Comparative Analysis of Long-Term Measles Immune Response After Natural Infection and Routine Vaccination in China

Abstract

Background: Given the significant impact of population immunity on the measles epidemic, understanding immunity differences among populations with varying immunity backgrounds is necessary for identifying immunity gaps and informing vaccination policies. In this study, we aimed to determine the distinct dynamics of vaccine-induced and naturally acquired antibodies, with specific focus on difference in vaccine-induced antibody responses across different birth cohorts. Methods: Based on two cohorts and one cross-sectional study conducted in Anhua County, Hunan Province, China, serum samples from children who followed China's routine measles vaccination schedule (i.e., two-dose schedule at 8/18 months) and adults who acquired immunity through natural infection were tested for measles IgG antibodies using an enzyme-linked immunosorbent assay. The generalized additive mixed model and a mechanistic model were employed to describe antibody dynamics following vaccination and infections. Wavelet analysis was used to investigate the temporal relationship between the measles epidemic and long-term antibody levels after natural infection. Results: A total of 408 children (0-12 years) and 222 adults (54-84 years) were included in the present study. Vaccine-induced antibody levels following 8 m/18 m vaccination were estimated to fall below the protective threshold of 200 mIU/mL by age of 15.8, whereas antibody levels following infections remained high. The decay rate of vaccine-induced antibodies was estimated at 3.0 × 10^-3 log-log mIU/mL per year, whereas naturally acquired measles antibodies persisted lifelong with a significantly lower decay rate of 2.30 × 10^-5 log-log mIU/mL per year. Moreover, vaccine-induced antibody levels in children born after 2010-a period of low measles incidence-declined more rapidly (duration of protective immunity: 12.5 years), compared to those born before 2010. Discussion: Our findings revealed immunity heterogeneity among individuals with difference measles immunity backgrounds. In particular, the birth-cohort specific differences in vaccine-induced immunity highlighted the key role of young generations born in settings with low measles incidence in contributing to population immunity gaps. This underlines that greater attention should be given to this group in future catch-up vaccination efforts.

Keywords: antibody dynamics; measles; natural infection; vaccination.

Figure 1

Selection procedure for study participants. (A) The geographic location of Anhua County, Yiyang City, Hunan Province, is displayed on the map. The density of the red color in each township represents the mean population density per 100 m². Red points in the three townships (Tianzhuang, Qingtang, and Jiangnan) indicate the locations of cohort participants, while red triangles mark the six hospitals where infant cohort recruitment was conducted. A bar plot illustrates the monthly age-specific measles incidence in Anhua from 2003 to 2021. Weekly vaccination numbers for cohort participants with available vaccination records are shown in the red rug plot at the top of the panel. The birthdays of included study children, categorized by birth cohorts, are represented in the yellow and blue rugs below. (B) Recruitment and follow-up visits for all participants of two longitudinal cohorts. (C) Selection procedure for study participants.

Figure 2

Measles antibody dynamics in study children. (A) Observed measles IgG antibody level (mIU/mL). (B) Predicted measles IgG antibody level (mIU/mL) based on the GAMM model. Box plots in (A) display the median (middle hinge), 25th and 75th quartiles (lower and upper hinges), and ± 1.5 times the interquartile range (lower and upper whiskers) of log-transformed measles IgG antibody levels. Points in (A) represent individual observed antibody levels. In (B), the solid line and orange shading indicate the mean of log-transformed antibody levels and the 95% CI predicted by GAMM. The horizontal dashed lines refer to the protective threshold of 200 mIU/mL. Numbers below the axes indicate the number of children in each age group and the sample size used to derive the CI.

Figure 3

Measles antibody dynamics in children born after 2010 receiving routine vaccination schedule. (A) Observed measles IgG antibody level (mIU/mL). (B) Predicted measles IgG antibody level (mIU/mL) based on the GAMM model. Box plots in (A) display the median (middle hinge), 25th and 75th quartiles (lower and upper hinges), and ± 1.5 times the interquartile range (lower and upper whiskers) of log-transformed measles IgG antibody levels. Points in (A) represent individual observed antibody levels. In (B), the solid line and orange shading indicate the mean of log-transformed antibody levels and the 95% CI predicted by GAMM. The horizontal dashed lines refer to the protective threshold of 200 mIU/mL. Numbers below the axes indicate the number of children in each age group and the sample size used to derive the CI.

Figure 4

Naturally acquired antibody levels in relation to measles epidemics. (A) Observed and predicted age-specific antibody levels by birth cohorts. Orange points and error bars are the estimated age-specific measles antibody levels. (B) Wavelet power spectrum of the measles incidence from 1953 to 1967 and observed measles antibody levels in participants born during this period. White transparent polygons indicate the cone of influence, and black lines represent the 95% confidence levels computed from 1000 bootstrapped series. (C) Phase angle for the reconstructed signal, restricted to 3 years.