Molecular genetic analyses of antipsychotic pharmacogenes in a South African first episode schizophrenia cohort

Higgins, Faatiemah (2015-12)

Thesis (MScAgric)--Stellenbosch University, 2015.

Thesis

ENGLISH ABSTRACT: Antipsychotic treatment of schizophrenia is often accompanied by distressing adverse effects and a high relapse rate. Although pharmacogenetic research has identified some promising candidate pharmacogenes, these remain poorly characterized in South African (SA) populations. This study investigated whether polymorphisms in several candidate pharmacogenes (COMT, CYP1A2, CYP2D6, DRD2, DRD3, HTR2A and SOD2) were associated with antipsychotic treatment outcome in a SA schizophrenia cohort. The cohort utilized consisted of 103 first episode schizophrenia (FES) patients from SA Coloured, Caucasian and Xhosa populations, clinically well-characterized and treated with a first generation antipsychotic, flupenthixol decanoate, for at least twelve months. Previously generated exome sequencing data of eleven of the FES patients were used to identify known and novel functional variants within the seven pharmacogenes, and the predicted functional effect of novel variants were assessed with SIFT and PolyPhen-2 algorithms. Based on these analyses and mining of the literature, a total of 33 variants were prioritized for genotyping by PCR-RFLP, long-range and allele-specific PCR, and TaqMan® assays. Minor allele frequencies in our cohort were compared to populations from the 1000 Genomes Project or the Human CYP Allele Nomenclature databases to distinguish unique genetic variation in SA populations. A mixed model for repeated measures analysis was used to determine whether any of the polymorphisms identified were associated with antipsychotic treatment response, as measured by a change in PANSS scores over time from the twelve-month longitudinal PANSS scores adjusted at baseline. Inheritance models to estimate the effect size (ES) of significant associations were assessed using 95% confidence intervals (CI). Exome data revealed a total of eighteen functional variants in six of the genes, predominantly present in CYP2D6. Novel variants were identified in COMT (rs373611092) and SOD2 (rs372173830), where rs373611092 was predicted to “damage” the COMT protein function (SIFT score = 0.015). Of the 33 variants, only variants in COMT and DRD2 were significantly associated with changes in PANSS negative scores after twelve months of treatment. Associations with improved treatment outcomes included COMT rs4633 [P = 0.0080 (T allele), ES = -0.17 95% CI (-0.30 to -0.04)], and DRD2 rs1799732 [P = 0.0004 (-C vs CC), ES = -0.19 95% CI (-0.38 to 0.00)]. COMT haplotypes (rs2020917-rs737869-rs6269-rs4633-rs9332377) also showed an improved treatment outcome for negative symptoms. Undesirable treatment outcomes were associated with COMT rs9332377 [P = 0.0006 (T allele), ES = 0.25 95% CI (0.11 to 0.39)], and DRD2 variants rs1799732 [P = 0.0004 (-- vs CC), ES = 0.50 95% CI (0.15 to 0.86)] and rs1079597 [P = 0.0020 (AA vs AG+GG), ES = 0.93 95% CI (0.34 to 1.53)]. This study confirmed the presence of unique variation in a SA FES cohort associated with antipsychotic pharmacogenetic traits, highlighting the value of re-sequencing in SA individuals. Several SNPs in COMT and DRD2 were identified to be involved in treatment response, specifically influencing negative symptoms. These results serve as a platform for future antipsychotic pharmacogenetic studies within the SA context that could aid the optimization of antipsychotic treatment in the uniquely diverse, but understudied populations.

