- Email: [email protected]
Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial
Vaccine xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial Shahrokh Izadi a,⇑, Seyed Mohsen Zahraei b, Masoud Salehi c, Mahdi Mohammadi a, Seyed Mehdi Tabatabaei a, Talat Mokhtari-Azad d a
Health Promotion Research Centre, School of Public Health, Zahedan University of Medical Sciences, Zahedan, Iran Centre for Communicable Diseases Control, Ministry of Health and Medical Education, Tehran, Iran c Research Center for Infectious Disease and Tropical Medicine, Zahedan University of Medical Sciences, Zahedan, Iran d National Reference Laboratory for Measles and Rubella, School of Public Health, Tehran University of Medical Sciences, Tehran, Iran b
a r t i c l e
i n f o
Article history: Received 15 September 2017 Received in revised form 15 December 2017 Accepted 18 December 2017 Available online xxxx Keywords: Measles Vaccine Infants Immunization Immune response
a b s t r a c t Background: A non-inferiority multi-centre parallel randomized double-blind trial was implemented in Zahedan district, Sistan-va-Baluchestan province, Iran, to compare the performance of the two measles vaccines which are in use in the National Immunization Programme of Iran and are of two different measles virus vaccine strains: Edmonston-Zagreb (EZ) strain vs. AIK-C strain. The main outcome measure was appearance of anti-measles antibody in sera. Methods: 200 infants, 8–12 months old, whose parents consented for their children to be included in the study, were randomized in permutation blocks of size 4–8 in four Urban Health Clinics. Having given a pre-vaccination blood sample, they received measles-rubella vaccine containing one of the vaccine strains mentioned before. After 60 days, the second blood sample was taken. The sera of the pre- and post-vaccination blood samples were tested for anti-measles antibodies in the National Reference Measles Laboratory. Parents, laboratory technicians and statistician were blind to groupings. Results: Of the 200 children equally randomized in the two arms, 185 who were seronegative before vaccination (88 in the EZ arm and 97 in the AIK-C arm) were entered in the final analysis. The seroconversion rate in the EZ arm was 76.1% (95% CI: 60.2–85.2%), and that in the AIK-C arm was 58.7%; (95% CI: 48.8– 68.7%). The absolute rate difference was 17. 4% (4.1–30.9%; P-value: .012), and the relative seroconversion rate of EZ to AIK-C was 1.3 (95% CI: 1.1–1.6; P-value: .012). No adverse events were reported during the study period. Conclusion: A considerable difference in the seropositivity of different measles containing vaccines could be demonstrated in the first year of life. Trial Registration: Iranian Registry of Clinical Trials Registration Number: IRCT2016032827144N1; May 10, 2016 (www.who.int/ictrp/network/irct/en/) Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Measles as one of the most contagious viral diseases, with the largest basic reproductive number ever known (about 20) and having a number of different vaccines produced by different laboratories around the world, remains an important preventable ⇑ Corresponding author at: P.O. Box: 98155-759, Zahedan, Iran. E-mail addresses: [email protected] (S. Izadi), [email protected] (S.M. Zahraei), [email protected] (M. Salehi), memohammadi@yahoo. com (M. Mohammadi), [email protected] (S.M. Tabatabaei), mokhtari@ hotmail.com (T. Mokhtari-Azad).
childhood disease. Most measles vaccines originate from the Edmonston strain of measles virus, including the EZ1 strain and the AIK-C2 strains that are in common use in Iran’s immunization program [1,2]. The first one is the product of Serum Institute of India (SII), and the second one is produced domestically in Razi Institute of Serum and Vaccine Production (RIS). AIK-C strain was the first live attenuated measles virus strain whose complete genomic sequence was published and the Edmonston-Zagreb strain is the most 1
Edmonston-Zagreb. A: America, I: Iran, K: The Kitasato Institute, and C: virus adapted to chick-embryo cells. 2
https://doi.org/10.1016/j.vaccine.2017.12.048 0264-410X/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
2
S. Izadi et al. / Vaccine xxx (2017) xxx–xxx
commonly used measles vaccine in the immunization programmes of the WHO. Both these virus strains have been developed from the Edmonston strain [3]. Since 2009, measles outbreaks have occurred in a gradually increasing trend in several parts of Iran [4,5]. Within the past decade, the measles elimination campaign in Iran has witnessed different levels of performance. A number of outbreaks were prevalent in the period of 2009–2014, and microcirculations were undermining the faith in the effectiveness of the immunization activities against measles [5–10]. To constrain an outbreak, the health system sometimes uses the imported measles vaccine SII and sometimes the products of RIS. In these activities, the target population is usually composed of a wide age range, from 6 months to 12 years [5,6,9,11]. The main objective of the present study was to compare the effectiveness of both types of vaccines, which are in common use in outbreak response immunization activities in Iran. The effectiveness of these vaccines has been evaluated in infants under 9 months of age in several studies; however, for the older age groups up to 12 months of age, the health authorities are seeking the best choice between the two vaccine types [12–14]. Hence, a randomized clinical trial was implemented in Zahedan, the capital city of Sistan-va-Baluchestan Province of Iran. 2. Methods 2.1. Trial design A multicentre double blind randomized non inferiority equally balanced parallel controlled trial, comparing measles vaccine strain AIK-C with EZ strain in 8-to-12-month children, was implemented in four Urban Health Clinics in Zahedan, the capital city of Sistan-va-Baluchestan Province of Iran, during spring and summer 2016. 2.2. Participants All infants aged between 8 and 12 months being brought to the health clinics involved in the study for immunizing purposes were examined; and if eligible, they were invited for participating in the study. The inclusion criteria for the participants were: (a) being in the age range 8–12 months (240 days to 12 months and 30 days); (b) having no contraindication for receiving measles containing vaccines; (c) negative history of receiving any kind of measles containing vaccine; (d) feasibility of blood sampling from the physical and medical points of view; (e) being permanent residents of Zahedan City; (f) being Iranian; Pakistani, Afghan refugees or travellers from other districts were not invited to be included in the study.
