Gene editing to eliminate HIV from mice: misleading and wrongly reported in media coverage
Many different news outlets have published articles about a study published on 2 July 2019 in Nature Communications,  which reports the possible elimination of HIV from a small number of humanised mice.
The research involved the use of a gene-editing technique, CRISPR/Cas9, which has been designed to try to remove HIV from infected cells. 
The intent is to perform what might be considered a sort of genetic surgery, slicing the HIV genome from where it has integrated into the genetic code of an infected cell.
The research group of Kamel Khalili at Temple University is developing the approach, and they have previously generated headlines after publishing preliminary results (see prior Media Monitor entries on the topic) . Khalili has also founded a company, Excision Biotherapeutics,which aims to commercialise the technology. 
The latest experiments involved combining a CRISPR/Cas9 construct targeting HIV with a souped-up form of antiretroviral therapy called LASER ART (sequential long-acting slow-effective release antiviral therapy), invented by study co-author Howard Gendelman and colleagues at the University of Nebraka.
The treatments were tested in a humanised mouse model – the mice are bred to be immune deficient, then have their immune systems reconstituted with transplanted human cells, which allows them to be infected with HIV (the virus cannot infect mouse cells). Notably, these models are imperfect because the human cells are eventually rejected as foreign, and don’t necessarily behave exactly how they would in the human body.
Out of a total of 23 humanised mice that received both LASER ART and CRISPR/Cas9 over the course of three different experiments, nine did not show evidence of HIV viral load rebound when LASER ART was stopped. HIV genetic material could also not be detected in multiple tissue samples from eight of the nine non-rebounders, leading to the claim that the virus may have been eliminated in these cases. In contrast, all animals that received LASER ART without CRISPR/Cas9 experienced HIV viral load rebound after LASER ART cessation.
The results appear encouraging, but there are still multiple reasons to be cautious about concluding that the work “paves the way to a human cure” as some headlines have stated.
Perhaps most important, it is not yet known if CRISPR/Cas9 targeting HIV can be safely administered to people.
A primary concern is the possibility of “off-target” effects, a scenario in which the CRISPR/Cas9 technology inadvertently edits and damages human genes. The researchers have not uncovered any significant off-target effects so far, but it’s a technically daunting task to assess the entire genetic code for evidence of problems.
Delivering CRISPR/Cas9 to all the cells in the human body that might be infected with HIV represents a major challenge. The delivery vehicle used in the new humanised mouse study was an adeno-associated virus (AAV) vector. AAVs are not known to cause harm and are a popular delivery method for gene therapies. However, the efficiency with which they might target CD4 T cells – the main location of persistent HIV in people on ART – is unclear. Indeed the AAVs used in this study would actually enter many cells in the body beyond HIV-infected CD4 T cells, raising issues of safety and off-target activity.
Adding to the challenge, CRISPR/Cas9 is derived from bacteria, meaning that the human body is likely to treat it as foreign and mount an immune response against it. Studies have found that most people have pre-existing immune responses to Cas9 due to exposure to the bacteria Staphylococcus aureus and Streptococcus pyogenes.  Using AAV as a delivery vehicle may also promote the development of immunity against the CRISPR/Cas9 cargo—this type of problem stymied the first attempt to use AAV to deliver an anti-HIV broadly neutralising antibody in a clinical trial. 
The genetic variability of HIV in humans means that CRISPR/Cas9 will need to target parts of the virus that mutate the least in order to be broadly effective across different populations (and recognise all the viral variants present in an individual). In the humanised mouse study, animals were infected with laboratory HIV strains.
Some scientists have expressed concern over what might happen if a cell harbours more than one copy of HIV. 
Rather than removing a single virus, the CRISPR/Cas9 technology could theoretically make cuts at sites in each HIV copy, causing the removal of all the cell’s genes that were in between. This would likely damage the cell, although the exact consequences would be unpredictable.
There are also some uncertainties regarding the technical aspects of the published study – as author Howard Gendelman notes in the Daily Mail article, the paper was rejected by ‘many different journals’. While Gendelman claims this was just due to scepticism about the results, it is likely that the independent scientific peer reviewers who recommended rejection had specific concerns about how the research was conducted and/or presented.
Notably, no other research groups have yet presented any similar results with CRISPR/Cas9 in animal models (at least to our knowledge). Independent confirmation would offer reassurance about the reliability of the findings.
The next step prior to human trials is conducting experiments in macaque monkeys infected with SIV (HIV’s counterpart in monkeys). Some preliminary results were presented in March 2019 at the Conference on Retroviruses and Opportunistic Infections (CROI), and were covered in detail on TAG’s HIV Basic Science, Vaccines, and Cure Project Blog. 
It will be critical to test whether targeting SIV with AAV-delivered CRISPR/Cas9 in macaques treated with antiretroviral therapy can prevent viral load rebound when therapy is stopped (as appeared to occur in some of the humanised mice). A successful outcome in the SIV/macaque model would offer far more convincing support for the potential efficacy of the approach than results in humanised mice.
A problem to be aware of with the current media coverage is that many of the headlines are inaccurate. For example, TIME: “For the first time, researchers eliminated HIV from the genomes of living animals” and the Daily Mail: “Scientists eliminate HIV in the entire genome of lab mice for the first time ever.” [9, 10, 11, 12]
The source of the confusion is the Temple University press release, which was titled “HIV eliminated from the genomes of living animals.” 
This is not true. Mice cannot be infected with HIV, so the virus was not eliminated from their genomes. The claim of HIV elimination only applies to the human cells that the mice had been transplanted with, and the genomes of those human cells belong to the human they came from, not the mice.
The Temple University research group and Excision Biotherapeutics hope to start human trials within a year or so, but that is likely to depend both on results in macaques and whether regulators at the U.S. Food and Drug Administration decide there is sufficient evidence that participants would not face undue risks.
Richard Jefferys, TAG. HIV eliminated from the genomes of living animals.
- Dash PK et al. Sequential LASER ART and CRISPR Treatments Eliminate HIV-1 in a Subset of Infected Humanized Mice/ Nature Communications, July 2, 2019.
- TAG.Gene Editing in HIV Research. (March 2019).
- Jefferys R. Cure for HIV possible “within three years”. TAG. April 2016.
- Excision Biotherapeutics website.
- Charlesworth CT et al. Identification of preexisting adaptive immunity to Cas9 proteins in humans. Nature Medicine (25) 249–254 (2019). (28 January 2019).
- Priddy FH et al. Adeno-associated virus vectored immunoprophylaxis to prevent HIV in healthy adults: a phase 1 randomised controlled trial. Lancet HIV (6):4, PE230-E239 (01 April 2019).
- Josefsson L et al. Majority of CD4+ T cells from peripheral blood of HIV-1–infected individuals contain only one HIV DNA molecule. Proc Natl Acad Sci U S A. 2011 Jul 5; 108(27): 11199–11204. doi: 10.1073/pnas.1107729108.
- Jefferys R. Selected highlights from CROI 2019. TAG (10 April 2019).
- TIME magazine. For the First Time, Researchers Eliminated HIV From the Genomes of Living Animals. (2 July 2019).
- Daily Mail. Scientists eliminate HIV in the entire genome of lab mice for the first time ever: Breakthrough paves the way to a human cure – with clinical trials set to start next year. (2 July 2019).
- New Scientist. Have mice really been cured of HIV using CRISPR gene editing?(2 July 2019).
- PBS Newshour.CRISPR gene-editing ‘eliminates’ HIV in some mice. What does it mean for humans? (2 July 2019).
- Temple University press statement. HIV eliminated from the genomes of living animals. 2 July 2019).