Adjuvant influenza vaccines

SESSION V FUTURE INFLUENZA VIRUS VACCINES Bull. Org. mond. Sant 1969, 41, 617-621 Bull. Wld Hlth Org. Adjuvant Influenza Vaccines C. H. STUART-HARRIS 1 Emulsified inactivated influenza vaccines have been in use for some 18 years and the goal ofenhanced serological response lasting 2 years or more has been attained. The safety of the method in relation to immediate pyrogenic reactions has been demonstrated and no carcinogenic effects are known to have occurred in man. However, the problem of delayed local reactions after the injection of mineral-oil vaccines has not been solved. British experience of adverse reactions to commercial adjuvant influenza vaccine is quoted. New methods for obtaining adjuvant action without the risk of local abscess formation are needed both for inactivated whole virus and for split haemagglutinin vaccines. Reversal of water-in-mineral-oil emulsion to oil-in-water emulsion reduces viscosity and permits diffusion ofthe depot injection. A trial in Britain has shown equally good adjuvant properties of the reversed emulsion incorporating influenza virus vaccine so far as serological response is concerned, although it has not yet been conducted on a scale that would allow ofadequate evaluation of the likelihood of delayed local reactions. The fact that adjuvant vaccines are listed for consideration in a session dealing with the future can only mean that available adjuvant materials possess drawbacks. As mineral-oil adjuvant vaccines con- taining Drakeol and Arlacel A are the only adjuvant influenza vaccines which have been used on a very large scale and for a long period of time, it is likely that at least with these materials the principal disadvantages are now known. In fact, 2 of the risks involved in the use of mineral-oil adjuvant vaccines still lack substance in relation to man. I refer to the theoretical possibility of carcinogenicity which was suggested by the experimental induction of tumours in BALB/C mice by Lieberman, Mantel & Humphrey (1961) and Potter & Boyce (1962), and in DBA/2 mice by Rask-Nielsen & Ebbesen (1965) using mineral oil intraperitoneally. The follow-up by Beebe, Simon & Vivona (1964) of some 44000 men in the US Armed Forces, of whom 18 000 received adjuvant influenza vaccine, is therefore of considerable importance. No suggestion of enhanced incidence of tumours was found in the mortality of these men during the 9 years following immunization. Secondly, sensitization to materials contained in 1 Professor of Medicine, University of Sheffield, The Royal Hospital, Sheffield, England. adjuvant influenza vaccine is a theoretical possibility in the case of extraneous substances such as penicillin and also with antigens of the blood group substances. Beebe and others (1964) found some evidence of increased sensitivity to penicillin leading to urticaria, particularly in those receiving vaccine in the early 1950s when it was manufacturing practice in the USA to add penicillin to the eggs from which vaccine was made. Iso-immunization by blood-group A substance has been a theoretical risk with all varieties of influenza vaccine since the demonstration of this antigen in commercial influenza vaccine by Springer & Tritel (1962). Springer & Schuster (1964), indeed, found that extracts of influenza vaccine induced increased iso-agglutinin titres in inoculated persons but others have failed to find such unwanted antibody induction after ordinary saline or commer- cial oil-adjuvant vaccines (Forsyth & Wilson, 1968). There remain the unwanted local reactions variously described as cysts (Bell et al., 1961) or chemical abscesses. Beebe, Simon & Vivona (1964) found evidence of these delayed local nodules in the follow-up of Army personnel but the incidence was too low to affect widespread use of mineral-oil vaccine in the US Armed Forces. Indeed, Davenport (1968) recently described 17 