Robert Yarchoan '71: “The AIDS Epidemic: The Personal Journey of a Physician-Researcher”

The following is an edited transcript of the talk Robert Yarchoan ’71 gave on May 25, 2013.

I first would like to thank the college and everyone who came out here this morning—and particularly the students who managed to get up in time to get out here. I know this is a bit early to be up and about for many of you.

In this session, I want to talk a bit about my research in AIDS over the years, going back to the beginning of the epidemic. It’s best to start, however, around 1967, when I was graduating high school; at that time, there was a substantial optimism in the country that infectious diseases were really a thing of the past. A number of vaccines had been developed, and in fact the surgeon general declared victory against the threat of infectious diseases and suggested that our nation turn its resources towards the more important threat of chronic disease. I went on to college here at Amherst, where I was a biophysics major. I did my honors thesis with Professor Peter Offenhartz and, as part of my honors research project, developed a crude computer program to model the energy of simple organic compounds, not thinking that this had any connection with any sort of research on diseases. I then went to medical school at UPenn, did my residency in internal medicine and then went to the NIH to do a fellowship in immunology in the National Cancer Institute (NCI).

One of the things I should say is that the science training that I received at Amherst was key in terms of the work that I did later on, not so much in the specifics of what I learned, although there are some resonances, but more in terms of the scientific approach. I obtained a good background in medical science in med school, but it was really the training at Amherst that helped prepare me for research; at least until tomorrow, I never obtained a formal Ph.D.

So I joined a group that was studying the immune system and especially rare genetic immunodeficiency diseases—often children that were born with some failure of their immune system. And we used to see patients that had unexplained immunodeficiencies. Around 1981, we saw a patient from New York who had severe immunodeficiency, and we couldn’t figure out what was going on with him; he died soon thereafter. This was the first AIDS patient seen at the NIH. At about that time, the Centers for Disease Control reported a group of mostly gay men from the West Coast that had Pneumocystis pneumonia. Soon after, a report came out about a cluster of cases of Kaposi sarcoma in gay men. This is a skin cancer that was very rare in the U.S. It had generally been seen in people living around the Mediterranean and was also known to exist in Africa. Suddenly this cancer was occurring in the same population that was getting Pneumocystis,and, in some cases, patients were getting both, and the medical community started to connect the dots that it was a new disease.

Now one of the things that we lose track of is that when AIDS first came on the scene, it didn’t explode. This slide shows an article from The New York Times, sort of buried on the middle of the paper, reporting on this disease; people really didn’t perceive right away that it involved anything more than a few cases. This coverage was in contrast to, for example, the Legionnaires’ disease epidemic that appeared in people attending the American Legion convention in Philadelphia in 1976 and that garnered huge media interest. This slide shows a front page from the L.A. Times at the time.

Actually, even by the end of 1983, when it was clear that the disease we now call AIDS was expanding in a relentless way, there was still a sense that this was relatively limited in scope; a number of gay men had manifestations that might be pre-AIDS but maybe were from other causes, and the medical community still wasn’t sure. It was still considered a rare disease: at this time, there had been 3,000 cases and about 1,200 deaths, but it really wasn’t appreciated as to how extensive it was. Starting with that first patient in 1981, a number of scientists in the NCI and other Institutes in the NIH started working on AIDS. At the NIH, the various Institutes all have disease-related missions, and to a certain extent AIDS falls under one of the other Institutes, NIAID, which deals with infectious diseases. However, a lot of the research in AIDS was done in the National Cancer Institute, in part because Kaposi sarcoma was an important part of this disease, and in part because there were scientists there interested in immunology and cancer-causing viruses. At about this time, I joined the two other people as shown here, Drs. Sam Broder and Hiroaki Mitsuya; Dr. Broder had been given the charge by NCI Director Vincent DeVita to develop some sort of therapy for AIDS. For a while we worked on a related virus, called HTLV-1, that caused leukemia-lymphoma and a kind of immunodeficiency in Japan. However, during this time we were trying to figure out a good way to approach AIDS.

At about that time, two scientists, Luc Montagnier and Bob Gallo, did work in which they discovered the virus that we now call human immunodeficiency virus, or HIV, and showed that this was the cause of AIDS. Luc Montagnier and his group in France were the first to identify the virus, but Bob Gallo and his group at the NCI, in a series of four papers, showed quite elegantly that this was the cause of AIDS, and with these papers, a number of scientists working on AIDS were convinced that this virus was in fact the cause.