the same health centre both for treatment and for the event to be recorded in the study checklists. After blood sampling, each participant’s allocation packet was opened, and vaccination with the defined vaccine was done. The packet and the allocation paper were pinned as attachments to the participant’s questionnaire. 2.4. Vaccines characteristics The participants in both arms of the study were vaccinated with a bivalent measles/rubella vaccine (that is in common use in outbreak containment activities in Iran). The vaccines used in this study had either been produced domestically by Razi Vaccine and Serum Research Institute (Batch Numbers: 01694016 and 01594026) or were the products of Serum Institute of India (Batch Numbers: 012N5011A and 012N5013) imported from India. The measles component of the domestically produced vaccine was an attenuated AIK-C strain of live measles virus propagated in human diploid cells, known as AIC-HDC strain, while the virus strain used in the imported vaccine was the Edmonston-Zagreb (EZ) strain, also produced in human diploid cells. The vaccines were injected subcutaneously in anterolateral aspect of the thigh, using standard needles 16 mm in length. The imported vaccines were packed in five-dose vials and the domestically produced vaccines in 10-dose vials. The vaccines used in the study were brought from the vaccine depot of the Ministry of Health from Tehran with the same vaccine carrier vehicle, and they were stored in the same vaccine storage system in the Province Health Centre. The vaccines were distributed to the four cooperating health centres using the same vaccine carriers by the same health staff on the first day of the study. Cold chain in the health clinic were checked by both the study supervisor and the physician in charge of each health clinic. 2.5. Outcomes The primary outcome measure was the detection of IgG-antimeasles antibody 60 days after vaccine inoculation, using ELISA technique. 2.6. Sample size calculations Regarding a reported seroconversion rate of 80–95% (effect size equal to 0.15) and considering type-one error equal to 0.05 and type-two error equal to 0.20, the sample size for each arm of the study was calculated as 76 persons. Considering an estimated lost to follow-up rate of about 25%, the sample size for each arm was increased to 100, which was equally distributed among the four Urban Health Clinics involved in the study. 2.7. Randomization
2.3. Interventions After an interview in the local language by a health expert about the study and its objectives, if the infant’s parent or guardian consented for their child to be included in the study, they were asked to sign a written consent form. A questionnaire containing questions about participants’ demographic characteristics and history of previous vaccinations and exanthematic diseases were filled out for them. Questionnaire comprised participants’ birth weight, mother’s age, age at the beginning of supplementary food, mother and father’s level of education, and anthropometric measurements, as well as their full contact information including phone numbers and addresses. Blood sample was taken before vaccination and 60 days after vaccination. The parents were instructed about the possible adverse effects of the vaccines. They were also advised that upon occurrence of any unexpected events, they should return to
After obtaining the informed consent (as described above), filling the questionnaire and doing the first blood sampling, the participants were randomly assigned, in a 1:1 ratio in order to receive either the AIK-C vaccine or the one containing EZ strain. Randomization was balanced with stratification according to the four health clinics. The names of the vaccine producers had been printed on small papers, placed in opaque packets and arranged based on computer generated permutation blocks of 8, 6 and 4 (produced by the first author). In order to prevent any disarrangement of the packets, in addition to printing the row number of the packets on them, they were threaded using a strong string; and in order to prevent reading the contents of the packets by keeping them against a strong light source, a layer of aluminium foil was placed inside each packet. On the participants’ blood and sera tubes, only the row number and the participants’ names had been
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
S. Izadi et al. / Vaccine xxx (2017) xxx–xxx
printed. Participants, lab technicians, and the study statistician were unaware of the study group assignments until the final report had been completed. However, vaccinators, opening the allocation packets, reading the lot papers, and inoculating the vaccines (all done by the same person), were not blind to the allocations. A special identification card was given to each participant’s parent or guardian (to remind them that they have to return to the same health clinic 60 days later). All of the participants’ exact contact details were both recorded in their questionnaires and booked in another document in order to recall them for the second blood sampling on the 60th day after the vaccination. All of the participants were contacted 60 days after vaccine inoculation and were invited to the same health clinic for doing the second blood sampling meant to check for seroconversion. For those participants who did not return on time, a health worker went to their addresses to ask them in person for coming to the clinic. The health workers cooperating in this study were experienced university-educated vaccinators adept in the local language, and all of them took part in two introductory training sessions two days before the beginning of the study. All the anthropometric measurements were done according to WHO recommendations and using high quality well-calibrated equipment [15]. The participants’ recumbent lengths were measured in supine position with an accuracy of half a centimetre. Their weights were measured with mechanical baby scales (SecaTM) with an accuracy of 10 g. Their ages were calculated by subtracting their birth dates from the date of vaccine inoculation. Anthropometric indices (Height for Age Zscore (HAZ), Weight for Age Z-score (WAZ), and Body Mass Index
3
for Age Z-score (BAZ)) were calculated using ‘WHO AnthroPlus’ software (a type of software for the global application of the WHO reference 2007; Department of Nutrition, World Health Organization; Geneva, Switzerland). Stunting and wasting were defined respectively as HAZ and BAZ equal to or below 2 standard deviations below the reference median value. From May 14th to May 18th, 2016, a total number of 200 participants, i.e. infants aged between 8 and 12 months, were recruited and blood sampled before the inoculation of the MR (MeaslesRubella) vaccine. All of the participants were invited for the second blood sampling 60 days later from July 14th to July 19th, 2016. There were 12 lost to follow-up participants (2 in the RIS arm and 10 in the SII arm) mostly due to parents’ refusal to bring their children for the second blood sampling. In total, 6 of the participants were categorized as seropositive (OD 0.1) based on the results of their blood samples taken before the vaccination. Three of these participants were among the twelve participants who did not take part in the second blood sampling on day 60. Out of the remaining three seropositive participants, one was in the RIS arm of the study and the other two were in the SII arm. Even though all of the three had increases in the OD levels of their second blood samples (which were taken 60 days after the vaccination), they were excluded from the analysis. This way, the total number of participants who were seronegative before the vaccination and whose second blood samples, 60 days after the vaccination, were considered appropriate for analysis added up to 185: i.e. 97 in the RIS arm of the study and 88 in the SII arm. Fig. 1 shows the flowchart of selecting and following the participants in the study.
Fig. 1. Flow of selecting and following the participants in the study.
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
4
S. Izadi et al. / Vaccine xxx (2017) xxx–xxx
2.8. Laboratory procedures The sera were separated from the blood samples in the same health clinic within an hour after blood sampling; and they were kept at 20° centigrade until delivery to the National Reference Measles Laboratory in the School of Public Health, Tehran University of Medical Sciences, Tehran, Iran, for the measurement of antimeasles IgG level. The mentioned laboratory is part of the WHO’s network of laboratories. In order to quantify the presence of IgG antibodies, indirect ELISA tests were used according to the manufacturer’s instructions (EnzygnostÒ Anti-Measles Virus/IgG; Siemens, Marburg, Germany). The sera of each phase of blood sampling were sent to the National Reference Measles Laboratory separately, but the laboratory did the test of both phases (i.e. the sera taken before vaccination and the sera taken 60 days later) at the same time after receiving all of the sera. Based on the manufacturer’s descriptions with Enzygnost Anti-Measles Virus/IgG, the samples containing approximately 150 mIU/mL were found to be within the range of 0.100–0.200 DA. Specimens below 0.100 were considered as negative [7,10].
