Therapeutic drug monitoring–guided treatment of XDR TB with an RpoB I491F mutation—a case report

This is the authors' abstract. We don't add key points for this paper.

XDR TB is characterized by resistance to isoniazid and rifampicin, alongside resistance to at least one fluoroquinolone and one additional WHO group A drug, such as bedaquiline or linezolid.1 Individualized XDR TB treatment regimens with a minimum duration of 18 months are recommended by the WHO.1 A recent concern is the emergence of XDR M. tuberculosis (Mtb) strains with a particular RpoB mutation, I491F, that is not detected by routine molecular WHO-recommended rapid diagnostics. Mtb strains with the RpoB I491F mutation are highly dominant in Eswatini and have been reported in several other countries.2–5 According to the WHO catalogue, the I491F mutation is classified as conferring borderline resistance to rifampicin in phenotypic drug susceptibility testing.6 In October 2023, a female patient from Ukraine presented with symptoms of cough and fever to a German hospital. The patient had previously been treated for pulmonary TB in Ukraine in 2015 with a four-drug regimen for 2 months; however, due to nausea, rifampicin was discontinued, and treatment was continued with isoniazid, ethambutol and pyrazinamide for an additional 4 months. In 2023, sampling of sputum and bronchoalveolar lavage fluid revealed acid-fast bacilli. The Xpert® MTB/RIF Ultra (Cepheid) test yielded positive results for Mtb, but negative results for rifampicin resistance, prompting the initiation of standard four-drug treatment in a German secondary level hospital. However, phenotypic resistance testing of the culture revealed borderline rifampicin resistance with an MIC at the critical concentration (CC) of 0.5 mg/L, along with resistance to isoniazid, pyrazinamide and ethambutol. Subsequent genomic and phenotypic analyses identified resistance to fluoroquinolones, bedaquiline, linezolid and clofazimine. Targeted sequencing using the Deeplex® Myc-TB assay identified an I491F mutation in RpoB. The isolate was classified as a lineage 2/Beijing strain. Additional phenotypic resistance testing revealed susceptibility to pretomanid, delamanid, amikacin and D-cycloserine at their respective CCs as recommended by the WHO.3 In December 2023, the patient was transferred to the University Hospital Cologne where a treatment regimen comprising terizidone, amikacin, meropenem with clavulanic acid, and pretomanid was initiated (Figure 1a). Pretomanid was chosen over delamanid due to recently published data showing pretomanid efficacy in non-BPaLM (bedaquiline, pretomanid, linezolid, moxifloxacin) regimens.7 Meropenem (2 g) was administered three times daily, together with oral amoxicillin/clavulanate (500/125 mg). A meropenem dosage exceeding the WHO recommendations1 was chosen in light of the limited availability of susceptible drugs and the higher bactericidal activity observed in regimens with dosages of 6 g daily.8 Therapeutic drug monitoring (TDM) was conducted for amikacin and for pretomanid as described recently.9 Pretomanid levels, at a starting dose of 200 mg daily, reached 2.89 mg/L 3 h after administration (Figure 1b). Since rifampicin resistance was borderline by phenotypic testing, we decided to include rifamycins in the regimen. Rifabutin was chosen because it is associated with a less pronounced CYP3A-mediated reduction of pretomanid concentrations than rifampicin.10 Rifabutin was initiated at 300 mg daily, resulting in a 10% reduction in pretomanid levels after 3 and 5 h, respectively. An increase of rifabutin to 600 mg daily, which was chosen to reach therapeutic drug levels in this borderline-resistant strain, further reduced the pretomanid trough levels by 54% to 0.87 mg/L (Figure 1d). According to EMA guidelines, this value is below the CC of 1 mg/L for lineage 1 strains and above the CC (0.5 mg/L) for strains of other lineages. Although our patient was infected with a lineage 2 strain, we decided to reduce the rifabutin dose to 300 mg daily to avoid the occurrence of pretomanid resistance in a regimen primarily based on WHO group C