Systems Immunology unravels the age/sex effects on the immune correlates of Tuberculosis (TB) progression

Systems Immunology unravels the age/sex effects on the immune correlates of Tuberculosis (TB) progression

Share this post

Choose a social network to share with, or copy the shortened URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

Tuberculosis, caused by Mycobacterium tuberculosis (Mtb), has plagued humans for thousands of years and is still one of the world’s deadliest infections1, due to the lack of effective diagnosis and prevention strategies. It’s estimated that up to a quarter of the global population has been infected with Mtb2. Among them, 5-10% of infected individuals will eventually develop active TB with clinical symptoms. Current diagnostic methods cannot predict TB progression. Therefore, an effective molecular signature for population screening to identify the individuals at risk of progression to active TB would be of major public health benefit.

In the past decade, many studies have established various clinical cohorts and generated high-throughput multi-omics datasets including transcriptomic3–5, metabolomic6, proteomic7, etc.. Numerous signatures have been successfully defined using each single omics dataset, highlighting increases in interferon (IFN) signaling and other inflammatory signals8. Antibodies and B cells have been underappreciated in TB immunology as Mtb is an intracellular pathogen but have received more and more attention in recent studies demonstrating their utility to differentiate TB disease states9–11. We therefore investigated the ability of Mtb antigen-specific antibody profiles to identify TB progression, hoping that the identified signatures might be used to develop an easily implemented, pathogen-specific test.

Our work employed a systems serology approach to comprehensively investigate Mtb antigen-specific antibody isotypes/subclasses and the binding of Fcγ receptors (FcγRs) expressed on various immune cells, in TB progressors and matched non-progressor controls from two well-established longitudinal clinical cohorts: the Adolescent Cohort Study (ACS), a large epidemiological study of TB, and Gates Grand Challenges 6-74 (GC6), a TB household contact cohort (Fig. 1). Using a Machine Learning (ML)-based feature selection and cross-validation framework, we extensively evaluated all the possible combinations of selected measurements and defined a multivariate signature capable of identifying TB progressors up to two years before the clinical diagnosis in ACS. Unfortunately, this signature poorly discriminated progressors from non-progressors in the validation cohort, GC6, which is more heterogeneous across age and geography. This was very discouraging to our hopes of developing a universal antibody-based biomarker for TB progression.

To understand why the cohorts looked so different, we delved into the clinical variables between ACS and GC6 cohorts. Both cohorts include similar proportions of male and female individuals, but all the individuals in the ACS cohort were adolescents from 12 to 18 years old, while in the GC6 cohort, there was a wide age distribution spanning young children to older adults. Several studies have demonstrated that these factors may affect the immune response and performance broadly in various diseases and immune systems12–14. We tested the ability of our defined humoral signatures to identify progressors in separate subpopulations defined by age and sex and acquired the satisfied performance in the validation cohort. The effects of age and sex in prediction performance were further validated in the transcriptomics dataset. Both transcriptomic and serologic signatures achieved better performance among males and adolescents (Fig. 1). We showed that demographic features like age and sex were key modulators of antibody and transcriptional signatures. This work lays the groundwork for the development of Mtb antigen-specific serologic diagnostics for TB progression and indicates the promise of embracing integrative information from multiple molecular layers.

Even though clinical variables including age, sex, geography, etc, have been noticed and reviewed in various biomedical research13–15, they are still not comprehensively evaluated in individual clinical studies. In the initial phases of this study, we encountered challenges with inadequate model performance on the validation cohort, indicating limited generalization and clinical applicability. Since the progressors and non-progressors were carefully paired by age and sex during the cohort establishment, we initially believed their effects would be negligible, albeit present to some extent. Despite grappling with the challenges of poor generalization for months and being advocated for exploration into the effects of age/sex, we remained hesitant to take action due to our initial impressions. Eventually, the results after considering these clinical variables dramatically changed our minds and paved the way for predictability. Overall, this work highlights the importance of these clinical variables, whose effects may be underestimated broadly in clinical studies.

Once Again, this work demonstrates the success of systems immunology in modern medicine and biology, requiring a variety of high-level skill sets as teamwork, including the clinical, omics experimental platforms, bioinformatics, and immunological (Fig. 1). We appreciate the enormous efforts from the clinical teams of ACS and GC6 cohorts to screen the populations and eventually establish these two advanced clinical cohorts. Thanks to their generosity in sharing these valuable serum samples, we could utilize the developed systems serology platforms to deeply profile the antibody profiles and simultaneously capture hundreds of measurements. Due to the high complexity of the omics dataset, advanced computational pipelines are essential to manipulate the large datasets to extract the biological meaning. Through close collaboration and active discussions from these pillars, we finally reached this milestone.

