Lapas attēli
PDF
ePub
[merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][ocr errors][ocr errors][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small]

a X, Y and Z represent the first, second and third principal components, respectively. The Z scale has been elevated 0.20 un above the X-Y plane to simplify the presentation of the "virus" points whose Z-coordinate is less than 0. The data are taken fro HI tests performed at the National Communicable Disease Center, Atlanta, Ga. Chicken antiserum was used.

[blocks in formation]

NUMERICAL TAXONOMY APPLIED TO THE FORMULATION OF INFLUENZA VIRUS VACCINE

[merged small][merged small][ocr errors][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][ocr errors][ocr errors][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]

267

1

land/59(V), which was a vaccine component for the influenza seasons from autumn 1965 to spring 1967, does not appear to be closely related to recently isolated 1966 strains. These data support the selection of B/Massachusetts/66(V12) as a replacement for B/Maryland/59 in the current vaccine because the former is antigenically more related to strains presently in the community.

Unlike the situation with the type-A influenza viruses, where there is evidence that immunization induces an effective immunity, there are no available data to attribute a protective effect to the influenza

* Civilian influenza vaccine 1965-67 formulation: A/PR/8/34, AI/AA/1/57, A2/Japan/170/62, A2/Taiwan/1/64, /Maryland/1/59.

[blocks in formation]
[blocks in formation]

159

(V3, V6, V, V, V V20, V21 V19) and (VS, V12, V10 V11 V13 V16, V18, V14, V17); at a correlation level of 0.35 there is 1 family. If 0.50 is an appropriate correlation level, 1 representative strain from cach family should be included in the vaccine; if the correlation level of 0.35 is significant, then 1 strain would suffice to represent the two "clusters". At present we do not know which correlation level would be more appropriate; clinical trials should reveal it. Aside from the numerical considerations, within any group of closely related virus strains, the selection of 1 for vaccine manufacture will depend on its suitability as a vaccine “production" strain and its immunogenicity in man.

Influenza A

Concerning type-A influenza virus strains, the dendrogram (Fig. 3) shows a high correlation (0.75 or greater) among all of the A2 strains. This high degree of correlation may be misleading because comparison with antigenically dissimilar A and Al strains tends to reduce the differences among the A2 strains. When only the A2 strains are analysed (Fig. 4 and 5) these differences become more apparent. A2/Japan/170/62 shows high correlations with 1966 strains A2/California/1/66 (0.90) and A2/UC/1/66 (0.73) and moderate correlations (0.53 or greater) with most of the other A2 strains analysed.

77-615 - 72-42

[ocr errors][merged small][merged small][merged small][subsumed][merged small][subsumed][subsumed][subsumed][subsumed][ocr errors][subsumed][subsumed][subsumed][merged small][graphic][ocr errors][subsumed][subsumed][subsumed][ocr errors][subsumed][subsumed][subsumed][subsumed][subsumed][subsumed][subsumed][subsumed][ocr errors][subsumed][ocr errors][ocr errors][subsumed][subsumed][subsumed][subsumed][merged small]

@X, Y and Z represent the first, second and third principal components, respectively. The Z scale has been elevated 0.20 unit Jove the X-Y plane to simplify the presentation of " virus "points whose Z-coordinate is less than 0.

Ahough this supports A2/Japan/170/62 as a civilian accine component, the data suggest that another rain closer to the centre of the cluster shown in he lower left of Fig. 5 might more adequately represent the current field strains.

A2/Taiwan/1/64 constitutes half of the A2 comwent in the civilian vaccines and all of the A2 ..omponent in the military vaccine. In Fig. 4 this ain appears closely correlated only to A2/Itsusachi/1/65 (0.84) and distantly related by an average relation coefficient of 0.19 to all other A2 strains alysed. Viewing in another way (Fig. 5), A2/Tai

Military influenza vaccine 1967-68 formulation: APR/S, 34. A Swine 33. A1/AA/1/57, A2/Taiwan/1/64, Lee 40 and B/Massachusetts/3/66.

wan/1/64 is distant from the main cluster of recently isolated A2 strains. If the numerical taxonomic data represent true antigenic relationships, our results suggest the advisability of either replacing the A2/Taiwan/1/64 vaccine component, especially in the military formulation which has only 1 A2 strain, with a strain more representative of the current field isolates, or adding this current strain to the A2 vaccine components as has been done with the civilian formulated vaccine. Investigations are planned to determine what is an appropriate correlation coefficient and, hopefully, once determined, numerical taxonomic methods might become more useful as a guide in the formulation of influenza virus vaccines.

