Unrecognized “AIDS” in Monkeys, 1969–1980: Explanations and Implications

Abstract

AIDS was recognized in humans in 1981 and a simian form was described in the years 1983 to 1985. However, beginning in the late 1960s, outbreaks of opportunistic infections of AIDS were seen in monkeys in the United States. This apparent syndrome went unrecognized at the time. We have assembled those early cases in monkeys and offer reasons why they did not result in earlier recognition of simian or human AIDS, including weaknesses in understanding disease mechanisms, absence of evidence of human retroviruses, and a climate of opinion that devalued investigation of infectious disease and immunologic origins of disease. The “epistemological obstacle” explains important elements of this history in that misconceptions blocked understanding of the dependent relationship among viral infection, immunodeficiency, and opportunistic diseases. Had clearer understanding of the evidence from monkeys allowed human AIDS to be recognized earlier, life-saving prevention and treatment interventions might have been implemented sooner.

AIDS was first recognized in humans in 1981.¹ In the years 1983 to 1984, a retrovirus (named HIV in 1985) was determined to be its cause. HIV has infected 78 million people, 39 million have died, and 37 million are living with HIV.^2,3

A form of AIDS in monkeys (caused by simian immunodeficiency virus [SIV]) was described between 1983 and 1985. However, outbreaks of opportunistic infections of AIDS were seen in monkeys in primate centers and laboratories in the United States starting in the late 1960s, with some evidence for virologic and immunologic etiologies. This apparent syndrome went unrecognized at the time. We have assembled those early cases in monkeys and offer reasons why they were not connected. We also speculate on what earlier understanding of these outbreaks in monkeys might have meant for the recognition of and response to AIDS in humans.

EARLY CASES OF “AIDS” IN MONKEYS

Between 1969 and 1980, reports from primate centers and laboratories documented outbreaks of unusual conditions in monkeys that were subsequently recognized as opportunistic infections of AIDS. A 1969 review listed 23 cases of lymphoma and leukemia in nonhuman primates scattered over the period 1923 to 1968 and stated that “reports from laboratories in which large numbers of young monkeys have been used for . . . experiments indicate that malignant neoplasms were rare.”^4(p161) In 1970, the first cases of lymphoma in monkeys were reported from the New England Regional Primate Center, a subsequently important center in the investigation of simian AIDS, in an article that also first associated the lymphomas with a herpesvirus.⁵

During the early 1970s, pathologists David Gribble and Boris Ruebner and veterinary student (and coauthor of this article) Roderick Bronson conducted necropsies of almost 300 monkeys at the National Center for Primate Biology at the University of California, Davis (UC Davis). The first published report from UC Davis, by Stowell et al. in 1971, documented an outbreak of lymphoma in 24 rhesus monkeys and concluded that this “rare” outbreak with “astonishing incidence” of disease could improve understanding of similar types of human cancer.⁶ They inoculated healthy monkeys with lymphoma cells and lymphomas appeared within three weeks, indicating transmission of an infectious agent.

In 1972, Bronson et al. reported nine cases of Yersinia pseudotuberculosis among macaques at UC Davis, an outbreak attributed to gastrointestinal disease.⁷ Avian tuberculosis, later seen as an infection associated with SIV, was found in numerous monkeys in the late 1960s and early 1970s, at both the UC Davis⁸ and New England centers.⁹ Monkeys at UC Davis were also diagnosed with Mycobacterium avium-intracellulare (MAI) and herpesvirus simiae infection, with additional diagnoses of MAI and oral candidiasis between 1976 and 1978.¹⁰

In a 1974 review, Manning and Griesemer assembled 59 cases of lymphoma among several types of monkeys since 1969 and reported that “temporal-clustering . . . suggests . . . an infectious etiology.”^11(p206) In at least one case, vertical transmission from mother to offspring was hypothesized.¹¹ Most of the cases were from UC Davis, but four were from the Southeast Asia Treaty Organization (SEATO) Medical Project colony in Bangkok, with possible etiologies related to malaria, splenectomy, cross-transfusion of blood, and prior herpesvirus hominis infections¹²; two were from the University of California Medical Center, San Francisco, and were thought to be associated with a C-type retrovirus^13,14; one was from the Naval Aerospace Medical Research Laboratory in Pensacola, FL¹⁵; and one was from the Yerkes Primate Research Center at Emory University.¹⁶ In 1980, Terrell et al. concluded that the outbreak at UC Davis “suggests an infectious etiology [but] to date no suspect agent has been isolated.”^17(p564)

Manning and Griesemer attributed the “appreciable increase” in case reports of lymphoma in monkeys to “a recognition by researchers of possible significant parallels in the etiology of this disease in non-human primates and man.” ^11(p204) They predicted that “an understanding of the factors responsible for the outbreak of spontaneous lymphoma in nonhuman primates will most certainly enhance an understanding of this disease in humans,”^11(p209) but this took almost 10 years to develop.

In 1975, Gribble et al. reported the first cases of progressive multifocal leukoencephalopathy (PML) in nonhuman primates—eight macaques at UC Davis. Several of these animals were also diagnosed with chronic colitis (shigellosis), avian tuberculosis, chronic anemia, and lymphoma.¹⁸ PML in humans had long been associated with immune deficiency from treatment by immunosuppressive agents and possibly with viral infection.^19,20 Gribble showed slides of the monkeys with PML to Edward Richardson of Harvard Medical School, who had discovered PML in humans and linked it to immunodeficiency. Author Roderick Bronson was present and recalled that the sole topic was the similarity in the pathologies of monkeys and humans with PML. Neither the relationship between immunosuppression and PML nor the potential cause of such immunosuppression came up. Gribble et al. concluded that further investigation of PML in monkeys “may give insight into PML in man.”^18(p604)

In 1977, Holmberg et al. reported a papovavirus (SV-40) in two rhesus monkeys with PML at UC Davis and commented on the “striking similarity” between PML observed in monkeys and humans.²¹ They suggested that immunosuppression could be a prerequisite for PML in monkeys but focused on the viral etiology of the disease without asking why some monkeys with SV-40 got PML but others did not.

Between the mid-1960s and the mid-1970s, Jarrett, Essex, and colleagues identified a feline leukemia virus, determined that it caused immune deficiency and was horizontally transmissible,^22,23 and characterized it as a retrovirus.^24,25 Prefiguring human AIDS, Essex showed that immune deficiency and opportunistic diseases killed more cats than did feline leukemia virus itself.²⁶ No connection was drawn at the time between the strands of investigation in monkeys and cats.

Possible immunologic causation of diseases in nonhuman primates was suggested in several reports from UC Davis in the 1970s, specifically in cases of PML,¹⁸ atypical tuberculosis,²⁷ lymphoma,^17,28 cryptosporidiosis,²⁹ cytomegalovirus (CMV), and retroperitoneal fibromatosis (now known to be caused by a herpesvirus similar to the cause of Kaposi’s sarcoma, an early and common opportunistic disease of AIDS in humans).^30–31

No further connection was drawn among viral infection, immune deficiency, and malignancies until 1978, when Francis and Essex hypothesized a relationship between horizontally transmitted retroviruses and lymphoma in animals and humans.³² They commented “that if human malignancies are caused by infectious agents, then their epidemiologic pattern should show the typical characteristics of infectious diseases, such as person-to-person transmission, case clustering, and seasonality.” In the context of “feline leukemia or lymphoma, clusters and cat-to-cat transmission of tumors are rarely observed,” but Francis and Essex urged further research in humans in which

investigators . . . examine both cancer patients and their contacts [and are] alert for unique occurrences of malignancies. . . . [T]he clue for an infectious agent in human malignancy might well be found in unusual clusters or unique clinical or epidemiologic occurrences in unlikely parts of the world. Surveillance systems and field laboratory capabilities must be ready to exploit these unusual situations.^32(p921)

Indeed, these were the strategies used to investigate human AIDS when it appeared several years later.

