EECA Regional SORT IT 9-12 months short treatment for patients with MDR-TB increases treatment success in Kyrgyzstan

Introduction: MDR/RR-TB is a growing problem in Kyrgyzstan. In 2005, the country introduced standard or individualized treatment for 20-24 months. Because of poor treatment outcomes, in 2017 a short treatment with strict eligibility criteria was introduced. The aim of this study was to compare characteristics and treatment outcomes of MDR/RR-TB patients receiving short (9-12 months) treatment in 2017 with those receiving standard or individualized (20-24 months) treatment in 2016/2017. Methodology: A comparative cohort study using routine programmatic data. Characteristics, sputum culture conversion and treatment outcomes were compared between those on short treatment with those on standard/individualized treatment using the chi-square test, crude and adjusted risk ratios (RR and aRR). Results: The study included 274, 82 and 132 patients on standard, individualized and short treatment, respectively. There were more females, fewer migrants/homeless and unemployed and more new TB patients on short treatment compared with the other two groups. A favorable outcome (cure and treatment completed) was significantly higher in short treatment patients (83%) compared with those on standard (50%) or individualized (59%) treatment ( p < 0.001). There was higher 1-month sputum culture conversion with short treatment (35%) compared with the other two groups (19% and 24%, p < 0.05). Short treatment (aRR 1.6, 1.4-1.8), female gender (aRR 1.2, 1.1-1.4), not being homeless (aRR 12.9, 4.5-17.3) and having new TB (aRR 1.3, 1.0-1.5) were independently associated with a favorable outcome. Conclusions: The treatment success was higher in selected MDR-TB patients given short treatment in Kyrgyzstan: this regimen should be scaled-up to all MDR-TB patients.


Introduction
Drug-resistant tuberculosis has become a major public health concern in many countries. Multidrugresistant TB (MDR-TB -resistance to at least isoniazid and rifampicin) is one of the most concerning types of drug-resistant TB because of its prevalence and associated difficulties with diagnosis and effective treatment.
The prevalence of MDR-TB has been gradually increasing in the last 10-15 years, especially in the countries of the former Soviet Union including Kyrgyzstan [1]. The diagnosis of MDR-TB has improved with the advent and scale-up of the automated molecular diagnostic assay, Xpert MTB/RIF, which enables a rapid diagnosis of both TB and rifampicin resistance within 2 hours [2,3]. Because most rifampicin-mono-resistant TB is also resistant to isoniazid, rifampicin-mono-resistant TB (RR-TB) is routinely regarded and treated as MDR-TB.
In 2018, it was estimated that 484,000 persons globally had MDR/RR-TB, with 78% having resistance to both rifampicin and isoniazid and the remainder resistance to rifampicin alone [1]. Traditionally, MDR/RR-TB has required a standardized "conventional" treatment for up to 24 months with second line anti-TB drugs which are less effective, more costly and associated with more adverse events compared with first-line drugs. Under programmatic conditions, successful treatment outcomes with these 67S conventional regimens have been poor and have been achieved in just over half of all patients [1].
The development and use of shorter MDR-TB treatment regimens, however, has given grounds for optimism. Under operational research conditions in Bangladesh, Vietnam and various African countries, short course regimens of between 9-12 months have shown successful treatment outcomes >75% with low or even zero relapse rates [4][5][6][7][8]. These findings from observational studies were confirmed in 2019 in a multicenter randomized controlled trial showing successful treatment outcomes >75% and noninferiority between short 9-12 months treatment course and standard 24-months treatment [9]. Based on operational research evidence, in 2016 the World Health Organization (WHO) recommended a short treatment regimen of 9-12 months for MDR-TB, provided patients had not been treated with second-line drugs and/or had resistance to fluoroquinolones and second-line injectable agents excluded [10]. Based on accumulating evidence of the frequency, severity and often permanency of ototoxicity from second-line injectable agents [11], the WHO has further recommended that fully oral short treatment regimens be considered [12,13].
In 2018, Kyrgyzstan in Central Asia notified 7,585 cases of TB [1]. It has a growing problem with DR-TB, with a prevalence of MDR-TB of 29% in new patients and 68% in previously treated patients (overall, the region has the highest resistance rates among previously treated patients which are above 50%) [1]. The proportion of MDR-TB cases with resistance to a fluoroquinolone and a second-line injectable agent among all MDR-TB cases in 2012 was 8% [1]. Conventional standardized MDR-TB treatment for 20-24 months was first started in 2005. Individualized treatment regimens of the same duration could also be given based on culture and drug susceptibility status. Short MDR-TB treatment of 9-12 months was introduced in 2017 for patients meeting eligibility criteria as outlined by the WHO [10]. The proportion of successful treatment outcomes using the conventional standard regimen in a recent cohort was reported to be 53% [1], which makes the 90% treatment success rate target of WHO practically unachievable.
There is no published data from Kyrgyzstan on how treatment outcomes of short treatment compare with those obtained with standard treatment or individualized treatment. This information is of importance to the National TB Program which is also engaged in discussions with WHO about the possibility of piloting and scaling-up fully oral short MDR-TB treatment from 2020 onwards. The aim of the current study therefore was to assess and compare characteristics, sputum culture conversion and treatment outcomes of MDR/RR-TB patients who received short (9-12 months) treatment in 2017 with those who received standard (20-24 months) treatment in 2016 or individualized (20-24 months) treatment in 2016 / 2017 in Kyrgyzstan.

