AJ Warr



Nominated From: University of Washington

Research Site: Kenya

Research Area: Infectious Disease

Primary Mentor: Grace John-Stewart

Research Project

Mycobacterium tuberculosis-specific T-cell response in breastmilk of HIV positive mothers

Approximately 1.5 million HIV infected women give birth each year (1). Kenya has contributed significantly to this figure, with over 800,000 women and 200,000 children currently living with HIV (2). The impact of maternal HIV infection upon infant health is grim. Infants who become infected with HIV have devastatingly high mortality rates of over 50% by 2 years of age if untreated, and infants who are HIV exposed but uninfected (HEU) have worse mortality and infectious morbidity than their unexposed peers (3-5). A common cause of death and poor outcomes among HIV infected women and their children is co-infection with Mycobacterium tuberculosis (MTB) (6,7). MTB infection is particularly detrimental to children, who progress from infection to active TB disease more rapidly than adults (50% of under 1 year olds with MTB progress to active TB disease versus only 10% of adults) (8,9).

Using MTB-specific interferon gamma release assays (IGRA) to detect MTB infection, a study in Kenya demonstrated that 42.7% of HIV positive pregnant women in Western Kenya had MTB infection and in infants born to these mothers annual incidence of MTB infection has been estimated to be 20.6%(10,11). Together these studies show the high prevalence of MTB infection in HIV infected women and their children.

Infants born to HIV infected mothers may have altered immune responses, even if they are not HIV infected. HEU infants have abnormal T-cell populations and altered immune response to bacille Calmette-Guerin (BCG) vaccinations (12,13). Breastmilk confers protection to infants from a variety of pathogens and includes antibodies and immune cells. HIV-specific cellular immune responses in breast milk of HIV positive mothers have been associated with a decreased risk of HIV acquisition in infants (14). Recently, maternal MTB antigen-specific T-cell responses have been detected in breast milk cells (BMC) using MTB-specific IGRAs (15). Breastmilk MTB-specific responses are correlated to systemic responses in the peripheral blood (Figure 1). However, this data was from a small pilot study (Less than 20 mothers). There are no data on the prevalence and correlates of MTB-specific T-cell response in breast milk of HIV infected mothers.

Specific Aims:

  • Determine the prevalence and magnitude of MTB-specific T-cell response in breast milk of HIV positive women. We hypothesize that prevalence will be substantial (>25%), reflecting the prevalence of latent TB infection in HIV infected women.
  • Determine cofactors of positive breastmilk MTB-specific T-cell response. We hypothesize that women with a TB history or exposure, HIV suppression on antiretroviral therapy, and higher CD4 count will be more likely to have breastmilk MTB-specific T-cell responses and will have higher magnitude responses.


Research Significance

Defining the prevalence and cofactors of breastmilk MTB-specific cellular responses is the first step in establishing associations between such responses and infant MTB infection. This information will enhance our understanding of immunity against MTB and contribute to TB vaccine and prevention efforts.





View on PubMed