原住民族潛伏結核感染短程3HP治療完成率與副作用風險之分析

Abstract

研究背景： 潛伏結核感染（latent tuberculosis infection, LTBI）治療是世界衛生組織終結結核病策略的重要措施之一。每週一次異煙肼（isoniazid）合併肺挺（rifapentine）共12週之短程療程（3HP）因療程短、治療完成率高，已成為國內外建議之LTBI治療方案。然而，藥物不良反應（adverse drug reaction, ADR）仍可能影響治療完成率，且臺灣原住民族接受3HP治療之完成率、ADR及其相關影響因子仍缺乏本土實證。此外，N-acetyltransferase 2（NAT2）基因多型性可能影響異煙肼代謝能力，但其與3HP治療結果之關聯仍有待進一步探討。 研究方法： 本研究採回溯性世代研究設計，收集2020至2024年桃園市復興區、楊梅區及花蓮縣秀林鄉三家衛生所接受3HP治療之LTBI個案，共納入126名受試者，比較原住民族與非原住民族的治療完成率、ADR發生情形及治療中斷相關因子。統計分析採用Student's t檢定、Pearson's chi-square test、Fisher's exact test、單變項及多變項邏輯斯迴歸分析，並進行探索性中介分析，評估ADR在族群別與治療中斷間之可能角色。此外，於桃園市復興區前瞻性招募受試者進行NAT2基因分型，完成品質控制後共納入37位受試者進行探索性藥物基因體學分析。 研究結果與討論： 126名受試者中，原住民族103人（81.7%），非原住民族23人（18.3%）。原住民族治療中斷率高於非原住民族（35.9% vs. 13.0%, p = 0.046），而ADR發生率雖較高，但族群差異未達統計顯著。多變項邏輯斯迴歸分析顯示，調整年齡、性別、體重及ADR後，原住民族仍為治療中斷之獨立相關因子（aOR = 5.13, 95% CI: 1.25–21.08），ADR亦為治療中斷的重要相關因子（aOR = 7.36, 95% CI: 2.67–20.28）。探索性中介分析顯示，ADR的中介比例為5.7%，未達統計顯著，顯示ADR並非族群別與治療中斷間的主要中介機制，原住民族較高的治療中斷風險可能仍受到健康識能、文化因素、醫療可近性及其他社會決定因子等多重因素影響。藥物基因體學分析中，慢速NAT2乙醯化代謝型雖呈現較高ADR與治療中斷趨勢，但與ADR、SDR及治療中斷均未達統計顯著。 結論： 本研究顯示，原住民族接受3HP治療具有較高之治療中斷風險，而ADR為治療中斷的重要相關因子，但未支持ADR為族群別與治療中斷間之主要中介機制。未來除持續強化ADR的監測與處置外，亦應發展符合原住民族文化脈絡及健康需求之照護策略，以提升LTBI治療完成率。本研究亦建立臺灣原住民族NAT2藥物基因體學之初步資料，可作為未來擴大樣本、多中心研究及精準公共衛生應用之基礎。

Background Treatment of latent tuberculosis infection (LTBI) is a key strategy of the World Health Organization (WHO) End TB Strategy. The 3-month once-weekly regimen of isoniazid plus rifapentine (3HP) has been widely recommended because of its shorter treatment duration and higher completion rate. However, adverse drug reactions (ADRs) remain a major challenge to treatment completion. Evidence regarding treatment outcomes, ADRs, and associated factors among Indigenous peoples in Taiwan remains limited. In addition, although polymorphisms of the N-acetyltransferase 2 (NAT2) gene may influence isoniazid metabolism, their association with 3HP treatment outcomes has not been well established. Methods This retrospective cohort study included individuals with LTBI who received 3HP treatment between 2020 and 2024 at three public health centers in Taiwan (Fuxing District, Yangmei District, and Xiulin Township). Treatment completion, ADRs, and factors associated with treatment interruption were compared between Indigenous and non-Indigenous participants. Continuous and categorical variables were analyzed using Student's t test, Pearson's chi-square test, or Fisher's exact test, as appropriate. Univariable and multivariable logistic regression models were performed to identify factors associated with treatment interruption. An exploratory mediation analysis was conducted to evaluate the potential mediating role of ADRs between ethnicity and treatment interruption. In addition, a prospective pharmacogenomic subgroup study was conducted in Fuxing District. After quality control, 37 participants underwent NAT2 genotyping and exploratory pharmacogenomic analyses. Results A total of 126 participants were included, comprising 103 Indigenous and 23 non-Indigenous individuals. Indigenous participants had a significantly higher treatment interruption rate than non-Indigenous participants (35.9% vs. 13.0%, p = 0.046). Although ADRs occurred more frequently among Indigenous participants, the difference was not statistically significant. In the multivariable logistic regression analysis, Indigenous ethnicity remained independently associated with treatment interruption (adjusted odds ratio [aOR] = 5.13, 95% confidence interval [CI]: 1.25–21.08), and ADR was also an independent predictor of treatment interruption (aOR = 7.36, 95% CI: 2.67–20.28). Exploratory mediation analysis demonstrated that the proportion mediated by ADR was 5.7% and was not statistically significant, suggesting that ADR did not substantially mediate the association between ethnicity and treatment interruption. The pharmacogenomic analysis showed no statistically significant associations between the predicted NAT2 acetylator phenotype and ADRs, systemic drug reactions (SDRs), or treatment interruption. Conclusions Indigenous ethnicity was independently associated with a higher risk of treatment interruption during 3HP therapy, and ADR was an important predictor of treatment interruption. However, ADR did not appear to be the primary mechanism linking ethnicity to treatment interruption. These findings suggest that additional factors, such as healthcare accessibility, health literacy, cultural context, and other social determinants of health, may contribute to treatment interruption among Indigenous populations. Furthermore, this study provides preliminary pharmacogenomic data on NAT2 acetylator phenotypes in Taiwanese Indigenous populations, which may serve as a foundation for future multicenter studies and precision public health approaches.