Table 1 The results of comparing some of the participants’ characteristics by the vaccine inoculated for them. AIK-C arm Number Mean (Standard Deviation)
EZ* arm Number Mean (Standard Deviation)
P-value**
Birth weight (g)
97 3513.8 (2005.0)
88 3548.0 (1962.8)
0.907
Mothers’ age (years)
97 25.8 (6.7)
88 25.0 (5.4)
0.394
Age at beginning of supplementary food (months)
97 6.1 (0.8)
88 6.2 (0.8)
0.279
Participants’ age
97 10.5 (1.1)
88 10.4 (1.1)
0.387
Fathers’ education level
97 6.0 (4.1)
88 5.7 (4.0)
0.581
Mothers’ education level
97 4.1 (4.0)
88 4.8 (4.0)
0.269
Body mass index for age Z-score
96 0.16 (1.0)
87 0.21 (1.2)
0.726
Height for age Z-score
96 0.9 (1.2)
87 0.7 (1.2)
0.317
Weight for Age Z-score
97 0.6 (0.9)
87 0.6 (1.1)
0.689***
2.9. Ethics The protocol of the study has been registered in Iranian Registry of Clinical Trials (IRCT), which is under supervision of World Health Organization (IRCT Registration Number: IRCT2016032827144N1; Date Registered: May 10, 2016). In addition, The Ethics Committee on Medical Research of Zahedan University of Medical Sciences has reviewed and approved the implementation of the study (Reference Number: IR.ZAUMS.REC.1395.26; Date Approved: April 25, 2016) (www.who.int/ictrp/network/irct/en/). *
2.10. Statistical methods
** ***
The collected data were transferred to a computerized data bank, and they were analysed using Stata (ver. 11.2). In the analysis of the data, the design of the study was brought into consideration by using the ‘‘svyset” command of Stata., i.e. the health clinics (where the randomizations and blood samplings were done) were defined as strata, and primary sampling units (i.e. participants) were defined within the health clinics. The data were analysed using independent sample t-test. In all instances, the 95% confidence intervals were calculated; and where applicable, the statistics were reported as ‘mean ± standard deviation’. 3. Results The mean follow-up time between the vaccination and the second blood sampling was 60.9 days (minimum 60 days and maximum 62 days). During this period, no adverse events were reported in any participants in the two arms of the study. Table 1 shows some of the characteristics of the participants. Total seroconversion rate was 67.0% (95% CI: 60.2–73.9%). The seroconversion rate in the SII arm of the study was 76.1% (95% CI: 60.2–85.2%), and that in the RIS arm was 58.7%; (95% CI: 48.8–68.7%). The absolute rate difference was 17. 4% (4.1–30.9%; P-value: .012), and the relative seroconversion rate of SII to RIS was 1.3 (95% CI: 1.1–1.6; P-value .012) (Table 2). For each one of the vaccine types used in the study, two vaccine batch with different batch numbers were in use. Since there was no insight in any probability of any difference between the different production series of the same vaccine within the same vaccine type, reasonably, no adjustment or control for balancing the number of vaccine for each serial number was made. However, after the
Edmonston Zagreb strain. Independent sample t-test. Unequal variances considered.
completion of the laboratory tests, when the two vaccine-series within each arm of the study were compared with each other (not with the other arm), surprisingly there were statistically significant differences between the capability of the two vaccineseries of each one of the vaccines in inducing antibody production. Table 2 shows the mean difference between the vaccines used in the study and within each arm of the study between the two serial numbers. The difference between the seroconversion rates of the two serial numbers used within each arm was statistically significant. In comparison of the baseline characteristics reported in Table 1 (using analysis of variance), except for the ‘Body Mass Index for Age Z-score’ (BAZ), there was no other statistically significant difference between the subgroups receiving the different vaccine batches. In pairwise comparisons (using Scheffe test), there were two statistically significant differences between the participants who had received the batch number ‘‘01594026” (AIK-C) and those who had received the vaccines with the batch numbers ‘‘01694016” (AIK-C) and ‘‘012N5013” (EZ).
4. Discussion This was in fact a head-to-head immunogenicity study of two different vaccine strains, i.e. the Edmonston-Zagreb strain (produced by SII) versus AIK-C (produced by RIS). The results of this study showed an obvious difference in the effectiveness of the two vaccines used in the study as well as a probable difference between the effectiveness of different production series of the same vaccines.
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
5
S. Izadi et al. / Vaccine xxx (2017) xxx–xxx
Table 2 Comparison of seroconversion rates between measles virus vaccine strains and between production series (marked by their batch numbers) within each one of the vaccine strains in the study. Measles vaccine strains
Comparison between arms
No.
Mean (SD)
Mean difference (CI)
P-value (independent sample t-test)
EZ* AIK-C
EZ* AIK-C Comparison within EZ* Arm 012N5011A** 012N5013** Comparison within AIK-C Arm 01694016** 01594026**
88 97
0.76 (0.43) 0.59 (0.49)
0.17 (0.04–0.31)
0.012
64 24
0.70 (0.46) 0.92 (0.28)
0.21 (0.01–0.41)
0.037
65 32
0.45 (0.50) 0.88 (0.37)
0.43 (0.23–0.62)
0.000
EZ*
AIK-C * **
Edmonston-Zagreb strain. Vaccine batch numbers.
Although the two vaccines used in this study were of two different measles virus strains, several studies have proved no significant differences between these two vaccines with regard to effectiveness and efficacy [12–14,16–19]. However, in this study, a statistically significant difference was observable. To make certain that the observed results (and seroconversion rates) have had no other reasons besides the qualities of the vaccines used in the study, the authors did their best to control all possible sources of variation, including the cold-chain practices, the skill and knowledge of health workers involved in the study, and appropriate randomization schedule implemented within each health clinic. In addition, the participants’ characteristics that could supposedly be the source of any kind of variations, including mothers’ ages, participants’ ages, feeding conditions, and presence of malnutrition, were brought into consideration and calculation; in no instances were there any statistically significant differences between the two arms of the study. Without any doubt, the presence of maternal antibodies in the first year of life might be mentioned as one of the main reasons for primary vaccination failure (i.e. failure of serologic conversion after the vaccination); and considering the common knowledge about antibody response in different age groups below 1 year of life and also considering the age range of the participants in the study (8–12 months), an antibody response of about 80% could be completely a reasonable expectation in this study [1,20,21]. What was surprising was a significant difference in the observed seroconversion rates between different serial numbers. This finding, that was found by chance, might deserve more attention in the present study. In more details, in each vaccine type used in the study, there was one vaccine serial number with seroconversion rates near or equal to 90%, which, as mentioned before, was completely acceptable and expected in the studied age range (8–12 months), while in the same vaccine type, there was another vaccine serial number with seroconversion rates lower than expectations [1,13]. In simpler words, the vaccines have not acted homogenously within each brand. Usually by default, the potency, efficacy and effectiveness of the different serial numbers of the same vaccine produced by the same vaccine producer are taken to be the same. This finding highlights the importance of paying attention to such details in future vaccine studies. As it was mentioned in the Results section, the only baseline characteristic that showed statistically significant difference between the subgroups receiving the different vaccine batches was BAZ. If this finding is a mere paly of chance or an implication of the presence of a causal relationship, it might not be found in the scope of the present study. As a last word, it should be reminded that the authors found this discrepancy in the action of the different serial numbers of the same vaccine type by mere chance and thoroughly accidentally during one of the exploratory sub-group analyses. Therefore, since this is an un-planned post hoc finding, it is quite natural that we
may not stress its importance more than a casual finding which deserves testing and consideration in future studies. 5. Conclusion It seems that there is considerable difference in the performance of different vaccine strains produced by different vaccine producers and even between different vaccine series produced by the same producer, which deserves more attention both in the selection of vaccines for future interventions and in future studies. Funding This work was supported by the Ministry of Health and Medical Education, I.R. Iran [D304/13893]. Conflict of interest All of the authors hereby declare not to have any potential conflicts of interest. Authors’ contributions SI, SMZ, MS and TMA conceived the idea; SI, SMZ, MM and TMA drafted the proposal; SI and SMZ approved the final proposal; SMT, SI and MS supervised the field stages and data collection; TMA supervised and conducted all of the laboratory work; and MM carried out data management and analysed the data through consultation with SI and SMT. All of the authors have read and approved the final manuscript. Acknowledgements Our due thanks are to the hard-working health workers, at all levels, from the managerial levels to those on the front lines of providing health services in the health facilities involved in the study. In addition, the authors would like to appreciate the hard work of all employees and staff of the National Reference Measles Laboratory in the School of Public Health, Tehran University of Medical Sciences, Tehran, Iran. The authors also wish to express their gratitude to Dr. Ali Beikian, the faculty member of English Language Department of Chabahar Maritime University, for editing the English text of this article. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.vaccine.2017.12. 048.