[blocks in formation]

Les relations entre les nouveaux isolats de virus de la grippe et les souches identifiées antérieurement n'apparaissent pas toujours clairement à l'examen des donnees fournies par les épreuves d'inhibition croisée de Themagglutination (IH). Des méthodes taxonomiques chilliées ont été utilisées pour préciser ces relations de façon que les caractères antigéniques puissent être déterminés plus facilement.

Les données fournies par deux séries importantes d'épreuves III exécutées par le Centre international de la grippe pour les Amériques au Centre national des Maladies transmissibles d'Atlanta, Ga. (Etats-Unis d'Amé rique) ont servi de base à l'analyse dont il est rendu compte dans cet article. Pour ces épreuves, qui portaient sur les souches A et B de virus de la grippe respectivement, on a employé des immunsérums de poulet et des antigènes de virus grippal propagé dans la cavité allantoïdienne d'embryons de poulet.

En utilisant la méthode d'analyse chiffrée de Sokal & Michener, fondée sur le rapprochement par grappes de moyennes, on a obtenu un dendrogramme ou arbre généalogique », qui indique l'existence de deux sousgroupes importants parmi les isolats (1964-1966) de virus B étudiés. Une seconde méthode, l'analyse des principaux constituants décrite par Anderson, qui représente les souches de virus par des points sur un diagramme à trois dimensions, a mis en évidence des groupements similaires à ceux du dendrogramme.

[ocr errors]

On a appliqué l'épreuve d'inhibition de l'hémaggluti nation à 20 souches grippales A, soit 18 isolats A2 ct! deux souches A/PR/8/34 ct AI/IM/1/47. D'aprè, ki dendrogramme, les souches A2 paraissent étroitement apparentées les unes aux autres, mais très distinctes,. en revanche, des souches PR/8 ct FM/I. Une analyse séparée des scules souches A2 semble indiquer l'existence de plusieurs sous-groupes parmi les isolats A2 examinés, į Le succès de l'immunisation contre la grippe dépend en grande partie de la présence dans le vaccin des souches virales appropriées. La structure antigénique des virus grippaux se modific constamment et il est nécessaire de contrôler sans cesse les propriétés des souches de virus! en circulation pour pouvoir, lors de l'apparition d'une variation antigénique notable, incorporer dans le vaccin les antigènes nécessaires. Les diagrammes établis partir des analyses taxonomiques chiffrées indiquent que ces méthodes pourront faciliter à l'avenir la sélection: des souches de virus grippal les plus représentatives.

Les résultats de la présente étude semblent indiquer qu'ils serait opportun de remplacer la souche B/Mary land/59 par la souche B/Massachusetts/66 (V12) dans k vaccin fabriqué aux Etats-Unis d'Amérique, cette der nière souche étant antigéniquement plus proche des souches de virus actuellement en circulation. On suggère en outre qu'un virus grippal A2 isolé couramment soit incorporé au vaccin administré aux forces armées au Etats-Unis ou remplace le constituant A2/Taiwan/1/6; actuellement utilisé.

REFERENCES

Anderson, T. W. (1958) Introduction to multivariate statistical analysis, New York, Wiley, pp. 272286

Sokal, R. R. & Michener, C. D. (1958) Kans. Univ. Sci. Bull., 38, 1409

United States Department of Health, Education, and Welfare, Public Health Service, National Communi cable Disease Center (1966) Influenza-respirator disease surveillance, Report No. 82, Atlanta, Ga. Lee, A. M. (1968) Nature (Lond.), 217, 620

« iepriekšējāTurpināt »