Drug-resistant tuberculosis (DR-TB) is a dreadful disease. It is extremely difficult to cure, often requiring 2 years of toxic medications. With >1.3 billion people in China, there are clearly many cases of DR-TB each year. Questions include exactly how many cases, where are they occurring, and what can be done to reduce DR-TB transmission? Genotyping technology helps epidemiologists understand the spread of infectious diseases in ways that were impossible a few years ago. In this blog post, I will review a recent CID research paper by Yang et al that uses VNTR genotyping to investigate transmission of TB in China.
First, a bit about tuberculosis. TB occurs along a spectrum from latent TB infection (LTBI) to symptomatic active disease. I describe LTBI to my patients as “sleeping TB.” The reason is that LTBI is taking a nap in your body but it can wake up at any time and make you very sick with active TB (coughing, weight loss, etc). That’s the reason we give months of chemoprophylaxis to LTBI patients (for example, people with a positive PPD/TST or IGRA).
In China, there are thought to be approximately 1 million new cases of active TB each year, but the true number of cases might be much greater because of difficulties of TB surveillance in such a large country, and a reliance on the AFB smear. AFB smear has been used for 120 years and has a sensitivity much lower than TB culture. There are though to be >60,000 cases of multi-drug resistant tuberculosis (MDR-TB) in China each year, but that number may be an underestimate because most cases of TB are not cultured and do not have drug susceptibility testing (DST) performed. Even when culture is performed in China, it seems to be done most frequently on Lowenstein-Jensen solid agar (LJ) which is not as sensitive as liquid culture.
In this CID paper, you will read that Yang and colleagues performed a population-based molecular epidemologic study in 5 study sites (in 5 provinces in China) over three years (2009-2012). The authors identified 2274 culture-positive TB patients over three years. Just hearing those numbers made me pause. China has over 1 million new cases of active TB each year. Are these 2274 cases representative of the enormous burden of TB across this vast country? Unlikely, if you know anything about sampling frames (although the authors don’t argue that their study is nationally representative. A nationally representative study would have been incredibly expensive).
Now, on to the laboratory methods. Yang et al used variable number tandem repeats (VNTR) typing which was optimized for the TB genotypes found in China. VNTR typing is a “conventional genotyping method” and is thought to have better resolution than older methods like RFLP or spoligotyping. VNTR allowed Yang et al to identify the “Beijing genotype,” the most prevalent family of TB in China. However, VNTR cannot detect genetic variations to the granular level of whole-genome sequencing (WGS). When you look at the most recent papers published in molecular TB epidemiology, most of them are using WGS (for example this paper from Switzerland and this one from California). WGS allows a higher resolution and helps researchers identify TB transmission hotspots and missing links in transmission chains. In the future, we will likely be reading papers from China which use WGS, just not this study.
Of the 2274 TB cases, using VNTR, Yang found that 705 (31%) had clustered isolates, whereas 1569 had unique isolates. This led them to conclude that recent transmission contributes significantly to the TB burden, as one in three TB patients was a secondary case due to recent transmission. In an accompanying editorial, Arend and van Soolingen explore this finding, explaining that 31% may underestimate the true percentage of clustered cases. (Yang relied on passive case finding, children were excluded, the study’s time frame was only 3 years, and migration may have led to an underestimation). Arend and von Soolingen ask the rhetorical question if 31% of clustered cases might represent the tip of the iceberg regarding recent TB transmission in China. It’s impossible to say based on this study but I suspect the answer is yes.
With such a large percentage of recent transmission, the implication is that China should focus on rapid molecular diagnosis and appropriate treatment of active TB, including DR-TB (i.e., FAST). This represents a major operational challenge in such an huge country. The cost of Xpert MTB-RIF alone (or other molecular TB tests) would represent a huge financial burden, along with the cost of drugs, programmatic support, etc.
TB clinicians often think of DR-TB as having reduced transmissibility when compared to drug-sensitive TB. However, Yang found that MDR-TB bacteria seemed to be more transmissible than drug-sensitive TB. This could be related to the genomic aspects of the Beijing genotype or MDR-TB in China, but much more research is needed. More transmissible or less, DR-TB should have our attention. DR-TB is here to stay, in China and throughout the world. We cannot afford to be complacent, as the study by Yang et al makes clear.