A controlled field trial of the effectiveness of monovalent classical and El Tor cholera vaccines in the Philippines*

Bull. Org. mond. Santi 1973, 49, 13-19 Bull. Wid Hith Org. A controlled field trial of the effectiveness of monovalent classical and El Tor cholera vaccines in the Philippines * PHILIPPINES CHOLERA COMMITTEE' Monovalent Ogawa and Inaba vaccines prepared from classical and El Tor strains of Vibrio cholerae were field-tested in Negros Occidental, Philippines, where cholera El Tor, serotype Ogawa, is endemic. The monovalent classical Ogawa and Inaba vaccines were of the same lots as those tested in East Pakistan in 1968-69. The results indicated that allfour types of vaccine tested offered significant degrees ofprotection varying from 58 % to 71 %. The Ogawa vaccines were slightly, though not significantly, more protective than the Inaba vaccines against disease caused by Ogawa. The biotypes of the vaccine strains were found to be of no consequence. A field trial in Negros Occidental, Philippines, in 1964 (Philippines Cholera Committee, 1965), indi- cated for the first time that protection against Ogawa El Tor cholera can be achieved by immunizing man with either classical or El Tor bivalent Ogawa and Inaba vaccines. The field trial in Dacca, East Pakistan (now Bangladesh), in 1968-69 (Mosley et al., 1970), where monovalent classical Inaba and Ogawa vaccines were compared, revealed that monovalent Inaba vaccine gave satisfactory protection against classical Inaba infection. On the other hand, classical Ogawa mono- valent vaccine conferred very poor protection against heterologous Inaba infection, indicating that the protective efficacy of cholera vaccine is serotype- specific as far as classical Inaba infection is con- cerned. Although in the same trial a purified Inaba cell-wall antigen obtained from an El Tor Inaba strain (Watanabe et al., 1970) gave a degree of protection against classical Inaba infection similar to that of the monovalent classical Inaba whole-cell * This study was performed as part of the joint Philip- pines-Japan-WHO Cholera Research Project in the Philip- pines. The members of the Philippines Cholera Committee were: Dr J. C. Azurin, Director of Quarantine, Manila (overall coordinator of the study), Dr A. Cruz, Secretary of Health, Dr J. Sumpaico, Director of Research Laboratories, and Dr J. Dizon, Chief, Disease Intelligence Center. Dr M. Alvero, Bureau of Health Services, is the Epidemiologist of this study. Requests for reprints should be sent to Dr J. C. Azurin. vaccine during the first 3 months, the question remained whether or not such serotype-specificity can be obtained against El Tor Ogawa infection, which has been endemic in the Philippines since 1961. This paper reports a field trial carried out in Negros Occidental in 1970-71, in which classical Ogawa and Inaba monovalent vaccines from the lots used in the 1968-69 Dacca trial, together with monovalent El Tor Ogawa and Inaba vaccines pro- duced in the Philippines, were tested against El Tor Ogawa infection. The principles applied were the same as in the 1964 Philippines trial. STUDY AREA AND POPULATION The general characteristics, environmental condi- tions, health facilities, resources, and population structure of the study area have been previously described (Philippines Cholera Committee, 1965). This area was included in the controlled field trial of cholera and cholera El Tor vaccines in 1964-65 and the controlled field trial of different doses of cholera El Tor vaccine in 1966-67 (Azurin et al., 1967; Philippines Cholera Committee, 1968). In the present trial, the estimated size of each of the 5 vaccine groups was 40 000 (adequate for statistically significant results), making a total of about 200 000 vaccinees. Barrios (villages) and com- munities with the highest incidence of cholera were 3080 13- 2 14 PHILIPPINES CHOLERA COMMITTEE Table 1. Vaccinated population by locality and coverage. Vaccinated pop. Locality Target pop. no. % Escalante 30 000 16 808 56.0 Sagay 26 000 21 434 