Immune response to measles vaccine in Peruvian children

Norma L. Bautista-López,¹ Abraham Vaisberg,² Rosa Kanashiro,³ Herminio Hernández,⁴ & Brian J. Ward⁵

OBJECTIVE: To evaluate the immune response in Peruvian children following measles vaccination.
METHODS: Fifty-five Peruvian children received Schwarz measles vaccine (about 10³ plaque forming units) at about 9 months of age. Blood samples were taken before vaccination, then twice after vaccination: one sample at between 1 and 4 weeks after vaccination and the final sample 3 months post vaccination for evaluation of immune cell phenotype and lymphoproliferative responses to measles and non-measles antigens. Measles-specific antibodies were measured by plaque reduction neutralization.
FINDINGS: The humoral response developed rapidly after vaccination; only 4 of the 55 children (7%) had plaque reduction neutralization titres + T-cells.
CONCLUSION: The 55 Peruvian children had excellent antibody responses after measles vaccination, but only 23% (8 out of 35) generated detectable lymphoproliferative responses to measles antigens (compared with 55– 67% in children in the industrialized world). This difference may contribute to the less than uniform success of measles vaccination programmes in the developing world.

Keywords Measles vaccine/immunology; Immunity, Cellular; Antibody-producing cells; Infant; Seroepidemiologic studies; Peru (source: MeSH).

Mots clés Vaccin antimorbilleux/immunologie; Immunité cellulaire; Cellule productrice anticorps; Nourrisson; Etude séro-épidémiologique; Pérou (source: INSERM).

Palabras clave Vacuna antisarampión/ inmunología; Inmunidad celular; Células productoras de anticuerpos; Lactante; Estudios seroepidemiológicos; Perú (fuente: BIREME).

Introduction

The correlates of immunity for natural measles and measles vaccination are not fully understood. Although high titres of neutralizing antibodies can protect against infection, evidence from case reports (1, 2), investigations of outbreaks in humans (3), and animal studies (4) suggests that some individuals who fail to develop or sustain high antibody titres after measles vaccination can nonetheless be protected against natural infection. Most people with limited or transient production of antibodies after revaccination have strong measles-specific lymphoproliferative responses (5) and patients who have had bone marrow transplants and have pre-existing lymphoproliferative memory of measles antigens make few or no antibodies in response to revaccination (6). When both humoral and lymphoproliferative responses were measured after measles vaccination, similar patterns were seen in both human and non- human primates (4– 9). In the absence of maternal antibodies, virtually all subjects mount a strong humoral response (>95%), and more than half (56– 66%) have detectable lymphoproliferative responses to measles antigens (4– 9).

With few exceptions (10), detailed studies of the immune response to standard-titre measles vaccination have been conducted in the industrialized world, where these vaccines have been successful. Although measles vaccination is highly effective in children in the developing world when performed in carefully monitored settings (11– 13), vaccination programmes in the developing world have not been uniformly successful (14). Among the many factors that may contribute to this discrepancy are the strain of the vaccine, the handling and quality of the vaccine, and host factors, such as age at vaccination, maternal antibody status, nutritional status, and whether there are intercurrent infections (15– 17).

In this study, the phenotype and activation state of peripheral blood mononuclear cells (PBMCs) were examined in Peruvian children at the time of vaccination, and immune responses to measles antigens were monitored at 1– 4 weeks (cellular and humoral responses) and at 3 months (humoral only) after vaccination.

Methods

Participants

Children were recruited during routine vaccination clinics associated with the Hôpital del Rimac (affiliated to Universidad Peruana Cayetano Heredia) in San Martin de Porras, Lima, Peru. Socioeconomic conditions in this sector of Lima vary widely; it is home to middle-class professionals and newly arrived migrants. The study was approved by the ethics committees of both the Montreal General Hospital Research Institute (McGill University, Montreal, Quebec) and Universidad Peruana Cayetano Heredia (Lima, Peru).

Fifty-five children (25 boys and 30 girls) undergoing routine measles immunization with Schwarz vaccine (10^3.0– 10^3.7 plaque forming units (pfu)) at 9 months of age were studied. After informed written consent was obtained from the parent or parents, heparinized blood samples (1– 5 ml) were obtained by venepuncture immediately before vaccination (n = 55) and at a single, randomly selected time between 1 and 4 weeks after vaccination: at week 1, n = 5; week 2, n = 15; week 3, n = 17; and week 4, n = 18. A sample was collected from all children 3 months after vaccination.