About two years after the identification of human AIDS in 1981, a simian version of the syndrome was announced. The first article referring to “acquired immunodeficiency” in monkeys came from the UC Davis center in 1983¹⁰ and identified “four separate outbreaks of apparent immunodeficiency” there between 1969 and 1981. The most recent, in 1980 and 1981, found cases of generalized lymphadenopathy, CMV infection, shigellosis, Y pseudotuberculosis, and cryptosporidiosis, some of which had appeared in earlier outbreaks. Roderick Bronson, who moved from UC Davis to Harvard and subsequently worked at the New England center, found CMV in a necropsy there in 1978. In 1980, noma, a rapidly progressive infection of the mouth or genitals associated with immune deficiency, had also been diagnosed at the New England center.³³

Later in 1983, two articles from the New England center reported simian AIDS in a colony of macaques. Letvin et al. commented that “the parallels of this disease to human AIDS are striking,”^34(p2721) and King et al. concluded that “the natural occurrence of AIDS in monkeys provides an extraordinary opportunity to . . . learn how a breakdown in immunosurveillance can lead to the development of opportunistic infections and tumors.”^35(p388)

Neither of the 1983 articles from the New England center referred to the earlier reports from UC Davis or other laboratories, but they included some diagnoses that had been reported elsewhere, including lymphoma, CMV infection, noma, and retroperitoneal fibrosis,³⁶ and some that had not been previously reported, including lymphoproliferative disorders and Pneumocystis carinii pneumonia, the first diagnosed and most common opportunistic infection of human AIDS.^34,35 A 1995 article by Mansfield et al. claimed that the first simian AIDS cases at the New England center occurred among a group of macaques that had been obtained from UC Davis in 1970.³⁷

None of the articles on simian AIDS published between 1983 and 1985 suggested that the evidence from animals going back to the late 1960s and early 1970s could or should have led to earlier recognition of simian or human AIDS. In a 1986 review of simian AIDS, Norval King of the New England center stated,

Since the early 1970s, veterinary pathologists at several Regional Primate Research Centers … noted sporadic occurrences or outbreaks of unusual opportunistic infections or neoplasms in their colonies of macaque monkeys. In most instances, these diseases were of the type commonly associated with profound immunosuppression.^38(p346)

In 2003, Gardner wrote that “in retrospect, we now know that SIV and [simian] AIDS were present, although unrecognized, at [UC Davis] in the late 1960s until the mid-1970s.”^39(p182) In his 1990 history of AIDS, Grmek referred to the early outbreaks among monkeys at UC Davis but did not explore the reasons that this pre-1981 history failed to prompt earlier vigilance for a human analog.⁴⁰

EXPLANATIONS FOR FAILURE TO RECOGNIZE THE PATTERN

Why was the pattern of cases in monkeys and its equivalent in humans not recognized at the time? Perhaps there were not enough cases to connect—of diseases among monkeys, diseases among humans, or transmission from animals to humans. We consider these and other possible explanations.

Cases Among Animals

The early reports were mostly of clusters or isolated cases among monkeys, some of which were attributed to accidental or deliberate transmission of cancer tissue during studies of other diseases. Some were also linked to generalized contact among animals in confined settings (M. Essex, oral communication, November 18, 2014).

The larger clusters at the UC Davis and New England centers should probably have prompted more etiologic investigation, but several factors may have intervened. At the time, many viruses, including simian virus 40, lymphoma (herpes) viruses, simian sarcoma virus, gibbon ape leukemia virus, Mason–Pfizer monkey virus, and Epstein–Barr virus, were hypothesized to cause diseases in monkeys. In the early 1970s, Mason–Pfizer monkey virus was found in macaques with severe immunosuppression.⁴¹ However, this retrovirus was unrelated to SIV or HIV, which could have delayed recognition of simian AIDS and its human version.

The many viruses under investigation made it difficult to understand the causal relationships among them and particularly between retroviruses and other viruses. In animals found to have both a retrovirus such as SIV and diseases linked to other viruses, was one the cause and the other the effect or did they simply occur coincidentally? In monkeys with herpesvirus-related lymphoma, was the herpesvirus the cause of the disease or the result of an underlying immunosuppression caused by a retrovirus such as SIV? Moreover, most of the cases reported in monkeys were of single rather than multiple diseases, which might have been more indicative of immunosuppression.

