台灣抗藥結核病流行病學及基因變異之臨床應用

Clinical applications of epidemiology and genomic alterations for drug-resistant tuberculosis in Taiwan

簡榮彥(Jung-Yien Chien)

指導教授 ： 余忠仁

共同指導教授 ： 薛博仁

國立臺灣大學/醫學院/臨床醫學研究所/博士(2016年)

https://doi.org/10.6342/NTU201600127

摘要

結核病(tuberculosis, TB)，是台灣地區人數最多的法定傳染性疾病，也是威脅全球的重大公衛問題。其中，抗藥性結核病的治療，包括單一抗藥結核病(mono resistant tuberculosis)、多重抗藥性結核病(multidrug-resistant tuberculosis, MDR-TB，指同時對isoniazid 和rifampicin抗藥之結核病)、廣泛抗藥性結核病(extensively drug resistant tuberculosis, XDR-TB，指對任一種fluoroquinolone，及任何一種注射型第二線抗結核藥物具抗藥性的MDR-TB) 及 超級抗藥性結核病(extremely drug resistant tuberculosis, XXDR-TB，亦稱為 totally drug resistant tuberculosis全抗藥型結核病)的防治更是棘手的問題。
了解抗藥性結核病的發生機轉、危險因子及有效之防治策略是一個十分重要的議題。結核菌的複雜抗藥機轉，病患不當的服用藥物，醫療照護機構不適當之處方及療程，導致藥物不足篩選出抗藥性結核菌株，加上實驗室診斷抗藥性結核病的延遲，抗藥性結核病的治療效果不彰，增加抗藥性病菌群聚傳播的機會，可能是抗藥性結核病不易控制的原因。
針對改善病人服藥順從性，世界衛生組織大力推薦的直接觀察治療法(directly observed therapy strategy, DOTS)是全世界防治結核的重要策略。然而，許多研究仍未證實DOTS策略具有顯著的益處。台灣疾病管制署則於2006年4月開始推動全國性DOTS計畫，更於2007年5月成立「MDR-TB醫療照護體系」，建立五個MDR-TB團隊，針對MDR-TB患者，進行分區集中治療照護並實施進階DOTS策略(DOTS-Plus)，期待有效控制台灣的MDR-TB疫情。
Isoniazid單一抗藥結核病是台灣地區人數最多的抗藥性結核病，但治療策略尚無共識。MDR-TB的治療預後不佳，台灣推動集中治療及DOTS-Plus策略的效果如何，仍待評估。Fluoroquinolones是治療MDR-TB最重要的藥物之一，但我們對於其抗藥基因突變及新一代fluoroquinolones,如moxifloxacin，於治療上之角色所知有限。因此，我們將分三個部份進行研究設計與分析：
在研究的第一個部份，我們將針對isoniazid單一抗藥結核病患，整合台灣北中南東多中心，進行治療策略分析。於2004 to 2011年間，共收集395名於臺大醫院、臺大雲林分院、衛生福利部胸腔病院、慈濟醫院接受治療的isoniazid單一抗藥結核病患者進行研究。
在研究的第二個部份，針對MDR-TB進行研究，分為兩個部分進行：(1)針對台灣地區MDR-TB、XDR-TB及XXDR-TB的相關危險因子及盛行率進行流行病學及防治成效研究。(2) 評估使用薄膜微陣列技術(BluePoint MycoID plus kit)，搭配液態培養基系統(BACTEC Mycobacterium Growth Indicator Tube, MGIT)，快速鑑定包括MDR-TB在內的抗藥性結核病的準確性。
在研究的第三個部份，針對fluoroquinolones抗藥結核病，將分為四個部分進行：(1)分析南臺灣fluoroquinolones抗藥MDR-TB的流行病學及防治成效。(2)分析北台灣及南台灣，fluoroquinolones抗藥及二線藥抗藥結核病之流行病學及防治成效。(3)探討基因體變異對ofloxacin/moxifloxacin抗藥程度的影響。(4)最後一部分，將探討新一代moxifloxacin抗藥程度及fluoroquinolones抗藥基因變異，對使用moxifloxacin治療ofloxacin抗藥MDR-TB之影響
第一個部份的研究發現：於isoniazid單一抗藥結核病，全程使用isoniazid和未使用isoniazid的治療處方，有相似的治療成功率(分別為83.1% 及 83.0%, P=1.000)。然而，中斷使用rifampin的患者較全程使用rifampin的患者有較高的不良預後發生率(分別為37.0% 及 14.3%, P<0.001)。若於中斷使用rifampin的患者使用新一代 fluoroquinolone，則可以顯著降低不良預後發生率(分別為60.0% 及 12.5%, P=0.003)。多變數存活分析也證實，中斷使用rifampin(hazard ratio=1.91)是不良預後的獨立危險因子。
第二個部份的研究發現：
(1)從2004年到2011年，在全面實施DOTS及DOTS-Plus 治療計畫後，南臺灣地區isoniazid、rifampicin抗藥及MDR-TB/XDR-TB發生率顯著的減少。我們發現，獲得性MDR-TB的比率下降，是造成MDR-TB的比率持續下降的主因，然而，原發性MDR-TB的比率並未顯著下降。