Vaccination booklets were not inspected, but vaccine coverage in this community exceeds 90% for BCG (bacillus Calmette– Guérin) and at least two DPT (diphtheria– pertussis– tetanus) immunizations at 9 months of age (18).

For comparison we have included unpublished flow cytometric data for PBMCs isolated from 57 Canadian children (30 boys and 27 girls) immediately before vaccination at about 13 months of age (416 ± 2.9 days). The recruitment of these children and their pattern of humoral and cellular immune responses following vaccination with MMR (measles– mumps– rubella) II (Merck, West Point, PA) have been reported (9).

Samples

Plasma samples and PBMCs were aliquoted and stored until used in assays as previously described (5). Briefly, the PBMCs were isolated by differential density centrifugation, enumerated by trypan blue exclusion, aliquoted, and cryopreserved in liquid nitrogen at – 180 °C. The cells were subsequently batched and sent to McGill University, Montreal, Quebec, in nitrogen vapour for further analysis. Samples taken at 3 months after vaccination were left to clot overnight at 4 °C before centrifugation (600 g for 10 minutes); the supernatant was then aliquoted and frozen at – 70 °C until use.

Measurement of neutralizing antibodies

Measles-specific neutralizing antibodies were measured in the Virology Laboratory of the Universidad Peruana Cayetano Heredia (Lima, Peru) by plaque reduction neutralization as described elsewhere (12, 19). The antibody titres are reported in milli- International Units (mIU)/ml, based on extrapolation from a standard curve generated using the WHO standard antiserum (5 IU of #66/202; WHO International Laboratory for Biological Standards, South Mimms, England) that was included in each assay run. The minimum antibody concentration detectable was 33 mIU/ml. Based on data suggesting that concentrations >200 mIU/ml protect against natural infection (20, 21), children with titres ³200 mIU/ml were considered to be seropositive.

Antigen preparations

Measles antigen was prepared from a wild-type virus (CHI-1, gift of W. Bellini, Centers for Disease Control and Prevention, Atlanta, GA) as previously described (5). A clarified and filtered Vero cell lysate served as the control antigen. The other antigens used were: tetanus toxoid (gift of B. Latham, Boston, MA), diphtheria toxoid, whole pertussis antigen (gifts of R. Wittes, Connaught Laboratories, Willowdale, ON), and whole, inactivated BCG (gift of D. Radziok, Montreal, Quebec). The protein content of the antigen preparations was estimated using a modified Bradford assay (BioRad Laboratories, Hercules, CA).

Lymphoproliferation assay

Antigen-specific cell-mediated immunity was measured using a solid-phase lymphoproliferation assay as previously described (5, 9). Briefly, flat-bottomed, 96-well plates (Nunc, Roskilde, Denmark) were coated with antigens (50 ml per well at 10 mg/ml); control and test antigens were also coated on the same plates. PBMCs resuspended in RPMI 1640 media containing 5% (v/v) heat-inactivated autologous plasma were distributed (2 x 10⁵ cells per well) and cultured for 6 days before the addition of 1 mCi of [³H]thymidine (ICN, Costa Mesa, CA). DNA was harvested on glass-fibre filters 24 hours later and counts per minute (cpm) were used to calculate stimulation indices (stimulation index = cpm for antigen-stimulated wells/cpm for the controls). A score of ³3 on the stimulation index was considered to indicate an important lymphoproliferative response. In North American children lymphoproliferative responses to measles antigen are first detectable 1– 2 weeks after vaccination and are well developed by 3– 4 weeks post vaccination (9).

Direct immunofluorescence staining

Cryopreserved cells were thawed at 37 °C, washed twice in HBSS (Hank's balanced salt solution (300 g for 10 minutes)), and resuspended in 200 ml PBS (phosphate buffered saline) containing 5% (w/v) bovine serum albumin (Boehringer Mannheim, Indianapolis, IN). Identical protocols for isolation of PBMCs, cryopreservation, and flow cytometric analyses were used for both the Peruvian and Canadian samples. Aliquots were triple stained, according to the manufacturers' instructions (Becton Dickinson, San Jose, CA; Dako, Carpinteria, CA; or Pharmingen, San Diego, CA), using combinations of monoclonal mouse antibodies conjugated with FITC (fluorescein isothiocyanate), PE (phycoerythrin), or perCP (peridinin chlorophyll A); the specificities of these antibodies are listed in Table 1. After incubation for 30 minutes at 4 °C, the cells were washed in cold PBS, resuspended in 150 ml PBS with 1% paraformaldehyde (v/v), and kept at 4°C until use. Three-colour fluorescence analysis was performed with a FACScan flow cytometer (Becton Dickinson). For each sample, 15 000 events were acquired using log-amplified fluorescence and linearly amplified side-scatter and forward-scatter signals. All samples were analysed by setting appropriate side-scatter and forward-scatter gates around the lymphocyte and monocyte populations.

