Using mathematical models to tell time of initial HIV infection in infants

The human immunodeficiency virus (HIV) has stubbornly stated that it is here to stay, causing the devastating HIV/AIDS epidemic. Fortunately, human stubbornness and resilience are not in short supply as shown by the breadth of research on HIV transmission, prevention, and treatment. Pre-exposure prophylaxis (PrEP) for HIV and anti-retroviral therapy (ART) are largely successful options for adults to prevent HIV infection and treat disease. Like adults, infants are also susceptible to HIV infection and related diseases. A pregnant mother living with untreated HIV can pass this virus on to her infant while the infant is within the womb or at later times during delivery or through breastfeeding. The mode of transmission and time of initial infection influences the abundance of virus, also known as viral load, which can impact disease outcomes for the infant. Therefore, estimating the time of initial infection and knowing how it occurs in each individual case is needed for continued research on HIV pathogenesis in infant cohorts with unknown timing of infection. Senior Staff Scientist Dr. Dara Lehman in the Human Biology Division and first author and graduate student Magdalena “Maggie” Russell from the Matsen Group in the Public Health Sciences Division at the Fred Hutchinson Cancer Center created and trained a mathematical model for estimating infant infection timing for cases in which the infection history of the infant is unknown. This work, that was recently published in PLoS Pathogens, will help define HIV disease in infants and aid research studies pinpointing when interventions should be initiated, and which types might be most effective. 

In adults, HIV diversity or the abundance of unique viral genomes increases over time. This diversity can be determined by sequencing essential HIV genes necessary for the virus to replicate and infect other cells. The increase in HIV diversity over time is attributed to the error prone machinery used to replicate viral genomes. Using this knowledge to their advantage, researchers have shown that mathematical models that use HIV diversity measurements as input data can help estimate the time of initial HIV infection for cases in which the infection history is unknown i.e., HIV sequence diversity measured in an individual at only one time point is needed to estimate time of infection. The Lehman lab sought to predict infection times specifically for infants as opposed to adults using a similar approach. To generate this model, they made use of a cohort of pregnant mothers living with HIV in Kenya who were enrolled between 1992 and 2002 – a time when ART was not yet standard of care in Kenya. After birth, HIV testing was performed routinely in their infants and following a positive HIV test, additional samples were collected, allowing HIV diversity to be measured over time by sequencing three separate regions within two essential HIV genes. In this group, half of the infants became infected within the womb and the other half, became infected after birth. Like HIV infection in adults, the HIV diversity in infants typically increased over time. Following additional comparisons with several variables including viral load, transmission type (infection in the womb, at time of delivery, or later through breast feeding), immune cell surveillance and others, the researchers found that “the rate of this accumulation [in HIV diversity] varies by individual, gene-region, and mode of infection, but not by set-point viral load or rate of CD4+ T cell decline”. 

The researchers also compared this infant specific data set and infant specific model to those developed for HIV infection in adults. “Our work suggests that rates of HIV sequence diversification appear to be higher in pediatric infections compared to adult infections,” commented Russell. “This finding raises the question of why there is such a difference in diversification rates between age groups. Understanding the underlying mechanisms driving this difference could provide valuable insights towards understanding differences in HIV pathogenesis in pediatric infection.” In other words, other factors may enhance HIV to generate more HIV diversity in infants over time as compared to adults.

Together, “this study makes several exciting contributions to the field,” shared Russell. “First, we provide novel insights into HIV sequence diversification over time in untreated pediatric infections, an area that has not been well studied, and differs from HIV sequence diversification in adults.” This feature of infant HIV infections is critical to understanding how to develop effective strategies to prevent and treat infection. Russell continued, “Second, we create models for estimating infection timing in children living with HIV. While there are HIV infection timing models for adults, our pediatric model fills a gap and will be useful for estimating time of HIV infection for children with previously unknown infection histories, which is important in pediatric HIV research studies.” Knowledge of infection timing also guides how HIV infection may be treated to limit disease in infants and why early infection of infants within the womb results in more pathogenic disease outcomes. These contributions aid the effort in understanding HIV infection in infants and the future generation of preventative and therapeutic options.

The Hutch Genomics & Bioinformatics Core performed Illumina sequencing on samples for this published work and Pritha Chanana provided bioinformatics assistance. The use and availability of these core facility resources were made possible by the support from the Fred Hutch, UW, and Seattle Children’s Cancer Consortium grant.

The spotlighted research was funded by the National Institutes of Health and the Howard Hughes Medical Institute.

Fred Hutch/University of Washington/Seattle Children's Cancer Consortium member Frederick A. Matsen, IV contributed to this work.

Russell ML, Fish CS, Drescher S, Cassidy NAJ, Chanana P, Benki-Nugent S, Slyker J, Mbori-Ngacha D, Bosire R, Richardson B, Wamalwa D, Maleche-Obimbo E, Overbaugh J, John-Stewart G, Matsen FA 4th, Lehman DA. 2023. Using viral sequence diversity to estimate time of HIV infection in infants. PLoS Pathog. 19(12):e1011861.