Book Review: all the wrong conclusions

Recently I’ve been reading The River, an 1100-page book about the origin of the HIV/AIDS pandemic. After its publication in 1999 this book’s hypothesis was effectively refuted, and I won’t dispute that. If reading this seems weird or conspiratorial, remember that I read about cryptozoology sometimes, too. What I gained from this story was something else: a different moment in time, an analysis of how science knows what they know, and the puzzling chain of research which led the author to one theory.

I first heard about the book in a Reddit thread about Robert Rayford, a St. Louis teenager who died of unexplained causes in 1969. Once HIV/AIDS had been identified and studied in the 1980s, several old medical files with similar symptoms were revisited, samples were unfrozen, and Rayford is one of the oldest cases globally (and oldest in the US) to test positive. The implication was that HIV infected others in the US for years before getting the attention of the CDC. These older tests were known to have a higher rate of false positives (especially in labs which were simultaneously testing modern blood samples), so no one knows for sure.

Having grown up in the 90s, I remember learning about AIDS through donation drives and safety warnings. We watched Tom Sawyer and our parents warned us not to be ‘blood brothers’. Once I could read, I remember a poster in a doctor’s waiting room which showed connected couples. It said something like ‘Bob has HIV… everyone is surprised’ (I’ve tried Googling for this poster, no luck), warning that people could appear healthy for 10 years. I had no awareness of the fear and activism and science which defined the 1980s and early 1990s for many people. So I have been trying to pick up some new knowledge.

What we know

One of the positive things about this book is that in proposing an unconventional theory, author Edward Hooper must dissuade other conspiracies and prove parts of the official story which he agrees with. Then and now, science has genetic proof that multiple HIV-1 strains crossed over to humans unintentionally from SIV in chimpanzees, first becoming epidemic in Congo.
A separate HIV-2 strain crossed more recently from a related virus in mangabey monkeys in Cameroon (this part was new info for me). It took time for medical research to point to an origin in Africa, and then to trace an origin in West Africa when the earliest, hardest-hit regions seemed to be communities around Lake Victoria.

Origins

The earliest HIV-positive sample known today was collected in Congo in 1959. There are limited medical records to work with, but thousands of blood samples show a slow rise in the infection rate in Congolese cities for years after. There were two unusual deaths which doctors later reported in New York and the UK (also in 1959) but neither case has a convincing HIV-positive test or explanations for where they would have encountered the virus. Western travelers did not start returning from Africa with HIV/AIDS until the late 1960s/early 1970s. The Rumble in the Jungle brought many outsiders through Kinshasa in 1974, so it was once thought to be a turning point from local to pandemic. But at this point the virus was already spreading in Europe, Haiti, and the United States.
This is a repeated problem in HIV/AIDS history: before the early cases and the long incubation period were known, it was easier to think one person or event was the cause. Similarly Gaëtan Dugas (the flight attendant named “Patient O.” and erroneously called “Patient Zero”) would be the focus of a cluster study which actually failed to find any one person as its origin.

Earlier origins?

The River covers the early cases and then starts to go fishing for other historical outliers. For example, when Kaposi’s sarcoma was first described in several patients in 1872 Vienna, was that actually AIDS? How did children in 1960s Czechoslovakia get infected with Pneumocystis carinii pneumonia? (The author indicates that researchers have tested samples from the children, finding all are negative for HIV.)
During part of the author’s research for the book an additional 1959 case in the UK (David Carr) was believed to be HIV-positive. The author goes into naval records and talks to former friends to see where Carr might have gotten HIV, a different virus, or possibly radiation poisoning from nuclear testing. In the end he could not provide a solid explanation, as Carr’s military service did not align with nuclear ships and ports. Later these samples were sequenced and found to match a 1990 HIV virus, meaning that they were contaminated in the lab.

The OPV Hypothesis

The main theory of the book, disclosed early on, is that a 1959 oral polio vaccination test by Hilary Koprowski in the Congo and Lake Victoria region could have accidentally infected many people with HIV-1. This theory appeared in the late 80s and was popularized in a 1992 Rolling Stone article.

Hooper (author of The River) corrects some scientific errors which were made in the Rolling Stone article. He interviews contemporary rivals and former colleagues who criticize the effectiveness of Koprowski’s vaccine and his ethics. He correlates vaccination sites with early hotspots for HIV. He interviews people involved in the vaccination program and testing done on chimpanzees in Congo. There are conference records, research papers, and even a quote from another scientist warning in 1959 that Koprowski’s vaccine is contaminated with an unknown monkey virus. If I read this much and no outside sources, I might have been convinced.