We obtained a pre-print of these four papers from Bob Gallo’s group before they came out. I remember reading them at that time and having a disturbing insight. As reported in one of these papers, Bob Gallo’s team had developed a blood test, a way of measuring antibodies to the virus, and then tested blood from people either with AIDS or who were at risk for AIDS. As you can see here, the people with AIDS were 87 percent positive; it wasn’t a perfect test, but it was pretty good. They then looked at groups of people at risk for AIDS. You can see the intravenous drug users were 60 percent positive, and gay men were 26 percent positive. I knew the population of gay men that they’d studied; it was a group that one of the other immunologists at the NIH had pulled together of relatively monogamous persons to study the immunologic responses to transplantation antigens. And in this population, about 26 percent were positive for exposure to (and presumptive infection with) HIV; doing some back-of-the-envelope calculations, I estimated that there were at least half a million people in the United States infected with this virus. At that time there was no cure, and there was increasing evidence to suggest that many or even all of the pre-AIDS patients would go on to develop AIDS and that this would lead to death. And it was clear, at least in my reading of it, that without us realizing it, this virus had weaved its way through the U.S. population and infected half a million people, who now essentially had a death sentence. This was quite frightening.

At the time that HIV was identified, based on the successes with other viruses such as polio, there was a lot of enthusiasm about a vaccine—there was a sense that there was a clear path to a vaccine. In fact, when the discovery of HIV was announced, Margaret Heckler, who was then secretary of health and human services, had a major press conference and stated that this would lead to an effective vaccine very soon, which was a prophecy that has not been realized almost 30 years later. As you probably know, there is no vaccine even now for HIV.

Our group took a different approach; we felt we should do something to try to treat the disease, and let other people work on the vaccine. There was at the time a lot of pessimism about whether anything could be done for people with AIDS. We thought the way to the disease was to go to attack the virus. HIV is a retrovirus, and there had been a lot of work on other retroviruses; many of these go on and cause cancers in various animals, and some of the steps in the life cycle of retroviruses were known at the time. This slide is more specific for HIV, but a number of these basic steps had been known for other retroviruses: The virus binds to a surface receptor on the cell and then fuses and inserts its genetic material into the cell. The genetic material in a retrovirus exists as RNA, and it is unique in that it has an enzyme that can convert the information from RNA into DNA. The DNA then can go and insert itself in the DNA of the infected cell, which is one of things that makes curing AIDS so difficult, because you’ve got cells that contain the DNA of the virus embedded in the DNA of the cell. At some later time, this viral DNA then gets activated and produces proteins and viral RNA, the proteins get cleaved by viral proteases, and a new virus is formed.

The one enzyme that really stood out as being unique and that we knew something about was reverse transcriptase. Other retroviruses have reverse transcriptases, and there had been some work done on ways of inhibiting them. Because it is a unique protein in HIV, we thought it was a good target to try to attack.

I should say a lot of people thought we were nuts at the time. There was a lot of reason to think that antiviral therapy for AIDS could not be developed. First of all, HIV integrates into the DNA of cells, and therefore skeptics felt it would not be amenable to treatment. Our thought was: Since HIV seems to have to move from cell to cell, it was still potentially attackable. The other thing is that the whole field of antiviral therapy was really new at the time. About the only antiviral drug that was really used at the time was acyclovir, and the idea of jumping in and developing an antiviral drug for this new disease was considered pretty far-fetched. Also, HIV, unlike the herpes virus that acyclovir works against, only had nine genes, so there were relatively few unique viral targets to attack. And then many scientists thought that there are so few infected cells in AIDS that most of the destruction of the CD4 cells and other cells of the immune system has to happen by indirect mechanisms, and even if you could stop the virus, you wouldn’t be able to do anything about this process. Finally there was pessimism that the immunological damage would not be reversible.

