Characteristic age distribution of Plasmodium vivax infections after malaria elimination on Aneityum Island, Vanuatu

Abstract

Resurgence is a major concern after malaria elimination. After the initiation of the elimination program on Aneityum Island in 1991, microscopy showed that Plasmodium falciparum disappeared immediately, whereas P. vivax disappeared from 1996 onward, until P. vivax cases were reported in January 2002. By conducting malariometric surveys of the entire population of Aneityum, we investigated the age distribution of individuals with parasites during this epidemic in the context of antimalarial antibody levels and parasite antigen diversity. In July 2002, P. vivax infections were detected by microscopy in 22/759 individuals: 20/298 born after the beginning of the elimination program in 1991, 2/126 born between 1982 and 1991, and none of 335 born before 1982. PCR increased the number of infections detected to 77, distributed among all age groups. Prevalences were 12.1%, 16.7%, and 6.0%, respectively (P < 0.001). In November, a similar age pattern was found, but with fewer infections: 6/746 and 39/741 individuals were found to be infected by microscopy and PCR, respectively. The frequencies of antibody responses to P. vivax were significantly higher in individuals born before 1991 than in younger age groups and were similar to those on Malakula Island, an area of endemicity. Remarkably low antigen diversity (h, 0.15) of P. vivax infections was observed on Aneityum compared with the other islands (h, 0.89 to 1.0). A P. vivax resurgence was observed among children and teenagers on Aneityum, an age distribution similar to those before elimination and on islands where P. vivax is endemic, suggesting that in the absence of significant exposure, immunity may persist, limiting infection levels in adults. The limited parasite gene pool on islands may contribute to this protection.

FIG 1

(a and b) Age-specific prevalence profiles for P. vivax infections in July (a) and November (b) 2002. Red bars represent the numbers of P. vivax infections detected by both microscopy and PCR, and blue bars represent the numbers detected only by PCR. All microscope-positive individuals were also PCR positive. (c) IgG antibodies to P. vivax antigens in 1998 on residents of Aneityum Island, where P. falciparum malaria transmission had been interrupted since 1991 and P. vivax malaria transmission had been interrupted since 1996. Blue bars represent seropositivity rates for antibodies to P. vivax erythrocyte-stage antigens, and orange bars represent those for antibodies to recombinant P. vivax CSPs, either VK210 or VK247.

FIG 2

Seroprevalence curves of IgG antibodies to schizont extracts for Malakula Island (an area of mesoendemicity), Futuna Island (with no endemicity), and Aneityum Island (where an integrated elimination program was implemented in 1991) in Vanuatu. Results from Malakula (1998) (a and b), Futuna (1997) (c and d), and Aneityum (1998) (e and f) for P. falciparum (a, c, and e) and P. vivax (b, d, and f) are shown. In each plot, the red triangles represent observed data points (divided into deciles), and the blue lines represent the predicted value from the maximum-likelihood model. For Aneityum, a model with 2 forces of infection was plotted, with a change set at 15 years before the survey. Seroconversion rates for both P. falciparum (SCR, 0.08 [95% CI, 0.07 to 0.10]) and P. vivax (SCR, 0.040 [CI, 0.035 to 0.050]) on Malakula are significantly higher than those on Futuna (P. falciparum SCR, 0.002 [CI, 0.000 to 0.004]; P. vivax SCR, 0.005 [CI, 0.000 to 0.008]). On Aneityum, current SCRs for P. falciparum (0.006 [CI, 0.003 to 0.010]) and P. vivax (0.002 [CI, 0.000 to 0.040]) are 10- to 20-fold lower than preelimination levels (P. falciparum SCR, 0.04 [CI, 0.03 to 0.06]; P. vivax SCR, 0.030 [CI, 0.020 to 0.035]).

FIG 3

Distribution of Pvmsp1 and Pvcsp haplotypes among P. vivax parasites in the Vanuatu archipelago. Shown are results for Aneityum Island during the outbreak in 2002, a decade after the beginning of the malaria elimination program (n, 27 and 25 for Pvmsp1 and Pvcsp, respectively), and for other islands with malaria transmission: Gaua (n, 33 and 20) (1997), Santo (n, 34 and 32) (1996, 1997, and 2001), Ambae (n, 22 and 11) (2002), Malakula (n, 14 and 12) (1998 and 2001), Pentecost (n, 20 and 14) (1998 and 2000), and Tanna (n, 15 and 11) (1999 and 2002). See Tables 1 and 2 for the detailed haplotype classifications for Pvmsp1 and Pvcsp, respectively. Parasite rates detected during the case selection surveys on these islands are presented on the map. The inset map shows the location of Vanuatu in Oceania.