AFRIKAANSE OPSOMMING: Antipsigotikumbehandeling vir skisofrenie gaan dikwels gepaard met ernstige negatiewe uitwerkings en ’n hoë terugvalkoers. Hoewel farmakogenetiese navorsing op ’n paar belowende kandidaatfarmakogene dui, is dit tot dusver nog swak getipeer onder Suid-Afrikaanse populasies. Hierdie studie het ondersoek ingestel na die verwantskap tussen polimorfismes in verskillende kandidaatfarmakogene (COMT, CYP1A2, CYP2D6, DRD2, DRD3, HTR2A en SOD2) en die uitkoms van antipsigotikumbehandeling by ’n kohort Suid-Afrikaanse skisofreniepasiënte. Die studiekohort het uit 103 skisofreniepasiënte met ’n eerste episode (sogenaamde FES-pasiënte) uit die Suid-Afrikaanse bruin, wit en Xhosa-bevolkingsgroepe bestaan. Die pasiënte is almal klinies goed getipeer en het reeds minstens 12 maande se behandeling met ’n eerstegenerasie-antipsigotikum, flupentiksol-dekanoaat, voltooi. Eksoomvolgorededata wat voorheen vir 11 van die FES-pasiënte verkry is, is gebruik om bekende én nuwe funksionele variante in die sewe farmakogene te identifiseer, en die voorspelde funksionele uitwerking van nuwe variante is met behulp van SIFT- en PolyPhen-2-algoritmes beoordeel. Op grond van hierdie ontledings en literatuurontginning is altesaam 33 variante vir genotipering met PKR-RFLP-, langreeks- en allelspesifieke PKR, en TaqMan®-toetse uitgewys. Alleelfrekwensies in die studiekohort is met populasies uit die projek “1000 Genomes” en die databasis van die alleelnomenklatuur vir menslike CYP gene vergelyk om unieke genetiese variasie by Suid-Afrikaanse populasies te bepaal. ’n Gemengde model van herhaaldemetingsontleding is gebruik om vas te stel of enige van die geïdentifiseerde polimorfismes verband hou met die reaksie op antipsigotikumbehandeling, soos wat geblyk het uit ’n verandering in PANSS-tellings oor tyd vergeleke met die longitudinale PANSS-tellings oor 12 maande wat op die basislyn aangepas is. Oorerwingsmodelle om die uitwerkingsgrootte (UG) van beduidende verwantskappe te bepaal, is met behulp van vertrouensintervalle (VI’s) van 95% beoordeel. Eksoomdata het altesaam 18 funksionele variante in ses van die gene, hoofsaaklik in CYP2D6, aan die lig gebring. Nuwe variante is in COMT (rs373611092) en SOD2 (rs372173830) geïdentifiseer, waar rs373611092 na verwagting die COMT-proteïenfunksie sal “beskadig” (SIFT-telling = 0.015). Variante in COMT en DRD2 hou sterk verband met veranderinge in negatiewe PANSS-tellings na 12 maande se behandeling. Verbande met beter behandelingsuitkomste sluit in COMT rs4633 [P = 0.0080 (T-allel), UG = -0.17 95% VI (-0.30 tot -0.04)], en DRD2 rs1799732 [P = 0.0004 (-C vs CC), UG = -0.19 95% VI (-0.38 tot 0.00)]. COMT-haplotipes (rs2020917-rs737869-rs6269-rs4633-rs9332377) het ook ’n verbeterde behandelingsuitkoms vir negatiewe simptome getoon. Ongewenste behandelingsuitkomste is verbind met COMT rs9332377 [P = 0.0006 (T-allel), UG = 0.25 95% VI (0.11 tot 0.39)], en DRD2-variante rs1799732 [P = 0.0004 (-- vs CC), UG = 0.50 95% VI (0.15 tot 0.86)] en rs1079597 [P = 0.0020 (AA vs AG+GG), UG = 0.93 95% VI (0.34 tot 1.53)]. Hierdie studie bevestig dus die unieke variasie in die farmakogenetiese kenmerke van antipsigotika by ’n Suid-Afrikaanse FES-kohort, en beklemtoon die waarde van hernude reeksbepaling by Suid-Afrikaanse individue. Die navorsing toon dat verskeie SNP-verwantskappe vir COMT en DRD2 by behandelingsreaksie betrokke is, wat veral negatiewe simptome beïnvloed. Hierdie resultate dien as ’n platform vir verdere studies oor die farmakogenetika van antipsigotika in Suid-Afrikaanse verband ten einde antipsigotikumbehandeling in eiesoortig diverse dog grootliks onbestudeerde populasies te help optimaliseer.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/99026
This item appears in the following collections: