation in the western Mediterranean area: implications for the peopling of the region Isolates in a corridor of migrations: a high-resolution analysis of Ychromosome variation in Jordan Clinical Pharmacogenetics Implementation Consortium guidelines for CYP2C19 genotype and clopidogrel therapy: 2013 update Complexity of genome sequencing and reporting: Next generation sequencing (NGS) technologies and implementation of precision medicine in real life Molecular inversion probes: a novel microarray technology and its application in cancer research A genotype calling algorithm for affymetrix SNP arrays Clinical pharmacology and pharmacogenetics in a genomics era: the DMET platform Exploring Diallelic Genetic Markers: The HardyWeinberg Package A Simple Sequentially Rejective Multiple Test Procedure The technical staff from Princess Haya Biotechnology Center at the Jordan University of Science and Technology provided their assistance with this study. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted March 8, 2021. ; https://doi.org/10.1101/2021.03.07.21253095 doi: medRxiv preprint All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted March 8, 2021. ; https://doi.org/10.1101/2021.03.07.21253095 doi: medRxiv preprint 2 9 6 All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted March 8, 2021. ; https://doi.org/10.1101/2021.03.07.21253095 doi: medRxiv preprint All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.