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
6
S. Izadi et al. / Vaccine xxx (2017) xxx–xxx
References [1] World Health Organization. Measles vaccines: WHO position paper – April 2017. Weakly Epidemiol Rec 2017;92:205–28. [2] Makino S. Development and characteristics of Live AIK-C measles virus vaccine: a brief report. Rev Inf Dis 1983;5:504–5. [3] Bankamp B, Takeda M, Zhang Y, Xu W, Rota PA. Genetic characterization of measles vaccine strains. J Inf Dis 2011;204(Suppl. 1):S533–48. [4] Moghadam M, Afsarkazerooni P, Ebrahimi M, Soltani M, Razmpoor A, Pirasteh E, et al. Measles outbreak in South of Iran, where vaccine coverage was high: a case-series study. Iran J Pub Health 2014;43:375–80. [5] Izadi S, Zahraie SM, Sartipi M. An investigation into a measles outbreak in southeast Iran. Jpn J Inf Dis 2012;65:45–51. [6] Esteghamati A, Gouya MM, Zahraei SM, Dadras MN, Rashidi A, Mahoney F. Progress in measles and rubella elimination in Iran. Pediat Inf Dis J 2007;26:1137–41. [7] Izadi S, Mokhtari-Azad T, Zahraei SM. Measles vaccination coverage and seroprevalence of anti-measles antibody in south-east Islamic Republic of Iran. Eastern Mediterranean health journal = La revue de sante de la Mediterranee orientale = al-Majallah al-sihhiyah li-sharq al-mutawassit. 2015;21:396–402. [8] Karimi A, Arjomandi A, Alborzi A, Rasouli M, Kadivar MR, Obood B, et al. Prevalence of measles antibody in children of different ages in Shiraz, Islamic Republic of Iran. Eastern Mediterranean health journal = La revue de sante de la Mediterranee orientale = al-Majallah al-sihhiyah li-sharq al-mutawassit. 2004;10:468–73. [9] Zahraei SM, Gouya MM, Azad TM, Soltanshahi R, Sabouri A, Naouri B, et al. Successful control and impending elimination of measles in the Islamic Republic of Iran. J Inf Dis 2011;204(Suppl. 1):S305–11. [10] Zahraei SM, Izadi S, Mokhtari-Azad T. Factors affecting the seroconversion rate of 12-month-old babies after the first injection of measles vaccine in the southeast of Iran. Hum Vaccines Immunother 2016;12:3118–24. [11] Personnaz D, Larson H. Iran and Turkey launch world’s largest measles campaigns. Press release. Geneva: Unicef; 2003.
[12] Mirchamsy H, Nilforoushan MA, Shafyi A, Razavi J, Ashtiani MP, Youssofi I, et al. Comparative evaluation of two combined measles-mumps-rubella vaccines based on AIK and Edmonston- Zagreb strains of measles virus. Kitasato Arch Exp Med 1991;64:141–7. [13] Tidjani O, Grunitsky B, Guerin N, Levy-Bruhl D, Lecam N, Xuereff C, et al. Serological effects of Edmonston-Zagreb, Schwarz, and AIK-C measles vaccine strains given at ages 4–5 or 8–10 months. Lancet 1989;2:1357–60. [14] Cutts FT, Grabowsky M, Markowitz LE. The effect of dose and strain of live attenuated measles vaccines on serological responses in young infants. Biol: J Int Assoc Biol Stand 1995;23:95–106. [15] WHO, UNICEF. WHO child growth standards and the identification of sever acute malnutrition in infants and children. Geneva, Switzerland: WHO and UNICEF; 2009. p. 12. [16] Mirchamsy H, Shafyi A, Bahrami S, Kamali M, Nazari P, Razavi F, et al. A comparative field trial of five measles vaccines produced in human diploid cell, MRC-5. J Biol Stand 1977;5:1–18. [17] Mirchamsy H, Shafyi A, Nazari P, Ashtiani MP, Sassani A. Evaluation of live attenuated measles vaccines prepared in human diploid cells for reimmunization. Epidemiol Inf 1988;101:437–43. [18] Rasmussen SM, Biering-Sorensen S, Byberg S, Andersen A, BjerregaardAndersen M, Rodrigues A, et al. The effect of early measles vaccination at 4.5 months of age on growth at 9 and 24 months of age in a randomized trial in Guinea-Bissau. BMC Pediat 2016;16:199. [19] Sasaki K. Studies on the modification of the live AIK measles vaccine. I. Adaptation of the further attenuated AIK measles virus (the strain AIK-L33) to chick embryo cells. Kitasato Arch Exp Med 1974;47:1–12. [20] Hoest C, Seidman JC, Pan W, Ambikapathi R, Kang G, Kosek M, et al. Evaluating associations between vaccine response and malnutrition, gut function, and enteric infections in the MAL-ED cohort study: methods and challenges. Clin Inf Dis: Off Pub Inf Dis Soc Am 2014;59(Suppl. 4):S273–9. [21] Pannuti CS, Morello RJ, de Moraes JC, Curti SP, Afonso AMS, Camargo MCC, et al. Identification of primary and secondary measles vaccine failures by measurement of immunoglobulin G avidity in measles cases during the 1997 São Paulo epidemic. Clin Diagn Lab Immunol 2004;11:119–22.