We thought it would still be worth a shot, and tried it. We divided our efforts. [Hiroaka] Mitsuya started working on an assay to test the drugs, and he developed an assay in which he used a T cell line that he made for some other purposes; this line was infected with another virus, HTLV-1, and it turned out to be very susceptible to being killed by the AIDS virus. And since we were in the same institute as Gallo’s group, we were able to go across the street, get the virus, bring over the vial, and start using it. This slide shows our initial primitive assay. This is a picture taken with a regular camera looking up at the bottom of a test tube in which these cells are growing, and you can see the size of the pellet of cells. This isn’t taken through a microscope. You can see that without any drug and without HIV, cells are growing quite happily. When you add HIV, the cells are getting destroyed, but when you add AZT, the cells are protected and not destroyed. The cells are also very happy with AZT by itself, showing that it isn’t toxic.

In terms of getting drugs to test, we looked at the literature, especially the work that had been done with the other retroviruses, and also went around to drug companies to see if they had anything that might work. We teamed up with the Burroughs Wellcome company, who had developed acyclovir, had expertise in antiviral drugs, and had started a program themselves trying to look at drugs to treat HIV. But they had no ability to use, or interest in using, HIV themselves. So we had a very nice relationship with them. They had some drugs and medicinal chemistry; we had the assay, and we had access to patients and a way of testing them.

We also identified some drugs based on earlier literature in mouse retroviruses, and one of them was ddI, as seen here; that drug was also quite active. We actually found fairly quickly a series of compounds that were active and shared one characteristic: they were all nucleosides, sort of the building blocks of DNA, but they had one modification, in that the 3’ OH group here is replaced by some other group, so they can’t form further bonds once used in a growing DNA chain. So the 3’ OH is either replaced by hydrogen or, in the case of AZT, an azido group. And all of these compounds are now AIDS drugs. Their long names are from the chemical names of these compounds.

The first of these compounds that we put into clinical trial was AZT. The reason we chose AZT was because we had a pharmaceutical company as a partner. They had done some animal toxicity testing, and in fact they were developing it as an antibacterial drug, so it was ready to go into humans, so we could very quickly start trials in AIDS patients. To give you a sense of the timeline here: The virus was discovered and shown to be the cause of AIDS in May of 1984, we found that AZT had activity in February of the following year, and we put it into the first patient in July of 1985, which is just about 13 months after the virus was discovered. If any of you are aware of the pace of drug development these days, this is sort of an indoor track record.

The initial trial was a Phase I trial to see if AZT could be given safely, but also to look at some parameters to see evidence of activity against AIDS. This slide shows the first patient that received AZT in our trial—he was a gay man who had very few CD4 cells, as seen here. We really didn’t have any way of measuring the amount of virus at the time, so we could only look at indirect parameters like the CD4 count. What we found is that there was a nice increase in the CD4 count and also the total T cell count. Before treatment, the patient had been anergic—in other words: When we did a skin test for tuberculosis (TB), the patient didn’t react, in spite of the fact that he had been exposed to TB. When we tested him a few weeks later, he had developed a very strong skin response, in this case indicating that his immune system had improved. We had a sense that something was going on with the patient, but it could also have been a fluke. We then continued to test further patients with increasing doses of AZT as the trial continued. The drug was first given intravenously, and then orally. As I recall, one of these patients lived in Aberdeen, Md., which is about 40 miles from the NIH, and would drive down every day to get his AZT by vein and drive back home again—it was really quite remarkable—until we were able to get him an oral form of the drug.

The FDA really worked with us during that time to try to expedite amending the protocol when needed and keeping things going as quickly as possible. And what we found, by the time we had treated about 12 patients or so, was that just about every patient had an initial increase in the CD4 count, and it was statistically significant. CD4 counts tend to bounce around a lot in AIDS patients, and for us this observation was huge. We were actually seeing a consistent increase in this key immunologic parameter … and we thought we had something.

And some of the patients were also feeling better. There was one female patient who was a nurse; she had become infected with HIV through a needle stick. When she started taking AZT, she had a fungus infection of her fingernail, and suddenly the fungus infection started clearing itself, and her normal nail started growing back.

