Mind the gap. Rolling out new drug resistant tuberculosis regimens with limited diagnostic tools

For the first time in decades, we are witnessing unprecedent progresses in Tuberculosis (TB) treatment. More than 15 new compounds are currently in clinical development and new regimens are under evaluation in several phase 2 and phase 3 clinical trials [1]. Moreover, shorter, all-oral treatment options for both Drug Susceptible (DS) and Drug Resistant (DR) TB are now available. In particular, the World Health Organization (WHO) has recommended the use of 4-month Rifapentine-Moxifloxacin regimen (given with Isoniazid and Pyrazinamide) for DS-TB and the 6-month Bedaquiline- Pretomanid- Linezolid (BPaL) without or with moxifloxacin (BPaLM) regimen for Multi Drug Resistant TB (MDR-TB) and Rifampicin Resistant TB (RR-TB) [2].

In this vibrant and exciting scenario, the data reported by the latest Global Tuberculosis Report, even if not unexpected, have been received with utter dismay by the TB research community [3]. For the first time in almost 20 years, we have observed a raise in TB numbers, including an upsurge of MDR-TB and RR-TB cases (from 437.000 to 450.000 incident cases) [2].

The reasons for these upsetting results have been found in the disruption of the TB- related services during the COVID-19 pandemic, further exacerbated by the ongoing conflicts in Ukraine and across Africa and the Middle-East [4]. Since 2020, access to TB diagnostic has paid a heavy price, further deepening the gap between estimated number of incident TB cases and diagnosed ones. Last year approximately 4 million people with TB remained undiagnosed [3], and this trend is not expected to improve in the next future, confirming that testing is still one of the weakest link in the TB cascade of care [4]. Moreover, according to the latest WHO data, in 2021 significant variations have been reported in the availability of testing for drug resistance among different countries. Of the WHO 30 high MDR/RR -TB burden countries, 20 reached a Rifampicin testing coverage >80 %, whereas global coverage for fluoroquinolones susceptibility testing remained below 50 %, with the lowest rates reported in the Western Pacific Region (<20 %) [3].

As we are observing the rapid emergence of TB strains resistant to the new drugs, in particular to Bedaquiline, the lack of availability of Drug Susceptibility Testing (DST) and widespread surveillance for TB drug resistance may represent a threaten to the newly introduced all oral Bedaquiline based shorter regimens [5].

Primary resistance to Bedaquiline was first documented in 2016 and since then several studies have reported emerging resistance against Bedaquiline during treatment. The main mechanisms associated to Bedaquiline resistance identified so far mainly involve three genes, atpE, mmpR (Rv0678) and pepQ [6], [7], [8]. Even if resistance to Bedaquiline appears to be limited in population not exposed to the drug [9], it seems to be acquired quite easily in groups who had previously received Bedaquiline treatment. A retrospective study on the microevolution of an MTB strain isolated from a person affected by TB who has received Bedaquiline for 18 months, describes the transient emergence of an atpE_Ala63Pro mutation related to high-level Bedaquiline resistance. Moreover, during the Bedaquiline treatment multiple Rv0678 mutations emerged independently and the atpE_Ala63Pro mutation disappears, probably because of its high fitness costs [10].

A recent study conducted in a high burden country (Moldova) has reported Bedaquiline resistance acquisition in more than 15 % of all MDR-TB patients who received this drug [11]. In South Africa and Taiwan Bedaquiline resistance in MDR/RR-TB population is currently reported around 3 % in patients with previous exposure to Bedaquiline or Clofazimine and solely associated to Rv0678 mutations in the South African study [12].

Worrisome results have been also published concerning Pretomanid resistance, another key drug of the new MDR-TB short regimens. Increased Minimal Inhibitory Concentration (MIC) values, probably linked to resistance to Pretomanid, have been identified in case of mutations in six non-essential genes ddn, fbiA, fbiB, fbiC, fbiD and fgd1 [13], [14] whose products are necessary for the activation of this drug. Notably, it has been reported that Lineage 1 (L1) M. tuberculosis (MTB) was less susceptible than other MTB lineages with a 4-fold increase of the Ecological Cut Off (ECOFF) in comparison with other tested MTB lineages (2,3,4 and 7) [15]. The implications of this finding may be particularly relevant as 28 % of all MTB strains belong to L1 and have been identified mostly in high MDR-TB burden countries as India, Philippines and Bangladesh [16], countries where the new shorter regimens could really represent a game changer for the local TB elimination strategy.

Given these results, improve surveillance policies and methods for the resistance to the new drugs and guarantee access to fast and reliable DST emerged as key needs. Nonetheless, according to a survey conducted among 44 TB National Reference Laboratories (NRLs) in the WHO European Region, only 66 % of them had capacity to test for Bedaquiline resistance [17] while Pretomanid DST is even more limited [18]. Hence, the inadequate availability and ease of use of susceptibility testing for the new drugs represents a major source of concern.

Genotypic DST can represent a rapid, efficient and cost-effective alternative to phenotypic DST. New rapid fully automated kits have been released to allow fast identification, directly from the sample of Fluoroquinolones resistance [19]. Other tools as Targeted Next Generation Sequencing (tNGS) may improve TB clinical management having a turnaround time that could range from 2 to 4 days according to the technology used [20]. A new tNGS kit should be released in 2023 allowing to identify mutations also associated to resistance to Bedaquiline and Delemanid.

Furthermore, the publication last year of the first WHO endorsed catalogue of MTB mutations associated to drug resistance can have significant impact in scaling up new drugs susceptibility testing using Whole Genome Sequencing [21]. In this document a candidate gene approach is used to grading mutations associated with resistance for 13 antituberculosis drugs. Due to the few phenotypic data available and the relatively low prevalence of phenotypic resistance only 2 resistance mutations have been identified for Bedaquiline, Delamanid and Linezolid [22]. Nevertheless, a new version of the catalogue is going to be published in 2023 including updated data on the new drugs.

The lack of commercial, easy to use phenotypic DST for new drugs plays an important role not only in the scaling up of testing, but also in clinicians trust and use of resistance probability into their clinical decision-making. A qualitative study involving 14 physicians with long lasting experience in treating RR and MDR-TB from different high and low burden countries has reported that they found challenging including Bedaquiline resistance probability into their decision making process [23]. This appears to be due to the participants limited experience with DST for Bedaquiline, unfamiliarity with genotypic DST interpretation and lacking understanding of WGS results.

In conclusion, to protect the results obtained so far and finally meeting the End TB programme goals, TB resistance diagnostic should keep up with the progresses made in the drug development field. There is a need to revise TB diagnostic algorithms and to establish a regular widespread program of DR surveillance. The current evolution of the TB diagnostic pipeline, now focusing on sampling strategies alternative to sputum and rapid targeted tests for resistance identification, will hopefully provide an improved patient centred approach to TB care.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.