Abstract
BACKGROUND AND OBJECTIVE: Accurate detection of Mycobacterium tuberculosis (MTB) in formalin-fixed paraffin-embedded (FFPE) tissues remains a significant challenge in routine pathological practice. This study aimed to evaluate the performance of the first National Medical Products Administration (NMPA)-approved quantitative polymerase chain reaction (qPCR) kit specifically optimized for FFPE tissues. We compared its efficacy with acid-fast bacilli (AFB) staining for MTB detection, while simultaneously assessing its utility for drug resistance profiling and non-tuberculous mycobacteria (NTM) identification in pathological diagnosis. METHODS: We analyzed 1,050 FFPE tissue specimens that were histopathologically diagnosed as granulomatous inflammation suggestive of tuberculosis. All specimens underwent parallel testing with both qPCR (using the NMPA-approved kit) and AFB staining. Drug resistance testing was conducted on qPCR positive samples (Ct ≤ 35; n = 143), while Non-tuberculous Mycobacterial (NTM) identification was indicated for AFB positive/qPCR negative cases (n = 16). RESULTS: The cohort, with a median age of 52 years and comprising 43.71% males, included 631 surgical specimens and 419 biopsy specimens, predominantly sourced from lung tissues (37.05%) and lymph nodes (24.67%). qPCR demonstrated a higher positive rate compared to AFB staining (63.43% vs. 26.29%, p < 0.001). Furthermore, qPCR exhibited a higher positive rate in surgical specimens (70.36%) compared to biopsy specimens (52.98%; p < 0.001) and significantly outperformed AFB staining in both lung tissues (70.44% vs. 31.62%) and lymph nodes (64.86% vs. 23.17%). The two methods displayed moderate overall concordance (58.09%; κ = 0.257), with the highest concordance observed in intestinal specimens (83.95%). Resistance was noted in 18.18% of cases, with 7.0% classified as multidrug-resistant tuberculosis (MDR-TB), peaking in fallopian tubes (33.33% drug-resistant TB (DR-TB), 16.67% MDR-TB). Among the 25 discordant cases, 16 successfully underwent NTM identification, revealing 4 (25.0%) cases of NTM infection. CONCLUSION: This study demonstrates that the qPCR kit approved by the NMPA and optimized for FFPE tissue samples exhibits high effectiveness in the pathological diagnosis of tuberculosis when culture methods are not feasible. Its performance significantly surpasses that of AFB staining. Using DNA extracted from a single FFPE tissue section, we established an integrated molecular testing strategy. This approach enables the sensitive detection of MTB through qPCR, demonstrating a higher detection rate compared to AFB staining in FFPE tissues. Furthermore, it facilitates subsequent drug resistance analysis and the differentiation of NTM species from the same DNA sample, thereby providing a practical and efficient solution for comprehensive pathological diagnosis.