Evaluation of a Mask Aerosol Sampling System (MASS) as an Active Case Finding Approach Focused on Infectious TB in Low-Resource Settings.

Globally, tuberculosis (TB) is the most important cause of death and disease due to a single bacterial agent. TB is transmitted by aerosols (tiny droplets expelled by coughing and other respiratory efforts) generated by infected individuals. The bacteria-containing droplets are breathed deep into the lungs where, over months and years, disease is produced in about one in ten of those who get the initial infection. Over one quarter of humanity is infected with the bacterium that causes TB, Mycobacterium tuberculosis (Mtb), but only people who go on to develop lung disease can spread the infection.
Mtb must cause lung disease and be spread by aerosols to survive in the long term; if we prevent transmission, eventually the organism will die out and the disease will be eliminated. Thus, detection and treatment of infectious cases is of paramount importance.
TB is a disease associated with poverty and, in resource constrained settings, many cases of TB go undiagnosed. This is particularly so in sub-Saharan Africa where healthcare is resource limited. Here, detection of individuals with suggestive symptoms (e.g. prolonged cough) and Mtb positive sputum (coughed up mucus) provides the main means of diagnosis. However, many symptomatic individuals do not access healthcare, even when enabled to do so by community health workers (CHWs) and a significant proportion (20-50%) are not able to produce sputum.
The Leicester team have been developing the use of adapted face masks to capture aerosols in a way that allows detection of Mtb in the World Health Organisation-recommended Mtb detection system, called Xpert MTB/RIF. The mask samples allow detection of Mtb positive individuals at a similar frequency to sputum analysis. However, unlike sputum samples, everyone who can wear a mask produces a sample in our mask aerosol sampling system (MASS). MASS is very low cost and requires only modest expertise to take and analyse the samples. In this project we shall evaluate use of the MASS as a low cost community screening tool.
Our study is based in the Tshwane district of South Africa (around 1% are estimated to have lung TB) where the Pretoria team have developed a community oriented primary care (COPC) programme (http://www.up.ac.za/en/family-medicine/article/2077127/launch-video) directed to patient and household (HH) centred health promotion. Building on our established TB aerosol collaboration (MRC and Wellcome Trust funded) and, critically, with the support of the 1670 COPC CHWs, we now plan to evaluate application of MASS as a TB screening tool in Tshwane. If favourable, this evaluation will enable us to develop a large scale intervention trial where we will compare the in-practice capacity of MASS-based screening to detect infectious cases with the conventional approach of symptom and sputum screening. If a high proportion of such cases can be detected and treated, the approach has clear potential to reduce the overall frequency of TB in this and other poor communities.
We will first test MASS in the CHWs to optimise the 15 minute procedure and train them for the study. We will then use the standard CHW review programme to detect HHs with individuals showing TB-related symptoms. Following informed consent, all members of such HHs aged 5 years and above will be mask sampled. We aim to sample 120 HHs with symptomatic individuals and expect to detect at least 20-40 positives. In parallel, through data collected from the near 300,000 individuals registered on the COPC programme we will identify 120 HHs matched for key socioeconomic indicators. All members of these HHs will be sampled as above. Symptomatic individuals and MASS positive individuals will all be referred to local clinics for standard diagnosis and treatment.
We believe MASS could enable resource-constrained health systems to focus control efforts on the most infectious TB cases and their contacts, thereby reducing transmission and consequently reducing disease.

