In conversation with four experts who are applying lessons from the decades-long battle against HIV/AIDS to the ongoing COVID-19 pandemic.
By Michael Dumiak and Kristen Jill Kresge
Forty years ago, the U.S. Center for Disease Control and Prevention’s (CDC) Morbidity and Mortality Weekly Report noted an unusual cluster of five cases of Pneumocystis pneumonia among gay men in Los Angeles, heralding the HIV/AIDS pandemic. Since then, more than 77 million people have been infected with the virus and nearly 35 million have died from AIDS-related illnesses.
From the time of that first case report in 1981, it took a couple of years for HIV to be identified as the retrovirus that caused AIDS. It wasn’t until four years later, in 1985, that the U.S. Food and Drug Administration licensed the first test to detect the virus in blood. Ten years after the first report of what came to be known as AIDS, scientists discovered combinations of drugs that could help keep the virus in check. Then it took several years until these life-saving medicines made their way to some of the hardest hit countries in sub-Saharan Africa.
Those timelines stand in stark contrast with the scientific and medical progress in combatting COVID-19. Despite several missteps, within a year of when SARS-CoV-2 was identified as the cause of this new human disease, tests, antibody treatments, and vaccines were already available. In a recent perspective article, Kevin DeCock and Harold Jaffe of the CDC and James Curran of Emory University reflected on 40 years of AIDS. They wrote that: “As a result of technologic advances such as whole-genome sequencing, scientific progress on COVID-19 has been breathtakingly rapid compared with early laboratory research on HIV.”
The speed with which scientists were able to develop COVID vaccines can also be attributed, in part, to the remarkable scientific advances by researchers who have spent decades untangling HIV’s thorny traits and applying creative strategies to counteract them. Although none of these vaccine strategies have so far been successful for HIV — one of, if not the, most difficult viral target researchers have ever faced — this work facilitated the swift progress in tackling SARS-CoV-2.
Now, a major challenge is ensuring COVID vaccines are available globally. Here too, HIV/AIDS can offer valuable lessons. “Although initially slow, the HIV/AIDS response over the years has been a beacon in global health for respect for individuals and their rights and for health equity,” write DeCock, Jaffe, and Curran.
But this is still a work in progress. Despite best efforts, several proven HIV prevention strategies and a highly effective armamentarium of antiretroviral drugs, four decades into the HIV/AIDS pandemic, the virus continues to spread and kill. In 2020, while the world faced the second pandemic in half a century, 1.5 million people were newly infected with HIV and 690,000 people died from HIV/AIDS-related causes, according to the latest figures from UNAIDS. Vulnerable populations still bear the greatest burden when it comes to HIV.
Data from UNAIDS also suggest progress in treating and preventing HIV is slowing. Disruption of treatment and prevention programs are just one of the many consequences of the COVID-19 pandemic, and UNAIDS warns there may be lingering effects on HIV/AIDS programs if COVID vaccines aren’t made more widely accessible in developing countries.
In the end, the sustained global response to HIV, whether measured through financial investment, community engagement, or scientific progress, may offer important lessons not just for how the world handles COVID, but for pandemic preparedness overall. As Jaffe, DeCock, and Curran conclude: “More reflection is required with regard to what the responses to HIV and Ebola have taught us and how they might be relevant to COVID-19 and other future epidemics.”
We spoke with four experts to explore how the past four decades of HIV/AIDS science have prepared us for ongoing and future pandemics.
A giant in the vaccine field
There are several heroes in the rapid development of COVID-19 vaccines. Without a doubt, one of them is Barney Graham.
Graham is deputy director of the Vaccine Research Center (VRC) at the U.S. National Institute of Allergy and Infectious Diseases (NIAID) and chief of the Viral Pathogenesis Laboratory and Translational Science Core. In this position, he oversaw the design of the modified SARS-CoV-2 Spike protein and worked with the company Moderna to develop one of the first COVID vaccines authorized in the U.S.
This was the culmination of decades of work, and it started with HIV. “In the late 90s/early 2000s, people started figuring out how to make human monoclonal antibodies. Monoclonal antibody isolation was largely motivated by HIV,” recalls Graham. Technological advances in the early 2000s allowed scientists to more readily isolate and clone human monoclonal antibodies.