When we had determined the tolerable dose of AZT on this Phase I trial, Burroughs Wellcome organized what’s called a placebo-controlled trial. They enrolled a number of people with AIDS or with severe AIDS-related complex—sort of pre-AIDS. Half of them received AZT, and half received placebo. And then they followed them for a period of time. There was a lot of controversy about whether this approach was ethical. But by the same token, if they hadn’t done that, the approval of the drug would likely have been delayed fairly substantially. This is the gold standard in clinical trials and the quickest way to see if a drug is really working. As you can see from the blue line on this slide, in the patients who got placebo, there were a series of AIDS-related events; the arrows show those that died. The red line shows the patients who got AZT; there was only one death and very few AIDS-related events. It was highly statistically significant. When it was clear that AZT was working, everyone on the trial was given AZT, and AZT was made available to patients all over the country on an expanded access program. It was approved in 1987 by the FDA, which is only 25 months after the demonstration of its in vitro activity and less than three years after the virus was first discovered. We were very pleased about that.

There was substantial interest in these results. President Reagan, who apparently had never uttered the name of AIDS before that, came to visit the NIH soon thereafter and, as part of his visit, toured our lab. This picture shows me at the time, shaking hands with Reagan. You can see that I was nervous about doing anything that would get the Secret Service suspicious—my other hand is held very cautiously in the back. I should also say the Secret Service agents were quite concerned about coming into our lab where we were dealing with the AIDS virus. They made us clean it up. And they also made us take down all our Far Side cartoons, which we were very displeased about. [Laughter] But anyway, President Regan then went and made his first speech about AIDS, and we were quite proud of the work that we had done.

While AZT was a start, it was soon apparent that it was not a cure and that the benefits to patients were often short-lived. Patients were still dying, and better treatments were urgently needed. A year or two later, the fear and concern of the AIDS patients bubbled up into a protest at the NIH. Led by the group ACT UP, they stormed the NIH, as you can see here on their slogan: “10 years, $1 billion, one drug.” Here is a picture of them at the NIH campus. Some administrators who were around at that time were taken aback by the protest—if you mention “pink smoke,” even today, it brings back vivid memories. We understood the concerns of the patients from talking to them privately, and felt that we shared a common goal in developing better therapy. And we continued to work at this.

Unlike AZT, in which we had a drug company as a partner, the next two drugs, ddC and ddI, were first developed here in the NCI and then licensed out to other drug companies. ddI was the next drug approved, and then ddC. ddC had been discovered first and put in trial first, but because we didn’t think it was as good, it was tested in a randomized trial and approved a bit later. So at this point we had three drugs that were working. I should say that the patents on these latter two drugs were held by the NIH, and we licensed them out to drug companies. We take some pride in the fact that our group at the Cancer Institute has actually shown a profit since the beginning of the AIDS epidemic on royalties from the drugs.

Once we had developed more than one drug, they could be combined; to test this, we conducted the first trials of combination therapy for AIDS. One of the problems with giving AZT was that the increases in CD4 counts were relatively transient. They would only last a few months, and then fall. We now know that this is because the virus becomes resistant. It was thought that combination therapy could help address this, and might also reduce the toxicity from high doses of any one drug. This is a slide showing results of a trial in which we tested a regimen of alternating AZT and ddI compared with a regimen in which the two drugs are given simultaneously. As you can see, people were having sustained increases in their CD4 counts for several years when we used the drugs together rather than using them in an alternating regimen.

One other thing to point out is that most pharmaceutical companies initially were very reluctant to get involved in developing AIDS drugs. They had the sense that this was an orphan disease, that there were not enough cases, that it was not worth it financially. We went around and gave little pep talks at various drug companies, trying to get them interested. Most companies passed. Burroughs Wellcome, to their credit, was just about the only one that seemed to be interested in those early days.

But after AZT was approved, it became apparent the drug wasn’t only useful for people with AIDS but also patients with pre-AIDS and that there was actually a big market. Other companies suddenly started getting involved. Since that time there have been a number of drugs approved. You can count them in various ways. I think it’s fair to say there are about 26 separate drugs. A number of them are in the class of nucleoside reverse transcriptase inhibitors shown in yellow here; others are chemicals with various structures that hit the same enzyme, shown in green. Another class of drugs, shown in blue, are HIV protease inhibitors, which block an essential viral enzyme that cleaves some of the proteins of HIV. There are additional drugs that hit other HIV targets.