Statistical analysis

The data were first reviewed for distribution patterns, and skewed data were log transformed to approximate a normal distribution. Comparisons between groups were performed using Student's t test. When data remained skewed after log transformation, the Mann– Whitney U test was used. Correlations were calculated using the Spearman Rank correlation coefficient. All analyses were performed using Statview 512 software (SAS, Cary, NJ).

Results

Results are expressed as mean values ± standard error of the mean.

Humoral response: neutralizing antibody titres

Low concentrations of maternal neutralizing antibodies were detectable in 14 of the 55 samples (25%) taken before vaccination (plaque reduction neutralization (PRN) titre 36 ± 8 mIU/ml, range 8– 235). Protective concentrations of measles-specific neutralizing antibodies developed rapidly after vaccination (Table 2). For example, 11 of 15 samples (73%) obtained 2 weeks after vaccination had titres > 200 mIU (564.2 ± 196 mIU/ml). Three months after vaccination, 51 of 55 children (93%) had titres ³200 mIU/ml (1214 ± 145 mIU/ml, P

Cellular response: measles-specific lymphoproliferation

As expected, lymphoproliferative responses to measles antigens were undetectable before vaccination (stimulation index 1.1 ± 0.1) (Table 2). Significant lymphoproliferative responses to measles antigens first became apparent in the second week after vaccination (4 of 15 samples had stimulation indices ³3). By 3– 4 weeks after vaccination 8 out of 35 children (23%) had a stimulation index ³3 (3.6 ± 0.8 vs 1.0 ± 0.1 in the remaining children). Children who had maternal antibodies present at the time of vaccination had slightly lower lymphoproliferative responses at 3– 4 weeks than those without maternal antibodies (2.4 ± 0.7 vs 3.8 ± 1.0), but this difference did not reach statistical significance. There was no difference between boys and girls in lymphoproliferative responses to measles antigen.

Cellular responses to other antigens

Despite the weak cellular response to measles antigens, the Peruvian children had vigorous lymphoproliferative responses to other antigens. Roughly half of the children had significant responses (stimulation index ³3) to BCG (56% (31/55), index 9.8 ± 2) and tetanus toxoid (49% (27/55), index 11.7 ± 1.8). Fewer children had a stimulation index ³3 in response to whole pertussis antigen (29% (14/49, index 4.5 ± 1.1) and diphtheria toxoid (18% (10/55), index 2.4 ± 0.5). Although children with a measles-specific index ³3 generally had stronger lymphoproliferative responses to other antigens (Table 3), many of the children with little or no cellular response to measles antigens after vaccination had readily detectable lymphoproliferative responses to other antigens. Correlations between the cellular responses to measles and other antigens were barely significant or of borderline significance (e.g. for BCG, r = 0.24, Pr = 0.16, P = 0.1).

Relationship between humoral and cellular responses

At 3– 4 weeks after vaccination, high antibody titres and strong cell-mediated immune responses (Ab^H CMI^H) were detected in only 8 of 35 children (23%) (Table 4). Almost three-quarters of these children (25/35, 71%) had strong humoral responses without evidence of measles-specific lymphoproliferation (Ab^HCMI^L); and 2 of the 35 (6%) had neither significant humoral nor lymphoproliferative responses to measles antigens (Ab^LCMI^L). These two children remained Ab^L at 3 months after vaccination (PRN titres 125.5 ± 67.5 mIU/ml). Antibody titres tended to be lower in Ab^HCMI^H children, but this difference did not reach statistical significance (Ab^HCMI^H: 1287 ± 176 vs Ab^HCMI^L:1483 ± 355).