This work, in combination with advances in determining the structures of viral proteins, eventually led Graham to pursue a vaccine against respiratory syncytial virus (RSV) alongside Jason McLellan, a structural biologist who started out in Peter Kwong’s lab at the VRC and is now an associate professor at the University of Texas at Austin. “Jason was working with Peter on this HIV structure, and he wanted to work on something else. He asked me whether I had any ideas for him. I said, ‘Well if you’re willing to work on RSV with me, nobody else really cares about it.’ And so, Jason and I built a friendship and we started working on different RSV epitopes.”
The turning point in their work on RSV came when they obtained the crystal structure of the prefusion form of RSV’s F protein using stabilizing mutations they identified in collaboration with Kwong. “This turned out to be a much, much better vaccine antigen,” Graham says. “It gave you a 20-fold boost in neutralizing activity. We’ve done a clinical trial and proved that it really was also good in humans.” This vaccine candidate is now being developed by Pfizer and GSK.
From the work on RSV, scientists were able to develop more stable protein structures for several viruses, including parainfluenza, Nipah, measles, mumps, metapneumovirus, and coronaviruses.
“As we were bringing that first part of the RSV story to a close, that’s when MERS was happening,” Graham recalls. MERS, Middle East Respiratory Syndrome, was first reported in Saudi Arabia in 2012. Its cause was a novel coronavirus. “Our lab made some MERS vaccines, but we were never very satisfied with what we were able to do. Part of the problem was there was no structural information on any coronaviruses.
“Jason [McLellan] was headed to Dartmouth for his first faculty position in late 2013 and he was looking for an area where there wouldn’t be quite so much competition as in HIV. We talked about it and thought coronaviruses would be the perfect place to work because there were no structures. And so that’s what we agreed to work on together.”
Every pandemic threat, including HIV, began as a regional problem that wasn’t recognized and dealt with in time. If we really want to get on top of pandemic threats, we need to use global resources for regional problems.
Initially, the pair came up empty handed in their efforts to stabilize the Spike protein structures of either MERS or SARS-1. “It wasn’t until we serendipitously started working on the endemic betacoronavirus HKE1, which we did because a post-doc in my lab had gotten sick and he ended up having it, that we got this to work,” says Graham.
He and McLellan engaged Andrew Ward, a professor at Scripps Research and an expert in applying cryogenic electronic microscopy to obtain atomic-level protein structures. Ward was immediately able to obtain the structure of Spike for the HKE1 betacoronavirus. This led Graham and McLellan to try to stabilize the Spike protein.
They identified stabilizing mutations for HKE1, and these same mutations ended up also working for MERS and SARS-1. “It happened to work in almost every coronavirus we tested,” says Graham. “Now, we thought we had a generalizable approach to stabilizing the Spike protein that could make it a better vaccine antigen. Interestingly and unexpectedly, those stabilizing mutations also improved protein expression from transduced cells, which means if you’re delivering the antigen in a gene-based vector, you’re going to make a lot more protein and it’s going to end up being a better vaccine.”
This was a major turning pointing in vaccine discovery. “All of that came from RSV, which all came from HIV technology,” he adds.
And all of it led to the fastest vaccine development in history — Graham and McLellan were able to use this same approach to develop a COVID-19 vaccine within a year.
Graham and his NIAID colleagues had been working with Moderna on paramyxoviruses since 2017, so when they suspected that a new human coronavirus was behind the cases of atypical pneumonia being reported in China, they were ready to use mRNA technology to try to rapidly develop a vaccine. As soon as the genetic sequence for this new virus was published, Graham and McLellan began designing proteins.
“We were confident enough to apply that same method [of protein stabilization] without any additional experimentation to improve the structure, and it worked,” Graham says, allowing his signature modesty and gentle nature to shine. “We’re very fortunate and happy that it worked out.”
It’s hard to imagine ending your research career on a higher note — at the end of August, Graham is retiring from the federal government. But he still has much to contribute. “I really want to try to help move some of this technology to lower- and middle-income country settings and help people understand that, like we’ve said a million times, a problem anywhere is a problem everywhere.
“Every pandemic threat, including HIV, began as a regional problem that wasn’t recognized and dealt with in time. If we really want to get on top of pandemic threats, we need to use global resources for regional problems,” he says.
And he’s not giving up on the problem that started all of this work in the first place. “I’m still hoping that eventually we can get back around and figure out how to get HIV taken care of.” —KJK
Drawing on hard-won experience
Talking to Glenda Gray is a good reminder that political and social rancor tied to health emergencies aren’t a feature unique to the moment. Here is a person who came of age in medicine in South Africa during the decline and fall of apartheid while, at the same time, HIV was rising to its terrible heights.