In addition, other causal theories were offered, such as environmental toxins, which may have been correct in some cases.^39,42 This was later the case when some thought that exposure to nitrite inhalants (“poppers”) was a cause of AIDS in gay men.

All the pieces of the puzzle were present, but no one grasped the logic that monkeys were dying from a viral infection that caused immunodeficiency, which was then manifested in opportunistic infections. Stowell et al. inoculated monkeys with lymphoma cells to demonstrate the infectious etiology of that disease.⁶ Further transmission of tissues and fluids from sick, immunosuppressed monkeys to healthy monkeys could have established the existence of a transmissible, immunosuppressive condition, but these studies were not done.

Cases Among Humans

In 1978, Francis and Essex recommended vigilance for clusters of unusual diseases in humans that might reflect disease mechanisms similar to those seen in nonhuman primates.³² But what if there were no human case clusters of sufficient size to attract attention until 1981?

Unrecognized human AIDS cases occurred in multiple locations prior to 1981. Grmek asserted that “an HIV-type virus seems to have been present in human populations since at least the middle of the twentieth century.”^40(p120) He assembled 16 reports (12 from the continental United States) of cases meeting the clinical definition of AIDS between 1940 and 1981, but the risk factors were unknown. Grmek concluded that the existence of these cases “challenges the neat scenario whereby AIDS [was] imported into the U.S. at the end of the 1970s, where it was rapidly recognized.”^40(p124) Indeed, an analysis of archival blood samples from the earliest AIDS cases in Haiti suggests that HIV-1 originated in Africa and moved to Haiti in the mid-1960s and thence to the United States in about 1969. The virus circulated “cryptically” in the United States for 12 years, infecting “several thousand” individuals there by 1978.^43(p18568) Recent research by Worobey et al. on the genomes of the earliest cases in the United States strongly indicates that New York City was the original site of HIV subtype B in 1970.⁴⁴

Retrospective testing revealed that by 1978, prevalence of human T-cell lymphotropic virus 3 (HTLV-III, later HIV) among men who have sex with men was 4.5% in San Francisco, California,⁴⁵ and 6.6% in New York City.⁴⁶ In 1989, Des Jarlais et al. concluded that HIV first entered the population of people who inject drugs (PWID) in New York City in the mid-1970s and spread rapidly starting in 1979.⁴⁷ In retrospect, Des Jarlais believes that this rapid spread began in the mid-1970s, a finding also suggested by Biggar et al.⁴⁸ and Moore et al.⁴⁹ Des Jarlais recalled that numerous cases of P carinii pneumonia were seen in New York City hospitals among PWID and infants born to addicted mothers beginning in 1976. This was reflected in increased prescriptions for pentamidine, the orphan drug used to treat Pneumocystis pneumonia (D. Des Jarlais, oral communication, March 13 and 19, 2015).

Cases retrospectively identifiable as AIDS were seen as early as 1979 by Donald Craven, an attending physician at Boston City Hospital. These were in a non-drug-using Haitian taxi driver and several PWID (D. Craven, oral communication, February 5, 2015).

Sharp and Hahn concluded that HIV first emerged in colonial West Africa and then spread for 50 to 70 years before being recognized. The “cradle” of AIDS is believed to have been Leopoldville, Belgian Congo, with rapid spread as urban populations grew, transportation expanded, and sex work increased.^50,51 Retrospective testing of plasma samples drawn in Belgian Congo in 1959 found at least one individual who had been exposed to a virus similar to HTLV-III.⁵² Worobey et al. demonstrated that several strains of HIV-1 were present in Leopoldville (now Kinshasa) by 1960.⁵³ Peter Piot recalled reviewing case files from several hospitals and clinics in the Kinshasa area in 1983 and finding about one AIDS case per year in each facility between 1975 and 1979. Starting in 1980, each of these hospitals had more than 30 cases per year, indicating a rapid growth in infections roughly coterminous with that in the United States. However, according to Piot, until 1980 these Kinshasa cases reflected “a very slow rate of transmission, like a barely flickering candle”^54(p165) and thus did not produce clear clusters that might have demanded more attention. Poor general health and health care certainly played a role in the nonrecognition of the meaning of these early cases.

