THE DEVELOPMENT OF A VACCINE AGAINST AFRICAN HEMORRHAGIC FEVER
GERALD A. EDDY, FRANCIS E. COLE, JR.
United States Army Medical Research Institute of Infectious Diseases Fort Detrick, Frederick, Maryland 21701, U.S.A.
In addressing the subject of the preparation of a vaccine against Ebola or Marburg viruses, it might be instructive to consider the experience with vaccines against other hemorrhagic fevers. In terms of human immunizations, however, such data are very limited. Indeed, the experience of the Argentine workers with an attenuated Junin virus used against Argentine hemorrhagic fever represents the only significant human immunization trial against a severe hemorrhagic virus disease (1).
While the vaccine used in Argentina was attenuated, there are some experimental animal data which suggest that an inactivated vaccine may be satisfactory The advantages of an inactivated vaccine, from the developmental standpoint, are so numerous, that we must first consider that type of immunogen in any vaccine program. The purpose of this paper is to describe briefly the experience in our laboratory with passive and active immunization of rhesus monkeys against Bolivian hemorrhagic fever (BHF) using the etiologic virus, Machupo virus, in experimental challenge. Further, we will discuss the feasibility of developing a vaccine against Ebola or Marbrug viruses.
IMMUNIZATION AGAINST HEMORRHAGIC FEVERS
Part of our concern about inactivated vaccines arose from data, some of which were presented in an earlier publication (2), suggesting that Machupo virusinfected monkeys passively protected against hemorrhagic disease with specific immune globulin were more likely to suffer late neurologic complications than monkeys treated with lower doses of globulin. We have recently expanded these data and the results are shown in Table 1.
In conducting the research described in this report, the investigators adhered to the "Guide for the Care and Use of Laboratory Animals," as promulgated by the Committee on the Revision of the Guide for Laboratory Animal Facilities and Care of the Institute of Laboratory Animal Resources, National Research Council. The facilities are fully accredited by the American Association for Accreditation of Laboratory Animal Care.
The views of the authors do not purport to reflect the positions of the Department of the Army or the Department of Defense.
OCCURRENCE OF LATE NEUROLOGIC DISEASE IN PASSIVELY PROTECTED MACHUPO VIRUS-INOCULATED MONKEYS
Immune globulin dose(x) ml /kg
Hemorrhagic disease/ Total (Deaths)
Neurologic disease/ Total (Deaths)
(x) Monkeys inoculated with 1000 PFU Machupo virus, 4 hr later received indicated dose of immune globulin (human origin).
(xx) Mean day of death.
The data represent a composite of two passive immunization experiments as compared to our broader experience in untreated control monkeys. In both instances the monkeys were inoculated subcutaneously with Machupo virus followed 4 hr later by the indicated dose of human immune globulin administered intramuscularly. Those which developed acute hemorrhagic disease became ill within a week after virus inoculation and died approximately 3 weeks after virus challenge. The monkeys protected with immune globulin exhibited a different response. Although there were some hemorrhagic signs and deaths in the lowest immune globulin dosage group, most of the monkeys, including all in the two highest dosage groups, exhibited no hemorrhagic disease.
Nevertheless, several of these monkeys developed late neurologic disease without hemorrhagic signs and without detectable viral antigen in their tissues when examined by immunofluorescence. There was a clear relationship between immunoglobulin dose and the development of late neurologic disease.
It is important to emphasize that the neurologic phenomenon was distinct from the hemorrhagic disease. The neurologic signs of illness appeared abruptly and the monkeys died within 4-5 days, or recovered. There were none of the typical hemorrhagic signs in the controls as described previously for Bolivian hemorrhagic fever in monkeys (3,4), and the pattern of acute neurological deaths were clinically dissimilar from the chronic neurological sequellae of the disease in monkeys (3,4).