And even far along into the struggle to make lifesaving antiretroviral therapies affordable and accessible in South Africa, public health workers, including Gray, were having to counter complacency and outright opposition. “Remember, we were still largely denialist at a government level, and patients were being given multivitamins, garlic, beetroot, and ginger,” Gray says tartly. “We were in a situation that, in hospitals, these were prescribed.”
Glenda Gray is a pediatrician at heart, and some of her early research focused on curbing HIV transmission from mother to child. In just five years, from the late 1980s to early 1990s, HIV prevalence in pregnant women in South Africa shot up from one in 100 to 30 in 100. As a result, more babies were becoming infected. “A lot of women only found out they were infected when their babies got ill,” she says. “It just got worse and worse. One in three children who were admitted to the hospital were dying from HIV.” In the face of considerable opposition, Gray drove research on the use of antiretroviral therapies that eventually became a mainstay of curbing mother-to-child transmission and found ways to get around what at the time were exorbitant costs for HIV treatment.
Gray is now president of the South African Medical Research Council, a co-principal investigator at the HIV Vaccine Trials Network (HVTN), and an influential academic researcher. She led the research committee advising the South African Health Ministry on the COVID-19 pandemic. The country, like many others, had to make difficult and contentious decisions on lockdowns during the COVID-19 pandemic, with the country at times a global hotspot.
Gray also led Uhambo, HVTN 702, the large, Phase IIb/III HIV vaccine trial that was stopped in 2020 for lack of efficacy. “It’s 10 years of hard work that goes into a trial that has a negative finding. It’s devastating,” she says, with high stakes and stress at every board meeting.
But Gray and her colleagues keep moving ahead. “We tend to be optimists.”
Today, access to antiretrovirals in South Africa is night and day compared to the times of beetroot and garlic. But that didn’t happen by itself, Gray says. It took grass roots and legal action to bring more affordable therapies to South Africa, the country hardest hit by HIV/AIDS. It took pressure to convince leery ethics committees to agree to trials in resource-limited settings. “Women drove their agenda. They shook the tree to make sure we were not dispassionate about things like that,” Gray says.
Even trials that got negative results created possibilities. “We created these centers that could look after mothers and children, which naturally opened up an avenue for us to do clinical research in children, mothers, and fathers,” she says. These became sites that could roll out antiretrovirals. Lab infrastructure followed. “The clinical research and the lab ability allowed us to start working in HIV prevention with microbicides and on monoclonal antibodies — it’s a tour de force in Africa.” Developing this infrastructure gives opportunity to the African research talent already present.
It also helped with the response to COVID-19. Gray was able to help organize an open-label trial of Janssen’s/Johnsons & Johnson’s COVID vaccine that inoculated 500,000 South African health care workers in under 16 weeks. “No one’s ever done that before here,” she says. “When we needed to get the funding to roll this out, people said, ‘Have you ever done something like this before?’ And we said, well, we’ve rolled out antiretrovirals, and we rolled out mother-to-child transmission.”
In other words, yes, Gray says. “We harnessed the decades of HIV experience to support the COVID response here in the country.” —MD
Peeling back the layers on pan-coronavirus vaccines
Coronaviruses have likely been infecting humans for centuries. Most of them cause some variety of the pesky, common cold. But as the world has now seen firsthand, they can be far more dangerous. SARS-CoV-2 is the third coronavirus to cause life-threatening disease among humans in the past 20 years. It probably won’t be the last.
SARS-CoV, the virus that caused severe acute respiratory syndrome (SARS), emerged in China in 2002, followed by the MERS (Middle East Respiratory Syndrome) coronavirus, which was first detected in Saudi Arabia in 2012. SARS-CoV spread easily and killed almost 10% of those who were infected. While less transmissible, MERS-CoV was much more fatal, killing nearly 35% of those infected. The case fatality rate for SARS-CoV-2 is estimated to be around 2%.
Genomic analysis of SARS-CoV-2 show the virus is approximately 80% similar to SARS-CoV-1 and 50% similar to MERS-CoV. As indicated by the rapid and exponential spread of SARS-CoV-2 around the globe, this is a highly transmissible virus. Fortunately, it is not as deadly as either SARS or MERS.