Study design
This was a comparative cohort study using routine programmatic data.

General setting
The Republic of Kyrgyzstan is a landlocked country in Central Asia bordering Kazakhstan, Tajikistan, Uzbekistan and China. The population of the country is about 6.3 million with 36% living in urban areas. According to the World Bank, Kyrgyzstan is a lowermiddle income country with GDP of USD$1281 per capita [14].

MDR/RR-TB care in Kyrgyzstan
The Directly Observed Treatment Short Course (DOTS) strategy recommended by WHO has been implemented in Kyrgyzstan since 1998. MDR-TB treatment under routine conditions in accordance with WHO guidelines was initiated in 2005.
Inpatient treatment of TB is organized in seven regional and two city TB centers while outpatient treatment is managed in facilities called "TB cabinets". There are 61 TB cabinets located in primary health care (PHC) units across the country. The laboratory network is composed of a single National Reference Laboratory (NRL) which performs drug sensitivity testing (DST) and 6 regional laboratories which perform sputum smear microscopy, of which 24 have installed GeneXpert machines (Cepheid, Sunnyvale, CA, USA) for rapid bacteriological diagnosis of TB along with information about RR-TB. Sputum collected at regional sites (or in some cases a traditional culture of Mycobacterium tuberculosis) is transported to the NRL for further processing and DST utilizing various methodologies including line probe assays.
Diagnosis of DR-TB is made through Xpert MTB/RIF and Genotype MTBDRplus (Hain Lifesciences GmbH, Nehren, Germany). Specimen transportation is carried out every other week to the nearest Xpert MTB/RIF laboratory with large numbers of samples arriving at the same time. The reports of the 68S Xpert MTB/RIF results are transported back to the centers using the same car that transports the specimens, all of which results in delayed deliveries by up to two weeks. Treatment is initiated and managed by "Medical Conciliums" (MCs) composed of experienced medical professionals located at regional TB centers. The MCs discuss treatment options on a case-by-case basis and prescribe appropriate treatment regimens (standard, individualized or short treatment) to the patients. The three different treatment regimens and their durations of treatment are shown in Table 1. Patients are followed up during treatment at the nearest TB center or cabinet by clinical assessment and by monitoring sputum culture conversion for the first six months and at the end of treatment. There is a patient support programme to ensure proper adherence to treatment. This includes transportation cost coverage and monthly provision of food and hygiene product packages for patients. Final treatment outcomes are reported in line with WHO guidelines [15].
Pilot project for short course treatment Starting from 2017, short treatment for MDR/RR-TB was prescribed to patients meeting eligibility criteria in accordance with WHO recommendations [10]. Exclusion criteria included: i) MDR/RR-TB cases with resistance to second-line drugs (presented in Table  1) used in the short treatment regimen; ii) contact with patients who had documented resistance to second-line drugs; iii) cases who received treatment with secondline drugs for more than one month; iv) a history of allergy or intolerance to the medications used in the regimen; v) disseminated types of pulmonary TB; vi) a combination of pulmonary and extra-pulmonary types of TB; vii) TB of the nervous system, spondylitis or any form of extra-pulmonary TB with HIV co-infection; viii) pregnancy; and ix) absence of antiretroviral therapy (ART) for HIV co-infected cases.

Study population
All MDR/RR-TB patients who were enrolled in the pilot program for short treatment in 2017 as well as those given standard treatment or individualized treatment in 2016/2017 in Kyrgyzstan were included in the study.

Data sources and variables
Routine programmatic data from the National TB Programme registers and from the pilot project registers for short treatment were used for data collection which took place between February and May 2020. Data variables included: treatment regimen; registration date; gender (male, female); age in years; TB type (pulmonary or extra-pulmonary); TB category (new or previously treated); sputum smear result (smear-positive, smear-negative, smear not done); Mycobacterium tuberculosis culture result (positive, negative, not done); drug susceptibility testing (DST) results if available; treatment start date; sputum culture conversion (Yes, No); date of sputum culture conversion; final treatment outcome (cured, treatment completed, treatment failed, died, lost to follow up and not evaluated) as defined by WHO [15]; and date of final treatment outcome. A favorable outcome was defined as cured and treatment completed and an unfavorable outcome as death, failure and lost to follow-up.