The OPV Debunking

The story falls apart when you put all of its elements together and mix in modern science:

• Everyone involved in the vaccination program agrees that the vaccines were manufactured (‘attenuated’ to be precise) with green and rhesus monkey kidneys, and not chimpanzee kidneys.
  Hooper records one retired scientist saying ‘chimpanzee’, but he quickly corrects it, blames it on a language issue, and provides supporting evidence.
• Scrutiny is placed on Koprowski for not detailing his manufacturing process or which monkeys were used in the vaccine process, but for this to be a conspiracy, he would need to start covering up his work in the 1950s and 60s, before anyone knew its significance. More likely this was due to competition with other vaccine manufacturers.
• A lab in the Congo did test the effectiveness of polio vaccine by inoculating chimpanzees, but Hooper says that Koprowski manufactured the vaccine in Congo. Everyone involved denies this.
• The made-in-Congo allegation is necessary because HIV/AIDS did not appear in Western countries where the vaccine was tested, and chimpanzees were not used. Hooper now searches for proof of one bad chimpanzee batch.
• HIV-2 is explained away as a secret, unreported test by an unrelated lab.
• An oral polio vaccine would have a small chance of spreading SIV.
• In recent years, advances in DNA-sequencing and ‘molecular clocks’ compare genomes and estimate when viruses’ evolution diverged. When The River was written, the author was informed that both HIV-1 and HIV-2 appeared in humans in the mid-20th century.
  An HIV-positive sample from 1960 was found in 2008, and that virus is genetically different from the 1959 sample. It’s enough to convince scientists that HIV-1 must be older, possibly crossing over to humans around 1908. At that time Congo was a Belgian colony with devastating human rights abuses, and maybe chimpanzee bushmeat.
• So why did polio vaccine sites and early HIV hotspots line up? A man who participated in the vaccinations describes finding health partners and flying out to sites where the vaccine was requested. These locations were not random — they were responding to polio outbreaks, or were recommended to the program by Western doctors. These sites would be likely to be hit by other diseases, and have good reporting on those outbreaks.

Radiation, chemical weapons, and baboons

When the author raised the nuclear testing hypothesis for early AIDS cases, I started to get concerned about how many conspiracy-themes the book was buying into. The author attributes other cases to mining and shipping uranium. Rayford’s immune system failure in 1969 is blamed on chemical weapons tests which did occur in St. Louis, but seem out-of-place here. There are strong warnings (even in the preface) against transplanting baboon organs into people. This is a real practice which needs safety guidelines, but is not believed to cause any spreadable virus.

I’m also frustrated by how the preface, the author, and a source cited by the author all use the same rhetorical pattern. They describe how they first heard about the OPV hypothesis from someone with more apparent knowledge, got past their initial disbelief, asked for more evidence, and believed it was necessary to further confirm and promote their source. The author was in part informed by an independent researcher (Louis Pascal) whose narrative starts with hearing about the theory on the radio (Dr. Eva Lee Snead)… here it pulls the classic ‘appeal to a well-informed but un-biased non-authority’ of other conspiracies.

One conspiracy which I did think about later (but cannot corroborate anywhere) is that nuclear testing contributed to the HIV/AIDS pandemic by damaging immune systems. Atmospheric radiation fell in high concentrations in the Congolese rainforest, which would have made it easier to transfer SIV to HIV, or for HIV to mutate in humans.

Why now?

The impetus for the radioactive-rain theory and this whole book was finding out how HIV-1 and HIV-2, which were so different from any virus seen before, could both appear and become epidemics in such a short and recent timeframe.
The new ‘molecular clock’ research places them at 1908 and 1960, which is minutely recent on a human timescale.

I’ve had a few ideas which don’t need as much explaining and storytelling:

• Prion diseases also appeared in public consciousness in the 1980s and 90s due to mad cow disease, they have been around and causing other unexplained and misdiagnosed symptoms for a long time, and we don’t understand them very well either.
• Retroviruses have been with us for a long time. Wikipedia tells me that human DNA is 5–8% retroviruses. There’s ongoing research here, but these include harmful viruses which may have been contagious and serious outbreaks before recorded history.
• As has been pointed out by many other people, Africa was going through major transitions in the 1900s and 1950s. African communities and behaviors and interactions changed in ways that weren’t documented well enough for us to detect and trace early cases of HIV/AIDS.
• Scientists only recently observed health issues in chimpanzees with SIV. Wikipedia tells me that SIV is over 32,000 years old. It’s hard to think about these viruses crossing over only in the past 100 years. Consider though, how many times has Ebola crossed over? We know about outbreaks since the 1970s, but ebolaviruses and marburgviruses have been around for thousands of years, too. It’s possible that viruses a) take thousands of years to fully emerge, b) more typically cause less-serious health issues, or c) are more likely to cross over today due to ‘coddled’ immune systems.
  In the early 20th century polio was thought to be a ‘new’ virus, and it took time to realize that improved hygiene had protected babies but increasing the risk of older children and adults to be sickened and paralyzed. HIV/AIDS can infect any immune system, but this might play a role in viruses crossing over from chimpanzees and monkeys.