But the next coronavirus may combine the worst characteristics of both. To be prepared for that eventuality, researchers are calling for development of pan-coronavirus vaccines — vaccines that will be broadly effective against various coronaviruses. “It’s like the layers of an onion,” says Dennis Burton, a professor of immunology and microbiology at Scripps Research in La Jolla, California, and scientific director of IAVI’s Neutralizing Antibody Center. “As you go in, each layer becomes more and more difficult.
“If you begin with the easiest, that would be pan-SARS-related viruses or sarbecoviruses. These would be vaccines that would work against SARS-CoV-1, SARS-CoV-2, and probably viruses in between those two, SARS-3 and SARS-4 if you like, depending on how similar they were. We know that this is possible in principle because we have antibodies that neutralize SARS-1 and SARS-2 very well. And it is the same principle as with HIV or with a universal flu vaccine — if you have the antibodies that are cross neutralizing in hand from natural infection, then, in principle, you should be able to design immunogens or vaccine candidates that induce those sorts of antibodies.”
Like HIV, the starting point for this effort is broadly neutralizing antibodies — those that can act against many different strains. For HIV, the virus mutates at such an alarming rate that the diversity of the virus is a huge obstacle to vaccine development. “HIV is the king of antibody avoidance,” says Burton, which is why researchers have been stymied so far in their ability to make a broadly effective vaccine. Researchers estimate that a single HIV-infected person may harbor as many as 100,000 different HIV strains.
“Influenza is also pretty sophisticated,” he notes, which is why annual jabs are required against whichever strains researchers anticipate will dominate from season to season.
But SARS-CoV-2 and other coronaviruses, particularly SARS-CoV-1, are similar enough that antibodies to one can also knock out the other. “Some of the first neutralizing antibodies to SARS-2 that were identified were from SARS-1 infected individuals,” says Burton. “But you can also find SARS-2 infected individuals who make antibodies that are cross-neutralizing to SARS-1. And in animal models, some of these cross-reactive antibodies do protect.”
Fishing out SARS-specific antibodies has proven much easier than it was for HIV. Part of this is because of technological advances, and part of it is just because you aren’t searching for a needle in a haystack. “The technologies are in place to screen sera in donors more rapidly using pseudovirus assays, and the process of isolating monoclonals from single B cells is also well established now so it is much easier to get the antibodies than it was 10 or 15 years ago,” Burton says.
“However, SARS-2 is also a much, much easier virus than HIV so there are lots more cross-reactive antibodies around. You don’t have to search through literally thousands of individuals to find what we’ve termed elite neutralizers as we had to do with HIV. There are many more of those sorts of people around with SARS-CoV-2.”
I think the enormity of this pandemic has woken folks up to the dangers of infectious disease so I think that there will be substantial investment in pandemic preparedness, and rightly so because you can see that the economic and health impacts are huge.
The virus itself, particularly the receptor-binding domain portion of the SARS-CoV-2 Spike protein, is more exposed to antibodies than HIV. “There are large, exposed surfaces on the receptor-binding domain that are very easily recognized by antibodies,” Burton says. This is likely why vaccination with almost any S-protein preparation induces fairly reasonable neutralizing antibody levels, he adds. “You can make antibodies that are very potent against SARS-2 very easily, with a minimal amount of maturation.” And though cross-reactive SARS antibodies require some level of maturation, “it’s still not anywhere near as difficult a problem as it is for HIV.”
This suggests that developing a pan-SARS-related virus vaccine looks feasible, even though SARS-CoV-2 is beginning to show some signs of mutating to avoid antibodies. “The virus is beginning to evolve mechanisms of escape or avoidance of antibody responses,” Burton says. “The variants of concern are the indication that the virus is hitting back.”
The next layer of the onion would be to go from a vaccine that could protect against pan-SARS-related viruses to protecting against all betacoronaviruses, which would include MERS and some of the seasonal cold-causing coronaviruses. Even at this level, Burton says there is already evidence of some degree of cross-reactivity for the antibodies that researchers have identified.
The final layer would be pan-coronavirus vaccines more generally, which would include both the alpha- and betacoronavirus families. “That, we would guess, would be very difficult because at least so far we’ve not seen antibodies that are cross-reactive between all alpha- and betacoronaviruses.”
However, even achieving the first step would be an accomplishment. “Having antibodies and vaccines that were active against even the SARS-related viruses would be very valuable.
