Ўзбекистон РеспубликасиСоғлиқни сақлаш вазирлиги

Effectiveness of a simplified short regimen for

Multidrug Resistant Tuberculosis treatment in Karakalpakstan,

Uzbekistan

Joint MSF - Operational Centre Amsterdam / MOH Uzbekistan Research Protocol

Research Protocol - amendmentApril 2014

Philipp du Cros1

Khamraev Atadjan (deputy of Chairman of Supreme Council of Karakalpakstan)2 Tillashaikhov Mirzagalib (MoH Uzbekistan, Director of specialized medical center of TB and pulmonology) 2

Parpieva Nargiza (MoH Uzbekistan, chief TB specialist)2

Tigay Zinaida (Chief of TB2)2

Sholtaeva Marjan (Chief TB doctor of Kegeily rayon)2

Abdrasuliev Tleubergen (Chief TB doctor of Shumanay rayon)2

Marttje Betlam (Doctor Kegeily Rayon)1 Emily Wise (Doctor Shumanay)1

Teshome Ashagre (Medical Coordinator)1

Johanna Kuhlin1

Andrii Slyzkyi(laboratory manager) 1

Alia Taye Epidemiologist Uzbekistan1

Sebastian Dietrich1

Bern-Thomas Nyangwa1

Jane Greig1

Pamela Hepple1

Graham Cooke3

Krzysztof Herboczek1

Muhammad Asif1

Jay Achar1

Catherine Berry1

1 Médecins Sans Frontières, Operational Center Amsterdam 2 Ministry of Health Uzbekistan3 Imperial College London

Main contactsCo-Principal Investigator MSF Co-Principal Investigator MoHDr. Philipp du Cros Dr. Khamraev A. KarimovichMédecins Sans Frontières–UK atadjan@yandex.ru 67 Saffron HillLondone-mail: philipp.ducros@london.msf.org

Pilot CoordinatorCatherine BerryMédecins Sans Frontières, Nukuse-mail: karakalpastan-research@oca.msf.orgPhone: +998913778652

TABLE OF CONTENTS

LIST OF ABBREVIATIONS............................................................................................................4SUMMARY..................................................................................................................................5INTRODUCTION..........................................................................................................................6Context...................................................................................................................................... 7Current Standard of Care for MDR TB........................................................................................8Evidence for short course MDR TB treatment...........................................................................9HYPOTHESIS............................................................................................................................. 10RESEARCH OBJECTIVES............................................................................................................ 10Primary objective.....................................................................................................................10Secondary objectives...............................................................................................................10PATIENTS AND METHODOLOGY...............................................................................................111. Study design.....................................................................................................................112. Study sites........................................................................................................................113. Partnership/study coordination.......................................................................................114. Study participants and sample size..................................................................................125. Study endpoints...............................................................................................................146. Case Definitions............................................................................................................... 157. Treatment options...........................................................................................................168. Monitoring of adverse events/reactions..........................................................................189. Assessment of treatment effectiveness...........................................................................1910. Study procedure and patient flow...................................................................................2211. Laboratory tests and electrocardiogram..........................................................................2412. Variables..........................................................................................................................2513. Data collection................................................................................................................. 2514. Statistical Analysis............................................................................................................26ETHICAL ISSUES........................................................................................................................261. Ethical Committee............................................................................................................262. Consent forms..................................................................................................................263. Expected risks and advantages........................................................................................274. Confidentiality..................................................................................................................285. Community Involvement..................................................................................................28BUDGET....................................................................................................................................28TIMELINE..................................................................................................................................29DISSEMINATION OF STUDY RESULTS.......................................................................................29REFERENCES.............................................................................................................................30

LIST OF ABBREVIATIONS

ALT Alanine aminotransferaseART Antiretroviral TherapyBMI Body Mass IndexCfz ClofazimineCm CapreomycinDST Drug Sensitivity TestDR TB Drug resistant tuberculosisE EthambutolFLD First line drugsGfx GatifloxacinH IsoniazidHb HemoglobinHIV Human Immunodeficiency VirusIUATLD International Union Against Tuberculosis and Lung DiseasesKm KanamycinMfx MoxifloxacinMSF Médecins Sans Frontières/Doctors without BordersMDR TB Multidrug resistant TBMTB Mycobacterium tuberculosisMoH Ministry of HealthMGIT Mycobacteria Growth Indicator TubeNRL National Reference LaboratoryOCA Operational Centre AmsterdamPto ProthionamideSLD Second line drugsTB TuberculosisTSH Thyroid stimulating hormoneVL Viral loadWHO World Health OrganizationXDR TB Extremely drug resistant tuberculosis Z Pyrazinamide

SUMMARY

Multidrug resistant tuberculosis (MDR TB) is a growing problem and few people have access to adequate diagnosis and treatment. The current recommended treatment regimen for MDR TB has a minimum of 20 months duration with high toxicity. Scale up of MDR TB treatment is associated with high default rates, and experience in the MSF programme in Uzbekistan shows that the current standard treatment greatly limits the ability to scale up to meet the high rates of MDR TB in the region. Evidence from Bangladesh in 2010 showed that a 9-month short-course regimen could achieve a relapse free-cure rate of 88%. Several countries in West Africa started implementing similar regimens with similar outcomes. Evidence of effectiveness of this shortened regimen amongst regions with high second line drug use and resistance is still limited. We propose an observational study under programmatic conditions to evaluate the effectiveness of a shortened course MDR TB regimen in the high MDR/XDR TB prevalence setting of Karakalpakstan.

A prospective observational study has been designed. All patients with presumptive MDR TB identified with rapid molecular testing (MTBDR plus or Xpert® MTB/RIF assay) and later on culture confirmed MDR TB without the history of the previous treatment for MDR TB for longer than one month will be included in the study. The study regimen is composed of an intensive phase of at least 4 months duration of Pyrazinamide (Z) + Ethambutol (E) + Isoniazid (H) + Moxifloxacin (Mfx) + Capreomycin (or Kanamycin/Amikacin) (Cm/Km/Am) + Prothionamide (Pto) + Clofazimine (Cfz) and a continuation phase of oral drugs Z-E-Mfx-Pto-Cfz. Patients will be followed up until the end of treatment and during 12 months after treatment completion in order to evaluate the rate of relapse.

The primary outcome measure is the success rate at the end of treatment, and relapse and re-infection rates during 12 months of follow up after completion of treatment. Secondary outcome measures include rate of adverse events, interim outcomes with sputum smear microscopy and culture conversion rates at 4 and 6 months and time to conversion.

Data will be recorded in patient’s clinical files and electronic databases and analyzed with Stata 11.0.

This study is a result of ongoing collaboration of MSF with the MoH in Uzbekistan; results will be shared with the national health authorities, World Health Organization and the rest of the scientific community and aim to influence and improve treatment and care of patients with MDR TB.

INTRODUCTIONMultidrug-resistant tuberculosis (MDR-TB) is defined as Mycobacterium tuberculosis resistant to at least isoniazid and rifampicin, the two most effective anti-tuberculosis drugs. An estimated 440,000 new cases of MDR-TB occur globally each year with an estimated 150,000 deaths (WHO 2010). The highest proportions of MDR-TB globally are in countries of Eastern Europe and Central Asia, including Uzbekistan.

MDR TB / XDR TB is an increasing problem as highlighted in the 2012 WHO global tuberculosis report. (WHO 2012) MDR-TB treatment requires complex, expensive treatment with less effective second line drugs, and is often associated with significant side effects. The minimum treatment duration is 20 months. Treatment outcomes for patients with MDR-TB are worse than for drug sensitive TB, with low success rates and high rates of treatment failure and default (WHO 2011). The implementation of MDR TB treatment worldwide is stagnating with very limited scale up predicted between 2012-2015 with current treatment, leaving a large gap between those in need of treatment and those diagnosed and commenced on treatment. (WHO 2012) In addition, a recent systematic review showed that increased cohort size is associated with increased rates of default suggesting that as programmes scale up, results worsen (Toczek 2012).

In 2010, a cohort study of MDR TB patients treated with a 9 month short course regimen achieved a relapse-free cure rate of 88% in 206 patients (Van Deun 2010). Following the publication of the Bangladesh experience, several countries in Africa are piloting implementation of similar regimens and early results show similarly impressive outcomes. The Union for tuberculosis and lung disease has included this regimen in its recently published tuberculosis programmatic guidelines (Ait-Khaled 2010). MSF is currently using the 9-month regimen in Chad, South Sudan and the Central African Republic, and will commence implementation of a prospective cohort study in Swaziland.