82.4 Victorias 22 000 18 156 82.5 Total 78 000 56 398 72.3 Saravia 9 000 6 551 72.8 Silay 21 000 14 503 69.1 Talisay 17 000 14 338 84.3 Bacolod 37 000 33 802 91.4 Total 84 000 69 194 82.4 Bago 28 000 23 201 82.9 Pulupandan 13 000 10 597 81.5 Valladolid 11 000 7937 72.2 San Enrique 8 000 4 233 52.9 Total 60 000 45 968 76.6 Hinigaran 20 000 17 206 86.0 Binalbagan 15 000 11 838 78.9 Himamaylan 30 000 16 122 53.7 Kabankalan 13 000 6 840 52.6 Total 78 000 52 006 66.7 Total 300 000 223 566 74.5 selected for inclusion. These are mainly coastal communities, belonging to the lowest socioeconomic classes, where factors such as high population den- sity, insanitary waste disposal, and poor water supply favour the spread of disease. The number of communities included in the trial was considered sufficient to yield an estimated population of 300 000, since previous studies had shown that 70-75% of the population was the maximum that could be covered in a vaccine trial. Table 1 shows the target populations and number of vaccinees in the towns and cities. The total coverage was 74.5% of the population. MATERIALS AND METHODS Vaccines Four monovalent cholera vaccines were tested against a control typhoid vaccine. The vaccines were: (1) Vaccine A: monovalent El Tor Inaba fluid vaccine prepared from El Tor strain Inaba 8273 and 6973, and containing 8 x 109 organisms per ml (2) Vaccine B: monovalent El Tor Ogawa fluid vaccine prepared from El Tor strain Ogawa 299 and 1418, and containing 8 x 109 organisms per ml (3) Vaccine C (control): monovalent typhoid vac- cine containing 1 x 109 organisms per ml (4) Vaccine D: freeze-dried monovalent classical vaccine prepared from Vibrio cholerae, Ogawa, strain NIH 41, containing 8 x 109 organisms per ml (when restored with sterile isotonic phosphate- buffered saline containing 0.25% phenol) (5) Vaccine E: freeze-dried monovalent classical vaccine prepared from V. cholerae, Inaba, strain NIH 35A3, containing 8 x 109 organisms per ml (when restored with sterile isotonic phosphate- buffered saline containing 0.25% phenol). Vaccines D and E were of the same lots as those used in the Dacca trial in 1968-69 (Mosley et al., 1970). The organisms were grown on the surface of casamino acids agar for 18 hours, har- vested in phosphate-buffered saline, and killed with formalin at a final concentration of 0.25%. The formalin was neutralized with sodium bisulphite and the vaccine was lyophilized. The 5 vaccines used in the trial were all bottled in the same type of 50-ml container, had a similar appearance, and were identified only by code letters. Each vaccine was administered subcutaneously in a dose of 0.5 ml, irrespective of the age of the vaccinee. The mouse-protective potencies of vaccines D and E have been described by Mosley et al. (1971), but those of A and B were not available. Vaccination The 223 566 volunteers were immunized between 2 and 29 July 1970. The campaign was conducted on a house-to-house basis, rather than by mass vaccina- tions in schools and other public places, so as to 1 Vaccines A, B, and C were prepared by the Bureau of Research and Laboratories, Manila. Vaccines D and E were kindly supplied by Dr J. R. Seal, National Institute of Allergy and Infectious Diseases, Bethesda, Md., USA. FIELD TRIAL OF CHOLERA VACCINE 15 ensure the maximum possible coverage of the com- munities. Random allocation of vaccines to volunteers was achieved by strict adherence to a random ordering of vaccine code letters. To determine the comparability of the 5 groups of vaccinees, an analysis was made of a 2% sample of the immunization cards (Table 2). The 5 groups were also shown to be similar in other respects, such as environment, conditions of work, and the risk of exposure to disease; consequently their susceptibility to cholera should have been the same. Surveillance The vaccinees were kept under observation by members of the surveillance teams. Patients with diarrhoea reporting to health centres, hospitals, or private clinics, and those found during house-to- house