Letter: Optimising Targeted Next‐Generation Sequencing for Intestinal Tuberculosis Diagnosis—Methodological Considerations

Ye et al. [1] evaluated targeted next-generation sequencing (tNGS) for the rapid differentiation of intestinal tuberculosis (ITB) from Crohn's disease with latent tuberculosis infection. The authors provided compelling evidence that tNGS, with 83% sensitivity and 100% specificity, outperforms conventional methods like acid-fast bacillus staining and tuberculosis polymerase chain reaction in a tuberculosis (TB)-endemic region [1]. This addresses a critical diagnostic challenge where misdiagnosis can delay treatment [2]. The study's reliance on positive interferon-gamma release assay (IGRA) results as an inclusion criterion may have introduced selection bias. Ye et al. acknowledged IGRA's limited specificity and factors like immunosuppression that can lead to false negatives [1], but did not discuss the exclusion of patients with IGRA-negative ITB. Such cases, potentially with lower Mycobacterium tuberculosis burdens, may be less detectable by tNGS, possibly overestimating its sensitivity [2]. This could limit applicability to patients with atypical immune responses, such as those with comorbidities [2]. Future studies should include IGRA-negative patients with suspected ITB to evaluate tNGS performance across a broader clinical spectrum, ensuring a more representative cohort. Ye et al. reported that 17% of patients with ITB had negative tNGS results, necessitating empirical anti-tuberculosis therapy [1]. However, the study did not explore reasons for these false negatives, such as low bacterial burden, biopsy sampling variability or technical limitations (e.g., sequencing depth or primer specificity) [3]. Understanding these factors is crucial for clinicians interpreting negative tNGS results in high-stakes diagnostic scenarios, where false negatives could delay treatment [3]. Future research should investigate biological and technical contributors to false negatives, potentially through analysis of biopsy quality or strain characteristics, to enhance tNGS's reliability as a frontline diagnostic tool. Finally, the single-centre design in a high TB-endemic region limits generalisability [1]. The performance of tNGS may differ in low TB-prevalence settings, where its positive predictive value could decrease due to lower disease prevalence, or in regions with diverse strains or healthcare systems [4]. Ye et al. did not address how these contextual factors might affect tNGS applicability [1]. Multi-centre studies encompassing diverse TB prevalence settings and patient demographics would strengthen the evidence for tNGS and support its global adoption [4]. Haleema Rahman: conceptualization, writing – original draft, writing – review and editing, methodology. The author has nothing to report. The author declares no conflicts of interest. This article is linked to Ye et al. paper. To view this article, visit https://doi.org/10.1111/apt.18522. The data that support the findings of this study are openly available in Lingna Ye, Yushu Cao, Yujuan Fu, Chuwen Tian, Qian Cao at https://doi.org/10.1111/apt.18522.