Title: In-depth analysis of a mixed Mycobacterium tuberculosis infection involving a multidrug-resistant strain and a susceptible strain
Abstract: Within a mycobacterial interspersed repetitive units–variable number tandem repeats (MIRU-VNTR) -based molecular epidemiology programme run on prisoners transferred from Peruvian prisons to a Spanish prison, we identified a mixed infection in one individual [[1]Abascal E. Herranz M. Acosta F. Agapito J. Cabibbe A.M. Monteserin J. et al.Screening of inmates transferred to Spain reveals a Peruvian prison as a reservoir of persistent Mycobacterium tuberculosis MDR strains and mixed infections.Sci Rep. 2020; 10: 2704Crossref PubMed Scopus (6) Google Scholar]. After segregating the two strains by serial dilutions on agar plates of the cultured isolates and single-colony analysis supported by MIRU-VNTR analysis, we detected the coexistence of a susceptible and a multidrug-resistant (MDR) strain (see Supplementary material, Table S1). The patient was transferred from a Peruvian prison in 2017. Active tuberculosis (TB) was diagnosed on reception into the Spanish penitentiary system. He reported a previous anti-TB treatment history during his stay in the Peruvian prison. The patient was a 50-year-old, human immunodeficiency virus-negative man, and had compensated cirrhosis due to hepatitis C. Although no symptoms were identified, the radiological findings indicated bilateral cavitary lesions. Auramine staining indicated smear positivity (2+) in all three sputa at diagnosis. The Xpert® MTB/RIF test (Cepheid, Sunnyvale, CA, USA) confirmed Mycobacterium tuberculosis infection (high load) and determined resistance to rifampicin. The M. tuberculosis infection was culture-confirmed 5 days later. First-line and second-line phenotypic drug susceptibility tests revealed resistance to all five first-line drugs plus rifabutin, and adjusted treatment was prescribed (amikacin, bedaquiline, protionamide, ethambutol, moxifloxacin and linezolid) for 24 months. Thirteen sequential cultures were obtained during the 5-month period in which the patient had positive cultures (despite full adherence to anti-MDR-TB treatment). MIRU-VNTR analysis enabled us to differentiate between three sequential phases (Fig. 1a): (a) the first five cultures, with the presence of both strains; (b) the following five cultures, with the MDR strain under-represented; and (c) the last three cultures before negativization, when the MDR strain was cleared and only the susceptible strain persisted. We determined the relative proportion for each strain using MIRU-VNTR analysis in standard agarose gel electrophoresis of 50 independent single colonies (Fig. 1b). The results were consistent with the previous qualitative MIRU-VNTR analysis. The susceptible strain was the minority strain at diagnosis (around 30:70). This observation was confirmed by capillary electrophoresis of the MIRU-VNTR products obtained directly from the diagnostic sputum (see Supplementary material, Fig. S1), so ruling out bias due to differential growth of each strain after culture [[2]Martin A. Herranz M. Ruiz Serrano M.J. Bouza E. Garcia de Viedma D. The clonal composition of Mycobacterium tuberculosis in clinical specimens could be modified by culture.Tuberculosis (Edinb). 2010; 90: 201-207Crossref PubMed Scopus (38) Google Scholar], which was also demonstrated by the equivalent pattern of growth units of the MDR and susceptible strains (see Supplementary material, Fig. S2). The susceptible strain remained less represented than the MDR strain for the following four cultures. In the sixth sequential culture, 10 weeks after diagnosis, the representativeness of the MDR strain was reduced; this reduction was observed for the four subsequent cultures (range, 22%–38%). In the last three cultures, only the susceptible strain was detected, which persisted even once the MDR had been cleared. Capillary electrophoresis confirmed clearance of the MDR strain in the last three cultures (see Supplementary material, Fig. S3). The patient was following a personalized MDR-TB regimen, without first-line drugs, because of resistances. Although the susceptible strain did not benefit from the most efficient drugs included in the standard regimens for susceptible TB, the drugs received should have had an equivalent effect on the MDR and susceptible strains. However, because second-line drugs have less bactericidal effect, some authors suggest administering a combination of drugs in heteroresistant mixed infections, targeting both