visits were swabbed once. The rectal swabs were inoculated into alkaline peptone water and sent to the central laboratory in Bacolod for examination as previously described (Philippines Cholera Com- mittee, 1965). Surveillance was carried on for a period of 7 months after vaccination, or longer if cholera cases were occurring in the study area. RESULTS During the surveillance phase, a total of 223 El Tor Ogawa cases were confirmed bacteriologically. There was no El Tor Inaba case or carrier. The distribution of the cases among the vaccine groups is shown in Table 3. The incidence of cholera in the control group from August 1970 to February 1971 was exactly 2 per 1 000. The incidence in the groups immunized with monovalent cholera vaccines was much lower. The distribution of the 223 cases by month of occurrence is shown in Table 4. The pattern of occurrence conforms to the previously observed incidence curve of cholera in Negros province, except for a rise during the month of January 1971. Usually, the peak incidence is observed late in November, and is followed by a gradual waning until late January or early February. This time, an explosive outbreak occurred in the northern town of Sagay during the second half of January 1971, explaining the unusual rise for that month. Table 5 shows the distribution of the cases by month (or 30-day period) from the date of vaccina- tion to the date of onset of the disease. This tabulation has greater validity than Table 4 for the purposes of this study, since all persons were not immunized on the same day. Thus, whereas the greatest number of cases occurred during January 1971, the sixth month of surveillance, most of the vaccinated persons fell sick in the seventh month after vaccination. Because of the Sagay epidemic, the efficacy of each vaccine was analysed separately for that city and other areas.1 The results show that the attack rate was low among the control group during the first 3 months (3.4 per 10 000 population), and higher during the second 3 months (8.8 per 10 000). The low rate during the first 3 months makes it difficult to judge the effectiveness of the vaccines during that period. It may be noted, however, that El Tor Ogawa and classical Inaba and Ogawa monovalent vaccines conferred a certain degree of protection against El Tor Ogawa infection (47-67%) in all age groups during this period. During the second 3-month period, all the vaccines showed a similar degree of effectiveness (63-68 %), which was virtually uninflu- enced by age, although there were some variations among the vaccine groups. Each vaccine showed a higher degree of effectiveness during the second 3- month period than during the first, with the excep- tion of the classical Ogawa vaccine, which showed similar degrees of protection in the two periods. In general, it appears that both classical and El Tor monovalent Inaba vaccines can protect man against heterologous El Tor Ogawa infection, although such heterologous protection may be a little lower with the El Tor Inaba vaccine tested in the trial. It is, however, interesting to note that the classical Inaba vaccine and the El Tor Ogawa vaccine revealed similar protective efficacy. In the Sagay City epidemic, which occurred during the sixth and seventh months after vaccination, the attack rate in the control vaccine group (74.4 per 10 000 population) was much higher than that ob- served in other areas. In this epidemic the classical Ogawa vaccine appeared to be the most effective of the vaccines, followed by El Tor Ogawa, classical Inaba, and El Tor Inaba vaccines, in that order; the differences between the groups, however, were not significant. Better protection was generally observed in the age groups above 5 years than in the 0-4-year 1 Space did not permit inclusion of two tables illustrating the results that follow. These tables have been deposited in the WHO library, and copies may be obtained on request to: Chief Librarian, WHO, 1211 Geneva 27, Switzerland. PHILIPPINES CHOLERA COMMIEE w 6 a w- Co co Co Co to 0 CV) 0 co 0 CV Co let N4 CIO Ni 04 1- 0N4 N O) Nl CV) CO- LO r- 6 CIO Co O) ai O) CV CO o_ i o. 6 oo C 6 N CO C.:O ) O _ o v-0 0 0N] - . 