PBMCs and co-stimulatory molecules

With the exception of basic subset measures of PBMCs (for example, enumeration of T-cells and CD4⁺/CD8⁺ ratios), little is known about activation and co-stimulatory molecule expression on PBMCs in healthy children in the developing world. As a result, much of the data is descriptive. Basic subset proportions (40% CD4⁺ T-cells, 27% CD8⁺ T-cells, 20% B-cells, and 6% monocytes) as well as CD4⁺/ CD8⁺ ratios (1.7 ± 0.1) were similar to those reported for children in the developing world at 9– 12 months of age (22– 26). Measles vaccination had no impact on the subset distribution of PBMCs as has been reported (27) and it had little overall effect on the expression of activation and co-stimulatory molecules (data not shown).

The most striking aspect of the cytometric analysis was the marked state of activation of PBMCs at the time of vaccination. Compared with 13-month-old Canadian children at the time of vaccination, the Peruvian children had diffuse activation of CD4⁺ T-cells, CD8⁺ T-cells, B-cells, and monocytes (Fig. 1). None of the Peruvian children was seriously ill at the time of vaccination.

Compared with the Canadian children, the Peruvian children also had markedly higher expression of CD45RO on CD4⁺ T-cells (18 ± 4% vs 31 ± 2%, P + T-cells (94.8 ± 0.9% vs 96.0 ± 99%, P = 0.09) and on CD8⁺ T-cells (59.4 ± 2.3% vs 85.8 ± 1.2%, P P P P = 0.06) on monocytes. Unfortunately, antibodies for CTLA-4 (CD152), the other T-cell ligand for CD80/86, were not commercially available at the time this study was performed.

Maternal antibodies and lymphocyte
activation state

Compared with children who had no maternal antibodies at the time of vaccination, children with detectable maternal antibodies had significantly higher expression of CD45RA on CD4⁺ T-cells (91.6 ± 1.3 % vs 88.4 ± 1.3 %, PP + T-cells (9.4 ± 0.9% vs 12.6 ± 1.2%, P

PBMC markers and immune responses

The presence of a detectable lymphoproliferative response to measles antigens among the Peruvian children was strongly correlated with the expression of CD45RO on CD4⁺ T-cells (r = 0.46, P HCMI^L children, those who developed both humoral and cell-mediated immune responses (Ab^HCMI^H) had significantly greater expression of CD45RO on CD4⁺ T-cells (33.9 ± 1.8% vs 28.8 ± 2.1%, P+ T-cells was in turn associated with increased expression of a constellation of activation markers on other PBMC subsets, including CD38 (r = 0.487, P r = 0.24, P = 0.07) on CD8⁺ T-cells, CD69 on natural killer cells (r = 0.258, P = 0.06), CD80 (r = 0.226, P = 0.09) and CD23 (r = 0.504, P r = 0.351, P

Discussion

Vaccine-induced immunity in the developing world

Although a substantial number of immunological investigations after natural measles infection have been conducted in the developing world (27, 28–32), few detailed studies of vaccine-induced immunity have been performed (11, 12, 26). One of the most striking observations in this study was the relative weakness of the lymphoproliferative responses induced by vaccination in Peruvian children (23% scored ³3 on the stimulation index) compared with the reported values for North American children vaccinated at either 13 months of age (61% reached stimulation index ³3) (9) or 6 months of age (63% reached ³3) (8). However, the Peruvian children mounted excellent humoral responses to vaccination with antibody titres equal to or superior to those reported for North American children (8, 9). Although lymphoproliferation was measured at only one point after vaccination, our recent description of the kinetics of this response in Canadian children suggests that measles-antigen-specific lymphoproliferation is fully developed by 3–4 weeks after vaccination and persists at detectable concentrations in many children for at least 5–10 years (9).

This combination of strong antibody and weak lymphoproliferative responses to measles antigens has also been described 4–5 years after vaccination in children originally enrolled in trials of high-titre measles vaccines in Haiti, Senegal, and Sudan (33–35). The vaccines administered in these studies varied by strain (Moraten, Schwarz, Edmonston-Zagreb, and Biken-CAM), viral titre (10^3.7 to 10^6.2 pfu), and age at vaccination (5–24 months); this suggests that factors other than strain, dose, and age at vaccination contribute to the relatively poor cellular response to measles antigens among children in the developing world. However, other aspects of cellular responsiveness (such as antigen-specific cytokine production and cytotoxic T-lymphocyte activity) were not measured in this or most of the previous studies (8, 9, 33–35). Measles-specific cytotoxic T-lymphocyte (11) and cytokine responses (36) are certainly measurable parameters after vaccination in young children in both the developing and industrialized world. The limited lymphoproliferative responses in this study were not caused by international transport of cells or a general inability of Peruvian children to mount such responses, since the stimulation index for non-measles antigens (such as tetanus toxoid and pertussis) were comparable to those seen in North American children (9, 37).

