Research Article: First-time detection and identification of the Mycobacterium tuberculosis Complex members in extrapulmonary tuberculosis clinical samples in south Tunisia by a single tube tetraplex real-time PCR assay

Date Published: May 5, 2017

Publisher: Public Library of Science

Author(s): Mariam Siala, Salma Smaoui, Wafa Taktak, Salma Hachicha, Asma Ghorbel, Chema Marouane, Sana Kammoun, Dhikrayet Gamara, Leila Slim, Radhouane Gdoura, Férièle Messadi-Akrout, Gerd Pluschke. http://doi.org/10.1371/journal.pntd.0005572

Abstract: IntroductionTunisia has one of the highest burdens of extrapulmonary tuberculosis (EPTB) among tuberculosis (TB) cases but the contribution of MTBC-mediated human EPTB is unknown. EPTB diagnosis is challenging due to the paucibacillary nature of clinical samples. Therefore, a need of a simplified molecular method for sensitive and specific TB detection and differentiation of MTBC members caused EPTB remains a priority to an early diagnosis, optimize successful anti-TB treatment and minimize transmission. We evaluated the performance of a single tube tetraplex Taq Man real time PCR for EPTB detection and differentiation between MTBC members directly on extrapulmonary samples.Materials and methodsExtrapulmonary samples obtained from clinically suspected EPTB patients from 2013 to April 2015 were tested by Ziehl Neelsen Staining, mycobacterial culture and qPCR assay for RD1, RD9, RD12 and ext-RD9 targets (MTBC-RD qPCR). The performance of qPCR was compared to a reference standard based on MTBC culture and/or at least two criteria of a composite reference standard (CRS) including clinical, radiological, histopathological and therapeutic findings.ResultsEPTB was identified in 157/170 (92.4%) of included patients of whom 99 (63%) were confirmed by culture and 58 (36.9%) by CRS criteria. The sensitivity and specificity of qPCR, in comparison to the reference standard were 100% (157/157) and 92.3% (12/13), respectively. The sensitivity of qPCR was statistically significant as compared to culture and smear microscopy (P< 0.001). QPCR results showed M. bovis identification in 77.1% of extrapulmonary samples in occurrence to lymphadenitis infection. M. tuberculosis and M.bovis BCG were detected in 21.6% and 1.3% of cases, respectively.ConclusionsMTBC–RD qPCR proved to be a rapid and sensitive assay for simultaneously TB detection and MTBC members identification on extrapulmonary samples within 1.5 days after sample receipt. Its high sensitivity could make this method a useful tool in diagnosing TB in addition to routine conventional methods and TB clinical parameters.

Partial Text: Tuberculosis (TB) remains a leading cause of morbidity and mortality worldwide [1]. According to annual surveys conducted by the World Health Organization (WHO), 10.4 million new active TB cases and 1.8 million deaths occurred in 2015 [1]. Although pulmonary TB is the most common presentation of this disease, it can involve any organ in the body [2]. Extrapulmonary Tuberculosis (EPTB) is defined as the isolated occurrence of TB in any part of the body other than lungs [2]. The prevalence of EPTB is highly variable which is essentially attributable to the geographic origin of the patient. A high incidence is observed among the immunocompromised HIV co-infected patients [3]. In Tunisia, EPTB makes up to 57% of TB cases despite the low prevalence of HIV which is higher compared to other countries [4]. EPTB can be caused by Mycobacterium tuberculosis complex (MTBC), such as Mycobacterium tuberculosis, Mycobacterium bovis, Mycobacterium bovis BCG and Mycobacterium africanum [5]. An early and rapid TB diagnosis as well as distinction between the different MTBC members are essential to determine the EPTB etiology and to optimize efficient anti-TB treatment since Mycobacterium bovis and Mycobacterium bovis BCG are intrinsically resistant to pyrazinamid (PZA), an important first-line anti-TB drug [6, 7]. Indeed, the natural mode of infection and surveillance measures for EPTB differ between complex members. For example, early diagnosis of M. bovis might prompt questions to ascertain the risk factors of zoonotic exposure or a contamination of derived food/dairy products from diseased cattle, the primary routes of EPTB infection [2, 5].

In this study, a single tube tetraplex MTBC-RD qPCR assay for the simultaneous detection and identification of MTBC species directly on extrapulmonary specimens was evaluated and was compared to the conventional methods. Here, we have chosen a qPCR test based on the amplification of specific mycobacterial RD motifs. Their presence or absence indicates a specific molecular profile that could differentiate between different MTBC species [6, 10, 17]. Though the identification of MTBC members based on the detection of RD patterns by PCR has been suggested previously [10], the majority of the published data used syber green detection and melting curve or conventional PCR or focused solely on positive culture materials [10, 18, 19, 20, 21]. There is only one study which evaluated the MTBC-RD qPCR directly on clinical specimens [6]. However, this latter work relied essentially on TB specimens initially positive for MTBC by IS6110 qPCR of which 5.7% were extrapulmonary samples compared to 94.3% from pulmonary TB which can be diagnosed more easily than EPTB [6]. Thus, our current study is the first to evaluate the MTBC-RD qPCR for the presence of MTBC DNA directly on 170 clinical non-respiratory specimens from patients with suspected EPTB. Thus, we could demonstrate that MTBC RD qPCR detected MTBC DNA in 100% (99/99) of samples from patients that were microbiologically confirmed EPTB using MTBC culture. MTBC RD qPCR was positive also in 58/58 (100%) of specimens from patients with EPTB diagnosis based on the CRS criteria. Upon comparison with the reference standard, the sensitivity of the MTBC-RD qPCR was 100% (157/157) which is statistically significant compared to culture (99/157) (p = <0.001) and smear microscopy (29/157) (p = <0.001). Twelve out of 13 specimens from patients who had a reference standard negative for EPTB (true negative patients) were negative by MTBC-RD qPCR indicating a specificity of 92.3% of this molecular assay. When comparing qPCR accuracy to culture known as the basic gold standard, the specificity was much lower (16.9%, 12/71) due to the lower sensitivity of the culture. Thus, our findings emphasize that the MTBC- culture most likely underestimates the mycobacterial detection i.e the diagnosis of EPTB. This could be essentially due to the paucibacillary nature of the extrapulmonary specimens and especially those from childhood TB or to the presence of other microorganisms in the same culture that have overgrown MTBC [22, 23, 24, 25, 26]. As reported previously, EPTB diagnosis requires an elaborated diagnostic algorithm based on the use of molecular methods such qPCR (e.g. GeneXpert) which is critically dependent on the CRS based on clinical diagnosis TB parameters [25, 26]. Accordingly, our data also showed and extend previous studies that the use of qPCR on non respiratory materials could be the method of choice for a rapid, specific and sensitive EPTB detection. Indeed, the current study raises the issue of the reference standard based on culture and/or CRS criteria to be used in the comparative evaluation of our tetra-plex qPCR test rather than considering only culture positivity or conventional PCR as a basic standard for EPTB detection. Therefore, its high sensitivity, reliability and ease of use could make this method a useful tool in diagnosing TB in addition to routine conventional diagnostic tests and TB clinical parameters. The high sample size of extra-pulmonary lymphatic specimens used in our study could be an additional advantage to evaluate the post decision of the clinical utility of this assay in EPTB clinical settings. The MTBC-RD qPCR seems to be useful regardless of the specimen type especially in lymph node biopsies and aspirates which constituted 84.7% of all samples. However, the low sample size of analyzed specimens from cases without lymphadenitis infection is the weak point of this work. Source: http://doi.org/10.1371/journal.pntd.0005572