MRI imaging used to find areas of active HIV replication in the body

Roman Chazov/

A research team led by Dr Denis Beckford-Vera of the University of San Francisco, California, have undertaken a first in-human study to identify areas of HIV replication using a new scanning technique that combines MRI imaging and radioactively labelled monoclonal antibodies. Viral reservoirs were identified not only in the gut and lymph nodes, but also the nose and bone marrow.

Though often described as a blood borne virus, ‘viral reservoirs’ are areas outside of the blood stream but within the human body where HIV lays dormant or replicates at very low levels. Prevented from disinhibited replication by ART these reservoirs are believed to act as a safe house for the virus – allowing it to evade destruction by the immune system and to restart viral production if ART is stopped.

Scientists have focused on locating these viral reservoirs in order to target them in a quest for HIV eradication. Other than direct biopsy and measurement of virus within tissue – an invasive procedure and one that would need to be repeated at multiple sites including possibly the brain and intestines – there are no ways of identifying viral reservoirs.


lymph nodes

Bean-sized structures throughout the body's lymphatic system, where immune cells congregate to fight infections. Clusters of lymph nodes are found in the underarms, the groin, and the neck.

magnetic resonance imaging (MRI)

A non-invasive, non-x-ray diagnostic technique that provides computer-generated images of the body's internal tissues and organs.

detectable viral load

When viral load is detectable, this indicates that HIV is replicating in the body. If the person is taking HIV treatment but their viral load is detectable, the treatment is not working properly. There may still be a risk of HIV transmission to sexual partners.


The process of viral multiplication or reproduction. Viruses cannot replicate without the machinery and metabolism of cells (human cells, in the case of HIV), which is why viruses infect cells.


The ‘HIV reservoir’ is a group of cells that are infected with HIV but have not produced new HIV (latent stage of infection) for many months or years. Latent HIV reservoirs are established during the earliest stage of HIV infection. Although antiretroviral therapy can reduce the level of HIV in the blood to an undetectable level, latent reservoirs of HIV continue to survive (a phenomenon called residual inflammation). Latently infected cells may be reawakened to begin actively reproducing HIV virions if antiretroviral therapy is stopped. 

Beckford-Vera and colleagues have circumvented the need for biopsy by combining MR-PET (Magnetic Resonance – Positron Emission Tomography) technology with a monoclonal HIV antibody. With traditional PET imaging a radioactive form of glucose is injected into the body via a vein, the radioactive glucose is deemed a tracer. The body is then scanned using CT or MRI (MRI doesn’t emit radiation thus is safer than CT but takes considerably longer). Areas in the body where uptake of tracer is highest can then be seen on the images produced from the scan. This is helpful in identifying areas of cancer or infection: cancerous cells, multiplying bacteria and immune cells responding to infection will have a higher glucose uptake. For example, PET-imaging is used to identify abscesses and bacterial infections of the heart valves (infective endocarditis).

Instead of using radioactive glucose, the team in San Francisco used a commercially available broadly neutralising antibody (VRC01) that targets multiple sites of the HIV virus. They combined this with a radioactive form of the element Zirconium (Zirconium-89). Zirconium has a half life of around 72 hours – its radioactivity will reduce by half every 72 hours – therefore can be tracked over an extended time period following injection. The Zirconium-antibody tracer will therefore bind to areas where HIV persists and this can then be seen on the MR-PET images.

The researchers wanted to compare people living with HIV with a detectable viral load, people with an undetectable viral load on ART and HIV-negative controls to see if there was a difference in how the radioactive antibody was taken up, with the aim of establishing whether this may lead to confirmation of areas where HIV was present.

Five detectable individuals, five with undetectable viral loads on ART and five controls were given the marker and then scanned four times over the following three days (at 2,6, 24 and 72 hours). In order not to confound the results, no participants in the study had a history of systemic malignancy, recent opportunistic infections, systemic autoimmune or inflammatory disorders, or were taking immune-modulating medications. Studies in mice had been undertaken by the research team prior in order to hypothesise a safe dose of tracer for use in humans.

All the participants were men other than one woman each in the viraemic and negative groups. The mean age of the uninfected group was slightly higher at 58.4 years compared with 52.2 and 53.2 for the viraemic and ART groups respectively.

Following injection, uptake of tracer is detected by the scanner and its measurement is analysed using a software programme that calculates the standardised uptake value (SUV). As tracer is injected into the bloodstream through a vein it enters the circulation, travels to the heart and is then delivered to the tissues. The concentration at the aorta (the main vessel leaving the heart) is therefore used as a baseline to compare the concentration in tissues, giving a ratio of standardised uptake value or  rSUV. This allows for easier comparison between tissues and also between individuals. A higher number means a higher concentration in the tissues relative to the blood. A mean value (average uptake of tracer into an area) and a max value (maximum uptake of tracer into an area) are calculated. Their application is broadly similar.


First the team looked at the inguinal lymph nodes (the lymph nodes in the groin area). The lymph nodes are small bundles of tissue that contain many different types of immune cells in large amounts. Waste tissue from cells throughout the body is drained through lymph nodes and when an infection or other abnormality (i.e. a cancer) is detected they will instruct the immune system to attack the abnormality through a variety of means, one of which is rapid replication of immune cells such as CD4+ lymphocytes. In those living with HIV, the virus infects the CD4 cells directly and these lymph nodes begin to harbour large amounts of HIV virus. 