00 O C OO CO q 0)M b!t C- R C - N-t N - N - 1 0 6 C co 6C ae 6 a at 6 C C- v- Co, V.' cO Co a) Co co 0 00 r- N4 V." C14. CV) N 0 n oi 0 O) co .: N 0 O) N4 N CV) a) 0 co 0 co ,CO a) CV 00 CV) co CD N1 L6 CD N1 co uNcs 00 0 CV 0 N1 Ne 00 co cV) C'.' N O) Co CV) C- 0 0. v- OY) 'V N4 N1 co r- CV) C-) co Mt C-V Co O)- r- - CV Cli Co N~ CV) 0 - La N N- N- 0O 00 mO) 0t O) 0C LO 0 Co .: LO Co C.'. CD 006 rC CIO I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C. 0 CV) O) 0 Co 0a c; co co Co 0 o0 NCo4 O O)n 00 'Vt O) LO) Vtt 0 0 0 _ 0) N _ 0o CO rlr 0o _ O) o ae m re n Lo qt w co Lo co c 4 a) LO O Lo N N 0 o 0) ) O) Co S 0 Co CO N N O Co CV X Co c LO N 0 I N Mt r- 0 O N Co CC- CV 0 CO CO Co Co 0 6t 0OC14 ole Nv- C1_ N N 0 -) N Co LO M 0 C)O %X N LON N 6 0 C ON O) > i C N N _ O 0 o N O)£ a O) Co CV 0 O) rh N Co -C _ O).t 0I" O 6 C co w- CO co N1 N~ N1 CD 0 w- Co 0Nq CV CV) CD Co 1- 0 O) 0C5 Lo 0 cV 6 'Vt m le m I, m t 0 t m Cot 0 Ce Ut r--o - N N mo 0 o; M O CD 1- %- N N co mcs w w c)o A\ 0) CD 'Vt 0 16 0 0 U. 0 0 C 0) .7m a) .0 C.)c C x0) cn 0) 0) cn .-O0)0) C 0 -o .5 CO) 0 .0 C .- +a 0 .2 0 0.0 0 N .CEE4- co ae FIELD TRIAL OF CHOLERA VACCINES Table 3. Cholera El Tor among vaccine groups, August 1970 to February 1971 Vaccine No. vaccinated No. of cases Rate per Effective-group 10000 ness % a2 A 44 500 38 8.5 58 B 45750 33 7.2 65 D 44 450 36 8.1 60 E 44700 26(1) b 5.8 71 C 44 200 90(1)b 20.3 0 total 223 600 223 49.9 a Effectiveness calculated as the percentage reduction in the case rate in the vaccinated groups compared with the control group. b The numbers in parentheses indicate fatal cases. Table 4. Monthly distribution of cholera cases among vaccine groups, August 1970-February 1971 Vaccine group Period Total A B D E C August 1 2 0 0 2 5 1970 September 12 6 6 5 13 42 October 1 2 4 1 7 15 November 6 8 7 6 23 50 December 4 3 6 5 9 27 January 14 11 13 9 34 81 1971 February 0 1 0 0 2 3 Total 38 33 36 26 90 223 Table 5. Cholera cases by 30-day periods from vaccina- tion to onset of illness, August 1970 to February 1971 Days from Vaccine groups vaccination Total to illness A B D E C 0-30 0 0 0 0 0 0 31-60 2 2 1 1 4 10 61-90 11 6 7 4 11 39 91-120 3 5 3 1 10 22 121-150 6 8 8 8 22 52 151-180 6 1 5 5 13 30 181-210 10 10 12 7 29 68 211-240 0 1 0 0 1 2 Total 38 33 36 26 90 223 age group. This observation confirms previous find- ings, including the latest Dacca field trial using the same classical Ogawa and Inaba monovalent vac- cines (Mosley et al., 1970). It was also observed that the efficacy of all the vaccines was still high (60-87% for the homologous Ogawa vaccines) during the sixth and seventh months after vaccination. The statistical significance of the differences in incidence rates for the various vaccine groups is shown in Table 6, which incorporates all cases occurring during the trial. Only the classical Ogawa monovalent vaccine conferred significant protection during the first 3 months after vaccination. From the fourth to the seventh months the four monovalent vaccines gave significant protection when compared with the control vaccine group. The differences in protection among the vaccine groups, however, were not significant, although the figures for the first 3 months suggest a difference between classical Ogawa and El Tor Inaba vaccines. Table 6. Incidence rates per 1000 by vaccine group and time interval between vaccination and onset of illness, and significance of the difference between incidence rates for various vaccine groups Time interval (days) Vaccine 31-90 91-150 151-210 of Incradtence of Incradtence of Incradtence cases cases ~~cases rt A 13 0.2921 9 0.2022 16 0.3600 B 8 0.1749 13 0.2842 11 0.2404 D 8 0.1780 11 0.2475 17 0.3842 E 5 0.1118 9 0.2013 12 0.2684 C 15 0.3394 32 0.7240 42 0.9502 Difference between vaccine groups C and A 0.39 3.70 a 3.43 a C and B 1.54 2.93a 4.35a C and D 1.49 3.22 a 3.28a C and E 2.25a 3.63a 4.16a B and A 1.15 0.80 1.04 B and D 0.04 0.34 1.20 B and E 0.79 0.80 0.26 A and E 1.90 0.01 0.77 D and E 0.81 0.46 0.94 a Significant at 5 % level. 