One of the things that I’ve been bemused by is that this ties in with the work I did at Amherst. Scientists now design drugs based on the molecular structure of the target enzyme. For example, this is the structure of HIV protease, and medicinal chemists try to design compounds that bind to a cleft inside of it and block its action. They use x-ray crystallography to define the structure of the protease, and then model it on the computer with different drugs. When they study HIV protease that has mutated to become resistant to a given drug, they use computer modeling programs of free energy to try to identify which confirmation the protease is going to take. This is actually the same sort of approach that I was working on in my college thesis, although they now use much more powerful computers and better programs. But it has been satisfying to see that the approach I worked on with Professor Peter Offenhartz here at Amherst has been found to have wide utility.

Once protease inhibitors were developed, we had several drugs in two classes, and researchers could start to combine three drugs. It had become apparent that one of the big challenges in treating HIV is that reverse transcriptase makes a lot of mistakes, and so the virus can develop resistance very quickly. There are some drugs that, if used by themselves, can induce the development of HIV resistance in as little as eight weeks. And that’s obviously a big problem. What we have learned is that if you give three drugs, you can block the viral replication in patients down to almost zero so it doesn’t replicate enough to mutate and develop resistance. With this approach, you can achieve long-term effective therapy. These are the anti-AIDS drug cocktails that go under the name of “highly active antiretroviral therapy” (HAART).

This first really happened on a large scale in 1996, when protease inhibitors were approved in the U.S. and elsewhere. The introduction of these drugs had a dramatic effect on the AIDS epidemic. Once HAART was introduced, the number of deaths from AIDS, which is shown in blue here, went down fairly dramatically, and the number of people that became sick enough to meet the criteria for AIDS (because this therapy was given to people before they developed full-blown AIDS) also went down. This was very encouraging.

In fact, there was a sentiment, echoed by the media, that the AIDS epidemic was over. For example, this is a cover of Newsweek saying “The End of AIDS.” HAART did in fact have a dramatic effect and converted AIDS from essentially a death sentence to a manageable disease that one could treat over a period of time. A few years ago, it was estimated that, in the United States, 3 million years of life have been saved as a result of these sorts of treatments. And worldwide it’s estimated that about 14 million life-years have been saved as a result of AIDS therapies. However, as I will discuss below, the epidemic is evolving, and we are seeing new problems as patients live for years with HIV infection and the population of persons living with AIDS ages overall.

Periodically people ask me, “Why are we still studying AIDS in the National Cancer Institute?” One reason is the association of AIDS and HIV with certain cancers. Early on, as the Centers for Disease Control (CDC) was trying to track the course of this new epidemic, they developed and periodically revised the definition of AIDS. They considered that certain tumors, when they developed in an HIV-infected person, conferred the diagnosis of AIDS. These “AIDS-defining tumors” are Kaposi sarcoma, certain aggressive lymphomas and cervical cancer. One of the things that the ability to treat the virus did is to reduce the incidence of those cancers, like Kaposi sarcoma, that tend to occur in people whose immune systems are really depleted by the AIDS virus. Overall, we saw a fairly dramatic drop in the number of AIDS-defining cancers. This was quite encouraging. But the story here, too, is not so simple.

I’m now going to talk a bit more about some of the cancers associated with AIDS, because I’ve been focusing most of my recent research in that area. One of the questions that many people wondered was, “Why is it that the cancers that tend to develop in AIDS patients aren’t all cancers but these very particular cancers?” Kaposi sarcoma just exploded on the scene with the AIDS epidemic and had been a really rare tumor before that. It was one of those cancers that, before AIDS, would cause a lot of excitement on the wards—you would pull all the medical students in to see a case. For years, it really wasn’t clear why Kaposi sarcoma was so associated with AIDS. One curious observation was that certain groups of AIDS patients tended to develop Kaposi sarcoma, but others did not. Gay men tended to get it, while intravenous drug users didn’t. People who had contracted HIV infection from a blood transfusion did not get it. Researchers had a sense that there was some other virus or other infectious agent involved. And for years people looked for and could not find it. Finally the team of Patrick Moore and Yuan Chang, a husband-wife team who were at the time working in Columbia University, discovered a new virus that they called “Kaposi sarcoma-associated herpes virus,” abbreviated as KSHV. It is in the same family as Epstein-Barr virus (EBV), which causes mononucleosis. They also showed that this new virus was the cause of Kaposi sarcoma. If you had this virus and you had HIV, you were reasonably likely to develop Kaposi sarcoma. With that observation, it became clear that most cancers that are highly associated with HIV are caused by other viruses. The cancers in orange here are the AIDS-defining cancers; most are caused by EBV, KSHV or human papilloma virus. Also shown below are some other cancers that are associated with HIV; as can be seen, many of these cancers are also associated with other viruses.