For those with detectable viral loads, the rSUV max in the inguinal lymph nodes was 3.24 fold greater at day three as in the HIV-negative control group (statistically significant, p. < 0.01). For the ART group the difference wasn’t statistically significant. However when looking at rSUV mean (the average amount of tracer uptake in an area), there was almost two fold (1.99, p < 0.01) greater uptake in the inguinal lymph nodes between the ART group and the controls.

In nearly all areas, by day 3 there was significantly more tracer uptake in the viraemic group as compared with the control group. This was especially pronounced in the axillary lymph nodes (under the arm pit), the gut and the descending sigmoid colon (area of large bowel just before the rectum). Other areas that were slightly less statistically significant (p < 0.05) included the ano-rectal junction, parts of the nose and also the bone marrow of the hip.

With regard to the ART group as compared with controls, there was very significant ( p < 0.01) uptake only at the descending colon, with significant uptake (but to a lesser degree, p < 0.05) at the inguinal nodes, the gut, the nose and the bone marrow of both femur and hip.

There was inconsistency in some areas, for example there was significant uptake (rSUV mean) at day 3 in the femur and hip marrow in the ART group (2.92, p < 0.05) when compared with the controls, but somewhat less so in the viraemic group compared with controls (1.40, p < 0.05). The authors do highlight how inconsistent and varied uptake was in the bone marrow for all those participants with HIV. They advise further study into this as bone marrow has not often been thought of as a large viral reservoir.

The team therefore established an imperfect but significant relationship between tracer uptake and viraemic status. Those with detectable virus in the blood seemed to have more uptake than those on ART, with those on ART having more uptake than control participants.

In all participants there were high levels of tracer seen in the liver – this is to be expected as this is where the tracer will go to be metabolised. Those without HIV had higher peaks of concentrations of the tracer in the liver, possibly because in those with HIV the tracer is more distributed to all the tissues where HIV is present, i.e. the antibody in the tracer is attaching itself to lots of places in the body so there won’t be as much of it in the liver. Nonetheless, testing this theory on other tissue areas such as muscle and adipose tissue (where there is little to no active HIV replication) showed no significant difference between groups, so increased liver accumulation can’t be explained by a distributive phenomenon alone.

Another question the authors asked was if time on ART influenced the uptake of tracer, as this could indicate the antibody tracer was being taken up proportionally to the size of the reservoirs, which can decay in response to prolonged ART exposure.The team found that rSUV mean in the inguinal lymph nodes, axillary lymph nodes and gut all decreased inversely to time spent on ART. Firm conclusions are hard to establish as the sample size was so small, but this could support the idea that the tracer is actively binding to HIV virus – theoretically those on ART for longer time periods should have smaller reservoirs and thus less HIV to bind to.

Given that uptake was so much greater in the inguinal lymph nodes in viraemic groups compared to HIV-negative controls the authors wanted to correlate this with biopsy results. Five participants, four with detectable viral loads and one on ART had samples taken via biopsy of the inguinal lymph nodes. These were then assessed for levels of HIV (using flow cytometry to look for p24 antigen). The authors found a significant positive correlation between levels of HIV detected and uptake on imaging (p < 0.02). There was no clear relationship between blood levels of HIV and uptake, or levels of HIV in the inguinal nodes and uptake in other tissues. This suggests uptake is correlated in areas with tissue specific burdens of HIV, in other words potential HIV reservoirs.

Previous studies have shown that the brain, as part of the central nervous system (CNS), is probably an important site of ongoing HIV replication, even with effective ART. Unfortunately none of the study participants in any of the three groups demonstrated any uptake in the brain.. There is a significant tissue barrier between the circulation and the CNS which acts to protect the CNS from infection and/or toxins in the circulation. Some ART regimens are able to get past this blood-brain barrier whilst others are less successful in doing so. In this study it appears that the antibody-Zirconium tracer was unable to penetrate the blood-brain barrier and this is why there was no uptake demonstrated.


The team behind this study have demonstrated that MR-PET using a radioactively labelled, broadly neutralising, HIV-specific antibody may highlight areas where HIV persists. Lymph nodes and gut were already known to be long-lasting viral reservoirs but the study has demonstrated potential areas in the nose and bone marrow, both of which warrant further study.

The number of people in the study is small and the methods are both time and resource intensive. It is unlikely that MRI-PET for HIV will be of great use in the doctor’s office anytime soon. Further research applications could be found when looking for differences in viral reservoirs between demographic groups or varying ART regimens.

Potentially more important will be its use in research into strategies aimed at reducing the viral reservoir directly, as MRI-PET may provide a way of non-intrusively measuring the size of these viral reservoirs. Due to the non-invasive nature of MRI-PET, study participants may be more willing to undergo this rather than biopsy. However MRI scans are more time consuming (30 minutes in the scanner) and people often find them quite claustrophobic.

This proof-of-concept study lays the groundwork for what could potentially be an exciting academic tool to find the more elusive HIV reservoir sites that have so far evaded detection. Though a precise role of PET-MRI is yet to be defined, Beckford-Vera and colleagues have shown how combining cutting-edge technologies can progress our understanding of HIV. This may not be the breakthrough that allows us to eliminate the reservoir, but it may well prove a useful stepping stone.


Beckford-Vera DR et al. First-in-human immunoPET imaging of HIV-1 infection using 89Zr-labeled VRC01 broadly neutralizing antibody. Nature Communications 13: 1219, 2022 (open access).