From Animals to Humans

Disease transmission from animals to humans comprises actual zoonotic transmission and equivalents in humans of conditions seen in animals. In 1975, Essex hypothesized the transmission of viruses from monkeys to humans, but he believed that such transmission would not cause large outbreaks in humans (M. Essex, oral communication, November 18, 2014).

Primate investigators were generally aware of the potential links between animal and human diseases, but the limited state of knowledge may have prevented their clear establishment. Francis and Essex raised the possibility in 1978, as did several of the early articles on lymphoma and other diseases in monkeys. However, lymphoma, the most common disease seen in these early monkey cases, was not a major opportunistic infection in human AIDS. Further complicating the picture was evidence that some viruses were pathogenic in humans but not in animals or vice versa.⁵⁰

Absence of Evidence of Human Retroviruses

Jarrett and Essex found retroviruses causing immune system dysfunction in cats, but there was no evidence of human retroviruses until Robert Gallo’s work published in 1980⁵⁵ and no evidence of human retroviruses causing immune deficiency until articles published in May 1983 by Gallo’s group and that of Montagnier in France.^56,57 These were, respectively, HTLV-III and lymphadenopathy-associated virus (LAV), soon determined to be the same and renamed HIV.

In 1970, Baltimore and Temin showed how retroviruses replicate, and Gallo began studying animal retroviruses in the early 1970s because they had been found to cause leukemia.⁵⁸ In 1972, Kawakami et al. reported a retrovirus they called gibbon ape leukemia¹⁴ and Gallo’s group found several cases suggestive of retroviruses in humans but were unable to isolate a virus.⁵⁸

Gallo’s report on the isolation of HTLV-1 appeared in December 1980⁵⁵ and the scientific landscape dramatically changed. Anders Vahlne “sincerely doubt[ed] that anyone would have been looking for a retrovirus as the etiologic agent for AIDS had HTLV-1 not previously been isolated.”^59(p41) Still, during the period of the early outbreaks in monkeys, there were no known transmissible, immunosuppressive viruses in humans.

Climate of Opinion in Science and Public Health

During this period, many scientists and public health officials believed that the major infectious diseases had been conquered. The eradication of smallpox was a recent dramatic example. US Surgeon General William H. Stewart has been widely quoted as declaring in 1967 that “It’s time to close the books on infectious diseases, declare the war against pestilence won, and shift national resources to such chronic problems as cancer and heart disease.”^60(p2) Although Stewart never spoke these exact words,⁶¹ the fact remains that this belief was widely held at the time.⁶² A colleague of prominent historian of medicine Erwin Ackerknecht told him in the early 1970s that drugs and antibiotics would soon eradicate all diseases so that the only professors needed in medical schools of the future would be historians!⁶³

In about 1978, author Bronson told Nobel laureate Thomas Weller, the discoverer of CMV, about necropsies of monkeys in which he had discovered CMV and other diseases associated with immunosuppression. Weller replied that he was not interested because no one was interested in infectious diseases anymore.

During this period, JaRue Manning of the UC Davis center received a National Cancer Institute (NCI) grant to isolate a virus that may have been causing lymphoma in monkeys. Manning found viral particles in some lymphoid cells and evidence of immunodeficiency. However, the NCI found nothing significant in these findings and canceled the grant because they “were not interested in immune deficiencies” (B. Osburn, written communication, September 12, 2014). In fact, in the early 1970s the scientific consensus was that human cancer-causing retroviruses did not exist. Reports of them were derided as “human rumor viruses.” In 1980, Gallo’s first article on HTLV-1, which presented the “first clear-cut [evidence of] viral origin of a human cancer,” was rejected by the journal Virology on the ground that there were no such viruses.⁵⁹

Gardner concluded that the failure to recognize simian AIDS earlier “dramatically illustrate[s] the relative [lack of] interest in infectious disease research in the 1970s.”^39(p182) It was a time of greater interest in cancer than in infectious disease or immune deficiency,³⁹ let alone a link among all three.