Children have characteristically paucibacillary tuberculosis and thus they may require shorter duration and fewer drugs than adults to treat MDR TB infection and they could benefit from this regimen; however evidence of MDR TB treatment in children is scarce. We aim to include children in this study.

Further data from the Bangladesh short course programme have consistently shown relapse free treatment success of 85.5% in 476 patients with 2 years follow up. Reported adverse events are lower than those reported with standard MDR TB treatment. However in patients with fluoroquinolone resistance the relapse free success rate was lower at 73.2%, and amongst this group those who failed had high rates of amplification of resistance (Van Deun 2012). There

is still much contention about the 9 month MDR TB regimen owing to the fact that the data available is of low quality evidence from single arm cohort studies, and in the case of the Bangladesh programme there were a large number of patients eligible for enrolment who did not get included which suggests the possibility of biased selection. It is unclear what would allow this short course regimen to work, and there is no data about the impact of baseline resistance to drugs used in the regimen such as pyrazinamide, ethambutol, injectables or prothionamide. WHO has requested that countries piloting this regimen collect data under operational research conditions, and have cautioned that there is not enough data to know whether this regimen will be effective in all contexts.

We propose an observational study to evaluate the effectiveness of this short-course MDR TB regimen in the high MDR TB prevalence and high second line drug resistance setting of Karakalpakstan, Uzbekistan.

ContextUzbekistan is included in WHO’s 27 high burden MDR-TB countries, with a recent National TB drug resistance survey reporting 23% of new TB patients and 62% of previously treated patients were infected with MDR-TB (WHO 2012). Karakalpakstan is an autonomous region of Uzbekistan, situated in the North West of the country, covering approximately one third of the country, with a population of approximately 1.2 million. The Karakalpakstan population suffers from high rates of health problems, associated with severe environmental degradation as well as economic, social and health care services deterioration (Crighton 2011). The prevalence of TB in Karakalpakstan was reported to be 464/100,000 population, considerably higher than in the rest of Uzbekistan (227/100,000) (WHO 2009, MoH Uzbekistan 2009).

Médecins Sans Frontières (MSF), together with the Ministry of Health, initiated a DOTS-plus project to treat MDR-TB in 2003 in 2 rayons (districts) in Karakalpakstan, Uzbekistan. Drug sensitivity testing (DST) was initially provided for DR-TB suspects, but was expanded to include all smear positive patients from 2006. Treatment was based on international treatment guidelines, initially with individualized treatment regimens. In 2010, the programme commenced scaling up to other districts. Simplified treatment guidelines for comprehensive TB care for diagnosis and treatment of all forms of TB were developed with standardized side effect management. Psychosocial support, health education and enablers/incentives are provided in accordance with international best practice. Initially patients were hospitalized, but in 2011 ambulatory treatment from day 1 was introduced as an option for patients who did not require hospitalization. Comprehensive TB care has now been introduced to 8 districts, and in 2013 treatment will commence in a further 2-3 districts.

Sputum smear microscopy, culture and DST are conducted according to international standards in the Nukus mycobacteriology laboratory. Smears are assessed using fluorescence microscopy. Rapid molecular tests were introduced in 2007 (Hain MTBDR plus) and Xpert MTB/RIF was introduced for 2 districts and pediatric cases in 2012. BACTEC™ MGIT™ 960 liquid culture system was introduced to the laboratory for first line drug testing in September 2007 and for second line drug testing in 2011 Quality control tests of the performance of the Nukus laboratory are evaluated by the Gauting laboratory.

By September 2012, 2670 patients with TB showing any form of drug resistance had completed treatment in Karakalpakstan. Treatment success rates are in line with experience reported internationally, with 62% success rate for all MDR-TB patients treated between 2003-2008 (Lalor 2011). Resistance rates to ofloxacin (fluoroquinolones) and capreomycin at baseline are low as <3%, while resistance rates to kanamycin have ranged between 10-20%. Resistance rates to ethambutol and pyrazinamide are >60%, and to prothionamide approximately 15% (Nukus epidemiologist personal communication).

Current Standard of Care for MDR TB

The current standard of care for MDR TB treatment involves a long duration of treatment (minimum 20 months) with at least 5 anti-TB medications including an injectable for a minimum of 8 months. The treatment recommendations are based on low quality of evidence with some recommendations being conditional, mostly from evidence from single arm retrospective cohort descriptions, expert opinion and anecdotal evidence (WHO 2011). Treatment success rates are less than ideal with systematic reviews showing success rates of 62-64% (Orenstein, 2009; Johnston, 2009), but a recent individual patient data meta-analysis of over 9000 MDR TB patients reporting success rates of only 54% (Ahuja, 2012).

Standard MDR TB treatment is associated with a high incidence of side effects. More than one third of patients will suffer from significant nausea and vomiting despite prescription of anti-emetics, and more than 10% of patients will suffer from each of the following side effects: diarrhea (21%), arthralgia (16%), dizziness/vertigo (14%), hearing disturbances (12%), headache (11%), sleep disturbances (11%), electrolyte disturbances (11%) and abdominal pain (11%). In addition, serious life threatening or life impinging side effects are reasonably frequent including: psychosis (3%), peripheral neuropathy (8%), depression (6%), allergic reaction (5%), seizures (4%), hypothyroidism (3.5%), hepatitis (2%) and renal failure (1%). (WHO 2008) One of the major issues discussed by all patients participating in online blogging about their experience with MDR TB in several countries was the major impact on their lives of the side effects of the current treatment regimen (TB and me blogs: http://blogs.msf.org/tb/).

Default rates from MDR-TB treatment are high with 23% default reported in a recent meta-analysis(Ahuja 2012). Reported factors influencing defaulting from TB treatment include lengthy treatment, side effects, high cost of treatment for patients in settings where patients must pay, indirect costs such as loss of wages, increased poverty and gender discrimination, dissatisfaction with health care worker attitudes, limited knowledge, and negative beliefs and attitudes to treatment, challenges with drug procurement and sustained supply of second-line drugs, substance abuse, and psychiatric disorders. In addition, a recent meta-analysis showed that as programme size increases default rates also increase, highlighting the difficulties of scaling up effective MDR-TB treatment with current international gold standard treatment (Toczek 2012). Patients defaulting from treatment are at higher risk of mortality and morbidity, and may contribute to further spread of MDR-TB in the community.

In summary, the current standard of care for MDR TB treatment is complex, requires a long duration, results in over 1 in 5 patients starting treatment defaulting and the majority of patients experience significant and sometimes life threatening side effects. The regimens are based on low quality scientific evidence and as experienced in the MSF/MoH programme in Uzbekistan the current regimen is not feasible to rapidly scale up in order to meet the scale of the global problem. In response to the new WHO global TB report in 2012 a recent Lancet editorial stated that “The response to the problem is woefully insufficient” and concluded that it is time to re-examine the current tuberculosis control approach as “The status quo is unacceptable” (Lancet editorial 2012).

Evidence for short course MDR TB treatment

WHO cites that the current recommendation for the duration of MDR TB regimens is based on low quality evidence. The guidelines acknowledge that a shorter regimen would be preferable, but due to a lack of evidence for an optimal alternative, did not endorse short-course therapies for MDR-TB treatment. The short course regimen used in the Bangladesh study would provide benefits in terms of patient tolerability and improved feasibility for programme scale up. A standardized short-course treatment regimen for MDR TB including Kanamycin (Km), Gatifloxacin (Gfx), Clofazimine (Cfz), Prothionamide (Pto), Pyrazinamide (Z), Ethambutol (E) and Isoniazid (H) for 4 months (intensive phase) and 5 months of E, Z, Gfx, Cfz (continuation phase) has demonstrated effectiveness in a prospectively followed cohort of patients in Bangladesh. The duration for some patients slow to sputum smear convert was extended by 1-2 months. This 9-11 month regimen was reported to have a relapse-free cure rate of 87.9% (95% CI, 82.7-91.6) among 206 patients (Van Deun et al 2010) and a culture conversion rate of 85% at 2 months of treatment. Updated data in this cohort of patients enrolled until 09/2010 show a

relapse-free cure rate of 85% among 493 patients (Van Deun, 2012). In addition, adverse events recorded are significantly less frequent and less serious than those that have been reported for the 20+month WHO regimen (Shin, 2007).