Despite this complicating aspect of an immunologic imbalance resulting from passive antibody protection, we found that an experimental, inactivated, cellculture grown, BHF vaccine was safe and efficacious for monkeys. Data in Fig. 1 show the antibody responses of monkeys given inactivated BHF vaccine and challenged 3 weeks later with Machupo virus. The monkeys to be vaccinated were divided into two groups: one received 0.3 ml of vaccine and the second received 3 ml. Following the single dose of vaccine the monkeys were challenged 3 weeks later. Both groups of monkeys mounted relatively brisk antibody responses following challenge, despite the low levels or absence of detectable antibody prior to challenge. These responses should be compared to antibody patterns of control monkeys which may occasionally survive long enough to develop neutralizing antibodies (4). Antibodies in unvaccinated monkeys do not usually appear until between days 21 and 28 after virus challenge. Only one of the monkeys in this experiment became ill. It was viremic through day 17 of the study and did not develop antibody until day 21. The monkeys died 38 days after challenge.
These data suggest that immunity against hemorrhagic disease can develop in the absence of detectable neutralizing antibody and that an inactivated vaccine can induce an effective immune response. Although one of the monkeys suffered a relatively late death with neurologic manifestations, his overall disease course was not particularly different from that of approximately 20% of Machupo virus-infected control monkeys as reported previously (4), and shown in Table 1.
DEVELOPMENT OF AN EBOLA VIRUS VACCINE
If it is possible to immunize against arenavirus hemorrhagic fevers with inactivated vaccines, it would seem feasible to immunize also against Ebola or Marburg viruses with specific inactivated immunogens. The major problem would be to find the appropriate combination of virus strain and acceptable substrate to yield a sufficient concentration of viral antigen. Our data with inactivated Venezuelan encephalitis vaccine 5 indicate that a concentration of approximately 9.5 log 10 of alphavirus virions per milliliter may be required if an inactivated vaccine is to induce an adequate antibody response to an inactivated viral product. Nevertheless, lower concentrations may be effective through the use of repeated inoculations. Electronmicrographs of Vero cells infected with Ebola virus (6) suggest a large amount of viral antigen production. It would therefore seem feasible to obtain adequate antigen concentrations approaching immunogenic levels in substrates acceptable for vaccine use.
The scheme shown in Figure 2 describes the general procedures that might be used to prepare an inactivated Ebola virus vaccine. The substrate for this would be selected from among the limited number available, i.e., diploid fetal human lung, diploid fetal rhesus lung, diploid fetal cercopithecus lung or perhaps primary chick embryo cells. Most of the procedures shown are standard 5 with respect to those outlined in referenced vaccine production methods Although Figure 2 indicates the use of formalin as an inactivating agent, this may not be a realistic choice. Betapropiolactone might be used, but the recent development of the psoralen compounds for vaccine inactivation holds considerable promise (7). In summary, we believe that the production of limited quantities of an inactivated vaccine against either Marburg or Ebola viruses is a distinct, realistic possibility within the next 2 or 3 years. While the difficulty of preparing a sufficiently potent product in an acceptable cell culture system may prove to be a formidable obstacle, the preliminary indications suggest that it will not be impossible.
fig. 1 : Monkeys inoculated with one dose of inactivated BHF vaccine were challenged with 1000 PFU of Machupo virus 21 days later (day 0). The points indicate the geometric mean antibody response for groups receicing either 3 ml or 0.3 ml. Numbers in parentheses indicate the number viremic (n=4 per group) on the day tested.
fig. 2 : A projected plan for the production of an experimental lot of Ebola vaccine. This plan assumes that the virus seed stock has been characterized and certified and that an acceptable cell culture substrate has been selected.
1. de Guerrero, L.B., Weissenbacher, M.C., Parodi, A.S. (1969) Immunizacion contra la fiebre hemorrhagica con una cepa atenuada de virus Junin, Medicina (B. Aires) 29, 1-5.
2. Eddy, G.A., Wagner, F.S., Scott, S.K., Mahlandt, B.J. (1975) Protection of monkeys against Machupo virus by the passive administration of Bolivian haemorrhagic fever immunoglobulin (human origin), Bull. WHO, 52, 723-727.
3. Kastello, M.D., Eddy, G.A., Kuehne, R.W. (1976) A rhesus monkey model for the study of Bolivian hemorrhagic fever, J. Infect. Dis., 133, 57-62.
