Pharmacodynamics of anti-HIV gene therapy using viral vectors and targeted endonucleases

Objectives: A promising curative approach for HIV is to use designer endonucleases that bind and cleave specific target sequences within latent genomes, resulting in mutations that render the virus replication incompetent. We developed a mathematical model to describe the expression and activity of endonucleases delivered to HIV-infected cells using engineered viral vectors in order to guide dose selection and predict therapeutic outcomes.

Methods: We developed a mechanistic model that predicts the number of transgene copies expressed at a given dose in individual target cells from fluorescence of a reporter gene. We fitted the model to flow cytometry datasets to determine the optimal vector serotype, promoter and dose required to achieve maximum expression.

Results: We showed that our model provides a more accurate measure of transduction efficiency compared with gating-based methods, which underestimate the percentage of cells expressing reporter genes. We identified that gene expression follows a sigmoid dose-response relationship and that the level of gene expression saturation depends on vector serotype and promoter. We also demonstrated that significant bottlenecks exist at the level of viral uptake and gene expression: only ∼1 in 220 added vectors enter a cell and, of these, depending on the dose and promoter used, between 1 in 15 and 1 in 1500 express transgene.

Conclusions: Our model provides a quantitative method of dose selection and optimization that can be readily applied to a wide range of other gene therapy applications. Reducing bottlenecks in delivery will be key to reducing the number of doses required for a functional cure.