Three West African countries (Benin, Cameroon and Niger) have implemented the same or a very similar regimen; unpublished evidence so far has shown excellent treatment outcomes. Cameroon is implementing a 12-month course Gfx-based treatment with Pto for the entire treatment duration. This cohort has reported a cure rate of 93% amongst 74 patients and no short term relapse. Benin has similar successful outcomes from implementation of a 9-month MDR TB treatment regimen where Gfx is replaced by Moxifloxacin (Mfx) as a 4 th generation quinolone and Pto used for the entire regimen. Gfx was replaced by Mfx due to the problems and availability of Gfx and concerns about its safety. According to personal communication with The International Union Against Tuberculosis and Lung Diseases (IUATLD), Benin and Niger are reducing treatment length to 9 months due to significant findings of its efficacy. Other countries like Central African Republic are looking into piloting this regimen. None of these countries have reported any major tolerance problems or adverse events in their study populations. However, those cohorts are still very small and the availability of second line TB drugs outside of TB programmes is limited, with the rates of second line drug resistance likely much lower than in Uzbekistan. As a result of the above evidence, the 9-month MDR TB regimen has been formally recommended by the IUATLD for confirmed MDR TB cases (Ait-Khaled, 2010).

Cost of drugs is a major limiting factor for scale up of MDR TB programs. The drugs alone for the 20+ month regimen cost approximately 4,000 Euros. The cost of the 9 month regimen used in the Bangladesh study was approximately 200 Euros.

It is uncertain whether the findings in the Bangladesh study are applicable to a different setting such as Uzbekistan. Due to the small size of the cohorts available and the single arm study designs, there is debate about the generaliseability of the findings in the Bangladesh study and subsequent cohort studies in Africa. There is conjecture that the results may be due to population selection bias with use in contexts with low levels of second line drug resistance. The role of isoniazid and first line drugs in the regimen has also been postulated to have an effect where there is likely to be low level isoniazid resistance or low rates of resistance to ethambutol and pyrazinamide. However, personal communication with the authors of the Bangladesh study reports that rates of high level isoniazid resistance were very high in the cohort, and ofloxacin resistance was approximately 10%. MSF and MoH Uzbekistan have decided based on the available evidence and the Union recommendation to commence implementation of the short course regimen with ongoing programme monitoring and have implemented this in 2 districts. A third district will be added to increase recruitment with the

aim of completing the recruitment phase and meeting the planned sample size by the end of 2014.

HYPOTHESISA standardized short-course MDR TB treatment regimen (of 9-11 months) is effective under routine programme conditions in a context with high MDR TB prevalence and high second line drug resistance rates.

RESEARCH OBJECTIVES

Primary objective

To describe outcomes at end of treatment and relapse rate at 1 year following treatment completion of a short course (9-11 month) MDR TB treatment in programmatic conditions in Karakalpakstan, Uzbekistan.

Secondary objectives

1. To describe the adverse events of the treatment regimen2. To evaluate risk factors for unfavorable outcomes (death, lost-to-follow up and failure) and

relapse as a combined cohort.3. To describe the rate of resistance amplification amongst patients with outcome failure.

PATIENTS AND METHODOLOGY

1. Study designWe aim to conduct a prospective observational cohort study in three districts in Karakalpakstan, Uzbekistan.

2. Study sitesKegeily Rayon Comprehensive TB care was implemented in Kegeily in March 2012. Kegeily rayon1 is a large rayon (district) with a population of 83,000 and the estimated number of MDR TB patients is 85 patients per year. The rayon has 2 Polyclinics (outpatient clinics for TB care) and 21 SVPs (general practice surgeries with staff trained in TB care), there are 3 TB doctors in the rayon who supervise the care provided by the general practitioners (SVP doctors) and nurses. Both TB doctors and all SVPs doctors in Kegeily have been trained on the TB diagnostic algorithm, management of side effects and when to refer patients as per the Karakalpakstan comprehensive TB care protocol.

1 Rayon is the Russian word describing an administrative region, most easily translated as district, with health services planned around the populations living in Rayons.

Shumanay RayonThe comprehensive TB care programme will be implemented in Shumanay in 2013. Shumanay is a small rayon and close to Khodjeily rayon where comprehensive TB care is already implemented. Shumanay has a population of 52,000 and an estimated number of MDR TB patients of 50 patients per year. There is 1 Polyclinic and 9 SVPs in the rayon and it has 3 TB doctors, where one of these TB doctors has experience working with MDR TB diagnosis and treatment having worked in the main referral hospital for MDR TB (TB2) in Nukus City.

Nukus CityNukus city is the largest district in Karakalpakstan with a population of 268 000. There are 9 polyclinics serviced by 12 MoH TB doctors. The TB doctors are supervised by Sapaev Nurmukhamed in TB1 and Zinaida Tigay via the TB2 consilium. There were around 250 new diagnoses of DR tuberculosis in Nukus City in 2013; we expect 15 new MDR TB admissions per month.

3. Partnership/study coordinationThe study will be conducted in a partnership between the Ministry of Health of Uzbekistan and Médecins Sans Frontières/Doctors without Borders (MSF) – Operational Center Amsterdam (OCA). The National TB programme Uzbekistan and the Ministry of Health have actively participated in protocol development, and will participate in study implementation and interpretation and publication of results.

The study will be implemented by the caregiver team in each site supported by the epidemiologist and data management team. An additional doctor (study coordinator) will be responsible for supervision of the training, implementation and monitoring. The study coordinator will be responsible for the implementation of the study procedures in all clinical sites in collaboration with the MoH and MSF staff working in each Rayon. Two additional nurses will be provided as “pilot project nurses” to ensure documentation is filled in, side effects recorded and graded, proper information given to patients/DOT-nurses, collecting data and performing ECG. In addition, 3 MoH doctors will be responsible for monitoring of the implementation. The study team will ensure correct implementation and follow-up of the study procedures and will discuss study developments and all complex cases with the reference person (HIV/TB adviser) at the head office level (MSF OCA).

4. Study participants and sample sizeThe study population will constitute all consecutive presumptive MDR TB patients (adults) identified by standard diagnostic protocols in Uzbekistan screening with molecular tests or with MGIT-culture/first line drugs DST (in case of initial Xpert® MTB/RIF negative). Children suspected of MDR TB diagnosis who are close contacts of a confirmed MDR TB patient at the

three participating study sites will also be enrolled. Participants will fulfill the inclusion/exclusion criteria outlined below as eligible for taking the short MDR TB treatment regimen and fill an informed consent form.The following patients will be offered inclusion in the study (as described in the research protocol): New presumptively diagnosed MDR TB patients (adults and children) with Xpert® MTB/RIF

or Hain MTBDR, or confirmed with Hain MTBDR plus on positive cultures if initial molecular tests negative or confirmed from MGIT culture/DST if initial molecular tests negative;

Children (<14 yo) suspected of MDR TB without bacteriological confirmation but documented as a close contact of a confirmed MDR TB patient;AND

Informed consent to participate in the study signed by the patient or the responsible caretaker for patients <16 years old (as per national legislation).

Only patients with a history of prior treatment with second line anti-TB drugs for less than one month will be eligible for inclusion. Patients will be included regardless of HIV status.Exclusion criteria at baseline: Baseline contraindications to any medications of the study regimen medications, where

benefits of the regimen do not outweigh the risks as judged by treating physician; Severe renal insufficiency with estimated creatinine clearance of <30 ml/min at baseline

(calculated with Cockcroft-Gault formula); Patients with extrapulmonary TB only (without involvement of lung parenchyma) Patients with documented ofloxacin resistance Patients with XDR TB (additional resistance to SLD kanamycin (or capreomycin) AND

ofloxacin); Patients with resistance to both Km and Cm. Critically ill and in the judgement of the treating physician unlikely to survive more than 1

week (these patients may still be commenced on standard MDR TB treatment according to the Karakalpakstan comprehensive TB treatment guidelines)

Has one or more of the following risk factors for QTc prolongation:

o A confirmed prolongation of QTc interval (Fridericia formula), e.g., repeated demonstration of QTcF (Fridericia correction) interval > 500 ms in the screening ECG (i.e., retesting to reassess eligibility will be allowed once using an unscheduled visit during the screening phase);

Of note, pregnancy and breastfeeding are not exclusion criteria. Consideration to treatment initiation after the first trimester (12 weeks of pregnancy) as it is done with the standard 20+ month regimen and comprehensive information and counselling of risks and benefits will be offered to pregnant women. We decided to include pregnant women because the alternative is a longer regimen with similar toxicity, risks and similar safety class drugs and with limited evidence of safety as well. All cases of pregnancy /breastfeeding and pediatric cases will be

discussed with the TB / HIV adviser and a mini-consilium consisting of the medical team leader, the Head pediatric TB doctor for Karakalpakstan and the chief physician for TB2 inpatient facility

Withdrawal from study and analysis:All patients commenced on the 9 month regimen and thereafter with full DST found to be either: Non MDR TB (eg drug sensitive, mono or poly drug resistant TB) or resistance to Ofx (with or without injectable resistance) with MGIT culture and FLD/SLD-DST

or Showing resistance to both Km and Cm. In case of resistance only to Km the regimen will be continued with the use of Cm at the same dose as Km and this adjustment of regimen will not be considered as a change of regimen and therefore will not be a cause of withdrawal.