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial Shahrokh Izadi a,⇑, Seyed Mohsen Zahraei b, Masoud Salehi c, Mahdi Mohammadi a, Seyed Mehdi Tabatabaei a, Talat Mokhtari-Azad d a
Health Promotion Research Centre, School of Public Health, Zahedan University of Medical Sciences, Zahedan, Iran Centre for Communicable Diseases Control, Ministry of Health and Medical Education, Tehran, Iran c Research Center for Infectious Disease and Tropical Medicine, Zahedan University of Medical Sciences, Zahedan, Iran d National Reference Laboratory for Measles and Rubella, School of Public Health, Tehran University of Medical Sciences, Tehran, Iran b
a r t i c l e
i n f o
Article history: Received 15 September 2017 Received in revised form 15 December 2017 Accepted 18 December 2017 Available online xxxx Keywords: Measles Vaccine Infants Immunization Immune response
a b s t r a c t Background: A non-inferiority multi-centre parallel randomized double-blind trial was implemented in Zahedan district, Sistan-va-Baluchestan province, Iran, to compare the performance of the two measles vaccines which are in use in the National Immunization Programme of Iran and are of two different measles virus vaccine strains: Edmonston-Zagreb (EZ) strain vs. AIK-C strain. The main outcome measure was appearance of anti-measles antibody in sera. Methods: 200 infants, 8–12 months old, whose parents consented for their children to be included in the study, were randomized in permutation blocks of size 4–8 in four Urban Health Clinics. Having given a pre-vaccination blood sample, they received measles-rubella vaccine containing one of the vaccine strains mentioned before. After 60 days, the second blood sample was taken. The sera of the pre- and post-vaccination blood samples were tested for anti-measles antibodies in the National Reference Measles Laboratory. Parents, laboratory technicians and statistician were blind to groupings. Results: Of the 200 children equally randomized in the two arms, 185 who were seronegative before vaccination (88 in the EZ arm and 97 in the AIK-C arm) were entered in the final analysis. The seroconversion rate in the EZ arm was 76.1% (95% CI: 60.2–85.2%), and that in the AIK-C arm was 58.7%; (95% CI: 48.8– 68.7%). The absolute rate difference was 17. 4% (4.1–30.9%; P-value: .012), and the relative seroconversion rate of EZ to AIK-C was 1.3 (95% CI: 1.1–1.6; P-value: .012). No adverse events were reported during the study period. Conclusion: A considerable difference in the seropositivity of different measles containing vaccines could be demonstrated in the first year of life. Trial Registration: Iranian Registry of Clinical Trials Registration Number: IRCT2016032827144N1; May 10, 2016 (www.who.int/ictrp/network/irct/en/) Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Measles as one of the most contagious viral diseases, with the largest basic reproductive number ever known (about 20) and having a number of different vaccines produced by different laboratories around the world, remains an important preventable ⇑ Corresponding author at: P.O. Box: 98155-759, Zahedan, Iran. E-mail addresses: [email protected] (S. Izadi), [email protected] (S.M. Zahraei), [email protected] (M. Salehi), memohammadi@yahoo. com (M. Mohammadi), [email protected] (S.M. Tabatabaei), mokhtari@ hotmail.com (T. Mokhtari-Azad).
childhood disease. Most measles vaccines originate from the Edmonston strain of measles virus, including the EZ1 strain and the AIK-C2 strains that are in common use in Iran’s immunization program [1,2]. The first one is the product of Serum Institute of India (SII), and the second one is produced domestically in Razi Institute of Serum and Vaccine Production (RIS). AIK-C strain was the first live attenuated measles virus strain whose complete genomic sequence was published and the Edmonston-Zagreb strain is the most 1
Edmonston-Zagreb. A: America, I: Iran, K: The Kitasato Institute, and C: virus adapted to chick-embryo cells. 2
https://doi.org/10.1016/j.vaccine.2017.12.048 0264-410X/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
2
S. Izadi et al. / Vaccine xxx (2017) xxx–xxx
commonly used measles vaccine in the immunization programmes of the WHO. Both these virus strains have been developed from the Edmonston strain [3]. Since 2009, measles outbreaks have occurred in a gradually increasing trend in several parts of Iran [4,5]. Within the past decade, the measles elimination campaign in Iran has witnessed different levels of performance. A number of outbreaks were prevalent in the period of 2009–2014, and microcirculations were undermining the faith in the effectiveness of the immunization activities against measles [5–10]. To constrain an outbreak, the health system sometimes uses the imported measles vaccine SII and sometimes the products of RIS. In these activities, the target population is usually composed of a wide age range, from 6 months to 12 years [5,6,9,11]. The main objective of the present study was to compare the effectiveness of both types of vaccines, which are in common use in outbreak response immunization activities in Iran. The effectiveness of these vaccines has been evaluated in infants under 9 months of age in several studies; however, for the older age groups up to 12 months of age, the health authorities are seeking the best choice between the two vaccine types [12–14]. Hence, a randomized clinical trial was implemented in Zahedan, the capital city of Sistan-va-Baluchestan Province of Iran. 2. Methods 2.1. Trial design A multicentre double blind randomized non inferiority equally balanced parallel controlled trial, comparing measles vaccine strain AIK-C with EZ strain in 8-to-12-month children, was implemented in four Urban Health Clinics in Zahedan, the capital city of Sistan-va-Baluchestan Province of Iran, during spring and summer 2016. 2.2. Participants All infants aged between 8 and 12 months being brought to the health clinics involved in the study for immunizing purposes were examined; and if eligible, they were invited for participating in the study. The inclusion criteria for the participants were: (a) being in the age range 8–12 months (240 days to 12 months and 30 days); (b) having no contraindication for receiving measles containing vaccines; (c) negative history of receiving any kind of measles containing vaccine; (d) feasibility of blood sampling from the physical and medical points of view; (e) being permanent residents of Zahedan City; (f) being Iranian; Pakistani, Afghan refugees or travellers from other districts were not invited to be included in the study.