years' experience with adjuvant influenza vaccines and urged their adoption generally in the USA. 2413 617- C. H. STUART-HARRIS BRITISH EXPERIENCE WITH MINERAL-OIL ADJUVANT VACCINES This has been of two sorts. First, the Medical Research Council's Committee on Clinical Trials of Influenza Vaccine (1955) and Committee on Influenza and Other Respiratory Virus Vaccines (1964) have carried out both small- and large-scale trials of mineral-oil vaccines since 1953. Evidence has been obtained both of the high and sustained antibody responses which they produce and also of their protective efficacy. In these trials local and general febrile reactions have always been less in frequency than after the use of ordinary saline vaccine. But delayed nodular reactions occasionally leading to liquefaction and tracking through the tissues have been seen in a low percentage of inoculated persons. The figure in the later trials was 3.3 per 10 000 per- sons but only a proportion of these developed nodules which required surgical aspiration or inci- sion. The second experience in Britain arose from com- mercial sale of the vaccine between 1963 and 1965, when some 1.3 million doses of mineral-oil influenza vaccine containing Drakeol and Arlacel A were released. An estimated 900000 doses were admi- nistered and during 1964 and 1965 some 40 delayed adverse reactions came to the notice of the Safety of Drugs Committee in the United Kingdom and to manufacturers. It is possible that further unreported reactions actually occurred but unlikely that serious reactions were missed. In fact, only 9 instances of reactions requiring local surgical procedures were known. However, this number was sufficient for the vaccine to be voluntarily withdrawn by the manu- facturers except for use in special-risk groups of the population. This experience, coming on the heels of the reactions reported from the use of adjuvant tetanus toxoid in New Guinea (Pittman, 1967) and of mineral-oil adjuvant cholera vaccine in the Philippines (Ogonuki, Hashizume & Abe, 1967), led to a halt in the exploitation of mineral-oil vaccines in Britain. The mechanism of these adverse local reactions has, however, remained obscure. In spite of many attempts to induce delayed local reactions in experi- mental animals, no success has been obtained. Histological examination of 4 human nodules in Britain showed granulomata resembling sarcoid lesions. Others (Ogonuki, Hashizume & Abe, 1967) have regarded the granulomata as being due to reactions of the foreign-body type. REVERSED OR MULTIPLE ADJUVANT VACCINE In 1965, Herbert (1965) described a method of treatment of mineral-oil adjuvant vaccine containing Drakeol and Arlacel A which caused the previous water-in-oil emulsion to pass into an oil-in-water dispersion. Herbert found that Tween 80 and ultra- sonic dispersion caused the emulsion to become less viscous and consequently to disperse in the sub- cutaneous tissues more readily. Experimental immu- nization suggested no impairment of the antigenicity of the vaccine in its new phase and Herbert (1967) reported that the emulsions were very stable. A clinical trial under the auspices of the Medical Research Council's Influenza Vaccine Committee was organized in 1966 and Taylor and others (1969) have just reported on its results. The vaccine con- tained 2 A2 and 2 B influenza viruses. Table 1 shows that the trial was a serological one involving 302 persons and using 2 mineral-oil emulsion vaccines- one water-in-oil (simple) emulsion, one prepared by TABLE I LOCAL AND SYSTEMIC REACTIONS UP TO 72 HOURS AFTER VACCINATION, ACCORDING TO VACCINE GROUPa Number t' showing reactions after Type of Simple Multiple Rhinovirusreaction Aqueous emulsion emulsion vaccine vaccine vaccine vaccine (control) Local 6 (8) 6 (8) 20 (26) 0 Systemic 21 (27) 27 (36) 29 (38) 10 (14) No reaction 51 (65) 41 (55) 27 (36) 64 (86) Total 78 74 76 74 a Reproduced, by