Problems connected with live influenza virus vaccines

Discussion Papers-Session IV Problems Connected with Live Influenza Virus Vaccines by GORDON MEIKLEJOHN a Twenty-eight years have passed since the first large field trial of inactivated influenza vaccine, and the product is, by general agreement, still far from per- fect. There is therefore ample reason to explore the possibility of developing attenuated but immunogenic strains suitable for administration via the respiratory tract. We are told that the local antibody which is produced, even though low in titre, will be a major factor in preventing illness. There is still much to be learned about the relative role of local and humoral antibody in influenza and other viral respiratory diseases. The paper by Tyrrell & Beare b excellently reviews several approaches. Extensive studies were begun many years ago by Professor Smorodincev and others in the USSR. They used serial egg passage at 32°C-34°C to obtain attenuation. When attenuation was achieved, growth of the virus in the vaccinee and antibody response were less than desired and large amounts of virus and 2 or 3 inoculations were required. The degree of protection observed was not as great as with some inactivated vaccines. Other papers have reported on other methods: (1) passage in newly hatched chicken kidney tissue culture at 25°C; (2) passage in increasing concentrations of horse serum to select inhibitor-resistant variants; (3) passage in bovine kidney cells at 25°C; (4) selec- tion of temperature-sensitive mutants from virus grown in bovine kidney cells in the presence of 5-fluorouracil. Several approaches are promising, but the ideal strain is not yet in hand. The stability of such strains and the development of reliable markers still require much study. Many theoretical and practical questions might be raised, but one will suffice here. We are now in the 13th year of the A2 era, and almost everyone has now had one or more infections. With potent, inactivated vaccines, particularly with adjuvant, the humoral antibody response should be of such a Department of Med University of Colorado Medical Center, Denver, C 80220, USA. b See the paper by D. A. J. Tyrrell & A. S. Beare on page 581 of this issue. magnitude that it would cover the relatively minor antigenic shifts that we see from year to year Attenuated strains, so far, have not produced com- parable rises in humoral antibody, and local anti- body titres are relatively low. While they may protect against homologous challenge, I do not know how well they would perform if the infecting virus were somewhat different. Perhaps the goal should be not an avirulent strain which grows just a little and produces a little antibody, but one which grows luxuriantly and evokes a maximal antibody response. Some participants at this Conference may have been bewildered by reports of vaccine trials in which the protective effect against Hong Kong influenza was reported to vary from nil to 98 %. It is well to point out that determining the precise degree of protection depends on separating cases of influenza from those due to other respiratory diseases, particularly when the base-line of the latter is high. This is best done when a specific serological diagnosis is made on all cases, using both com- plement-fixation and haemagglutination-inhibition tests. When diagnosis is based only on clinical diagnosis or on questionnaires, the effect of vaccina- tion is frequently diluted because other cases of respiratory diseases are classified as influenza. Finally, a comment directed to those who like to prognosticate about influenza. In 1957, we were in the 11th year of the influenza Al period when a variant designated "Denver " or " Nederland " appeared which was very different from earlier Al strains. It was not as far removed from the earlier Al strains as the Hong Kong strains are from the other A2, but it was an oddity. It caused sharp local outbreaks, but no major epidemics. While we were still puzzling about why it did not do more damage, the A2 strains appeared in the summer of 1957 and all Al strains vanished. I recall this simply as a warning that reputations for accurate prognostica- tion may suffer if predictions are made that there will be little influenza in a given season and then influenza A3 appears. 2411A 607