“I think this is a process that will probably be taken in stages and vaccine designs will probably arise with more and more difficult targets in mind,” he says.
Developing pan-coronavirus vaccines will require a long-term investment and research commitment. Burton and his Scripps Research colleague Eric Topol suggest an investment of US$100 million to $200 million and several years of work is required to take these concepts from basic research to Phase I trials. But Burton thinks that COVID-19 has convinced almost everyone about the importance of preparing for the next pandemic, no matter what the cost.
“I think the enormity of this pandemic has woken folks up to the dangers of infectious disease so I think that there will be substantial investment in pandemic preparedness, and rightly so because you can see that the economic and health impacts are huge.”
It’s also likely that efforts to develop pan-coronavirus vaccines will return some of the favors HIV research has offered. “There will always be advantages going both ways. We have already been working on the mRNA platform for a number of years with HIV immunogens and they will be coming to clinical trials quite soon, but SARS-CoV-2 has dramatically demonstrated the potential of mRNA vaccines and reduced some of the barriers to using these vaccines, so I think that’s going to be an enormous help to HIV vaccinology for sure.” —KJK
The signal finder
Perhaps never has so much of the general public paid attention to clinical trials the way they did in the last 18 months, as scientists raced to develop vaccines to prevent the spread of SARS-CoV-2. The interest in clinical trials — and the unprecedented speed through which vaccine candidates passed through them — was remarkable.
As biostatistician and clinical trial designer for the HIV Vaccine Trials Network (HVTN) based at the Fred Hutchinson Cancer Research Center, Peter Gilbert played an influential role in creating the clinical trial protocols for evaluating some of the vaccines that are proving successful against COVID-19. “It felt like it was reasonably easy and straightforward to pivot from HIV to COVID as a statistician because many of these statistical methods had been worked on for many, many, many years for HIV vaccines,” Gilbert says.
Decades of HIV vaccine research have given statisticians like Gilbert ample opportunity to develop flexible clinical trial protocols. “Many of them really carried over very well, some without any modification at all. They were just ready. We had an opportunity to use these tools that we built for a very pressing situation.” As the pandemic took hold, HVTN could also quickly repurpose some of its clinical trial infrastructure to help test vaccines developed under the U.S. government’s Operation Warp Speed program. Moderna, Johnson & Johnson, and AstraZeneca all made use of the HVTN.
Gilbert calmly recalls those intense days, pausing to think about his answers and showing the kind of sudden insight you see sometimes in those with higher order math skills. Having come to the HIV field in the late 90s as a math major at the University of Washington — following in the footsteps of his parents, prominent statisticians at the Seattle institution — Gilbert has both played a part in and witnessed increasing precision in data analysis over the years, developed to match more sensitive assays and advances in computing power.
“The data get richer and richer over time. Ages ago when I came into the field, we didn’t even have good viral load assays for HIV,” he says. “I remember being a student and sitting in Jim Mullin’s lab meetings. I remember the big nut they were trying to crack was how to make sure they could quantify viral load. That was the big problem. Now, of course, viral load is a well-accepted surrogate endpoint for HIV treatments.”
Faster lab analytical tools and more robust assays are also helping researchers understand the diversity of HIV — a virus that changes its structure at a dizzying pace after infection — and to do so closer to time of infection, Gilbert says. During HVTN studies Gilbert worked on in the past, single-genome amplification measured perhaps 10 viral sequences per infected individual. Newer deep sequencing can now deliver 200. “The new technology allows us to get deeper biological insights. Technology tends to drive the statistical methods, so the statistics have to catch up and become more complex to capture something about that new biology.” Applying those more complex statistics seems to invigorate Gilbert.
The efficacy of COVID-19 vaccines was in no way ambiguous, but that isn’t always the case. Gilbert points to results from the AMP trials, which tested the ability of an intravenously delivered broadly neutralizing antibody to prevent HIV infection. Even though this single antibody was not effective overall, the trials were large enough, and the analytical tools used by Gilbert and his colleagues were precise enough, that they were able to find a signal of efficacy. “We were able to piece together everything and get an insight that it actually does work if the virus is sensitive — and we can measure that.
“The exciting part is that the planning of the lab work and then the planning of the stats mesh in a cohesive way with the clinical work. It all worked. It was just enough to get an answer.”
And even if that answer isn’t as positive as it was for COVID-19 vaccines, it is still helping to advance the field. —MD