We aim to evaluate use of our mask aerosol sampling system (MASS) to detect individuals with pulmonary TB (pTB) in a community oriented primary care programme in Tshwane district (RSA). We will test the capacity of MASS combined with Xpert MTB/RIF analysis to detect pTB at similar or greater rates than conventional symptom and sputum review. Mask samples are obtained in 15 min from all who take part. Because many individuals with pTB either do not produce sputum or attend clinic when requested, we believe MASS may increase detection rates.
Recent evidence clearly indicates that analysis of patient aerosols for Mtb provides a more accurate assessment of infectivity than sputum. Our unpublished MASS studies show that up to 20% of individuals may be discrepant for aerosol and sputum positivity at any one sampling time. Thus, in addition to its potential to increase community-based pick up rates, we suggest that MASS will preferentially detect people who are actively spreading TB. If diagnostic rates in this study are favourable and if the results of our ongoing transmission related studies support the capacity of MASS to detect infectious pTB then we will use these results to develop an intervention study. This will test, in a defined target community, whether active case finding by MASS compared to conventional screening can a) increase the number of individuals diagnosed and treated; b) allow focusing of limited resources on individuals and contacts most likely to be involved in transmission and c) reduce the prevalence of TB within 4-5 years.
Following optimisation of MASS sampling for implementation in Tshwane, CHWs will sample entire HHs triggered by detection of symptomatic members. They will then sample an equal number of HHs matched for key indicators without symptom assessment. Screening of 240 HHS and ~1,200 individuals in this way will allow feasibility and design of the follow on intervention trial to be determined.

Through the mask aerosol sampling system (MASS) we aim to deliver a new approach to the assessment of TB in communities. The simplicity and low cost of the approach indicate its feasibility in the proposed setting and this evaluation will establish whether it can deliver results that are of value in TB control. Favourable evaluation here will deliver:
1.A means of detecting pulmonary TB cases in communities that does not require individuals to produce sputum or attend a clinic.
2.An assessment of the relative value of MASS against symptom and sputum-based screening.
3.Detection of Mtb aerosol positivity in individuals who do not report symptoms.
While this will be the first substantial evaluation of the MASS approach, the Leicester team expect to be in a position to support other groups who wish to explore sampling this way by the end of this project.
Comments below all assume a positive outcome in the evaluation, development towards the impacts described is outlined in the Pathways to Impact document.
We emphasise that TB control is just one of the applications of MASS. The approach can and is being developed to assess all airborne microbes in a format compatible with routine clinical practice.

WHO WILL BENEFIT AND HOW?
ORGANISATION LEVEL PUBLIC HEALTH AUTHORITIES; THE WORLD HEALTH ORGANISATION; DEPARTMENT OF HEALTH; HEALTH PROTECTION ENGLAND; GOVERNMENTS; THE HEALTHCARE INDUSTRY.
Identification of infectious individuals within a community will enable refocussing of control efforts on these individuals and their contacts. This approach could rapidly impact on transmission and prevalence of disease. Evidence supporting this approach would lead to significant changes in TB control strategies funded and implemented by the organisations listed. In addition to production of the sampling system and the downstream analysis of samples, the approach creates new opportunities for the healthcare industry to develop products directed to limiting transmission

MEDICAL COMMUNITY AND PATIENTS AND THOSE AT RISK OF EXPOSURE TO MTB.
MASS will enable practitioners to recognise and manage infective patients and to obtain microbiological confirmation of diagnosis in those who do not produce sputum. A pilot study to detect the effect of treatment in patients with fully sensitive and drug resistant infection (10 each) has ethical permission and will commence shortly in Pretoria. A simple test recognising infectivity based on samples that can be obtained from everyone who can wear a mask will be of great value to practitioners, hospital managers, patients and their contacts. Currently, in the UK patients considered infectious, particularly those with drug resistant infections are maintained in isolation at considerable cost until they are sputum culture negative; if they don't produce sputum then they are evaluated by bronchoalveolar lavage. While much work will needs to be done, to establish the negative predictive value of MASS for transmission risk, the potential is clear.

PHARMACEUTICAL INDUSTRIES AND CHARITIES E.G. - GLOBAL ALLIANCE FOR TB DRUG DEVELOPMENT.
While there are no obvious opportunities for new drug development here, the capacity to recognise infectivity will allow evaluation of different treatments in this regard. In the case of resistant infections, existing drugs may differentially affect infectivity. Moreover, our parallel studies on the biological basis of Mtb aerosol transmission (MR/PO23061/1) have potential to identify and transmission-specific bacterial targets.