(2)藉由收集950個MGIT陽性臨床檢體，我們發現，相較於傳統使用subculture的方法鑑定MGIT陽性培養管，採用薄膜微陣列技術的BluePoint MycoID plus kit有較佳的敏感度，能較準確的鑑定出非結核分枝桿菌(92.9% v.s. 96.3%, respectively; P=0.001)。同時，BluePoint MycoID plus kit鑑定出rifampicin及isoniazid抗藥MTB的敏感度分別可達100%及82.6%。然而，因為死菌的干擾，對於結核菌的鑑定，雖然BluePoint MycoID plus kit較subculture有較高的敏感度，但特異性較差。
在第三個部份的研究發現：
(1)從2007年DOTS-Plus計劃的執行後，原發性(P=0.019)與續發性(P=0.047) fluoroquinolones抗藥MDR-TB的發生率與XDR-TB(P=0.055)的發生率都顯著下降，兩者皆和DOTS-Plus計劃的執行涵蓋率呈現顯著的負相關。
(2)在2004至2011年期間，即使含fluoroquinolones及其他二線藥的處方逐漸增加，fluoroquinolones及其他二線藥抗藥結核病的發生率仍逐年降低。
(3)在55株ofloxacin抗藥，56株ofloxacin敏感之MDR-TB菌株進行gyrA及gyrB基因全區段定序發現：出現gyrA基因 D94G/D94N突變及gyrB基因 G512R突變和ofloxacin 及moxifloxacin 抗藥有強烈關聯。然而，部分fluoroquinolones敏感的East-Asian (Beijing) 及 Indo-Oceanic 基因型菌株，亦帶有這些突變。因此，若在East-Asian (Beijing) 及 Indo-Oceanic 基因型菌株盛行的地區，運用分子檢測技術偵測fluoroquinolones抗藥的特異度會較差，需要特別加以注意。除此之外，我們進一步發現gyrA基因D94G、D94N突變 及 gyrB G512R基因突變和高度moxifloxacin抗藥有相當大的相關，相反的，gyrA基因D94A突變，反而經常導致低度moxifloxacin抗藥。進一步運用全基因體定序及藥物輸出幫浦基因變異分析，我們發現，ofloxacin抗藥及敏感菌株間，efflux pump 基因變異的分布顯著不同，暗示efflux pump相關基因的突變，可能和ofloxacin抗藥相關，但仍需進一步進行確認。
(4) 2006年4月到2013年12月間，我們從衛生福利部胸腔病院，篩選81位罹患ofloxacin抗藥MDR-TB患進行研究。以傳統瓊酯比例法發現的ofloxacin抗藥MDR-TB，69.1%仍然是moxifloxacin敏感或低度抗藥(minimum inhibition concentration, MIC≤2.0 mg/L)。在moxifloxacin敏感及低度抗藥的患者，使用moxifloxacin治療，能顯著縮短培養陰轉所需要的時間，但在MIC>2.0 mg/L的高度抗藥患者，moxifloxacin治療無法提高治療成功率及縮短培養陰轉所需要的時間。
持續使用rifampicin是成功治療isoniazid單一抗藥結核病的關鍵，面對無法忍受rifampicin的患者，應使用替代藥物，如新一代fluoroquinolones，以改善預後，是否使用rifabutin取代rifampicin 能達到更佳的效果，還須加以研究。我們也發現，採用DOTS及DOTS-Plus治療計畫，能有效降低fluoroquinolones抗藥結核病，MDR-TB, XDR-TB及XXDR-TB的發生率。同時，我們發現雖然對二線藥的暴露可能會增加對這些藥物抗藥的風險，然而，藉由持續進行治療處方的監測及檢討，搭配執行DOTS/DOTS-plus計畫，確保適當用藥及治療療程，仍可以有效預防二線藥抗藥的產生和減少XDR-TB的風險。
最後，我們發現特定gyrA/gyrB基因突變和ofloxacin及moxifloxacin抗藥有強烈關聯，可以做為設計快速分子檢定的標的，但此一標的在East-Asian (Beijing) 及 Indo-Oceanic 基因型菌株盛行的地區，診斷的特異度會較差，需要加以注意。我們更發現，以傳統瓊酯比例法發現的ofloxacin抗藥MDR-TB，高達69.1%仍然是moxifloxacin敏感或低度抗藥的患者，給予moxifloxacin治療，仍有助於改善治療預後，因此，建議在進行moxifloxacin抗藥檢驗時，應同時檢驗低濃度(0.5 mg/L)及高濃度(2.0 mg/L)抗藥，於判斷是否給予moxifloxacin治療時，可以使用MIC≤2.0 mg/L當作參考的切點。
總之，我們的系列研究發現，rifampicin在isoniazid單一抗藥結核病的治療上具有舉足輕重的角色。證實DOTS及DOTS-Plus治療計畫確實能降低抗藥性結核病的發生率，建議未來台灣公共衛生政策的擬定上，應持續加以落實。在治療MDR-TB時，建議未來實驗室，能提供新一代fluoroquinolones，如moxifloxacin的最小抑菌濃度及抗藥相關突變的分子檢測，做為臨床醫師處方藥物的重要參考，以達到良好的治療效果，避免進一步的抗藥發生。期待經過大家的努力，未來的台灣，不再是一個抗藥性結核病盛行的國家。