drugs. Amikacin serum levels remained within target ranges (trough 6.8 µg/mL; peak 24.3 µg/mL). From March 2024 onward, sputum microscopy consistently yielded negative results. Over the course of treatment, the sputum time-to-positivity (TTP) increased from 5 days to 36 days over the first 6 months of therapy (Figure 1c). Culture conversion was achieved in May 2024; however, intermittent positive results recurred in June 2024. (a) Course of TB treatment. (b) Radiological improvement observed on low-dose chest CT imaging during treatment. (c) Time to culture positivity (TTP) in sputum samples. The arrow indicates start of the XDR TB treatment. (d) Therapeutic drug monitoring (TDM) of PTM before and after the introduction of RFB and after dose escalation of RFB. AMK, amikacin; DLM, delamanid; EMB, ethambutol; INH, isoniazid; PTM, pretomanid; PZA, pyrazinamide; RFB, rifabutin; RIF, rifampicin; TZD, terizidone. In March 2024, resistance testing demonstrated a slight increase in the MIC of pretomanid, rising from ≤0.5 mg/L to 1 mg/L without evidence of cross-resistance to delamanid, which maintained an MIC ≤0015 mg/L. A second isolate collected in April exhibited the same elevated MIC pattern for pretomanid and full susceptibility for delamanid. WGS of this strain did not reveal any known mutations in relevant genes that could have led to the emergence of cross-resistance to delamanid. As a result, pretomanid was replaced by delamanid, and rifabutin was substituted with rifampicin at 1200 mg daily to overcome borderline rifampicin resistance. Rifamycins are also known to impact delamanid plasma levels.11 Nevertheless, a decision was made to use high-dose rifampicin, as recommended for strains with low-level rifampicin mono-resistance.12 Delamanid TDM was not performed because the results would not have influenced further treatment decisions, given the lack of alternative therapeutic options for this highly resistant strain. Follow-up chest CT scans showed regression of the cavitary lesions (Figure 1b). After 6 months of amikacin therapy, high-frequency hearing loss was detected in routine audiometry. Despite repeated recommendations from the treating physicians to discontinue amikacin therapy, the patient insisted on continuing treatment for a year. Amoxicillin/clavulanate as well as meropenem were discontinued in February 2025. Since July 2024, the sputum culture results have remained negative. At the date of submission of this manuscript, the patient was being treated with an oral regimen that includes high-dose rifampicin, terizidone and delamanid. Amikacin-induced ototoxicity was the only notable side effect observed. This case underscores the growing relevance and therapeutic challenges posed by XDR TB in general and by XDR TB strains harbouring the RpoB I491F mutation specifically.3,13 Borderline rifampicin resistance due to this mutation evades detection by the GenXpert test, highlighting the need for an extension of the diagnostic workup in patients at risk for drug-resistant TB.14 As demonstrated by us and others, rifamycins reduce pretomanid serum concentrations, making TDM essential when co-administering these drugs.10 TDM is becoming increasingly important also with other drugs in TB treatment to avoid both subtherapeutic levels and toxicity.15,16 The increase in pretomanid MIC that developed after 5 months of treatment necessitated changes in the treatment regimen. This underscores the importance of regular WGS combined with phenotypic resistance testing throughout the course of treatment, especially when novel drugs such as pretomanid and delamanid are used. Relying solely on sequencing may overlook emerging, unknown mutations in relevant genes. The study was internally funded by Research Center Borstel; German Center for Infection Research; and National Reference Laboratory for Mycobacteria. J.R., A.K., I.S. and S.N. receive funding from the German Center for Infection Research (DZIF; TTU 02.814). S.N. and J.R. are funded by the European Union Innovative Medicines Initiative 2 Joint Undertaking program grant no. 853989 ERA4TB. All authors declare no conflicts of interest.