It is likely that many factors contribute to the discrepancy between the general success of measles control efforts in the industrialized world and the ongoing struggle with the virus in the developing world. These include straightforward issues of resources and infrastructure (for example, the maintenance of the cold chain (38) and the purchase of suboptimal products (15)) but may also include more subtle limitations related to the target populations themselves, such as maternal antibody status, nutritional status, and whether there are intercurrent infections. Certainly, maternal antibody status has long been recognized as a major confounding factor in measles vaccination campaigns (39–41).

Age at immunization

The choice of 9 months of age for routine measles immunization in the developing world reflects a compromise forced by the opposing pressures of high attack rates and high mortality in children younger than 1 year of age and the risk of failure of measles vaccine in children who have high maternal antibody titres (41). In this study low titres of maternal antibodies were detected in only about 25% of the children at the time of vaccination and these titres had little impact on antibody production or lymphoproliferative responses (with the probable exception of the two children who had maternal antibodies and who had neither a cellular nor humoral response after vaccination).

There were, however, striking associations between maternal antibody status and the expression of CD45RO on CD4⁺ T-cells and, in turn, between this marker of prior antigen exposure and a broad range of activation markers on other PBMCs. Although all of the 9-month-old Peruvian children had evidence of diffuse PBMC activation compared with 13-month-old Canadian children, the Peruvian children with the highest expression of CD45RO (and the most activated PBMCs) were significantly more likely to respond to measles vaccination with a "balanced" humoral and lymphoproliferative response.

T-cell expression of the mutually exclusive isoforms of CD45R (RA, RO) is thought to reflect cumulative antigen exposure, such that 90– 95% of cord blood CD4⁺ T-cells are CD45RA⁺ while 40– 60% of adult CD4⁺ T-cells are typically CD45RO⁺ (42, 43). It is tempting to speculate how the progressive loss of maternal antibodies during the first year of life could permit a more controlled acquisition of antigen exposure (such as through infectious agents). Children with low maternal antibody titres at birth or who lose maternal antibodies rapidly might reasonably be expected to have a precocious transition to an adult pattern of CD45RO expression on CD4⁺ T-cells. It is less clear how such an early transition could influence the pattern of immune response to a live attenuated viral vaccine. Furthermore, the correlation between maternal antibody titres at the time of vaccination and the serological response is far from perfect (44–46). It is certainly possible that CD45RO expression is simply a marker for the development of a capacity to respond to any antigen with a proliferative response. In this study, children with strong lymphoproliferative responses to measles antigens had higher lymphoproliferative responses to non-measles antigens, but there was no clear relationship between CD45RO expression on CD4⁺ T cells and cellular responses to non-measles antigens.

Unfortunately, in this study a detailed clinical evaluation was not performed at the time of vaccination. Vaccination was deferred in children with serious illnesses in accordance with local practice, but infectious episodes in the recent past, mild current illnesses, and prodromal states were not identified: any of these might have been associated with evidence of activation of PBMCs. Although it is not possible to distinguish between these states, our data support the current recommendation to administer measles vaccine despite mild intercurrent infection (47); the Peruvian children with evidence of ongoing immune activation generated excellent antibody titres as well as mounting detectable lymphoproliferative responses to measles antigens.

Importance of cell-mediated immunity

The precise implications of a limited lymphoproliferative response after vaccination are not known, but evidence is accumulating that cellular immunity needs to be considered in evaluating vaccine-induced protection (1–5, 9). "Experiments of nature" (1), case reports (2), seroepidemiology, outbreak investigations (3), and animal studies (4) all suggest that the cellular response is sufficient to protect against measles and may be necessary for such protection in many circumstances. These observations and our findings suggest that investigations directed towards identifying factors that limit the capacity of children in the developing world to mount lymphoproliferative responses to vaccination may shed light on the suboptimal efficacy of measles vaccine in this setting (48). Such factors might include nutritional status, specific micronutrient status (for example, of zinc, selenium, or vitamin A), and chronic infections with immunomodulating organisms such as Epstein-Barr virus, human herpes virus 8, and malaria parasites (49–52). A more detailed understanding of the immune response to measles vaccination in children in the developing world and the factors that limit or modify this response may be critical to the success of the drive towards eradicating measles.