關鍵字

抗藥結核病 ； 多重抗藥性結核病 ； 廣泛抗藥性結核病 ； 超級抗藥性結 ； fluoroquinolone抗藥性結核病 ； 都治計畫

並列摘要

Infection caused by Mycobacterium tuberculosis (MTB) continues to be a major public health burden worldwide. Drug resistance rates are one of the most important aspects in the national tuberculosis (TB) control program, especially multidrug-resistant (MDR), which is defined as TB that is resistant to isoniazid and rifampin, fluoroquinolone resistant MDR-TB and extensively drug-resistant (XDR) TB, an MDR-TB isolate that is resistant to any fluoroquinolone and any injectable SLDs (capreomycin, kanamycin, or amikacin).
Isoniazid (INH) mono-resistance is the most common first-line drug resistance in TB, but its treatment outcome remains unclear. In the first part of study, we identified 395 INH mono-resistant TB from four hospitals in Taiwan during January 2004 to October 2011. The treatment success rate was similar in patients with high-level and low-level INH resistant TB and among those taking anti-TB treatment with and without INH. Patients without rifampicin interruption had lower risk of unfavorable outcome and supplementation with a new-generation fluoroquinolone improved treatment success. The presence of cavitary lesions was significantly associated with a higher relapse rate and extended treatment of 7-9, 10-12, and >12 months had less relapse than six-month treatment. Multivariate Cox proportional hazards analysis revealed that co-morbidity with cancer and rifampicin interruption were independent factors associated with unfavorable outcomes. We found treatment throughout with rifampicin and extended treatment for cavitary disease are crucial for improving outcomes in patients with INH mono-resistant TB.
In the second part of study, we investigated the impact of the Directly-Observed- Therapy-Strategy (DOTS) and DOTS-Plus strategies on resistance profiles among tuberculosis. We performed a retrospective analysis of resistance profiles among isolates of MTB obtained from 2160 consecutive patients with culture-confirmed pulmonary TB during 2005 to 2011 at chest hospital in southern Taiwan. Trend analysis revealed that the rates of acquired MDR-TB were significantly lower after implementation of the DOTS and DOTS-Plus programs (P<0.01). The rates of resistance to rifampicin, isoniazid, ofloxacin, moxifloxacin, levofloxacin, and para-aminosalicylic acid also significantly decreased during the study period. However, the rates of primary MDR-TB remained stable (P =0.11). Multivariate logistic regression analysis showed that age ranging from 45 to 64 years, positive acid-fast stain results at the initiation of treatment, and treatment without DOTS were independent risk factors associated with acquired MDR-TB.
We further conducted a study to investigate the resistance of second-line drugs (SLDs) among 6,035 non-duplicated MTB isolates from 2004 to 2011 at National Taiwan University Hospital and Chest hospital. There was a significant decrease (all P-values <0.01) in the prevalence of isolates that were resistant to fluoroquinolons, injectable SLDs, and orally administered SLDs, and MDR and XDR isolates over time. There was a significant increase in the coverage rate of DOTS/DOTS-Plus programs and that of administering appropriate first-line and second-line regimens (all P <0.01). The decline in prevalence of resistance to SLDs was negatively correlated with the rise in rates of administering appropriate regimens as well as the DOTS/DOTS-Plus programs, but not with the increase in usage of second-line regimens, indicating the implementation of DOTS/DOTS-Plus programs with appropriate regimens was effective at preventing SLD-resistant and XDR-TB.
Fluoroquinolones (FQs) are used as second-line drugs for the treatment of TB and exert their bactericidal effects by inhibiting mycobacterial DNA gyrase activity, which prevents bacterial DNA from unwinding and replicating. Moxifloxacin (MFX), a fourth-generation fluoroquinolone, has been shown to have better activity against MTB than ofloxacin (OFX) and is recommended by the World Health Organization (WHO) for the treatment of MDR-TB. Unfortunately, the widespread use of fluoroquinolone to treat bacterial infections has led to the emergence of fluoroquinolone-resistant MDR-TB and XDR-TB, thereby complicating patient care.