Withdrawal from study but included in intention to treat analysis: Withdrawal of consent to participate in the study. Severe adverse event (SAE) necessitating change of regimen. Transfer out patients to centers where short regimen cannot be provided.

Patients that withdraw from the study will be offered an adjusted treatment regimen according to DST pattern and following international WHO recommendations.

Sample sizeThe calculation of sample size is based on the expected rate of favorable treatment response of documented TB (defined as cured and absence of relapse at 1 year after treatment completion). The success rate with the current treatment (20+ months) in Karakalpakstan is on average 60% in each annual cohort. The relapse rate for MDR TB with current international standard treatment has been reported to be between 1.1 and 10% depending on duration of follow up – with an average of 2%. Therefore current standard MDR TB treatment has a favourable treatment response of 58%. We hypothesized that on intention-to-treat patients that the short course MDR TB treatment regimen would have a successful outcome (cure and complete) above 75% and that the relapse rate at 1 year post treatment would be about 3%, giving a favourable treatment response of 72%. We expect a study loss to follow up (study withdrawal plus patients lost to follow up during the 1 year post treatment follow up) of 15%.

Cohort sample size with 95% confidence interval for different expected response rates and power:

Favourable Treatment response

Power Alpha Cohort size + withdraw rate (15%)

0.7 0.80 0.05 128 1470.72 0.80 0.05 93 1070.75 0.80 0.05 62 71

Taking all that into account, we propose an initial sample size of at least 110 MDR TB patients assuming a withdrawal rate of 15% for an estimated favorable treatment response 1 year post treatment of 72%.

Currently, the average number of MDR TB patients is estimated to be 8-12 per month in all sites. Based on that, the estimated period for recruitment will be for a minimum of 10-12 months (from April 2013) with patients enrolled consecutively.

5. Study endpointsPrimary Rate of success (cure and treatment completion) at the end of short course MDR TB

treatment. Rate of relapse at 12 months after treatment completion.

Secondary Rates of cure and success per groups of MTB resistance patterns. Occurrence, type and severity of treatment-related side effects. Treatment interruptions and modifications of treatment. Unfavorable outcomes (default, death, failure) while on treatment. Risk factors for unfavorable outcome (death, default, failure) including rate of inhA and katG

mutations in MDR TB samples and rates of correlation with DST for high dose isoniazid. Rate of correlation smear microscopy/culture during treatment monitoring. Rate of resistance amplification amongst patients with outcome failure.

Interim analysisThe program will be routinely monitored producing quarterly results on sputum culture conversion rates and outcomes. We define a minimum acceptable level of culture conversion below which we will consider study interruption which is 60% for the overall cohort at 4 months, based on culture conversion rates at 4 months with standard MDR TB treatment in similar cohorts (Holtz 2006, Kurbatova 2012). Sputum culture conversion rate at 4 and 6 months after commencement of treatment for the first cohort of 50 patients will offer interim analysis with acceptable 95% confidence intervals.

6. Case Definitionsa) Diagnosis of MDR TB

Diagnosis of MDR TB will be done according to current program algorithms. In summary all patients suspected of TB will be referred for sputum sample screening with Xpert MTB/RIF. Patients with rifampicin resistance detected by the Xpert MTB/RIF assay, or patients with Xpert MTB-negative but culture positive isolates which show rifampicin resistance on Hain MTBDR plus or phenotypic DST will be considered as cases. One sputum sample will be tested with Xpert MTB/RIF per patient. If a suspected patient is smear negative and Xpert® MTB/RIF negative, the diagnostic investigation will follow the comprehensive TB guidelines algorithm for Karakalpakstan for TB diagnosis (algorithm included in short course protocol: see Annex 1). All patients identified with rifampicin resistance by Xpert® MTB/RIF will be further evaluated with Genotype® MTBDR sl2 and 2 sputum samples will be referred for confirmation with reference standard conventional liquid culture (MGIT) and first line and second line DST. Culture and phenotypic DST will be considered the reference standard for confirmation or exclusion of MDR TB diagnosis. Genotype® MTBDR sl given the lower sensitivity and specificity for injectable agents will only be used for ofloxacin resistance. All patients with ofloxacin resistance on Genotype® MTBDR sl if in stable clinical condition will be eligible for enrolment based on FLD/SLD DST results, otherwise they will be commenced on standard MDR TB treatment. In this context, it is predicted that 3 of 100 patients will have ofloxacin resistance and that the Genotype® MTBDR sl will detect 2 of these patients with 1 case of false negative test that will only be detected by conventional DST 2-4 weeks after commencement of the short course regimen.

Rifampicin resistance in rapid molecular testing (Xpert® MTB/RIF) will be used as a proxy for MDR TB diagnosis. After rapid molecular testing and presumptive diagnosis, patients will start an empirical short course MDR TB regimen. Given the high prevalence of rifampicin resistance in Karakalpakstan, rifampicin testing with Xpert® MTB/RIF has a positive predictive value for rifampicin resistance of over 90%, and rifampicin will not be included in the empiric regimen. (see treatment protocol in Annex 1).

Patients shall be classified as smear negative or smear positive. This classification should be based on a sputum result performed in the 30 days prior to treatment or within 7 days of commencement. Drug susceptibility testing should be less than 90 days old and should be repeated if the patient has had any treatment in the interim. If a patient has not submitted a sputum sample, then this should be repeated prior to treatment.

2 Rapid molecular test for ofloxacin and injectable resistance.

Diagnosis of TB in children may be very difficult and bacteriological confirmation is sometimes impossible, despite all available means, thus diagnosed in children will be made according to the following criteria (Annex 1):

1. Bacteriological confirmation from 2 sputum samples screened with Xpert® MTB/RIF and obtaining result: MTB+/Rif+ or with positive MGIT culture/DST of Xpert® MTB/RIF negative patients with clinical suspicion of TB. Sputum samples will be obtained by spontaneous expectoration or sputum induction.

2. Children with clinically suspected TB who are contacts of MDR TB phenotypically-confirmed patients.

Patients will be classified based on WHO criteria according to:1. History of previous TB drugs use as new or previously treated patients with first or second

line drugs.2. Previously treated patients will be classified according to previous treatment outcome as

relapse, treatment after default, treatment after failure of category 1, treatment after failure of category 2, transfer in or other.

3. Localization of the TB: pulmonary or extra-pulmonary; Patients with extra-pulmonary TB alone will be excluded from the short course

regimen and will be offered standard MDR TB treatment Patients with severe form of extra-pulmonary TB e.g TB meningitis, bone TB and

Pulmonary TB simultaneously will also be excluded and treated with standard MDR TB treatment.

However, patients who have less severe form of extra-pulmonary TB (eg Lymph node TB) and pulmonary TB are not excluded and will be offered short regimen.

b) Diagnosis of HIV infectionHIV infection will be diagnosed according to the current national protocol in Uzbekistan.

7. Treatment optionsa) Treatment of MDR TB

The MDR TB regimen prescribed will be:

Intensive phase: Pyrazinamide (Z) + Ethambutol (E) + Isoniazid (H) + Moxifloxacin (Mfx) + Capreomycin (Cm) + Prothionamide (Pto) + Clofazimine (Cfz) for at least 4 months and until one negative culture is documented with a maximum of 6 months duration.

Continuation phase:

Continuation phase of Pyrazinamide (Z) + Ethambutol (E) + Moxifloxacin (Mfx) + Prothionamide (Pto) + Clofazimine (Cfz) for fixed 5 months duration.