Acknowledgements

Financial support: this work was supported by grants from the WHO/UNDP Programme for Vaccine Development and the Mexican National Council for Science and Technology (personal support for NLB-L). Parts of this study were presented at meetings of the American Society of Tropical Medicine and Hygiene in Baltimore, MD, 16–18 December, 1996 and in San Juan, Puerto Rico, 18–22 October 1998. This work was carried out in accordance with the Helsinki Agreement as well as the guidelines for human experimentation of the United States Department of Health and Human Services and the authors' institutions.

Conflicts of interest: none declared.

Résumé

Réponse immunitaire au vaccin antirougeoleux chez des enfants péruviens

OBJECTIF: Evaluer la réponse immunitaire chez des enfants péruviens après vaccination antirougeoleuse.
MÉTHODES: Cinquante-cinq enfants péruviens ont été vaccinés contre la rougeole par le vaccin de souche Schwarz (environ 10³ unités formant plage) vers l'âge de 9 mois. Des prélèvements de sang ont été réalisés une fois avant la vaccination puis deux fois après la vaccination, la première au bout de 1 à 4 semaines et la deuxième au bout de 3 mois, pour l'évaluation du phénotype des cellules de l'immunité et de la réponse lymphoproliférative aux antigènes rougeoleux et non rougeoleux. Les anticorps spécifiques de la rougeole ont été mesurés selon la méthode de neutralisation par réduction des plages.
RÉSULTATS: La réponse humorale se développait rapidement après la vaccination, puisque seuls 4 enfants sur 55 (7 %) avaient un titre en anticorps +.
CONCLUSION: Les 55 enfants péruviens étudiés avaient une excellente réponse en anticorps à la suite de la vaccination antirougeoleuse, mais seuls 23 % (8 sur 35) avaient une réponse lymphoproliférative détectable vis-à-vis des antigènes rougeoleux (contre 55-67 % chez les enfants des pays industrialisés). Cette différence peut expliquer la réussite inégale des programmes de vaccination antirougeoleuse dans les pays en développement.

Resumen

Respuesta inmunitaria a la vacuna antisarampionosa en niños del Perú

OBJETIVO: Evaluar en niños peruanos la respuesta inmunitaria consecutiva a la vacunación contra el sarampión
MÉTODOS: Se administró la vacuna antisarampionosa Schwarz (unas 10³ unidades formadoras de placas) a 55 niños peruanos de unos 9 meses de edad. Se tomaron muestras de sangre antes de la vacunación, y otras dos veces después de la vacunación: una muestra entre 1 y 4 semanas después de la vacunación y una última muestra a los 3 meses de la vacunación, para evaluar el fenotipo de las células inmunitarias y las respuestas linfoproliferativas a los antígenos sarampionosos y no sarampionosos. Se midieron los anticuerpos específicos para el sarampión mediante neutralización por reducción del número de placas.
RESULTADOS: La respuesta humoral se produjo rápidante después de la vacunación; sólo 4 de los 55 niños (7%) presentaron títulos +.
CONCLUSIÓN: La producción de anticuerpos en respuesta a la vacunación antisarampionosa fue excelente en los 55 niños peruanos, pero sólo el 23% (8 de 35) presentaron respuestas linfoproliferativas detectables a los antígenos sarampionosos (en comparación con 55-67% en los niños del mundo industrializado). Es posible que esa diferencia contribuya a explicar el irregular éxito de los programas de vacunación en el mundo en desarrollo.

References

1. Good RA, Zak SJ. Disturbances in gammaglobulin synthesis as "experiments of nature". Pediatrics, 1956, 18: 109–149.        

2. Ruckdeschel JC, Graziano KD, Mardiney MR Jr. Additional evidence that the cell-associated immune system is the primary host defence against measles (rubeola). Cellular Immunology, 1975, 17: 11–18.        

3. Samb B et al. Serologic status and measles attack rates among vaccinated and unvaccinated children in rural Senegal. Pediatric Infectious Disease Journal, 1995, 14: 203–209.        

4. Van Binnendijk RS et al. Protective immunity in macaques vaccinated with live attenuated, recombinant and subunit measles vaccines in the presence of passively acquired antibodies. Journal of Infectious Diseases, 1997, 175: 524–532.        