It seemed that a main reason that these virus-induced cancers are more common in AIDS is because of the poor immunologic control. Some of these are chronic viruses that we all live with—for example, I would bet that almost everyone in this room is infected with EBV. Maybe we got mononucleosis when we were first infected, and maybe we didn’t, but either way, we live with the virus for the rest of our lives. When the immune system is compromised, EBV or other cancer-causing viruses start replicating more. Another key point is that these viruses live within cells, commandeer the cellular machinery and do some things to the host cells in order to survive better. They tend to push the cells in ways towards becoming cancer cells; if the cells get another nudge, they can become fully developed cancer cells, and a cancer develops. So this is the main principle. There some other cancers that develop in AIDS for reasons that are still not really understood. Lymphomas not caused by EBV are one example. And whether caused by other viruses or not, these cancers are continuing to be a problem.

The other thing I should point out is that people have this mistaken idea that AIDS has gone away or is not a major problem anymore. Shown here is some data from the CDC indicating the number of people living with AIDS in the United States. Look at 1996, when HAART came out; as you can see, the population of people with AIDS has not dropped since that time. In fact, just the opposite is true: the number of people living with AIDS continues to grow in this country, and in fact the number of cases has essentially doubled since HAART was developed. This is because the number of people newly infected with HIV every year has been pretty constant—at about 40,000 to 50,000—and they are living longer.

In addition, as a number of these patients are living for many years, the AIDS population is getting older. The yellow and the orange and red bars here show the age of the patients. As you can see, in the earlier days of the epidemic, it was mostly a disease of young people. Now the population living with AIDS is increasing, and it’s becoming an older population. In addition to the reasons I mentioned before, this is also because there is now a lot more HIV spread in the older population.

One of the other things that happened is that although the number of AIDS-defining cancers has gone down, the number of non-AIDS-defining cancers has increased. Now, some of this increase is from cancers that people develop anyway, particularly in the upper years of life—colon cancer, prostate cancer, and so on—and which are not associated with AIDS. But much of this increase is from cancers that are associated with HIV but can still take years to develop. As I mentioned before, many of these cancers are caused by other viruses—for example, anal cancer, hepatocellular carcinoma and nasopharyngeal carcinoma. Lung cancer has also increased radically in the HIV-infected population. We don’t know really why that is, to what extent it is due to increased smoking in this population, to what extent it may be due to increased inflammation, or whether there’s another virus involved—so there are still a number of things that we need to tease out.

In fact, if we look at cancers in AIDS overall, the number of cancers reached a nadir around 1997 and has been increasing since then, with most of the increase being these non-AIDS-defining cancers. However, we still see a number of cases of AIDS-defining cancers developing in this population. This next slide shows some data from France, where they have catchment areas for medical care and can more easily track the causes of death in a given region. As you can see, about a third of the deaths in HIV patients have been due to cancer, and cancer has now become the most frequent cause of death in these HIV patients. So this is an emerging problem in the AIDS population.

As I mentioned, lot of our work in my group in the National Cancer Institute has now switched over to addressing cancers associated with AIDS. It seems to me that, with the development of the initial drugs and demonstration that they could be profitable, the private sector was well poised to develop additional drugs for HIV and AIDS, and in fact that is what has happened.

By contrast, these HIV-associated cancers tend to be orphan diseases; there are many such cancers, but the number of cases of any particular cancer tends to be relatively small. Most pharmaceutical companies are thus not particularly interested in this area, and it seemed that it was an area where continued government research was needed. Shown here are some of the clinical trials that we’ve done either on new drugs or regimens to treat HIV-associated cancers. In some cases these were the first trials of drugs that were specifically developed for Kaposi sarcoma. In other cases, we have taken drugs developed for other tumors and explored their use in HIV cancers. For example, paclitaxel, which most people know by the name Taxol, was originally developed for other tumors; our group showed that it was very active in Kaposi sarcoma, and it is now approved for this cancer.