The regimen and dosage follows the regimen used in Bangladesh with no modifications, with the exception of including Pto during the entire duration of treatment as implemented in West African countries, and the replacement of Gfx with Mfx as 4th generation quinolones. We will not use Gfx as it is not broadly available in the market due to concerns about its safety. Documentation of dosages (adult and pediatric) is detailed in the clinical protocol (Annex 1). The dose of H is prescribed as in the regimen used in Bangladesh is a higher dose than conventional use. Pyridoxine will be added to the regimen for prophylaxis of peripheral neuropathy as described in the clinical protocol.

All patients will initiate this regimen and response will be monitored with monthly smear microscopy and culture. From the experience with this regimen in other settings, patients show earlier culture conversion. We aim for a short intensive phase of 4 months as in the regimen used in Bangladesh, but with the possibility to offer a prolonged intensive phase to a maximum of 6 months. Intensive phase duration will be a minimum of 4 months. If a patient is smear positive at 3 months the intensive phase will be prolonged until 5 months. If the patient remains smear positive and is culture positive at 5 months then the patient will be declared as a failure and will have a repeat DST and be enrolled in standard MDR TB treatment. The regimen used in Bangladesh based the switch to continuation phase on smear conversion. However, as monthly culture will be performed for monitoring, these results will be used during the study for patients who do not become smear negative and a guide of interpretation of smear and culture monitoring results is included in the clinical protocol (annex 1). During the study, further confirmation of correlation between smear microscopy and culture for monitoring purposes will be examined. In case a patient is unable to produce a sputum sample for monitoring due to good clinical evolution, switch to continuation phase will be done at 4 months after treatment initiation. In case a patient has 2 consecutive positive cultures after becoming culture negative after 4 months of treatment, then the patient will be given the outcome of failure (see below outcomes definitions) and the patient will stop the study regimen. In case of one isolated positive culture, if the patient shows clinical good evolution, the patient will continue the regimen awaiting results of a second culture. Patients with the outcome of failure will be counselled and offered continuation of treatment with a regimen adjusted based on repeat DST, previous drug treatment history, national guidelines and consilium, study team and TB/HIV adviser recommendations. Provision of drugs for treatment of those patients will also be assured. Outcomes of these patients will be documented.

In case of confirmed resistance to Km but sensitive to Cm, patients will continue with Cm and will continue a shortened regimen; outcomes will also be documented for this subgroup of patients. In case of confirmed resistance to Cm but sensitivity to Km the patient can be switched to Km and continue the shortened regimen. Such cases should be discussed in consilium and with the HIV/TB advisor to determine if a prolonged intensive phase is required.

Patients with confirmation of ofloxacin resistance or XDR TB (documented resistance to Km or Cm and Ofx) will be withdrawn from the short regimen and removed from the study to initiate a more appropriate regimen adjusted to the resistance pattern observed.

b) Treatment of HIVAll HIV/MDR TB co-infected patients who are ART naïve will be started on first line ART following national and international HIV care and ART recommendations. All included HIV co-infected patients not yet on ART will start ART as soon as possible, between 2 weeks and 2 months after tolerating TB treatment. The preferred first option for ART will avoid the use of tenofovir due to the increased risk of renal toxicity in co-administration with aminoglycosides. AZT will be preferred over D4T due to the risks of peripheral neuropathy, except in cases of severe anemia (Hb <8).

Patients already on ART at the moment of MDR TB diagnosis will be investigated for ART failure in collaboration with the MoH HIV-centre in Nukus. This will include HIV RNA VL and following the ART failure diagnostic algorithm following international and national recommendations. A second HIV RNA VL will be repeated after 3 months (after adherence counseling and effective MDR TB treatment initiation) and if ART failure confirmed patients will be initiated on second line ART.

HIV treatment and prevention/treatment of other opportunistic infections will be provided and recorded in the database following national and international recommendations.

8. Monitoring of adverse events/reactionsMonitoring for adverse events associated with the use of the proposed regimen in programmatic conditions in a high MDR /XDR TB context will be done. Adverse event data will be collected in a systematic, standardized format based on regular follow up visits and lab monitoring. The format used will be based on current programme side effect monitoring forms. (see Annex 3) The type, severity and possible relationship of side effects will be assessed according to the DAIDS grading score (Annex 2) adapted for the agents used in the current study and for side effects recorded at greater than 1% frequency in MDR TB treatment (current

standard treatment and experience reported from the cohorts using short course MDR TB treatment).

The following definitions will be used for adverse events and reactions:Adverse event (AE): any medical occurrence in a subject to whom a medical product has been administered including occurrences which are not necessarily caused by or related to that product. An AE can therefore be any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of a medicinal (investigational) agent, whether or not related to the medicinal (investigational) agent.Adverse reaction (AR): any unintended response to a medical product which is related to any dose administered to the subject.Serious Adverse Event (SAE) and Serious Adverse Reaction (SAR): any event or reaction that results in death, is life-threatening, requires hospitalization or prolongation of existing hospitalization or results in persistent or significant disability or incapacity.

The decision to postpone, modify or to interrupt/discontinue the regimen will be left at the discretion of the clinical team following the patient in discussion with advisors. Some safety rules will be pre-established and standardized in the protocol of MDR TB short regimen management (Annex 1). According to expected toxicities suggestions for alternative management will be provided based on the current evidence and recommendations from the WHO “Guidelines for the programmatic management of drug-resistant tuberculosis. Emergency update 2008”. Any alteration in the scheduled regimen will be thoroughly registered and justified in the patients’ file and database.

Contraception will be systematically offered to the female participants during the treatment period, and a pregnancy test will be performed systematically before starting MDR TB treatment and at each treatment visit if there is uncertainty about consistency of correct contraceptive use. Contraceptive choice will be documented at baseline. The need to pursue (or to stop) the treatment in case of pregnancy will be evaluated by the clinical team and advisors on an individual basis, according to the benefit/risk balance at that moment. Women who reject hormonal contraception are not excluded from the study.

9. Assessment of treatment effectivenessWe will use adjusted international WHO definitions to classify the type of response to TB treatment (see below). An assessment will be performed at the end of the treatment period, and a final assessment will take place at the end of the follow-up (12 months after completion of the treatment).

Sputum conversion will be defined as two consecutive sets of negative smears (smear conversion) and cultures (culture conversion), from samples collected at least 30 days apart. The date of the collection of the first set of negative cultures and smears will be used as the date of conversion. Both bacteriological techniques (microscopy and culture) will be monitored and documented throughout the treatment. The smear will be the primary means of determining the duration of the intensive phase as described above. For patients who remain smear positive, then culture results will also be taken into account. However, both smear and culture conversion will be analysed for the study secondary end point analysis on a quarterly basis, and correlation of both measures with primary outcomes will be done at the conclusion of the study.

The TB treatment outcome at the end of the treatment period is defined as follows (adjusted outcomes definition for MDR TB programs WHO 2008, emergency update of guidelines for the programmatic management of drug resistant tuberculosis):

Cure : An MDR TB patient who has completed the treatment according to programme protocol and has at least four negative cultures from samples collected at least 30 days apart within the final 5 months of treatment and if there is a positive culture it is followed by a minimum of three consecutive negative cultures taken at least 30 days apart in the final 3 months of treatment.

Treatment complete: An MDR TB patient who has completed treatment according to programme protocol but does not meet the definition for cure because of lack of bacteriological results (i.e. fewer than five cultures were performed in the final months of treatment) or otherwise, completion of treatment with documented bacteriological conversion persisting through the end of treatment, but fewer than five negative cultures. Treatment completion will only be an outcome for patients that are not able to produce sputum; in case of patients where the lack of bacteriological results is due to other reasons the outcome will be registered as “other” in order to avoid misclassification.

Treatment outcome “other”: An MDR TB patient who has completed treatment according to programme protocol but does not meet the definition for cure because of lack of bacteriological results due to programmatic reasons (reasons other than the lack of patient’s ability to produce sputum) such as culture contamination or no timely referral of sample by the clinician, the outcome will be registered as “other” in order to avoid misclassification. In case of contamination of the culture tube, new sputum samples for culture will be collected and culture tubes de-contaminated and re-inoculated following standard laboratory procedures.

Death: An MDR TB patient who dies for any reason during the course of MDR TB treatment and is not already classified as a treatment failure prior to death. Assumed causes of death will be recorded.