5. Ward BJ et al. Cellular immunity in measles vaccine failure: demonstration of measles antigen-specific lymphoproliferation despite limited serum antibody production after revaccination. Journal of Infectious Diseases, 1995, 172: 1591–1595.        

6. Pauksen K et al. Influence of the specific T cell response on seroconversion after measles vaccination in autologous bone marrow transplant patients. Bone Marrow Transplantation, 1996, 18: 969–973.        

7. Pabst HF et al. Kinetics of immunologic response to MMR vaccination. Vaccine, 1997, 15: 10–14.        

8. Gans HA et al. Deficiency of the humoral response to measles vaccine in infants immunized at age 6 months. JAMA, 1998, 280: 527–532.        

9. Bautista-López N et al. Development and durability of measles antigen-specific lymphoproliferative response after MMR vaccination. Vaccine, 2000, 18 (14): 1393–1401.        

10. Jaye A et al. Ex vivo analysis of cytotoxic T lymphocytes to measles antigens during infection and after vaccination in Gambian children. Journal of Clinical Investigation, 1998; 203: 1969–1977.        

11. Henderson RH, Keja J. Global control of vaccine-preventable diseases: how progress can be evaluated. Reviews of Infectious Disease, 1989, 11 (Suppl. 3): S644–S654.        

12. Berry S et al. Comparison of high titer Edmonston-Zagreb, Biken-CAM and Schwarz measles vaccines in Peruvian infants. Pediatric Infectious Disease Journal, 1992, 11: 822–827.        

13. Cutts FT, Markowitz LE. Successes and failures in measles control. Journal of Infectious Diseases, 1994, 170 (Suppl. 1): S32–S41.        

14. Special Program for Vaccines and Immunization, PAHO/ EPI. Confirmed measles cases, 1998. Bulletin of the World Health Organization, 1998, 14 (43): 1–2.        

15. Adu FD et al. Low seroconversion rates to measles vaccine among children in Nigeria. Bulletin of the World Health Organization, 1992, 70: 457–460.        

16. Albrech P et al. Persistence of maternal antibody in infants beyond 12 months: mechanism of measles vaccine failure. Pediatrics, 1977, 91: 715–722.        

17. Expanded Programme on Immunization. Safety of high titer measles vaccines. Weekly Epidemiological Record, 1992, 67 (48): 357–371.        

18. Hernandez H. Unpublished data.        

19. Albrecht P, Herrmann K, Burns GR. Role of virus strain in conventional and enhanced measles plaque neutralization test. Journal of Virology Methods, 1981, 3: 251–260.        

20. Markowitz LE et al. Duration of live measles vaccine-induced immunity. Pediatric Infectious Disease Journal, 1990, 9: 101–110.        

21. Chen R at al. Protective measles antibody titers in college students. Paper presented at the 35th Annual Epidemic Intelligence Service Conference, Atlanta, GA, 14–18 April 1986.        

22. Lisse IM et al. Evaluation of T cell subsets by an immunocytochemical method compared to flow cytometry in four countries. Scandinavian Journal of Immunology, 1997, 45: 637–644.        

23. Lisse IM et al. T-lymphocyte subsets in West African children: impact of age, sex and season. Journal of Pediatrics, 1997, 130 (1): 77–85.        

24. Griffin DE, Ward BJ. Differential CD4+ T cell activation in measles. Journal of Infectious Diseases, 1993, 168: 275–281.        

25. Ward BJ, Griffin DE. Changes in cytokine production after measles virus vaccination: predominant production of IL-4 suggests induction of a TH2 response. Clinical Immunology and Immunopathology, 1993, 67: 171–177.        

26. León ME et al. Immunologic parameters 2 years after high-titer measles immunization in Peruvian children. Journal of Infectious Diseases, 1993, 168: 1097–1104.        

27. Hussey GD et al. The effect of Edmonston-Zagreb and Schwarz measles vaccines on immune responses in infants. Journal of Infectious Diseases, 1996, 173: 1320–1326.        

28. Murley DC. Measles in the developing world. Proceedings of the Society for Experimental Biology and Medicine, 1974, 67: 1112–1115.        

29. Hull HF, Williams PJ, Oldfield F. Measles mortality and vaccine efficacy in rural West Africa. Lancet, 1985, 1: 972–975.        

30. Shaheen SO et al. Cell mediated immunity after measles in Guinea-Bissau: historical cohort study. BMJ, 1996, 313: 969–974.        