permission, from Taylor et al. (1969). b Figures in parentheses indicate percentages. the Herbert process-and for controls a saline influenza vaccine and a tissue culture rhinovirus vaccine. Local and systemic reactions were reported as shown but the systemic reactions were mild and no delayed local reactions were noted. Table 2 shows that the distributions of antibody titres before immunization were similar in the various groups though, as shown in Table 3, persons with a history of previous immunization with oil-adjuvant vaccines had higher titres than those previously unvaccinated. Table 4 shows the geometric mean titres in the 618 ADJUVANT INFLUENZA VACCINES 619 TABLE 2 ANTIBODY TITRES BEFORE VACCINATION ACCORDING TO VACCINE GROUPa Range of antibody titres (haemagglutination-inhibition) Vaccine group No. < 12 12-48 >48 Geometric mean titres A2 B A2 B A2 B A2 B Aqueous 41 20 27 15 11 6 3 16 10 Simple emulsion 39 17 24 12 12 10 3 20 12 Multiple emulsion 42 18 27 19 13 5 2 17 10 Rhinovirus (control) 40 16 24 12 12 12 4 25 13 a Reproduced, by permission, from Taylor et al. (1969). TABLE 3 ANTIBODY TITRES IN FIRST SERUM SAMPLE ACCORDING TO HISTORY OF PREVIOUS INFLUENZA VACCINATION a Hi antibody titres b Total no. Group < 12 12-48 >48 A2 B A2 B A2 B A2 B Previously vaccinated 53 53 5 (9) 18 (34) 22 (42) 24 (45) 26 (49) 11 (21) Not previously vaccinated 109 109 66 (61) 84 (77) 36 (33) 24 (22) 7 (6) 1 (1) a Reproduced, by permission, from Taylor et al. (1969). b Figures in parentheses indicate percentages. TABLE 4 GEOMETRIC MEAN OF ANTIBODY TITRES IN SERUM SAMPLES TAKEN IMMEDIATELY BEFORE VACCINATION (1st), APPROXIMATELY 3 MONTHS AFTER VACCINATION (2nd) AND APPROXIMATELY 12 MONTHS AFTER VACCINATION (3rd) a Geometric mean titre (HI) Vaccine group No. A2 B 1st 2nd 3rd 1 st 2nd | 3rd Aqueous 41 16 171 102 10 34 23 Simple emulsion 39 20 334 166 12 89 52 Multiple emulsion 42 17 461 206 10 132 70 Rhinovirus (control) 40 25 26 25 f 13 13 12 a Reproduced, by permission, from Taylor et al. (1969). 620 C. H. STUART-HARRIS various groups of persons before and 3 and 12 months after immunization. The superiority and comparability of the emulsion vaccines was clearly demonstrated and there was no evidence of reduc- tion of antigenicity after the redispersion of the mineral-oil vaccine. A larger trial is now being plan- ned, partly to obtain evidence concerning reactions. Because both Drakeol and Arlacel A are still present in the vaccine, the Safety of Drugs Com- mittee has refused to sanction other than limited trials. THE FUTURE It is quite obvious that so long as there is any question of using inactivated whole influenza vaccine or a split product subcutaneously, methods of enhancing the antibody response are still much to be desired. Hilleman's experiences with a meta- bolizable vegetable-oil adjuvant vaccine (Woodhour et al., 1964) containing Arlacel have been published and no additional comment can be made. Other adjuvant materials have been used, some success- fully and some unsuccessfully. Fukumi (1967) has used a vegetable oil (sesame) with influenza vaccine though he found it had less immunizing capacity than mineral-oil vaccine, as was the British experience with peanut oil.' Holt (1967) has success- ' Pollock, T. M.-unpublished report, 1967, to the Medi- cal Research Council's Committee on Influenza and Other Respiratory Virus Vaccines. fully used squalane (prepared from shark-oil) and Arlacel with triple diphtheria, tetanus and pertussis vaccine. The emulsions were absorbed on aluminium- phosphate gel. Such emulsions do not seem to have been tested with influenza vaccine. It seems that aluminium phosphate is not an adequate adjuvant for influenza haemagglutinin according to Davenport, Hennessy & Askin (1968), though Davenport & Hennessy (1967) were able to obtain adjuvant effects when haemagglutinin was suspended as a water-in-oil emulsion with mineral