the same health centre both for treatment and for the event to be recorded in the study checklists. After blood sampling, each participant’s allocation packet was opened, and vaccination with the defined vaccine was done. The packet and the allocation paper were pinned as attachments to the participant’s questionnaire. 2.4. Vaccines characteristics The participants in both arms of the study were vaccinated with a bivalent measles/rubella vaccine (that is in common use in outbreak containment activities in Iran). The vaccines used in this study had either been produced domestically by Razi Vaccine and Serum Research Institute (Batch Numbers: 01694016 and 01594026) or were the products of Serum Institute of India (Batch Numbers: 012N5011A and 012N5013) imported from India. The measles component of the domestically produced vaccine was an attenuated AIK-C strain of live measles virus propagated in human diploid cells, known as AIC-HDC strain, while the virus strain used in the imported vaccine was the Edmonston-Zagreb (EZ) strain, also produced in human diploid cells. The vaccines were injected subcutaneously in anterolateral aspect of the thigh, using standard needles 16 mm in length. The imported vaccines were packed in five-dose vials and the domestically produced vaccines in 10-dose vials. The vaccines used in the study were brought from the vaccine depot of the Ministry of Health from Tehran with the same vaccine carrier vehicle, and they were stored in the same vaccine storage system in the Province Health Centre. The vaccines were distributed to the four cooperating health centres using the same vaccine carriers by the same health staff on the first day of the study. Cold chain in the health clinic were checked by both the study supervisor and the physician in charge of each health clinic. 2.5. Outcomes The primary outcome measure was the detection of IgG-antimeasles antibody 60 days after vaccine inoculation, using ELISA technique. 2.6. Sample size calculations Regarding a reported seroconversion rate of 80–95% (effect size equal to 0.15) and considering type-one error equal to 0.05 and type-two error equal to 0.20, the sample size for each arm of the study was calculated as 76 persons. Considering an estimated lost to follow-up rate of about 25%, the sample size for each arm was increased to 100, which was equally distributed among the four Urban Health Clinics involved in the study. 2.7. Randomization
2.3. Interventions After an interview in the local language by a health expert about the study and its objectives, if the infant’s parent or guardian consented for their child to be included in the study, they were asked to sign a written consent form. A questionnaire containing questions about participants’ demographic characteristics and history of previous vaccinations and exanthematic diseases were filled out for them. Questionnaire comprised participants’ birth weight, mother’s age, age at the beginning of supplementary food, mother and father’s level of education, and anthropometric measurements, as well as their full contact information including phone numbers and addresses. Blood sample was taken before vaccination and 60 days after vaccination. The parents were instructed about the possible adverse effects of the vaccines. They were also advised that upon occurrence of any unexpected events, they should return to
After obtaining the informed consent (as described above), filling the questionnaire and doing the first blood sampling, the participants were randomly assigned, in a 1:1 ratio in order to receive either the AIK-C vaccine or the one containing EZ strain. Randomization was balanced with stratification according to the four health clinics. The names of the vaccine producers had been printed on small papers, placed in opaque packets and arranged based on computer generated permutation blocks of 8, 6 and 4 (produced by the first author). In order to prevent any disarrangement of the packets, in addition to printing the row number of the packets on them, they were threaded using a strong string; and in order to prevent reading the contents of the packets by keeping them against a strong light source, a layer of aluminium foil was placed inside each packet. On the participants’ blood and sera tubes, only the row number and the participants’ names had been
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
S. Izadi et al. / Vaccine xxx (2017) xxx–xxx
printed. Participants, lab technicians, and the study statistician were unaware of the study group assignments until the final report had been completed. However, vaccinators, opening the allocation packets, reading the lot papers, and inoculating the vaccines (all done by the same person), were not blind to the allocations. A special identification card was given to each participant’s parent or guardian (to remind them that they have to return to the same health clinic 60 days later). All of the participants’ exact contact details were both recorded in their questionnaires and booked in another document in order to recall them for the second blood sampling on the 60th day after the vaccination. All of the participants were contacted 60 days after vaccine inoculation and were invited to the same health clinic for doing the second blood sampling meant to check for seroconversion. For those participants who did not return on time, a health worker went to their addresses to ask them in person for coming to the clinic. The health workers cooperating in this study were experienced university-educated vaccinators adept in the local language, and all of them took part in two introductory training sessions two days before the beginning of the study. All the anthropometric measurements were done according to WHO recommendations and using high quality well-calibrated equipment [15]. The participants’ recumbent lengths were measured in supine position with an accuracy of half a centimetre. Their weights were measured with mechanical baby scales (SecaTM) with an accuracy of 10 g. Their ages were calculated by subtracting their birth dates from the date of vaccine inoculation. Anthropometric indices (Height for Age Zscore (HAZ), Weight for Age Z-score (WAZ), and Body Mass Index
3
for Age Z-score (BAZ)) were calculated using ‘WHO AnthroPlus’ software (a type of software for the global application of the WHO reference 2007; Department of Nutrition, World Health Organization; Geneva, Switzerland). Stunting and wasting were defined respectively as HAZ and BAZ equal to or below 2 standard deviations below the reference median value. From May 14th to May 18th, 2016, a total number of 200 participants, i.e. infants aged between 8 and 12 months, were recruited and blood sampled before the inoculation of the MR (MeaslesRubella) vaccine. All of the participants were invited for the second blood sampling 60 days later from July 14th to July 19th, 2016. There were 12 lost to follow-up participants (2 in the RIS arm and 10 in the SII arm) mostly due to parents’ refusal to bring their children for the second blood sampling. In total, 6 of the participants were categorized as seropositive (OD 0.1) based on the results of their blood samples taken before the vaccination. Three of these participants were among the twelve participants who did not take part in the second blood sampling on day 60. Out of the remaining three seropositive participants, one was in the RIS arm of the study and the other two were in the SII arm. Even though all of the three had increases in the OD levels of their second blood samples (which were taken 60 days after the vaccination), they were excluded from the analysis. This way, the total number of participants who were seronegative before the vaccination and whose second blood samples, 60 days after the vaccination, were considered appropriate for analysis added up to 185: i.e. 97 in the RIS arm of the study and 88 in the SII arm. Fig. 1 shows the flowchart of selecting and following the participants in the study.
Fig. 1. Flow of selecting and following the participants in the study.
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
4
S. Izadi et al. / Vaccine xxx (2017) xxx–xxx
2.8. Laboratory procedures The sera were separated from the blood samples in the same health clinic within an hour after blood sampling; and they were kept at 20° centigrade until delivery to the National Reference Measles Laboratory in the School of Public Health, Tehran University of Medical Sciences, Tehran, Iran, for the measurement of antimeasles IgG level. The mentioned laboratory is part of the WHO’s network of laboratories. In order to quantify the presence of IgG antibodies, indirect ELISA tests were used according to the manufacturer’s instructions (EnzygnostÒ Anti-Measles Virus/IgG; Siemens, Marburg, Germany). The sera of each phase of blood sampling were sent to the National Reference Measles Laboratory separately, but the laboratory did the test of both phases (i.e. the sera taken before vaccination and the sera taken 60 days later) at the same time after receiving all of the sera. Based on the manufacturer’s descriptions with Enzygnost Anti-Measles Virus/IgG, the samples containing approximately 150 mIU/mL were found to be within the range of 0.100–0.200 DA. Specimens below 0.100 were considered as negative [7,10].
