Routine clinical laboratory evaluation demonstrates limited utility of Fourier-transform infrared spectroscopy for subspecies identification of the Mycobacterium abscessus complex
Journal of Clinical Microbiology · 11.11.2026 · 64 (11)
DOI: 10.1128/jcm.01087-26 ↗Kurz gefasst
Leserbrief in der Zeitschrift Journal of Clinical Microbiology, erschienen am 11.11.2026. Verfasst von Mitsunori Yoshida und 5 weiteren Autor:innen. Beteiligte Einrichtung: Leprosy Research Center, National Institute of Infectious Diseases, Japan Institute for Health Security, Tokyo, Japan. Die Arbeit nennt 6 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Japan Agency for Medical Research and Development, Japan Society for the Promotion of Science.
Abstract (Originalsprache)
complex T he Mycobacterium abscessus complex (MABC) comprises three closely related subspecies whose accurate identification is clinically important because inducible macrolide resistance is largely determined by the functional status of erm(41) (1).However, reliable subspecies discrimination remains challenging in routine clinical microbiology.Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has improved species-level identification of non-tuberculous mycobac teria but remains unreliable for MABC subspecies discrimination.Although machine learning-based approaches have shown improved performance, they require custom ized analytical pipelines that limit routine implementation (2).Likewise, lipid fingerprinting using the MALDI Biotyper Sirius failed to reliably distinguish MABC subspecies (3).Fourier-transform infrared (FTIR) spectroscopy is a rapid phenotypic typing platform (4), and one previous study reported successful differentiation of M. abscessus subsp.abscessus and subsp.massiliense under optimized culture conditions, particularly on Löwenstein-Jensen medium (5).We evaluated whether FTIR could discriminate MABC subspecies under routine clinical laboratory conditions.A total of 137 clinical and environmental isolates, together with three reference strains (140 isolates total), previously assigned by whole-genome sequencing, were analyzed (Table S1).FTIR spectra were obtained from liquid cultures using the IR Biotyper and analyzed as described in the supplemental methods.Broad spectral regions and targeted spectral windows generated using multiple softwaredefined wavenumber thresholds (35%-85%) were evaluated by principal component analysis (PCA) and supervised machine-learning classification.FTIR consistently failed to discriminate the three MABC subspecies.No broad spectral region enabled separation either individually or in combination.PCA demonstrated substantial overlap among the three subspecies across all spectral profiles (Fig. 1A; Fig. S1).The selected windows consistently mapped to three carbohydrate-associated subregions, but none improved PCA discrimination (Fig. 1B; Fig. S2).Supervised machine learning classification likewise showed limited performance, with a maximum mean cross-validated accuracy of only 75.8%, insufficient for reliable clinical identification (Fig. 1C; Fig. S3).PCA also failed to separate isolates according to colony morphotype or erm(41) T28C status (Fig. S4).These findings demonstrate that FTIR-derived spectral information is insufficient for routine subspecies identification of MABC.Unlike previous proof-of-concept studies performed under optimized culture conditions (5), our study evaluated FTIR using liquid cultures in a workflow reflecting routine clinical laboratory practice.The inability of
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Leprosy Research Center, National Institute of Infectious Diseases, Japan Institute for Health Security, Tokyo, Japan.
- Bruker Japan K.K., Kanagawa, Japan.
ORCID-Profile: Mitsunori Yoshida, Yoshihiko Hoshino
Förderung
- Japan Agency for Medical Research and Development
- Japan Society for the Promotion of Science
Zitieren
APA
Yoshida, M., Doke, K., Komine, T., Fukano, H., Fujinaga, A., & Hoshino, Y. (2026). Routine clinical laboratory evaluation demonstrates limited utility of Fourier-transform infrared spectroscopy for subspecies identification of the Mycobacterium abscessus complex. Journal of Clinical Microbiology, 64(11). https://doi.org/10.1128/jcm.01087-26
Vancouver
Yoshida M, Doke K, Komine T, Fukano H, Fujinaga A, Hoshino Y. Routine clinical laboratory evaluation demonstrates limited utility of Fourier-transform infrared spectroscopy for subspecies identification of the Mycobacterium abscessus complex. Journal of Clinical Microbiology. 2026;64(11). doi:10.1128/jcm.01087-26
BibTeX
@article{Yoshida2026,
author = {Yoshida, Mitsunori and Doke, Kohei and Komine, Takeshi and Fukano, Hanako and Fujinaga, Azumi and Hoshino, Yoshihiko},
title = {Routine clinical laboratory evaluation demonstrates limited utility of Fourier-transform infrared spectroscopy for subspecies identification of the Mycobacterium abscessus complex},
journal = {Journal of Clinical Microbiology},
year = {2026},
volume = {64},
number = {11},
doi = {10.1128/jcm.01087-26}
}
EUVIMED