oil. Experimentation with aluminium-oxide or aluminium-hydroxide gel as adjuvants for inactive viral vaccines such as poliovaccine (Drescher, Grutzner & Godgliick, 1967) or influenza vaccine (Schmidt, 1967) suggest that there is still much to be learnt. Perhaps it would be as well to end on a note of caution. The history of the use of drugs in man indicates the great need for caution before risks are dismissed as negligible. Until a fuller understanding of the mechanism of adjuvant action has been obtained, it is clear that unusual and unwanted effects may continue to occur whatever the materials which are used. However, adjuvant vaccines seem certain to have a future in programmes of human immunization and further basic research work is necessary. REFERENCES Beebe, G. W., Simon, A. H. & Vivona, S. (1964) Amer. J. med. Sci., 247, 385-405 Bell, J. A., Philip, R. N., Davis, D. J., Beem, M. O., Beigelman, P. M., Engler, J. I., Mellin, G. W., Johnson, J. H. & Lerner, A. M. (1961) Amer. J. Hyg., 73, 148- 163 Davenport, F. M. (1968) Ann. Allergy, 26, 288-292 Davenport, F. M. & Hennessy, A. V. (1967) Comparison of antibody response to influenza viruses and to purified hemagglutinins emulsified in mineral oil. In: Regamey, R. H. et al., ed., International Symposium on Adjuvants ofImmunity, Utrecht 1966, Basel, Karger, pp. 283-293 Davenport, F. M., Hennessy, A. V. & Askin, F. B. (1968) J. Immunol., 100, 1139-1140 Drescher, J., Grutzner, L. & Godgluck, G. (1967) Immu- nogenic activity ofaqueous and aluminum oxide adsorbed poliovirus vaccines in Macaca mulatta. In: Regamey, R. H. et al., ed., International Symposium on Adjuvants of Immunity, Utrecht 1966, Basel, Karger, pp. 157-167 Forsyth, J. R. L. & Wilson, T. E. (1968) J. Hyg. (Lond.), 66, 1-6 Fukumi, H. (1967) Effectiveness and untoward reactions of oil adjuvant influenza vaccines. In: Regamey, R. H. et al., ed., International Symposium on Adjuvants of Immunity, Utrecht 1966, Basel, Karger, pp. 237-240 Herbert, W. J. (1965) Lancet, 2, 771 Herbert, W. J. (1967) Multiple emulsion adjuvants. In: Regamey, R. H. et al., ed., International Symposium on Adjuvants of Immunity, Utrecht 1966, Basel, Karger, pp. 89-90 Holt, L. B. (1967) Oily adjuvants. In: Regamey, R. H. et al., ed., International Symposium on Adjuvants ofImmu- nity, Utrecht 1966, Basel, Karger, pp. 131-135 Lieberman, R., Mantel, N. & Humphrey, W., Jr (1961) Proc. Soc. exp. Biol. (N.Y.), 107, 163-165 Medical Research Council, Comnittee on Clinical Trials of Influenza Vaccine (1955) Brit. med. J., 2, 1229-1232 Medical Research Council, Committee on Influenza and Other Respiratory Virus Vaccines (1964) Brit. med. J., 2, 267-271 ADJUVANT INFLUENZA VACCINES 621 Ogonuki, H., Hashizume, S. & Abe, H. (1967) Histo- pathological tests of tissues in the sites of local reactions caused by the injection of oil-adjuvant cholera vaccine. In: Regamey, R. H. et al., ed., International Symposium on Adjuvants ofImmunity, Utrecht 1966, Basel, Karger, pp. 125-127 Pittman, M. (1967) Some remarks regarding the " reaction provoking properties ": discussion. In: Regamey, R. H. et al., ed., International Symposium on Adjuvants of Immunity, Utrecht 1966, Basel, Karger, pp. 101-102 Potter, M. & Boyce, C. R. (1962) Nature (Lond.), 193, 1086-1087 Rask-Nielsen, R. & Ebbesen, P. (1965) J. nat. Cancer Inst., 35, 83-88 Schmidt, G. (1967) The adjuvant effect of aluminium hydroxide in influenza vaccine. In: Regamey, R. H. et al., ed., International Symposium on Adjuvants of Im- munity, Utrecht 1966, Basel, Karger, pp. 275-282 Springer, G. F. & Schuster, R. (1964) Vox sang., 9, 589- 598 Springer, G. F. & Tritel, H. (1962) Science, 138, 687-688 Taylor, P. J., Miller, C. L., Pollock, T. M., Perkins, F. T. & Westwood, M. A. (1969) J. Hyg. (Lond.), 67, 485- 490 Woodhour, A. F., Metzgar, D. P., Stim, T. B., Tytell, A. A. & Hilleman, M. R. (1964) Proc. Soc. exp. Biol. (N.Y.), 116, 516-523