Failure: Treatment will be considered failed when there is absence of bacteriological response that will be defined as follows: Patient fails to show culture negative by the end of month 5 of a prolonged intensive phase. Culture positive during the continuation phase: two cultures positive during the

continuation phase or one culture positive during the last 3 months of treatment. Treatment will also be considered to have failed if a clinical decision has been made to

terminate treatment early because of poor clinical or radiological response or adverse events (including due to prolonged QTc on ECG according to criteria defined in ECG section below) where the team decides the regimen is failing and treatment is changed. These latter failures can be indicated separately in order to do sub-analysis.

All failures with a documented positive culture will have DST and investigation of resistance to document the rate of resistance amplification. Patients with a sputum sample collected after the commencement of treatment that shows a positive culture and a DST demonstrating amplification of resistance with either ofloxacin resistance or with capreomycin/kanamycin resistance will be declared failures.

Default: An MDR TB patient whose treatment was interrupted for two or more consecutive months for any reason without medical approval and not meeting the criteria for failure.

Transfer out: An MDR TB patient who has been transferred to another reporting and recording unit and for whom the treatment outcome is unknown. Patients that require a transfer out will be informed that it is very unlikely that they can continue the same regimen and they will have to change to standard MDR TB regimen. In case that the treatment can be provided in the receiving center and the outcome documented, this will be recorded.

Favourable versus Unfavourable OutcomesOutcome will be considered as favorable in case of cure or treatment completion. All other outcomes at the end of treatment (default, death, failure) will be considered as unfavorable.

At the end of the study follow up period the final outcome will be assessed and defined as:

Relapse-free: An MDR TB patient who meets the criteria of cured or completed short course of treatment and remains asymptomatic at the end of the follow up period (one year after treatment completion).

Relapse: An MDR TB patient who meets the criteria of cured or completed short course of treatment and at any time during the follow up period (first year after treatment completion) is subsequently diagnosed with at least one sample of bacteriologically positive MDR TB by culture and DST.

10. Study procedure and patient flow The study protocol will be explained to all physicians and nurses in all facilities in the study districts (Shumunay, Kegeily and Nukus City). They will confirm the clinical diagnosis of patients and together with the pilot project coordinator and rayon’s team will decide on patients’ eligibility for inclusion in the study. Informed consent will be obtained at that moment.

The patient flow is described in the clinical protocol (Annex 1).

Informed consent procedure All patients at the moment of rifampicin resistance diagnosis and prior to MDR TB treatment initiation will be offered the informed consent with thorough explanation on the study and their invitation to participate in the study.

Baseline visit at MDR TB diagnosisAt baseline, the clinical history and physical examination will be evaluated and described in detail. Other baseline clinical data will consist of HIV information (negative or positive, date of HIV diagnosis, HIV WHO staging), co-morbidities, and ongoing therapies. A chest X-ray will be performed at baseline. Laboratory examinations will include molecular DST (Xpert® MTB/RIF or for patients with negative Xpert® MTB/RIF and a positive culture then Hain MTBDR plus on the positive culture; Genotype® MTBDR sl), sputum smear, sputum culture/DST and in addition basic parameters (full blood count, liver transaminases, electrolytes, TSH, glucose and kidney function), as well as HIV parameters (CD4 cell count baseline for all and HIV RNA viral load in patients that have been at least 6 months on ART) and hepatitis B and C tests. Pregnancy testing will be performed systematically, and contraception will be offered to female patients during the whole treatment period.

Daily Observed Therapy (DOT)Treatment will be administered and observed 7 days a week either at the selected outpatient clinics (polyclinic or DOT corner) or at the home of the patient. All patients with severe clinical conditions will be hospitalized in TB2 hospital if smear positive or the smear negative ward (TB1/3) in Nukus, and DOT will be continued in that setting. Data on adherence and pill intake

will be recorded. Nurses trained in the study protocol will assure daily observed therapy seven days per week. Support for adherence will be done according to standard protocols in Uzbekistan, with a nurse educator ensuring treatment education and adherence support.

Follow up visit during treatmentPatients will do clinical follow up with a doctor at 2 weeks after MDR TB treatment initiation and then monthly until treatment completion. At each visit, clinical assessment with evaluation of side effects and a set of laboratory tests will be performed according to schedule and clinical protocol (see table below and Annex 1). Evolution will be assessed by the doctor and/or nurse with a specific data collection form (drug-o-gram or treatment card) to assess treatment response and complications.

Any relevant clinical event (adverse events or reactions) and any required additional diagnostic testing and/or therapy will be recorded. In case of unexpected events, study participants will be advised where and to whom to present in emergency and they will have an emergency phone number to call.

Patients will receive the same incentives provided currently in the standard comprehensive TB protocol which includes some nutritional support for the patient and family. Patients will attend the designated DOT corner to receive daily medication, or if the patient is living far from the facility or unable to walk to the DOT corner then home based care will be provided.

Follow-up after treatment completion/outcome visitsAll patients completing the prescribed 9-11 months treatment will be followed up until 12 months after treatment completion. A follow-up visit will be planned at 6 months after treatment completion (or at any time earlier in case of re-occurrence of symptoms) for clinical assessment and a final visit will take place at month 12 post-completion to assess the final outcome (relapse).

The examinations during the study visits are summarized in the following table:Baseline visit1st visit

Follow-up during treatment After finalizing MDR TB treatmentFollow-up/outcome visits6 months after treatment completion

12 months after treatment completion

2weeks

1M

2M

3M 4M-end IP CP

Clinical assessmentHistory &Physical examination (including weight)

X x monthly x x

Evaluation side effects

At every clinical consultation x x

Outcome assessment At end of treatment xLaboratoryXpert® MTB/Rif X In case of

relapseIn case of relapse

Genotype MTBDR sl X In case of positive culture after culture conversion In case of relapse

In case of relapse

Smear X x x x Monthly Monthly

x x

Culture X x x x Monthly Monthly

In case of relapse

In case of relapse

DST (1st and 2nd line) X If +veCulture

If positive culture after conversion

In case of relapse

In case of relapse

ECG X x x If develops syncope or dizzinessFull Blood Count X If indicated clinically In case of

relapseIn case of relapse

Creatinine* X x x x End IP End CP If clinically indicatedPotassium X x x x Monthly

until end IP

ALT X x x x Monthly until end IP

Glucose X Monthly if elevated at baseline

x x 3-monthly

TSH Perform once at 6 months and if patient has symptoms/signs suggestive of hypothyroidismHepBs Antigen X

Repeat only if indicatedHep C Antibodies XHIV XPregnancy test X If negative, offer to repeat if indicatedIf HIV+, CD4 X At 6M and 12M after ART initiation xIf HIV+, RNA VL** X At 6M and 12M after ART initiation xOther complementary examsChest X-ray X Repeat only if clinically

indicatedEnd CP In case of

relapseIn case of relapse

Hearing test clinical X Monthly clinical assessment during IP – refer for audiometry if indicatedHearing audiometry X For patients with suspected hearing loss

*for patients with higher risk of renal insufficiency the monitoring of Creatinine may need to be more frequent. This includes diabetes, elderly and baseline renal dysfunction and in case of co-treatment with TDF.**if on ART for at least 6 months at MDR TB treatment initiation.

11. Laboratory tests and electrocardiogramHematology, biochemistry, HIV test, CD4 and molecular DST test (Xpert® MTB/RIF assay / Hain MTBDR plus/ Genotype MTBDR sl) will be performed. MGIT culture and DST for first and second line drugs will be performed at the Nukus Laboratory as per the current functioning of the programme. The Nukus Mycobacteriology laboratory is proficiency tested for microscopy, culture and first and second line TB drug DST through the supranational reference laboratory in Gauting, Germany, via the Tashkent NRL. All sputum smear examination is also performed by fluorescence microscopy at the Nukus laboratory prior to culture. All samples will be tested in the Nukus laboratory and no samples will be exported out of the country.

ElectrocardiogramBoth moxifloxacin and clofazimine may potentially prolong the QT interval as measured on an electrocardiogram. Prolonged QT interval in the presence of some drugs has been associated with torsade de pointes, which is a life threatening arrhythmia. However, a direct link between QT prolongation and torsade de pointe arrhythmia has not been established, and prolonged QT can occur without increased risk of arrhythmia (Giorgi 2010).