The “Epistemological Obstacle”

French philosopher Gaston Bachelard developed the concept of the “epistemological obstacle” in the 1930s to advance understanding of progress and challenges in the history of science. Bachelard theorized that progress was nonlinear and noncontinuous because of the need to overcome powerful misconceptions within accepted and dominant knowledge.⁶⁴ In this case, such misconceptions blocked the disentangling of the viral and immunosuppressive etiology of the diseases seen in monkeys and the understanding of the dependent relationship among these elements in causing the observed diseases. Figure 1 depicts alternative disease pathways in simian and human AIDS. A 1977 report by Holmberg et al.²¹ of SV-40 virus in monkeys with PML focused on the independent viral etiology of the disease, referred to in Figure 1 as “pathway presented in early articles.” Holmberg et al. suggested that “immunologic compromise is also necessary for this disease [PML] to become manifest in monkeys,” ^21(p596) but this referred to a separate role for immunosuppression (also shown in Figure 1) rather than the current understanding that immunosuppression caused by a retrovirus led to pathogenicity of a latent viral infection (SV-40) and that the combination produced an opportunistic disease (PML).⁶⁵

FIGURE 1—

Pathways of Disease in Simian and Human AIDS

POSSIBLE IMPLICATIONS OF EARLIER RECOGNITION

We have offered reasons why simian AIDS and its human analog were not recognized and perhaps could not have been recognized earlier. Nevertheless, the counterfactual remains interesting. Had the early cases of AIDS in monkeys been understood sooner, might there have been enhanced understanding of the relationship of animal disease to human disease and increased vigilance leading to earlier recognition of cases and clusters of cases in humans and thus earlier initiation of effective prevention and treatment responses to human AIDS?

Myron Essex, one of the world’s leading experts on retroviruses among animals and HIV/AIDS among humans, said that earlier recognition of AIDS in monkeys might have advanced understanding of human AIDS by “a few months to a year or two” (M. Essex, oral communication, November 18, 2014). AIDS was first recognized in humans in 1981. According to Brookmeyer’s estimated HIV infection rates for the United States from 1978 to 1990 (Figure 2), there were 50 000 cumulative infections by 1981, 250 000 by 1983, and 715 000 by 1986.⁶⁶ Given the extremely rapid spread of the virus in the late 1970s and early 1980s, even a slight acceleration in the progress of discovery and consequent deployment of prevention and treatment might have made a difference. Consider the implications of mapping the dates of key events in the response to AIDS on this epidemiological trajectory, moving the start of that sequence of events back one year to 1980. Assuming that the political, social, and other obstacles and delays that beset the early response to AIDS would have been overcome one year earlier, then all of the subsequent advances in prevention and treatment would also have occurred one year earlier.

FIGURE 2—

Reconstruction of HIV Infection Rates in the United States, 1978–1990

This story may offer some useful lessons for the future of public health practice. More rapid and efficient sharing and organization of information makes it less likely that a smoldering epidemic will go unrecognized for very long. We have vastly more effective and efficient tools in surveillance, communication, literature search, and knowledge management. Moreover, the climate of scientific and popular opinion has changed. We know that infectious disease is fully alive and likely to produce new and dangerous outbreaks, as evidenced by the recent outbreaks of Ebola and Zika viruses. However, to make proper use of the available tools and data, continued vigilance, creative thinking, and interdisciplinary collaboration remain essential. Cooperation and information exchange between veterinary and human disease researchers in a “One Health” approach are especially important.

We cannot precisely predict the future or definitively answer the questions posed by history. This story may be useful for looking both backward and forward. It may also offer a valuable case study of how scientific knowledge develops and is understood—or does not develop and is not understood. Nevertheless, we know with certainty that once the geometric spread of HIV began, it claimed the lives of thousands of people per month and tens of thousands per year. Even a slightly earlier recognition of the disease could have saved many lives. Let us hope that the attitudes, tools, and conditions now exist to avoid such public health tragedies in the future.