17 PHILIPPINES CHOLERA COMMITrEE Table 7. Confirmed cholera cases by place of confine- ment, August 1970 to February 1971 Vaccine No. of cases (%) Total no. group of cases Hospital Home A 8 (21.1) 30 (78.9) 38 B 6 (18.2) 27 (81.8) 33 D 4 (11.1) 32 (88.9) 36 E 5 (19.2) 21 (80.8) 26 c 7 (7.8) 83 (92.2) 90 Total 30 (13.5) 193 (86.5) 223 Table 7 shows the relative proportions of hospi- talized and home cases in each vaccine group, an indirect measure of the clinical severity of the cases in each group. The number of hospitalized cases is not large enough to allow valid comparison of vaccine groups. Nevertheless, the figures show that only 1 out of every 8 cholera cases was severe enough to need hospitalization. This feature, like the age predi- lection of the disease, is indicative of the endemic nature of cholera in this area. Analysis of the duration of diarrhoea in confirmed cases in each vaccine group agreed with earlier observations that once an individual contracts the disease, the clinical picture is the same irrespective of vaccination status (Philippines Cholera Committee, 1965; Azurin et al., 1967). DISCUSSION The present field trial shows that Inaba mono- valent cholera vaccine cross-protects against El Tor Ogawa infection irrespective of the biotype of clas- sical or El Tor vibrio strains used for vaccine production. The homologous classical Ogawa mono- valent vaccine showed a somewhat better degree of protection than the other vaccines, but the difference was not significant. This observation is quite differ- ent from those in the 1968-69 Dacca trial, where only classical Inaba monovalent vaccine gave appreciable protection against homologous classical Inaba infec- tion, and classical Ogawa monovalent vaccine failed to protect against the heterologous Inaba disease. To explain the type-specific protection revealed by the classical Inaba monovalent vaccine against classical Inaba infection, Mosley et al. (1970) assumed that immunity related to Inaba type-specific antigen C must play a greater role in human protection than immunity related to common antigen A. Their assumption was based on the observation that al- though the classical Ogawa vaccine induced appre- ciable heterologous vibriocidal antibody against Inaba vibrios, it did not offer any significant protec- tion. Since the same lots of freeze-dried Ogawa and Inaba monovalent vaccines used in Dacca were used in the present trial, it is conceivable that the poten- cies of both vaccines were virtually the same. The present finding that protection against Ogawa infec- tion was conferred by the homologous Ogawa as well as by the heterologous Inaba monovalent vaccines suggests the following possibilities. First, the immu- nities associated with both type-specific B antigen in the Ogawa monovalent vaccine and common A anti- gen in the Inaba monovalent vaccine are almost equally effective against Ogawa infection in man. Secondly, the common antigen A contained in both monovalent vaccines is responsible for the protection against Ogawa infection. Thirdly, the C antigen in the Inaba cell is responsible for protection against Ogawa infection, since Ogawa cell contains a small amount of C antigen according to the antigenic schema described by Kauffman (1950) and Sakazaki & Tamura (1971). Of these, the third assumption may be unlikely, since it is known that Inaba factor serum containing only anti-C antibody cannot act on Ogawa cells in vitro and in passive immunization in experimental animals. It is therefore most probable that anti-common A immunity produced by Inaba monovalent vaccine is responsible for the protection against the heterologous Ogawa infection, which indicates that, contrary to Inaba cells that are con- trolled only by the type-specific anti-C immunity, Ogawa cells may be almost equally susceptible to either anti-specific B or anti-common A immune mechanism(s). This hypothesis