Table 1 The results of comparing some of the participants’ characteristics by the vaccine inoculated for them. AIK-C arm Number Mean (Standard Deviation)
EZ* arm Number Mean (Standard Deviation)
P-value**
Birth weight (g)
97 3513.8 (2005.0)
88 3548.0 (1962.8)
0.907
Mothers’ age (years)
97 25.8 (6.7)
88 25.0 (5.4)
0.394
Age at beginning of supplementary food (months)
97 6.1 (0.8)
88 6.2 (0.8)
0.279
Participants’ age
97 10.5 (1.1)
88 10.4 (1.1)
0.387
Fathers’ education level
97 6.0 (4.1)
88 5.7 (4.0)
0.581
Mothers’ education level
97 4.1 (4.0)
88 4.8 (4.0)
0.269
Body mass index for age Z-score
96 0.16 (1.0)
87 0.21 (1.2)
0.726
Height for age Z-score
96 0.9 (1.2)
87 0.7 (1.2)
0.317
Weight for Age Z-score
97 0.6 (0.9)
87 0.6 (1.1)
0.689***
2.9. Ethics The protocol of the study has been registered in Iranian Registry of Clinical Trials (IRCT), which is under supervision of World Health Organization (IRCT Registration Number: IRCT2016032827144N1; Date Registered: May 10, 2016). In addition, The Ethics Committee on Medical Research of Zahedan University of Medical Sciences has reviewed and approved the implementation of the study (Reference Number: IR.ZAUMS.REC.1395.26; Date Approved: April 25, 2016) (www.who.int/ictrp/network/irct/en/). *
2.10. Statistical methods
** ***
The collected data were transferred to a computerized data bank, and they were analysed using Stata (ver. 11.2). In the analysis of the data, the design of the study was brought into consideration by using the ‘‘svyset” command of Stata., i.e. the health clinics (where the randomizations and blood samplings were done) were defined as strata, and primary sampling units (i.e. participants) were defined within the health clinics. The data were analysed using independent sample t-test. In all instances, the 95% confidence intervals were calculated; and where applicable, the statistics were reported as ‘mean ± standard deviation’. 3. Results The mean follow-up time between the vaccination and the second blood sampling was 60.9 days (minimum 60 days and maximum 62 days). During this period, no adverse events were reported in any participants in the two arms of the study. Table 1 shows some of the characteristics of the participants. Total seroconversion rate was 67.0% (95% CI: 60.2–73.9%). The seroconversion rate in the SII arm of the study was 76.1% (95% CI: 60.2–85.2%), and that in the RIS arm was 58.7%; (95% CI: 48.8–68.7%). The absolute rate difference was 17. 4% (4.1–30.9%; P-value: .012), and the relative seroconversion rate of SII to RIS was 1.3 (95% CI: 1.1–1.6; P-value .012) (Table 2). For each one of the vaccine types used in the study, two vaccine batch with different batch numbers were in use. Since there was no insight in any probability of any difference between the different production series of the same vaccine within the same vaccine type, reasonably, no adjustment or control for balancing the number of vaccine for each serial number was made. However, after the
Edmonston Zagreb strain. Independent sample t-test. Unequal variances considered.
completion of the laboratory tests, when the two vaccine-series within each arm of the study were compared with each other (not with the other arm), surprisingly there were statistically significant differences between the capability of the two vaccineseries of each one of the vaccines in inducing antibody production. Table 2 shows the mean difference between the vaccines used in the study and within each arm of the study between the two serial numbers. The difference between the seroconversion rates of the two serial numbers used within each arm was statistically significant. In comparison of the baseline characteristics reported in Table 1 (using analysis of variance), except for the ‘Body Mass Index for Age Z-score’ (BAZ), there was no other statistically significant difference between the subgroups receiving the different vaccine batches. In pairwise comparisons (using Scheffe test), there were two statistically significant differences between the participants who had received the batch number ‘‘01594026” (AIK-C) and those who had received the vaccines with the batch numbers ‘‘01694016” (AIK-C) and ‘‘012N5013” (EZ).
4. Discussion This was in fact a head-to-head immunogenicity study of two different vaccine strains, i.e. the Edmonston-Zagreb strain (produced by SII) versus AIK-C (produced by RIS). The results of this study showed an obvious difference in the effectiveness of the two vaccines used in the study as well as a probable difference between the effectiveness of different production series of the same vaccines.
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
5
S. Izadi et al. / Vaccine xxx (2017) xxx–xxx
Table 2 Comparison of seroconversion rates between measles virus vaccine strains and between production series (marked by their batch numbers) within each one of the vaccine strains in the study. Measles vaccine strains
Comparison between arms
No.
Mean (SD)
Mean difference (CI)
P-value (independent sample t-test)
EZ* AIK-C
EZ* AIK-C Comparison within EZ* Arm 012N5011A** 012N5013** Comparison within AIK-C Arm 01694016** 01594026**
88 97
0.76 (0.43) 0.59 (0.49)
0.17 (0.04–0.31)
0.012
64 24
0.70 (0.46) 0.92 (0.28)
0.21 (0.01–0.41)
0.037
65 32
0.45 (0.50) 0.88 (0.37)
0.43 (0.23–0.62)
0.000
EZ*
AIK-C * **
Edmonston-Zagreb strain. Vaccine batch numbers.
Although the two vaccines used in this study were of two different measles virus strains, several studies have proved no significant differences between these two vaccines with regard to effectiveness and efficacy [12–14,16–19]. However, in this study, a statistically significant difference was observable. To make certain that the observed results (and seroconversion rates) have had no other reasons besides the qualities of the vaccines used in the study, the authors did their best to control all possible sources of variation, including the cold-chain practices, the skill and knowledge of health workers involved in the study, and appropriate randomization schedule implemented within each health clinic. In addition, the participants’ characteristics that could supposedly be the source of any kind of variations, including mothers’ ages, participants’ ages, feeding conditions, and presence of malnutrition, were brought into consideration and calculation; in no instances were there any statistically significant differences between the two arms of the study. Without any doubt, the presence of maternal antibodies in the first year of life might be mentioned as one of the main reasons for primary vaccination failure (i.e. failure of serologic conversion after the vaccination); and considering the common knowledge about antibody response in different age groups below 1 year of life and also considering the age range of the participants in the study (8–12 months), an antibody response of about 80% could be completely a reasonable expectation in this study [1,20,21]. What was surprising was a significant difference in the observed seroconversion rates between different serial numbers. This finding, that was found by chance, might deserve more attention in the present study. In more details, in each vaccine type used in the study, there was one vaccine serial number with seroconversion rates near or equal to 90%, which, as mentioned before, was completely acceptable and expected in the studied age range (8–12 months), while in the same vaccine type, there was another vaccine serial number with seroconversion rates lower than expectations [1,13]. In simpler words, the vaccines have not acted homogenously within each brand. Usually by default, the potency, efficacy and effectiveness of the different serial numbers of the same vaccine produced by the same vaccine producer are taken to be the same. This finding highlights the importance of paying attention to such details in future vaccine studies. As it was mentioned in the Results section, the only baseline characteristic that showed statistically significant difference between the subgroups receiving the different vaccine batches was BAZ. If this finding is a mere paly of chance or an implication of the presence of a causal relationship, it might not be found in the scope of the present study. As a last word, it should be reminded that the authors found this discrepancy in the action of the different serial numbers of the same vaccine type by mere chance and thoroughly accidentally during one of the exploratory sub-group analyses. Therefore, since this is an un-planned post hoc finding, it is quite natural that we
may not stress its importance more than a casual finding which deserves testing and consideration in future studies. 5. Conclusion It seems that there is considerable difference in the performance of different vaccine strains produced by different vaccine producers and even between different vaccine series produced by the same producer, which deserves more attention both in the selection of vaccines for future interventions and in future studies. Funding This work was supported by the Ministry of Health and Medical Education, I.R. Iran [D304/13893]. Conflict of interest All of the authors hereby declare not to have any potential conflicts of interest. Authors’ contributions SI, SMZ, MS and TMA conceived the idea; SI, SMZ, MM and TMA drafted the proposal; SI and SMZ approved the final proposal; SMT, SI and MS supervised the field stages and data collection; TMA supervised and conducted all of the laboratory work; and MM carried out data management and analysed the data through consultation with SI and SMT. All of the authors have read and approved the final manuscript. Acknowledgements Our due thanks are to the hard-working health workers, at all levels, from the managerial levels to those on the front lines of providing health services in the health facilities involved in the study. In addition, the authors would like to appreciate the hard work of all employees and staff of the National Reference Measles Laboratory in the School of Public Health, Tehran University of Medical Sciences, Tehran, Iran. The authors also wish to express their gratitude to Dr. Ali Beikian, the faculty member of English Language Department of Chabahar Maritime University, for editing the English text of this article. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.vaccine.2017.12. 048.
