The natural history of TB disease-a synthesis of data to quantify progression and regression across the spectrum

Abstract

AbstractBackgroundPrevalence surveys have found a substantial burden of subclinical (asymptomatic but infectious) TB, from which individuals can progress, regress or even persist in a chronic disease state. We aimed to quantify these pathways across the spectrum of TB disease.MethodsWe created deterministic framework of TB disease with progression and regression between three states of pulmonary TB disease: minimal (non-infectious), subclinical, and clinical (symptomatic and infectious) disease. We estimated ranges for each parameter by considering all data from a systematic review in a Bayesian framework, enabling quantitative estimation of TB disease pathways.FindingsTwenty-four studies contributed data from 6030 individuals. Results suggested that, after five years, 24.7%(95% uncertainty interval, UI, 21.3%-28.6%) of individuals with prevalent subclinical disease at baseline had either progressed to clinical disease or died from TB, whereas 16.1%(95%UI, 13.8%-18.5%) had recovered after regressing to minimal disease. Over the course of five years 30% (95%UI, 27.2%-32.6%) of the subclinial cohort never developed symptoms. For those with clinical disease at baseline, 39%(95%UI, 35.8%-41.9%) and 10.3%(95%UI, 8.5%-12.4%) had died or recovered from TB, with the remainder in, or undulating between, the three disease states. The ten-year mortality of people with untreated prevalent infectious disease was 38%.InterpretationOur results show that for people with subclinical disease, classic clinical disease is neither inevitable nor an irreversible outcome. As such, reliance on symptom-based screening means a large proportion of people with infectious disease may never be detected.FundingTB Modelling and Analysis Consortium and European Research CouncilResearch in ContextEvidence before this studyIn recent years the existence of a spectrum of TB disease has been re-accepted. The classic paradigm of disease is one active state of symptomatic presentation with bacteriologically positive sputum, now referred to as clinical disease. Within the spectrum, a subclinical phase (where people do not report symptoms but have bacteriologically positive sputum) has been widely accepted, due to prevalence surveys using chest radiography screening in addition to symptom screening. On average these prevalence surveys have found around 50% of people with prevalent infectious TB had subclinical disease. There is also another state of minimal disease, or non-infectious disease, that is the earliest point on the disease spectrum after progression from infection. The likelihood or speed of natural progression, regression, or persistence of individuals across this spectrum remains unknown. As a consequence, the ability to accurately predict the impact of interventions has been limited. As individuals with bacteriologically-positive TB now receive treatment, contemporary data to inform the required transitions is highly limited. However, a large number of cohorts of patients were described in the pre-chemotherapy era. Until now, these data have not been synthesised to inform parameters to describe the natural history of TB disease.Added value of this studyWe synthesised data from historical and contemporary literature to explore the expected trajectories of individuals across the spectrum of TB disease. We considered a cohort of people with prevalent bacteriologically positive disease, with a 50/50 split of people with subclinical and clinical disease at baseline. We found that within five years, 13.3% of people recover from TB, with no chance of progressing to active disease without reinfection. However, we also find that 26.3% are still spending time infectious at the end of the five years. Our estimates for 10 year mortality and duration of symptoms before treatment aligned with the known and accepted values.We also show that regression from subclinical disease results in a large reservoir of people with minimal disease, from which they can permanently recover, but can also progress again to subclinical disease. The undulating pathways that lead to regression and progression mean that 30% (27.2%-32.6%) of individuals with prevalent subclinical disease do not experience symptoms over the course of five years. This shows that clinical disease is neither a rapid, nor inevitable outcome of subclinical disease.Implications of the available evidenceWith these data-driven estimates of parameters, informed projections of the relative value of addressing minimal, subclinical, or clinical disease can now be provided. Given the known reservoir of prevalent subclinical disease and its contribution to transmission, efforts to diagnose and treat people with “earlier” stages of TB are likely to have a larger impact than strategies targeting clinical disease, particularly on individuals who never would have progressed to clinical disease.