Statistical analysis
Data were analyzed and presented using descriptive statistics with frequencies and proportions. Baseline demographic, social and clinical characteristics, sputum culture conversion and final treatment outcomes between those on the short treatment regimen were compared with those on standard treatment and those on individualized treatment with the use of Pearson's χ 2 test (Chi square). Baseline characteristics and the different treatment regimens were assessed and compared with respect to a favorable outcome (cured and treatment completed) and results presented as risk ratios (RR) with 95% confidence intervals (95% CI). All these factors were included in an adjusted binomial regression model and presented as adjusted RR with 95% CI. Levels of significance were set at p < 0.05.

Ethics
The study was approved by the national ethics committee ("Preventive Medicine" ethical board at Ministry of Health, Kyrgyzstan) as well as the Ethics Advisory Group of the International Union against Tuberculosis and Lung Disease, Paris, France. As this study was a retrospective analysis of anonymized program data with no patient identifiers, informed patient consent was not required.

Results
Altogether, 274 patients were treated with standard treatment, 82 with individualized treatment and 132 with short treatment. Demographic and social characteristics of patients on the three different treatment regimens are shown in Table 2.

70S
For those on short treatment, there were significantly more females, fewer adults, fewer migrants, fewer homeless persons and more employed persons compared with those on standard treatment. There were no significant differences in characteristics between those on short and individualized treatment.
Clinical characteristics of patients on the three different treatment regimens are shown in Table 3. Altogether, between the three cohorts, there were more patients with smear-positive TB, with new TB, with confirmed MDR-TB and with negative HIV status. New TB was significantly more common in those on short treatment compared with those on the longer treatments, but other characteristics were similar between the treatment groups.
Cumulative monthly rates of sputum culture conversion over the first six months of treatment for the three different treatment regimens are shown in Figure  1. In the first month, significantly more patients on short treatment had sputum culture conversion compared with the two longer regimens, but thereafter cumulative sputum conversion rates were similar. Final treatment outcomes are shown in Table 4. Significantly more patients on short treatment had a favorable outcome compared with the other two treatment groups. Characteristics associated with a favorable outcome are shown in Table 5. After adjusting for confounders, the key characteristics associated with a favorable outcome were short treatment, being female, not being homeless and having new TB.