In the third part of study, we evaluated the association between level of resistance to OFX and MFX and mutations of the entire gyrA and gyrB genes in MTB isolates. A total of 111 isolates were categorized into OFX-susceptible (minimum inhibitory concentrations, MIC<=2 mg/L), low- (MIC 4-8 mg/L) and high-level (MIC>=16 mg/L) OFX-resistant isolates, and MFX-susceptible (MIC<=0.5 mg/L), low- (MIC 1-2 mg/L) and high-level (MIC>=4 mg/L) MFX-resistant isolates. Resistance-associated mutations inside the gyrA gene were found in 30.2% of OFX-susceptible, 72.5% and 72.2% of low- and high-level OFX-resistant isolates, and in 28.6% of MFX-susceptible, 58.1% and 83.9% of low- and high-level MFX-resistant isolates. Compared with OFX-susceptible, low- and high-level OFX-resistant isolates had a significantly higher prevalence of mutations at gyrA codons 88–94 and a higher prevalence of the gyrB G512R mutation. Similarly, compared with MFX-susceptible, low- and high-level MFX-resistant isolates had a significantly higher prevalence of mutations at gyrA codons 88–94 as well as gyrB G512R mutation. D94G and D94N mutations in gyrA and the G512R mutation in gyrB were correlated with high-level MFX resistance while the D94A mutation was associated with low-level MFX resistance. However, some OFX- or MFX-susceptible East Asian (Beijing) and Indo-Oceanic strains also harbored those particular mutations, implying that molecular techniques to detect FQ resistance may be less specific in areas with a high prevalence of those strains.
MFX has lower minimum inhibitory concentration (MIC) than older-generation fluoroquinolones, such as OFX or ciprofloxacin, and it has been proposed as potentially effective drug to treat OFX-resistant MDR-TB. However, there are still no clinical study to evaluate how levels of MFX resistance influence the treatment outcomes of using MFX in treatment of OFX-resistant MDR-TB. From April 2006 to December 2013, we investigated 81 patients with OFX-resistant MDR-TB and followed-up for two years after treatment. Treatment with MFX had significantly higher treatment success rate and earlier culture-conversion (P=0.004), especially among MFX-susceptible and low-level MFX resistant, but not among high-level MDR-TB. In conclusion, we found that most (69.1%, 56/81) OFX-resistant MDR-TB determined by conventional susceptibility testing were still susceptible or low-level resistant to MFX and benefited from treatment with MFX. The results suggested MFX could be used with other active antituberculosis agents to treat OFX-resistant MDR-TB based on a breakpoint of 2.0 mg/L and also support the new WHO recommendations for testing MFX susceptibility at 2 critical concentrations in the management of MDR-TB.

並列關鍵字

Drug resistance tuberculosis ； Fluoroquinolone-resistance tuberculosis

參考文獻

Chiang C Y, Hsu C J, Huang R M, Lin T P, Luh K T. Antituberculosis drug resistance among retreatment tuberculosis patients in a referral center in Taipei. J Formos Med Assoc. 2004 Jun;103(6):411-5.

Five-year follow-up of a controlled trial of five 6-month regimens of chemotherapy for pulmonary tuberculosis. Hong Kong Chest Service/British Medical Research Council. Am Rev Respir Dis. 1987 Dec;136(6):1339-42.

Abuali M M, Katariwala R, LaBombardi V J. A comparison of the Sensititre(R) MYCOTB panel and the agar proportion method for the susceptibility testing of Mycobacterium tuberculosis. Eur J Clin Microbiol Infect Dis. 2012 May;31(5):835-9.

Acar J F, Goldstein F W. Trends in bacterial resistance to fluoroquinolones. Clin Infect Dis. 1997 Jan;24 Suppl 1:S67-73.

Achtman M. Evolution, population structure, and phylogeography of genetically monomorphic bacterial pathogens. Annu Rev Microbiol. 2008 62:53-70.