finds support in the observation of Shrivastava (1965), who showed that Ogawa somatic antigen and its corresponding poly- saccharide reacted with Ogawa as well as Inaba antiserum, but that Inaba cell-wall antigen and its polysaccharide combined specifically with the ho- mologous Inaba antiserum only. Although the me- chanism(s) involved in the protection of man are not clear at present, the results of this field trial may support the conclusion of Shrivastava (1965) that the use of the Inaba vibrio for the manufacture of cholera vaccine should be adequate. The observation that the effectiveness of each of the monovalent vaccines, regardless of the biotype, 18 FIELD TRIAL OF CHOLERA VACCINES 19 persisted beyond 3 months after vaccination is very interesting, since the majority of the past field trials in Dacca and Negros Occidental, where bivalent cholera vaccines were tested, tended to show that the highest degree of protection was achieved during the first 3 months after vaccination and was followed by a gradual decline in protective effect. In the present trial, however, an appreciable degree of protection (60-87%) was observed during the Sagay City epi- demic, which occurred 6 and 7 months after vaccina- tion. This long-lasting immunity conferred by the monovalent vaccines would seem to require confir- mation in further trials comparing monovalent vac- cines with bivalent vaccine produced from the same batch of monovalent Ogawa and Inaba vaccines. Although the Inaba monovalent vaccine may be considered useful against both Inaba and Ogawa infection, the results from the Sagay City epidemic suggest that the homologous classical Ogawa mono- valent vaccine may be better than the heterologous classical Inaba monovalent vaccine, particularly in the 0-4-year age group, which is the group at highest risk, at least in endemic areas. There is also a possibility that the monovalent vaccines may en- hance antigenic shift, as has been demonstrated by Sack & Miller (1969) in experimental animals. R1tSUMm ESSAI PRATIQUE CONTR6Lt VISANT A EVALUER L'EFFICACITE DE VACCINS ANTICHOLERIQUES MONOVALENTS CLASSIQUES ET EL TOR AUX PHILIPPINES On a procede a un essai control6 de differents vaccins anticholeriques monovalents (classique Inaba, El Tor Inaba, classique Ogawa, El Tor Ogawa) dans la province du Negros Occidental aux Philippines, oiu le chol6ra El Tor serotype Ogawa est end6mique. Plus de 220 000 personnes y ont participe. Les deux vaccins prepar6s a partir des souches classiques appartenaient aux m&mes lots que les vaccins essay6s A Dacca (Bangladesh) en 1968-69, essai au cours duquel le vaccin classique Inaba avait fait preuve d'une efficacit6 tres satisfaisante (93% de protection) contre le cholera classique serotype Inaba alors que le vaccin classique Ogawa ne conf6rait prati- quement aucune protection contre la maladie. Dans l'essai actuel aux Philippines, les quatre vaccins ont confer6 une bonne protection (58 a 71 ,°) contre le chol6ra El Tor serotype Ogawa, les vaccins homologues Ogawa t6moignant d'une l6gere superiorit6. La nature du biotype n'a eu aucune influence sur l'efficacit6 compar6e des divers vaccins. REFERENCES Azurin, J. C. et al. (1967) Bull. Wld Hlth Org., 37, 703-727 Kauffmann, F. (1950) Acta path. microbiol. scand., 27, 283 Mosley, W. H. et al. (1971) In: International Symposium on Enterobacterial Vaccines, Berne, 1969, Symposia Series in Immunobiological Standardization, Basel, Karger, vol. 15, pp. 185-196 Mosley, W. H. et al. (1970) J. infect. Dis., 121, Suppl., 1-9 Philippines Cholera Committee (1965) Bull. Wld Hlth Org., 32, 603-625 Phillipines Cholera Committee (1968) Bull. Wld Hith Org., 38, 917-923 Sack, R. B. & Miller, C. E. (1969) J. Bact., 99, 688-695 Sakazaki, R. & Tamura, K. (1971) Jap. J. Sci. Biol., 24, 93-100 Shrivastava, D. L. (1965) In: Proceedings of the Cholera Research Symposium, Honolulu, 1965, Washington, D.C., US Government Printing Office, pp. 244-247 (PHS Publication No. 1328) Watanabe, Y. et al. (1970) Tex. Rep. Biol. Med., 27, Suppl. 1, 275-298