Discussion
This is the first study in Kyrgyzstan to assess outcomes of patients with MDR/RR-TB treated with 9-12 months short treatment compared with 20-24 months standard or individualized treatment. The findings were encouraging.
Over 80% of patients treated with a short regimen had a favorable outcome which was significantly better than patients treated with a standard or individualized regimen in whom favorable outcomes were observed in less than 60%. After adjusting for baseline characteristics, short treatment was still independently associated with a more favorable outcome. Mortality and treatment failure rates, as well as losses to followup were all lower in patients on short treatment. Sputum culture conversion was higher in those on short treatment at the end of the first month but thereafter rates of culture conversion were similar between the three groups for up to six months of follow-up. We speculate that this might be due to the high-dose isoniazid in the short treatment regimen: isoniazid is an effective rapidly acting bactericidal drug which can retain activity in patients with isoniazid resistance due to inhA or katG 315 Thr resistance mutations [16].
Baseline characteristics between the three treatment groups were slightly different with more females, fewer migrants, fewer homeless and unemployed persons and more patients with new TB on the short treatment regimen. Being female and not being homeless were  71S also independent factors associated with better treatment outcomes. Previous studies have found better outcomes in females compared with males [17,18], the latter frequently showing socio-economic or behavioral characteristics that lead to poor access to health services and unreliable follow-up while on treatment. The association between homelessness and TB, and particularly poor treatment success, has long been recognized with homelessness often associated with older age, injecting drug use and HIV infection [19,20]. New TB was an independent factor for having a favorable outcome, so this might have biased the short treatment patients towards better outcomes. Previous studies in Eastern Europe have shown similar findings with better treatment outcomes in those with new compared with previously treated TB [17,21,22]. This is not surprising as those with previous TB may have residual restrictive or obstructive lung disease consequent upon their initial TB or more severe acquired drug resistance [23], and this may adversely affect their prognosis if they get a further episode of TB.
Finally, the criteria that excluded patients from the short regimen, but not the longer regimens, may also have biased the short treatment group towards a better 72S outcome as patients with disseminated TB, with HIVpositive status not on ART and with resistance to other second-line drugs are known to have poorer outcomes [24]. The strengths of this study were the large nationwide sample, the implementation of the study within the national TB programme and the conduct and reporting of the study in line with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement [25]. There were some limitations. We had some important missing information in baseline socio-demographic and clinical characteristics such as smoking, alcohol abuse and diabetes mellitus which can directly impact on treatment outcomes [20], and we had missing information regarding drug-susceptibility testing. Given the exclusion criteria, patients receiving short treatment were a highly selective group of individuals and this limits the generalizability of the findings. We also had no data to present on adverse or serious adverse events, and this is an area needing further study in the country. Additionally, many factors such as changes in TB management guidelines and diagnostic algorithms may have influenced the treatment outcomes in the different treatment cohorts and biased the results. Hence, the before and after comparison may not adequately control for secular trends.
Limitations notwithstanding, there are three important programmatic implications from this study. First, the good treatment success from short treatment should encourage Kyrgyzstan to continue with and scale up short treatment and phase out the longer standard and individualized treatment regimens. This should lead not only to better individual treatment outcomes but also to better acceptability for patients for whom 20-24 months of treatment can be mentally and physically debilitating [26]. Modelling studies have also shown that an effective 9-month treatment regimen at the population level can double treatment access and reduce MDR-TB incidence by 23% over 8 years [27], and in neighboring Uzbekistan a 9-11 month regimen can reduce MDR-TB incidence from 15.2 to 9.7 per 100,000 population and MDR-TB mortality from 3.0 to 1.7 deaths per 100,000 per year [28]. Community benefits should therefore be expected as a result of scaling up short MDR-TB treatment.
Second, more attention in terms of further research and targeted strategies should be paid to improving the current treatment outcomes in males, the homeless and those who have been previously treated, and this may help towards meeting the 90% treatment success target set in the 2016-2020 Global Plan of the Stop TB Partnership [29].
Third, the country needs to consider switching to a fully oral short treatment regimen and eliminate the injectable component of drug-resistant treatment, which is associated with too high risk of ototoxicity and poorer treatment outcomes, including death [11,30]. The most recent WHO guidelines recommend a short all-oral bedaquiline-containing regimen of 9-12 months in eligible patients with confirmed MRD/RR-TB who have not been exposed to treatment with second-line TB medicines used in this regimen for more than 1 month and in whom resistance to fluoroquinolones has been excluded [13]. For patients who have resistance to fluoroquinolones and additional second-line drugs, other effective oral treatment regimens are being studied, one of which (bedaquiline, pretomanid and linezolid) has been shown to be effective in South Africa [31]. This is an exciting time for the treatment of drug-resistant TB with more effective and safer drugs becoming rapidly available [32]. However, these drugs come at a financial cost, and lower middle-income countries such as Kyrgyzstan will need financial support in the future to ensure they can deliver for their affected communities.

Conclusions
In Kyrgyzstan over 80% of patients with MDR/RR-TB treated with 9-12 months short treatment in 2017 had a favorable outcome (defined as cured or completed treatment) compared with less than 60% of patients treated with 20-24 months standard or individualized regimens in 2016/2017. After adjusting for confounders, independent factors associated with a favorable outcome were short treatment, being female, not being homeless and having new TB. These encouraging findings should pave the way for the scale up of short treatment in the country as well as support a move to fully oral treatment regimens for MDR-TB. coordinated by TDR (the UNICEF/UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases). The specific SORT IT programme that led to these publications included a partnership between TDR and the European Tuberculosis Research Initiative (ERI-TB) at the WHO Regional Office for Europe, and was implemented by: the Tuberculosis Research and Prevention Center, Armenia; The Alliance for 73S Public Health, Ukraine; The Centre for Operational Research, The Union, France; the Operational Research Unit (LuxOR), Médecins Sans Frontières, Brussels Operational Centre, Luxembourg; Sustainable Health Systems Sierra Leone; and TDR.

Funding
TDR and partners are able to conduct their work thanks to the commitment and support of a variety of funders. These include long-term core contributors from national governments and international institutions, as well as designated funding for specific projects within current priorities. A full list of TDR donors is available on our website at: https://www.who.int/tdr/about/funding/en/ This SORT IT programme was funded by the United States Agency for International Development (USAID) and supported by implementing partners. Funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Author Contributions
EZ: Conception of the study; designing the protocol; data collection, analysis and interpretation; writing first draft of the paper; critically reviewing the paper and giving approval for the final version to be published. OG: Designing the protocol; data collection, analysis and interpretation; critically reviewing the paper and giving approval for the final version to be published. HD: Designing the protocol; data analysis and interpretation; writing first draft of the paper; critically reviewing the paper and giving approval for the final version to be published. SA: Data analysis and interpretation; critically reviewing the paper and giving approval for the final version to be published. AS: Writing first draft of the paper; critically reviewing the paper and giving approval for the final version to be published. ADH: Designing the protocol; data analysis and interpretation; writing first draft of the paper; critically reviewing the paper and giving approval for the final version to be published. BM: Designing the protocol; critically reviewing the paper and giving approval for the final version to be published.