There are several research directions we could further pursue. First, all the individuals from these two involved cohorts are from the same geographical location, South Africa. Previous studies have already suggested that there are altered immune responses in epidemiologically distinct locations due to non-tuberculous mycobacteria exposure, environmental factors, and even co-morbidities including HIV co-infection. Therefore, more diverse cohorts with different countries, and even continents, could be included to evaluate the defined signatures. Secondly, we demonstrate that Mtb-specific humoral immune responses correlate with TB progression and provide complementary information to that captured by the transcriptomic dataset. These results indicate that the integration of available multi-omics datasets may provide novel interactive information across molecular layers to predict TB progression with increased performance. Thirdly, how to computationally integrate the clinical variables, including age/sex, with immune measurements to define the efficient signature to identify the TB progressor is still a promising task. Overall, using the systems immunology approach, this study emphasizes that the Mtb-specific antibody response during the latent period may provide critical insights to predict the TB progression and unravel the role of clinical factors including age, and sex, highlighting the possibilities of more comprehensive analysis in the future TB study for TB progression prediction.

Thanks to Dr. Sarah Fortune for the constructive feedback on this article.


  1. Global Tuberculosis Report 2023.
  2. The End TB Strategy.
  3. Penn-Nicholson, A. et al. RISK6, a 6-gene transcriptomic signature of TB disease risk, diagnosis and treatment response. Sci. Rep. 10, 8629 (2020).
  4. Zak, D. E. et al. A blood RNA signature for tuberculosis disease risk: a prospective cohort study. Lancet 387, 2312–2322 (2016).
  5. Suliman, S. et al. Four-gene pan-African blood signature predicts progression to tuberculosis. Am. J. Respir. Crit. Care Med. 197, 1198–1208 (2018).
  6. Weiner, J., 3rd et al. Metabolite changes in blood predict the onset of tuberculosis. Nat. Commun. 9, 5208 (2018).
  7. Penn-Nicholson, A. et al. Discovery and validation of a prognostic proteomic signature for tuberculosis progression: A prospective cohort study. PLoS Med. 16, e1002781 (2019).
  8. Scriba, T. J. et al. Sequential inflammatory processes define human progression from M. tuberculosis infection to tuberculosis disease. PLoS Pathog. 13, e1006687 (2017).
  9. Lu, L. L. et al. A functional role for antibodies in tuberculosis. Cell 167, 433-443.e14 (2016).
  10. Li, H. & Javid, B. Antibodies and tuberculosis: finally coming of age? Nat. Rev. Immunol. 18, 591–596 (2018).
  11. Rijnink, W. F., Ottenhoff, T. H. M. & Joosten, S. A. B-cells and antibodies as contributors to effector immune responses in tuberculosis. Front. Immunol. 12, 640168 (2021).
  12. Klein, S. L. & Flanagan, K. L. Sex differences in immune responses. Nat. Rev. Immunol. 16, 626–638 (2016).
  13. Holmes, C. B., Hausler, H. & Nunn, P. A review of sex differences in the epidemiology of tuberculosis. Int. J. Tuberc. Lung Dis. 2, 96–104 (1998).
  14. Gaya da Costa, M. et al. Age and sex-associated changes of complement activity and complement levels in a healthy Caucasian population. Front. Immunol. 9, 2664 (2018).
  15. GBD 2019 Tuberculosis Collaborators. Global, regional, and national sex differences in the global burden of tuberculosis by HIV status, 1990-2019: results from the Global Burden of Disease Study 2019. Lancet Infect. Dis. 22, 222–241 (2022).

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Follow the Topic

Life Sciences > Biological Sciences > Microbiology
Life Sciences > Biological Sciences > Microbiology > Medical Microbiology > Infectious Diseases > Tuberculosis
Life Sciences > Biological Sciences > Biological Techniques > Computational and Systems Biology > Bioinformatics
Computational and Systems Biology
Life Sciences > Biological Sciences > Biological Techniques > Computational and Systems Biology
Translational Immunology
Life Sciences > Biological Sciences > Immunology > Translational Immunology