Letter: TB-CD Puzzle—Is tNGS the Final Piece?

Abstract

We read with keen interest the recent article by Ye et al. [1] entitled “Crohn's Disease with Latent Tuberculosis Infection or Intestinal Tuberculosis (ITB): Rapid Discrimination by Targeted Next—Generation Sequencing”. We feel that this work represents important progress in discriminating ITB and CD. Not only do the authors demonstrate targeted next generation sequencing (tNGS) on intestinal tissue as a useful strategy to discriminate ITB from CD with latent tuberculosis, but they also demonstrate the validity of use of early mucosal response as early discriminator when the initial evaluation is unhelpful. This study further validates the approach of assessing early mucosal response at 2 months after initiation of anti-tubercular therapy (ATT) trial for patients having diagnostic dilemma between ITB & CD [2]. This approach of utilising a short duration ATT trial, reduces the impact of unnecessary ATT exposure on natural course of CD [3]. There are two tools which appear particularly impressive in discriminating ITB and CD at present, and apart from tNGS, the use of multiplex PCR appears to be a sensitive approach to diagnose ITB [4]. Future studies should assess the comparative performance of these two approaches in solving the TB-CD puzzle.

We would also like to point out some issues of concern in respect to the current study—the authors, while evaluating the tNGS as a novel tool for ITB diagnosis, used a positive result in tNGS as a component of defining the gold standard for ITB diagnosis. Obviously, this is methodologically fallacious, as a new test should be tested against a gold standard (a composite reference standard in this case). Furthermore, we are intrigued by the low sensitivity of TB-PCR [22% (95% CI 12%–36%)] in comparison to previous studies where IS6110 primer has a pooled sensitivity of 47% [5]. The authors have not specified which primers were used for TB-PCR in their study. It also raises a question of whether formalin fixed, paraffin embedded (FFPE) biopsy samples were used for PCR study, while fresh biopsy specimens have been used for tNGS. Could this be the reason behind the greater sensitivity of tNGS in detecting tuberculosis? Since there is time-dependent physical degradation of DNA in FFPE tissue, it could affect the success rate of the amplification. Further, ITB patients in this study were treated with ATT for 1-year duration, as per local guidelines, which is questionable since a Cochrane review [6] has found no evidence to suggest that six-month treatment regimens are inadequate for treating people that have ITB.

Abhishek Yadav: writing – original draft, writing – review and editing. Arpita Agrawal: formal analysis. Uday Kiran Mangipudi: conceptualization.

The authors declare 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.