the susceptible and the MDR subpopulations [[3]Cohen T. van Helden P.D. Wilson D. Colijn C. McLaughlin M.M. Abubakar I. et al.Mixed-strain Mycobacterium tuberculosis infections and the implications for tuberculosis treatment and control.Clin Microbiol Rev. 2012; 25: 708-719Crossref PubMed Scopus (121) Google Scholar]. Host-related factors may have been involved in the differential behaviour of the co-infecting strains. The patient had cavitary TB. Different strains/clonal variants can be found at different anatomical sites [[4]Perez-Lago L. Comas I. Navarro Y. Gonzalez-Candelas F. Herranz M. Bouza E. et al.Whole genome sequencing analysis of intrapatient microevolution in Mycobacterium tuberculosis: potential impact on the inference of tuberculosis transmission.J Infect Dis. 2014; 209: 98-108Crossref PubMed Scopus (99) Google Scholar]. If the two different strains were each located in separate cavities, the efficiency of the drugs for reaching the various sites might not be equivalent [[5]Prideaux B. Via L.E. Zimmerman M.D. Eum S. Sarathy J. O'Brien P. et al.The association between sterilizing activity and drug distribution into tuberculosis lesions.Nat Med. 2015; 21: 1223-1227Crossref PubMed Scopus (277) Google Scholar], which would explain the variations in persistence. Considering the cavitation found in this case, we can assume the disease to be prolonged. A previous episode of pulmonary TB was diagnosed in the Peruvian prison in December 2013. Resistance to isoniazid, rifampicin, pyrazinamide, ethambutol and streptomycin led to adjustment of the treatment, and kanamycin, levofloxacin, ethionamide, cycloserine, pyrazinamide and para-aminosalicylic acid were administered for 14 months. Cultures were negative after treatment. No isolates were available for this first TB episode in Peru to clarify whether the episode in Spain corresponded to a relapse of that infection or to re-infection with a totally different M. tuberculosis strain. Mixed infections are more frequently expected in settings where TB burden is high [[3]Cohen T. van Helden P.D. Wilson D. Colijn C. McLaughlin M.M. Abubakar I. et al.Mixed-strain Mycobacterium tuberculosis infections and the implications for tuberculosis treatment and control.Clin Microbiol Rev. 2012; 25: 708-719Crossref PubMed Scopus (121) Google Scholar], such as prisons [[6]Shamputa I.C. Jugheli L. Sadradze N. Willery E. Portaels F. Supply P. et al.Mixed infection and clonal representativeness of a single sputum sample in tuberculosis patients from a penitentiary hospital in Georgia.Respir Res. 2006; 7: 99Crossref PubMed Scopus (113) Google Scholar]. In the study identifying this case [[1]Abascal E. Herranz M. Acosta F. Agapito J. Cabibbe A.M. Monteserin J. et al.Screening of inmates transferred to Spain reveals a Peruvian prison as a reservoir of persistent Mycobacterium tuberculosis MDR strains and mixed infections.Sci Rep. 2020; 10: 2704Crossref PubMed Scopus (6) Google Scholar], two of 17 cases analysed from a single prison corresponded to mixed infections. Prisons may constitute settings where in-depth analysis of such complex infections should be performed, ideally with higher discriminatory methods, namely whole-genome sequencing, to overpass the limitations of MIRU-VNTR. We can gain greater insight into their evolution during anti-TB treatment and evaluate the outcome of treatment, especially when different drug-resistance profiles are involved. To conclude, the detailed analysis of the evolution of a mixed infection involving both resistant and susceptible strains highlighted the complex management of this clinical situation. More efforts are needed to properly diagnose mixed and/or heteroresistant infections, and optimal management procedures need to be clarified to treat and control these complex infections. EA contributed to conceptualization, methodology, investigation and to writing the original draft. MH, FF-G and EG contributed to investigation and MJRS contributed to resources and investigation. PM contributed to writing-review and editing. DGdV contributed to conceptualization, methodology, writing-review and editing, supervision and funding acquisition. There are no conflicts of interest for any of the authors. This work was supported by ERANet-LAC ( ELAC2015/T08-0664 ) and Instituto de Salud Carlos III ( AC16/00057 , FIS15/01554 , FIS13/01207 ) and co-funded by European Regional Development Funds from the European Commission: A way of making Europe. We are grateful to Thomas O'Boyle for proofreading this manuscript.