While the death rates in the cohort studies of the short course regimen that have been reported are low, it is not known whether this will be the case in other populations. In Uzbekistan, the combination of clofazimine and moxifloxacin has been used in drug regimens for several years for the treatment of XDR TB and some patients failing MDR TB treatment. Anecdotally there have not been cases of sudden death that could be due to arrhythmia, however there has not been specific monitoring with ECG and it is possible that if cases did occur they would have been missed. As there is no universal threshold for identifying drug-induced QTc prolongation, and given the experience of combining these 2 drugs together already (with a higher dose of clofazimine), the thresholds chosen for QTc will be when the QTc exceeds >60ms compared with the baseline ECG QTc or when the QTc prolongs beyond 500 ms, as these are thresholds that have been associated with the highest risk (Fenichel 2004). The QTc will be calculated using the Fridericia’s formula which corrects for the heart rate and has been shown to be more accurate at slower or faster heart rates than other correction formulae.

As one of the secondary objectives is to describe the rate of adverse events, a baseline ECG, an ECG at two weeks and at 1 month will be performed. The ECG machine will be portable allowing it to be performed in patients’ homes or DOT corners. Patients with an extension of the QT interval from baseline of greater than 60 ms or a QTc of greater than 500 ms will have

treatment withheld and will be discussed with the study coordinator and TB/HIV adviser. Reversible causes such as low potassium and magnesium will be checked for and corrected and a decision about stopping treatment or continuing with weekly ECG monitoring will be made based on whether the QTc persists above the thresholds despite correction of reversible factors.

12. VariablesThe following variables will be collected and monitored (see Annex 3 data collection forms). The follow-up treatment card (or drug-o-gram) is the standard monitoring tool for the follow-up of patients on MDR TB treatment. These variables will allow monitoring the treatment response, safety and risk factors for unsatisfactory response:

Demographic variables : age, gender. Physical examination : weight, height, BMI, hearing test (clinical and audiometry). TB : Date of diagnosis, date of treatment initiation, localization, dates of

laboratory tests, TB classification according to treatment history, adherence, previous TB and MDR TB regimen drugs.

HIV status and date of HIV diagnosis, cotrimoxazole and other prophylactic treatments, ART, type of ART, ART starting date, adherence, and ART outcome.

Co-morbidities including diabetes mellitus and opportunistic infections. Clinical diagnosis/complications . Chest X-ray abnormalities. Laboratory : TB smear, molecular test results (Xpert® MTB/RIF, Hain MTBDRplus/

Genotype MTBDR sl), MGIT culture, DST, Hb, CD4 cell count (if HIV infected), ALT, Creatinine, Potassium, pregnancy test, TSH, glucose, Hepatitis Bs Antigen, Hepatitis C antibodies.

Side effects : hematologic, peripheral neuropathy, optic neuropathy, gastrointestinal effects, fever, dermatologic events, renal, hepatic, audiologic, thyroid, psychiatric (including depression)

TB Outcomes : culture conversion, cure, completion, death, failure, default, transfer out and relapse as per standard WHO definitions.

Documented cause of death .

13. Data collectionData collection forms will be used (see Annex 3). Data will be collected at baseline and at each study visit. Data from these forms will be entered into a separate MSF MDR TB database (Koch’6) that is already in use in the project. HIV-related data is not foreseen to be collected in Koch’6, it will be exported and cross matched from a separate database that will collect all the HIV-related data during the study. Data will be exported from both sources and merged into a single dataset in Stata for statistical analysis. All databases are password protected, and will have backups without patient names stored by the epidemiologist in headquarters MSF OCA.

14. Statistical AnalysisStatistical analysis will be done in Stata 11.0 statistical package. During the analysis, withdrawal from study (as described above) will be handled in 2 ways: an intention-to-treat analysis where patients who are withdrawn from the study will be considered as treatment failures and a per-protocol analysis where withdrawn patients are excluded from the analysis.

The following statistical analyses will be done: 1. Analysis of rates of treatment outcomes.2. Rates of relapse.3. Survival analysis for mortality and relapse to obtain Kaplan-Meier survival graphs for

mortality and time to relapse.

For the secondary objectives the following analyses will be done:4. Rates of sputum conversion and survival analysis for time to culture conversion.5. Rates and severity of side effects.6. Univariate analysis to assess risk factors for unfavorable response and confounding

factors will be investigated and adjusted in a multifactorial model (Cox proportional hazards model). Hazards Ratio will be investigated for each risk factor with a 95% confidence interval and a level of significance of 0.05.

Interim analyses of culture conversion, unfavourable outcomes and adverse events will be done on the first cohort of 50 patients with completion of the intensive phase, this data will be reviewed by the WHO Europe epidemiology unit. In addition, quarterly data will be reviewed by the MoH of Uzbekistan monitoring group for the short course MDR TB treatment (Professor Nazirov from Tashkent TB Institute, Sapaev Nurmukhamed (TB Coordinator from DOTS Center, KK), Turaev Laziz (Lab.specialist, Tashkent).

ETHICAL ISSUES

1. Ethical CommitteeThe study will be conducted following approval of both ethical committees of Médecins sans Frontières and Uzbekistan.

2. Consent formsAll patients with suspected TB will undergo standard counseling with explanation of TB, TB treatment and drug resistant TB and patients with diagnosed TB are required to sign an informed consent form as per standard practice in the Karakalpakstan programme (annex 5). In addition, a written informed consent form in English and local languages (Karakalpak/Russian) will be given, explained and read to the patients prior to the enrolment in the study (Annex 4). Only patients from whom a signed (or fingerprint for those not able to write their name)

informed study consent form has been obtained will be included in the study. For children <16 years old, informed consent will be obtained from the legal guardian. The information will include the aim of the study, data collection procedures, potential benefits and risks, and assurance of confidentiality. The consent process will make clear that it is the individual’s decision to participate in the study or not, and this will not affect the quality of their care – however as the short course MDR TB treatment is only provided with the study setting a decision not to participate in the study will mean that the standard 20+ month treatment will be provided. Similarly they will be free to withdraw from the study at any time. Patients refusing participation in the study or use of data for the study will not be eligible for a shortened regimen of MDR TB, but will be able to access standard MDR TB treatment.

3. Expected risks and advantagesThe study regimen uses first and second line MDR TB drugs which are recommended for use in DR-TB treatment. These drugs are well known to cause side effects, some of which may be severe. However, Karakalpakstan already has senior MoH clinicians in the programme having experience since 2003 with these drugs and MDR TB treatment according to international guidelines. In addition, the current evidence we have is that the side effects occur less often with the study regimen than with the longer WHO recommended regimen. There is a risk of interaction of the TB drugs with ARVs; this risk applies to both the standard regimen and the study regimen and is judged to not be worse with the study regimen. There is a risk of amplification of resistance if the regimen is not effective and in those cases the patient will be given an outcome of failure and treatment duration will have to be 20+ months after adjustment of the regimen according to DST and expert advice (discussion with MSF TB adviser and Uzbekistan expert consilium); however, this risk also exists with the current standard of care with failure rates close to 10% requiring treatment regimen adaptation, in addition to high toxicity and long duration.

There is a risk that this shorter regimen is not as efficacious in this context with high second line drug usage and resistance rates; however the limited data available show that even in patients with ofloxacin resistance the regimen still has acceptable, albeit lower success rates, and relapse-free rates from over 200 patients are very low. This risk will be managed by interim monitoring of time to culture conversion and clinical response with overview by the data safety review team. In addition, one should not lose sight of the far from ideal outcomes of the current standard of care regimen with known high death, failure and loss to follow up rates. In discussion with international external experts, WHO Europe region and experts from MSF and MoH Uzbekistan it has been decided to exclude patients with ofloxacin resistance from this pilot despite reported rates of over 70% success in this group treated in Bangladesh due to the

very small numbers of patients and also the high rates of other first and second line drug resistance (Z,E, Pto, high level isoniazid, Km) in Uzbekistan.

Study participants will be required to undergo 2 additional visits after treatment completion in order to determine relapse-free survival time, however in total this will be less than the total visits and similar duration of follow up if the patient underwent standard of care in the current programme.

The individual patients who participate in the study may have substantial benefit with a greatly reduced MDR TB treatment duration with likely less chance of severe adverse events. The improved tolerability and shorter treatment time are presumed to lead to decreased defaulter rates, which under the current standard of care treatment in the MSF/MoH programme have consistently been 20% or higher despite increased attention and programme measures to improve patient support and adherence.