4. Eddy, G.A., Scott, S.K., Wagner, F.S., Brand, O.M. (1975) Pathogenesis of Machupo virus infection in primates, Bull. WHO, 52, 517-521.
5. Cole, F.E., Jr., May, S.W., Eddy, G.A. (1974) Inactivated Venezuelan equine encephalomyelitis vaccine prepared from attenuated (TC-83 strain) virus, Appl. Microbiol. 27, 150-153.
6. Murphy, F.A. (1978) Ebola and Marburg virus morphology and taxonomy, (This Colloquium).
7. Hearst, J.E., Thiry, L. (1977) The photoinactivation of an RNA virus, vesicular stomatitis virus, with the aid of newly synthesized psoralen derivatives, Nucleic Acids Res., 4, 1339-1347.
A.W. Woodruff : I think that it is important that this meeting expresses concern at the lack of more effort along the lines for producing a vaccine against these infections. It is now 9 years nearly since Lassa virus was isolated. It 's been constantly causing trouble to public health authorities, physicians and health authorities generally and we still have no possible sight of a vaccine.
R.A. Coutinho : who should be vaccinated and why ?
G.A. Eddy : in the first place, laboratory personnel working with these viruses. With respect to Lassa, for instance, there is a real need for imunizing medical personnel in hospitals in the field where there is a great risk of acquiring this infection. And finally, for eventual national emergencies.
J. Casals : I want to make some comments on the use of immune plasma in the treatment of these hemorrhagic fevers. As Dr. Russeb mentioned, first of all we have to know whether it helps or not. Then whether it does any harm. In the case of Lassa fever, up to a certain point we knew how many people had been given the plasma and out of 5 or 6 one followed a course which was harmful. The plasma was given very late in the course of that disease but at the same time came from an individual who, for all we know, may have had virus in his blood. Another question that has been raised too is when to give this plasma . It seems to me there is going to be a lot of wastage. In the case of Lassa, it has been given almost to as many people who needed it as to people who just had a sore throat and simply because they had been in Sierra Leone or Nigeria, created a panic. Faced with a very acutely ill patient, is plasma to be given or not ? We hear a lot of immune complexing and the damage that it can produce. Is there any evidence that following the administration of plasma either in Argentinian haemorrhagic fever, where practically 300 individuals were treated with it, or in Lassa or in Ebola fever ?
J.I. Maiztegui : In Argentina in a period of about twenty years, the endemoepidemic area has extended so much that today it covers over 100.000 km2 with a population at risk of approximately 1.200.000 people. To our present knowledge, the ecology of the reservoir is such that the only possibility for control is the development of a vaccine. In Argentina the first research priority is the development of a vaccine. It was mentioned that clone 3 had been successfully used in 636 volunteers, however I think that the follow-up of these people was not well conducted. It was also said that laboratory workers who received this experimental vaccine were protected. Let me remind you that in the laboratory where this was done, people were handling mainly the attenuated strain and not pathogenic strains, of Junin virus so that this is not conclusive evidence. As to the convalescent plasma, we do not have any evidence of immune complexes in AHF. More than 10 autopsy cases were specifically investigated for this with negative results. Of the 18.000 reported cases of AHF, 60-70 % were confirmed thus there are 10.000 to 12.000 proven cases of AHF. Probably 6.000 or 7.000 of these received convalescent plasma in the acute stage of the disease. In my personal experience, I have never had an acute accident. We did a controlled trial that took four years and will be published shortly, which shows that plasma we gave 500 ml, is indeed useful. We are also concerned about the conservation of convalescent plasma and the search for better preparations. However, I would not dismiss so quickly the possibility that whole plasma is useful. Dr. Johnson raised the question of its action mechanism. It may seem that the active principle of the plasma are the antibodies. However I want to remember that in lymphocytic choriomeningitis the neutralizing antibodies in the acute stage are complement dependent. Some of the physicians in Junin are convinced that whole plasma is superior to gaumaglobuines. I do not know if the analogy with Ebola is valid but convalescent plasma in AHF is useful. A last comment, I would strongly recommend to give convalescent plasma to anybody who has to work in the laboratory with this virus or has to handle patients with this disease.