31. Griffin DE et al. Peripheral blood mononuclear cells during natural measles virus infection: cell surface phenotypes and evidence for activation. Clinical Immunology and Immunopathology, 1986, 40 (2): 305–312.        

32. Aaby P et al. No persistent T lymphocyte immunosuppression or increased mortality after measles infection: a community study from Guinea-Bissau. Pediatric Infectious Disease Journal, 1996, 15 (1): 39–44.        

33. Bertley FMN et al. Long-term follow-up of Haitian high titre measles vaccine recipients: no evidence of immunologic damage 4–5 years after vaccination (submitted for publication, 2001).        

34. Aaby P et al. Five year follow-up of morbidity and mortality among recipients of high-titre measles vaccines in Senegal. Vaccine, 1996, 14: 226–229.        

35. Libman MD et al. No evidence of short- or long-term morbidity after high titer measles vaccination in the Sudan (submitted for publication, 2000).        

36. Gans HA et al. IL-12, IFN-gamma and T cell proliferation in immunized infants. Journal of Immunology, 1999; 162: 5569– 5575.         

37. B Ward. Unpublished data.        

38. Krugman RD, Meyer BC, Parkman PD. Impotency of vaccines as a result of improper handling in clinical practice. Journal of Pediatrics, 1972, 85: 512–514.         

39. Reilly CM, Stokes J Jr, Buynak EB. Living attenuated measles- virus vaccine in early infancy. Studies of the role of passive antibody in immunization. New England Journal of Medicine, 1961, 265: 165–169.        

40. Expanded Programme on Immunization. Measles immunization before 9 months of age. Weekly Epidemiological Record, 1990, 65 (2): 8.        

41. Caceres VM et al. Factors determining prevalence of maternal antibody to measles virus through infancy: a review. Clinical Infectious Diseases, 2000; 31: 110–119.        

42. Hassan J, Reen DJ. Cord blood CD4(+) CD45 RA(+) T cells achieve a lower magnitude of activation when compared with their adult counterparts. Immunology, 1997, 90: 397–401.        

43. Suen Yu et al. Dysregulation of lymphokine production in the neonate and its impact on neonatal cell mediated immunity. Vaccine, 1998, 16 (14/15): 1369–1377.        

44. Orenstein WA et al. Appropriate age for measles vaccination in the United States. Developments in Biological Standardization, 1986, 65: 13–21.        

45. Hayden GF. Measles vaccine failure: a survey of causes and means of prevention. Clinical Pediatrics, 1979, 18: 155–167.        

46. Black FL. Measles active and passive immunity in a worldwide perspective. Progress in Medical Virology, 1989, 36: 1–33.        

47. Peter G. Measles immunization: recommendation, challenges, and more information. JAMA, 1991, 265: 211–212.        

48. Foster SO, McFarland DA, John MA. Health sector priorities review: measles. In: Jamison DT, Mosley WH, eds. Evolving health sector priorities in developing countries. Washington, DC, World Bank, 1993: 161–183.        

49. Lymphocyte function in malnutrition. Nutrition Reviews, 1975, 33 (4): 110–111.        

50. Duncan CJ, Duncan SR, Scott S. The dynamics of measles epidemics. Theoretical Population Biology, 1997, 52 (2): 155–163.        

51. Semba RD. Vitamin A as "anti-infective" therapy, 1920– 1940. Journal of Nutrition, 1999, 129 (4): 783–791.        

52. Salaman MH. Immunodepression by mammalian viruses and plasmodia. Proceedings of the Royal Society of Medicine, 1970, 63 (1): 11–15.        

¹ Graduate student, McGill University Department of Parasitology, Montreal, Quebec. Currently a postdoctoral fellow at the University of Alberta, Edmonton, Canada.

² Professor of Biology and Dean of the Faculty of Science, Universidad Peruana Cayetano Heredia, Lima, Peru.

³ Research Associate, Department of Biology, Universidad Peruana Cayetano Heredia, Lima, Peru.

⁴ Professor and Chief of Paediatrics, Department of Pediatrics, Universidad Peruana Cayetano Heredia, Lima, Peru.

⁵ Director of McGill University Division of Infectious Diseases, and Associate Professor of Medicine and Microbiology, McGill University, Montreal General Hospital Research Institute, 1650 Cedar Avenue, Montreal, Quebec H3G 1A4, Canada (email: brian.ward@mcgill.ca). Correspondence should be addressed to this author.

Ref. No. 00-0941