This study also has the potential to improve treatment for the community which faces a high burden of MDR TB. The potential for policy change within Uzbekistan and other countries in the region is high if the results from this study and other accumulating evidence from other cohorts support the early current available evidence. There is also the potential to reduce programme treatment costs and therefore increase access to treatment for patients not able to be treated due to the cost of the current regimen which prevents adequate scale up of treatment in order to meet the overwhelming needs. The early engagement of the NTP at the concept stage, will help ensure that the MoH will review the results with a view to changing national protocol if the results are positive.

4. ConfidentialityAll staff working at each clinical site will be trained in the importance of patient confidentiality. Patient names will be recorded in the clinical forms for the purpose of the clinical follow up. Patients will be identified by a unique ID number from the outset of the study and this ID number will be used for all study databases and reports.

5. Advocacy and Community Involvement5.1 Advocacy

Meeting with the authorities and important stakeholders to provide information about the study that is going to be launched in the specified rayons. Deputy of Chairman of Supreme Councils of Karakalpakstan, Director/Deputy Director of TB Institute, Tashkent ,MoH Chief TB Doctor Tashkent, Minister of Health/deputy Minister of Health, Director of DOTS center, MoH Chief TB Doctor, MoH Chief TB Paediatrician, Rayon Chief Doctors, MoH Rayon Supervisors, Rayon TB Doctors, SVP Doctors, TB nurses.

The above mentioned authorities have been involved in the preparatory discussions for this project and will be informed of the launching of the study either through letter or through a launch announcement ceremony.

5.2 Community InvolvementThe duration of current treatment and the side effects are common complaints from patients within the treatment programme, and this pilot study represents a collaborative effort from MSF and MoH to try to respond safely to patient concerns. The rationale for the pilot project will be presented to community members including current and past MDR TB patients by MSF counsellors.

Key community stakeholders and influential members of the local community are the Hakim of the Rayon and the deputy of the Hakim on social issues, and community leaders of each village in the Rayon. It will be important that before the project is launched in the selected rayons, meeting sessions will be arranged with these people and they will be briefed by the Pilot Project Coordinator and Chief Doctor of the Rayon. It will be emphasized that this is a collaborative MSF/MoH effort. The importance of this pilot project will be shared along with future benefits and risks involved and their continuous support will be requested. Ongoing informal discussions with people who have previously or are currently undergoing TB in the project have been occurring and will continue to occur throughout the project.

Dissemination of a summary of the results in Karakalpak after completion of the study will occur through clinic sites. Final dissemination of results will be shared with study participants and patients with TB and other important community members

BUDGETThe majority of costs of implementation of the project will be covered by the available funding of the project. All drugs required for implementation of the protocol are already present in the mission, and extra quantity of certain drugs has been ordered. There are very few additional expenses related to the study that differ from the current program implementation are detailed in the table below. During the study period, MSF OCA will provide all the required resources (diagnostic tests, TB drugs and additional human resources) which will be required in addition to the ongoing services. Note that there may actually be some cost savings for patients treated with a shorter regimen, but we have not included that as part of the budget calculation. The listed extra budget is available and will be provided by MSF.

Nº Type of expense Quantity Cost Total Euro1 Second line drugs 130 1000 USD per treatment 130,0002 Presentation study results to

health authorities1 3000 USD 3000

3 Additional staff – additional pilot project nurse in each district

2 25,000 50,000

4 Dissemination results in international conference and journal

1 3000 USD 3000

Expected total budget/site 186,000

REVISED TIMELINE Oct 2012 – Jan 2012: Development of the study protocol. Jan 2013: Submission to ethical committees (MSF and Uzbekistan). September 2013: Commence patient inclusion until sample size achieved. May 2014: Anticipated enrolment of first Nukus City patient December 2014 – Feb 2015: Anticipated final enrolment for 9 month project January-March 2016: end of 12 months follow up after treatment completion of last patient

included. Jan-April 2016: data analysis. Mid 2016: dissemination of results in Uzbekistan and publication.

DISSEMINATION OF STUDY RESULTSThe preliminary results of this research will be presented as soon as possible in Uzbekistan and be used to discuss with key stakeholders how to further improve MDR TB treatment and management. The preliminary and final results of the research will be presented, shared and discussed with the MSF international TB working group. The results will be published in a peer-reviewed journal, preferably open access, to allow the widest possible access and benefit to health practitioners internationally.

REFERENCES1. Ait-Khaled N, et al. Management of tuberculosis: a guide to the essentials of good practice.

Paris, France: International Union Against Tuberculosis and Lung Diseases, 6th edition. 2010. Page 36. http://www.theunion.org/index.php/en/resources/technical-publications/tuberculosis/item/44-anagement-of-tuberculosis-a-guide-to-the-essentials-of-good-clinical-practice-

2. Crighton EJ, Barwin L, Small I, Upshur R What have we learned? A review of the literature on children's health and the environment in the Aral Sea area. Int J Public Health 56: 125-138. 2011.

3. Fenichel RR, Malik M, Antzelevitch C et al. Drug--induced torsade de pointes and implications for drugdevelopment. J Cardiovasc Electrophysiol, 2004; 15:475–495.

4. Giorgi MA, Bolanos R, Gonzalez CD, Di Girolamo G. QT interval prolongation: preclinical and clinical testing arrythmogenesis in drugs and regulatory implications. Current Drug Safety, 2010, 5, 54-57.

5. Timothy H. Holtz, Maya Sternberg, Steve Kammerer, Kayla F. Laserson, Vija Riekstina, Evija Zarovska, Vija Skripconoka, Charles D. Wells, and Vaira Leimane. Time to Sputum Culture Conversion in Multidrug-Resistant Tuberculosis: Predictors and Relationship to Treatment Outcome. Ann Intern Med. 2006;144:650-659.

6. Johnston et al. Treatment outcomes of Multidrug-resistant tuberculosis: a systematic review and Meta-analysis. PLoS ONE 2009; 4, 9, e6914.

7. Kurbatova EV, et al., Predictors of poor outcomes among patients treated for multidrug-resistant tuberculosis at DOTS-plus projects, Tuberculosis (2012), http://dx.doi.org/10.1016/j.tube.2012.06.003

8. M K Lalor, S Allamuratova, Z Tiegray, A K Khamraev, J E Greig, K Braker, P A du Cros, A Telnov. Treatment outcomes in multi-drug resistant TB patients in Uzbekistan. Poster PC-920-29. 42nd Union World Conference on Lung Health. Lille, France. 2011.

9. Lancet Editorial. Tuberculosis—from ancient plague to modern-day nemesis Published online October 17, 2012 http://dx.doi.org/10.1016/S0140-6736(12)61772-3

10. Ministry of Health Uzbekistan (2009) Tuberculosis data for Uzbekistan.11. Munsiff et al. Public-private collaboration for multidrug-resistant tuberculosis control in

New York City. Int J Tuberc Lung Dis 2006; 10 (6):639-648.12. Orenstein et al. Treatment outcomes among patients with multidrug-resistant tuberculosis:

systematic review and meta-analysis. Lancet Infect Dis 2009; 9: 153-61. 13. Shin SS et al. Adverse reactions among patients being treated for MDR-TB in Tomsk, Russia.

Int J Tuberc Lung Dis. 2007 Dec; 11(12):1314-20.14. Toczek A, Cox H, du Cros P, Cooke G, Ford N. Strategies for reducing treatment default in

drug resistant tuberculosis: systematic review and meta-analysis. IJTLD 2012 Dec 4 (epub)15. Van Deun et al. Short, highly effective and inexpensive standardized treatment of multidrug-

resistant tuberculosis. Am J Respir Crit Care Med Vol 182. Pp 684-692, 2010.16. Van Deun. 9-month Standardised MDR-TB Regimen in Bangladesh. 43rd Union World

Conference on Lung Health. Kuala Lumpur. 2012.

17. World Health Organisation (2008) Guidelines for the programmatic management of drug-resistant tuberculosis. WHO/HTM/TB/2008.402

18. World Health Organisation (2009) Tuberculosis country profile, epidemiology and strategy, Uzbekistan.

19. World Health Organisation (2010). Multidrug and extensively drug-resistant TB (M/XDR-TB): 2010 global report on surveillance and response. WHO/HTM/TB/2010.3

20. WHO 2011 update. Guidelines for the programmatic management of drug resistant tuberculosis.

WHO. Global Tuberculosis Report 2012. WHO/HT