In recent years we’ve been focusing on a form of multicentric Castleman disease that develops in AIDS patients. This is a very interesting disease that is caused by KSHV, the same virus that causes Kaposi sarcoma. Basically some immune B cells that are infected with KSHV start increasing in number and produce what are called cytokines. Cytokines are proteins that are released into the blood and rev up the immune system. The cytokines produced in Castleman disease make patients incredibly sick—basically they get fevers and weight loss and, in severe cases, look like patients with rip-roaring sepsis from bacterial infection. And patients used to die of Castleman disease within a couple of years. We had noticed that some people with Kaposi sarcoma seemed to get better when they were given high levels of AZT, and we always wondered if there was some direct effect of AZT on Kaposi sarcoma or the virus that causes it. In fact, as people started understanding KSHV, they discovered that one of the proteins encoded by this virus actually activated AZT: this protein puts phosphate groups on AZT and transforms it to the form that works against HIV. However, this form is also somewhat toxic to cells. We hypothesized that with high doses of AZT, enough activation to this toxic form of AZT might occur in KSHV-infected cells so that these virus-infected cells would be selectively killed. We recently have done a clinical trial using this approach, combining AZT with another drug, valganciclovir. In the body, valganciclovir is converted to ganciclovir, which is similarly activated to a toxic moiety by another protein of KSHV. We found that a good percentage of patients with KSHV-multicentric Castleman disease improved when given high-dose AZT and valganciclovir. Shown here are two parameters of disease activity, C-reactive protein and interleukin 6—as you can see, both of these improve fairly dramatically. Also, patients had clear clinical improvement. So it’s been interesting to see a drug that we worked with early in the AIDS epidemic now having this different use in a disease, multicentric Castleman disease, that, until a few years ago, had been fatal in most cases.

Over the past several years, my career has taken a new twist. About six or seven years ago, the director of the National Cancer Institute asked me to form and lead an office to oversee AIDS and AIDS malignancy research throughout the Institute.

While a number of people continue to live with AIDS in the U.S., the AIDS epidemic is much more of a public health problem in other parts of the world. This slide shows the estimated number of cases of AIDS in various regions. So, as you can see, the real epicenter of the epidemic is sub-Saharan Africa, where there are now an estimated 22 million or so people infected. There’s also a major problem in Russia right now, and India has some cases, but the most cases are in sub-Saharan Africa. And the other thing about Africa is that even before the AIDS epidemic appeared, KSHV, which is a virus that actually evolved with the human species, was quite prevalent. It’s not known why it’s prevalent in some populations and not others, but it is highly prevalent in Africa. As can be seen on this slide, the regions of Africa in which there is a lot of HIV overlap with those regions in which there is a lot of Kaposi sarcoma, and in some of these countries, Kaposi sarcoma is the most common cancer in men overall. Kaposi sarcoma is often very, very aggressive in Africa.

I should mention that Harold Varmus [’61], who is now the NCI director, is also an Amherst graduate. He’s been very interested in doing global health … and he has been quite interested in the AIDS epidemic around the world. So one of the things our Office in the NCI has been doing has been to start supporting research on AIDS-associated cancers in Africa and also to do studies to try and figure out the best way of preventing and treating these cancers in low-income countries where you don’t have all the medical infrastructure that we have in the United States. This has been an interesting phase of my career, and I am doing it while I continue to do some other work in the lab. And there is synergy between these two roles. For instance, there are only rare reports of Castleman disease in Africa. We believe that, given the high prevalence of KSHV in Africa, Castleman disease has to be relatively common there, and we’re trying to figure out why these cases are being missed and what to do about them.

I do want to give a few thanks and acknowledgements here, and these are not the ones that I usually give in medical talks. I, first of all, want to thank my parents, who gave me a good start. Also my wife, Giovanna Tosato, who’s been a colleague of mine—she is also a principal investigator in the Cancer Institute, and we have collaborated on a number of projects and talk science a lot. I also want to give thanks to my two sons, Mark [’07] and John [’13], who are here in the back—they have been an inspiration to me all these years. I want to thank Amherst College for my background in science; my Amherst thesis advisor, Peter Offenhartz; my other mentors; my colleagues at the NIH. I really should thank the National Cancer Institute for providing a supportive environment and the American taxpayers for funding the research. That really made it all possible. And finally the patients who volunteered for our trials.