A randomised controlled trial (RCT) to evaluate a scalable active case finding primary care-based intervention for tuberculosis using a point-of-care

结核病仍然是全球头号传染病杀手。一项令人震惊的统计数据是，全球每5例结核病病例中就有2例（40%）未得到诊断和治疗。这些“漏诊”或未确诊病例不成比例地集中在非洲和亚洲大城市的大型城郊“贫民窟”和非正式住区（它们通常症状轻微，但仍具有传染性）。除非消灭这一巨大的传播源，否则结核将永远得不到控制。与目前患者“自我寻求”护理（被动病例发现）的模式相比，缺乏敏感的低成本当日检测对主动社区病例发现（ACF）构成了重大挑战。最近，被称为基因Xpert （Xpert）的敏感结核dna检测方法已经可用。这些测试的小型便携式电池供电版本现在可用（OMNI）。我们在南非和津巴布韦进行了一项大型研究（发表于2016年），该研究表明，在配备发电机的小型卡车上使用旧的非便携式Xpert版本是可行的，并且对ACF非常有效。在随后由美国政府资助的一项研究（XACT II）中，我们证明了在小型低成本可扩展厢式货车（实际上是一个移动迷你诊所）后面使用便携式Xpert是可行的，并且在城郊社区有很高的结核病感染率（约10%的人接受了目标筛查；见初步数据）。在拟议的研究（XACT III）中，我们将使用相同的方法，但我们需要确保这种策略在挑战和条件不同的不同环境中是可扩展和可行的。更重要的是，我们需要在方法上优化ACF模型。因此，我们需要从物理位置的角度确定Xpert（诊断测试）的最佳放置位置，即我们是否真的需要将其安装在移动迷你诊所上，或者是否可以将其放置在集中实验室（就像现在一样），并将样本送到这些实验室？这是一个非常重要的问题：我们知道，将收集到的痰样本送到集中实验室要容易得多，因为它使用了现有的基础设施，然而，缺点是有20%至40%的患者没有回来收集他们的结果（治疗前的随访损失；PTLF）。在移动小型货车（在医疗点）中使用诊断可显著减少PTLF，从而实现快速诊断并阻断传播。为了明确解决这个问题，我们需要使用这两种不同的策略进行研究，以找出哪种策略最具成本效益，但可以迅速发现最多病例并最大限度地减少传播。该研究还将回答另外两个重要的子问题。可以识别结核病高危人群的胸部x光片现在可以由计算机算法（称为计算机辅助结核病诊断；CAD-TB）自动读取。了解使用CAD-TB进行大规模筛查是否可以对个体进行分类将非常重要，即缩小网络，以便我们将ACF的目标仅针对那些患有结核病的高风险人群。这可以节省更多的钱，但同样有效。其次，一个基本的未解决的问题是，为什么只有很少症状或没有症状的人可能具有高度传染性（传播疾病）？我们需要更详细地研究这一现象，使用咳嗽气雾剂读数、胸部x光片和观察结核菌株。这可能为医学科学提供设计诊断或治疗干预措施以解决这一重要问题所需的信息。然而，现在的重点是证明XACT方法在不同情况下是可行的，并阐明分子诊断应该如何最佳定位。回答这些问题将使许多国家能够启动ACF方案，并将向决策者提供关键数据，以便传播和执行关于ACF的准则。

TB remains the foremost infectious disease killer globally. A starling statistic is that 2 out of every 5 TB cases globally (40%) remain undiagnosed and untreated. These 'missed' or undiagnosed cases are disproportionately concentrated in large peri-urban 'slums' and informal settlements of large cities in Africa and Asia (they are frequently minimally symptomatic but remain infectious). TB will never be brought under control unless this large reservoir of transmission is wiped out. The lack of a sensitive low cost same-day test represented a major challenge to active community-based case finding (ACF) compared to the current model where patients 'self-seek' care (passive case finding). More recently, sensitive TB DNA-detection tests called Gene Xpert (Xpert) have become available. Small portable battery-operated versions of these tests are now available (OMNI). We conducted a large study in South Africa and Zimbabwe (published in 2016) that showed that using the old non-portable version of Xpert on a mini-truck equipped with a generator was feasible and highly effective for ACF. In a subsequent study funded by the American government (XACT II) we showed that using the portable version of Xpert on the back of a small low-cost scalable panel van (in effect a mobile mini clinic) was feasible and had a very high pick-up rate of TB in peri-urban communities (~10% of those undergoing targeted screening; see preliminary data). In the proposed study (XACT III) we will use the same approach, but we need to be sure that such a strategy is scalable and feasible in different settings where the challenges and conditions vary. More importantly, we need to methodologically optimise the ACF model. Thus, we need to determine where Xpert (the diagnostic test) should be optimally placed from a physical location point-of-view, i.e. do we really need to install it on the mobile mini clinic, or, can it be located in centralised laboratories (as it is now) with samples being sent to these laboratories? This is a very important question: we know that sending collected sputum samples to centralised laboratories will be much easier as it uses existing infrastructure, however, the downside is between 20 and 40% of patients fail to come back to collect their results (pre-treatment loss to follow-up; PTLF). Using the diagnostic in the mobile mini van (at point-of-care; POC) dramatically reduces this PTLF enabling quick diagnosis and interrupting transmission. To definitively settle the question we need a study using the 2 different strategies to find out which strategy is most cost-effective yet can rapidly pick up the most cases and minimise transmission.There are 2 other important sub-questions that the study will answer. Chest x-rays, which can identify people at high risk of having TB, can now be automatically read by a computer algorithm (called computer-assisted diagnosis of TB; CAD-TB). It will be very important to know whether mass screening using CAD-TB can triage individuals i.e. narrow the net so that we target the ACF only to those at high risk of having TB. This could save even more money yet be just as effective. Secondly, a fundamental unanswered question is why individuals with minimal or no symptoms can be highly infectious (transmit disease)? We need to study this phenomenon in greater detail using cough aerosol readouts, chest x rays, and looking at the TB strains. This might provide medical science with the information it needs to design diagnostic or therapeutic interventions to address this important problem. However, the key priority now is to show that the XACT approach is feasible in different settings and to clarify how the molecular diagnostics should be optimally located. Answering these questions will allow the initiation of ACF programmes in many countries and will contribute critical data to policy makers so that guidelines on ACF can be disseminated and implemented.