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048
6
S. Izadi et al. / Vaccine xxx (2017) xxx–xxx
References [1] World Health Organization. Measles vaccines: WHO position paper – April 2017. Weakly Epidemiol Rec 2017;92:205–28. [2] Makino S. Development and characteristics of Live AIK-C measles virus vaccine: a brief report. Rev Inf Dis 1983;5:504–5. [3] Bankamp B, Takeda M, Zhang Y, Xu W, Rota PA. Genetic characterization of measles vaccine strains. J Inf Dis 2011;204(Suppl. 1):S533–48. [4] Moghadam M, Afsarkazerooni P, Ebrahimi M, Soltani M, Razmpoor A, Pirasteh E, et al. Measles outbreak in South of Iran, where vaccine coverage was high: a case-series study. Iran J Pub Health 2014;43:375–80. [5] Izadi S, Zahraie SM, Sartipi M. An investigation into a measles outbreak in southeast Iran. Jpn J Inf Dis 2012;65:45–51. [6] Esteghamati A, Gouya MM, Zahraei SM, Dadras MN, Rashidi A, Mahoney F. Progress in measles and rubella elimination in Iran. Pediat Inf Dis J 2007;26:1137–41. [7] Izadi S, Mokhtari-Azad T, Zahraei SM. Measles vaccination coverage and seroprevalence of anti-measles antibody in south-east Islamic Republic of Iran. Eastern Mediterranean health journal = La revue de sante de la Mediterranee orientale = al-Majallah al-sihhiyah li-sharq al-mutawassit. 2015;21:396–402. [8] Karimi A, Arjomandi A, Alborzi A, Rasouli M, Kadivar MR, Obood B, et al. Prevalence of measles antibody in children of different ages in Shiraz, Islamic Republic of Iran. Eastern Mediterranean health journal = La revue de sante de la Mediterranee orientale = al-Majallah al-sihhiyah li-sharq al-mutawassit. 2004;10:468–73. [9] Zahraei SM, Gouya MM, Azad TM, Soltanshahi R, Sabouri A, Naouri B, et al. Successful control and impending elimination of measles in the Islamic Republic of Iran. J Inf Dis 2011;204(Suppl. 1):S305–11. [10] Zahraei SM, Izadi S, Mokhtari-Azad T. Factors affecting the seroconversion rate of 12-month-old babies after the first injection of measles vaccine in the southeast of Iran. Hum Vaccines Immunother 2016;12:3118–24. [11] Personnaz D, Larson H. Iran and Turkey launch world’s largest measles campaigns. Press release. Geneva: Unicef; 2003.
[12] Mirchamsy H, Nilforoushan MA, Shafyi A, Razavi J, Ashtiani MP, Youssofi I, et al. Comparative evaluation of two combined measles-mumps-rubella vaccines based on AIK and Edmonston- Zagreb strains of measles virus. Kitasato Arch Exp Med 1991;64:141–7. [13] Tidjani O, Grunitsky B, Guerin N, Levy-Bruhl D, Lecam N, Xuereff C, et al. Serological effects of Edmonston-Zagreb, Schwarz, and AIK-C measles vaccine strains given at ages 4–5 or 8–10 months. Lancet 1989;2:1357–60. [14] Cutts FT, Grabowsky M, Markowitz LE. The effect of dose and strain of live attenuated measles vaccines on serological responses in young infants. Biol: J Int Assoc Biol Stand 1995;23:95–106. [15] WHO, UNICEF. WHO child growth standards and the identification of sever acute malnutrition in infants and children. Geneva, Switzerland: WHO and UNICEF; 2009. p. 12. [16] Mirchamsy H, Shafyi A, Bahrami S, Kamali M, Nazari P, Razavi F, et al. A comparative field trial of five measles vaccines produced in human diploid cell, MRC-5. J Biol Stand 1977;5:1–18. [17] Mirchamsy H, Shafyi A, Nazari P, Ashtiani MP, Sassani A. Evaluation of live attenuated measles vaccines prepared in human diploid cells for reimmunization. Epidemiol Inf 1988;101:437–43. [18] Rasmussen SM, Biering-Sorensen S, Byberg S, Andersen A, BjerregaardAndersen M, Rodrigues A, et al. The effect of early measles vaccination at 4.5 months of age on growth at 9 and 24 months of age in a randomized trial in Guinea-Bissau. BMC Pediat 2016;16:199. [19] Sasaki K. Studies on the modification of the live AIK measles vaccine. I. Adaptation of the further attenuated AIK measles virus (the strain AIK-L33) to chick embryo cells. Kitasato Arch Exp Med 1974;47:1–12. [20] Hoest C, Seidman JC, Pan W, Ambikapathi R, Kang G, Kosek M, et al. Evaluating associations between vaccine response and malnutrition, gut function, and enteric infections in the MAL-ED cohort study: methods and challenges. Clin Inf Dis: Off Pub Inf Dis Soc Am 2014;59(Suppl. 4):S273–9. [21] Pannuti CS, Morello RJ, de Moraes JC, Curti SP, Afonso AMS, Camargo MCC, et al. Identification of primary and secondary measles vaccine failures by measurement of immunoglobulin G avidity in measles cases during the 1997 São Paulo epidemic. Clin Diagn Lab Immunol 2004;11:119–22.
Please cite this article in press as: Izadi S et al. Head-to-head immunogenicity comparison of Edmonston-Zagreb vs. AIK-C measles vaccine strains in infants aged 8–12 months: A randomized clinical trial. Vaccine (2017), https://doi.org/10.1016/j.vaccine.2017.12.048