Travel-associated international spread of Oropouche virus beyond the Amazon

Abstract

Oropouche virus (OROV), first detected in Trinidad and Tobago in 1955, was historically confined to the Brazilian Amazon Basin. However, since late 2022, an increasing number of OROV cases have been reported across various regions of Brazil as well as in urban centers in Bolivia, Ecuador, Guyana, Colombia, Cuba, Panama, and Peru. In collaboration with Central Public Health Laboratories across Brazil, we integrated epidemiological metadata with genomic analyses from recent cases, generating 133 whole-genome sequences covering the virus's three genomic segments (L, M, and S). These include the first genomes from regions outside the Amazon and from the first recorded fatal cases. Phylogenetic analyses show that the 2024 OROV genomes form a monophyletic group with sequences from the Amazon Basin sampled since 2022, revealing a rapid north-to-south viral movement into historically non-endemic areas. We identified 21 reassortment events, though it remains unclear whether these genomic changes have facilitated viral adaptation to local ecological conditions or contributed to phenotypic traits of public health significance. Our findings demonstrate how OROV has evolved through reassortment and spread rapidly across multiple states in Brazil, leading to the largest outbreak ever recorded outside the Amazon and the first confirmed fatalities. Additionally, by analysing travel-related cases, we provide the first insights into the international spread of OROV beyond Brazil, further highlighting the role of human mobility in its dissemination. The virus's recent rapid geographic expansion and the emergence of severe cases emphasize the urgent need for enhanced surveillance across the Americas. In the absence of significant human population changes over the past two years, factors such as viral adaptation, deforestation, and climate shifts-either individually or in combination-may have facilitated the spread of OROV beyond the Amazon Basin through both local and travel-associated transmission.

Keywords: Amazon basin; Brazil; Orov; genomic surveillance.

Figure 1

Distribution and Epidemiological Insights of OROV Clinical Cases Detected Beyond the Amazon Basin. (a) Weekly notified OROV PCR tests and positive cases normalized per 100 K individuals per region from 2020 to 2024. The inset panel shows the positive rate (ratio of positive cases to PCR tests) over the same period, with the y-axis scaled from 0.05 to 0.1 to highlight the variations. Error bars in the inset panel represent 95% confidence intervals calculated using Wilson's method, providing statistical support for the positive rate estimates; (b) Map of Brazil showing the number of new OROV sequences by state. The map depicts the spatial distribution of newly generated OROV genomes across Brazilian states. The size and colour of the circles represent the number of genomes generated in this study, while the shading of each state indicates the total genome count. States such as Acre (AC), Rondônia (RO), and Amazonas (AM) are shown within the Amazon Basin, while Minas Gerais (MG), Bahia (BA), and Santa Catarina (SC) represent regions beyond the Amazon Basin where sequencing efforts have expanded. This highlights the broad geographic scope of the study and the increasing efforts to monitor OROV in diverse regions of Brazil; (c) Number of new OROV genomes per state obtained in this study (Santa Catarina n = 3; Acre n = 6, Mato Grosso n = 9; Bahia n = 34; Minas Gerais n = 81); (d) Number of genomes generated in this study compared to the number of Brazilian OROV sequences available on GenBank up to 31st July, 2024. Bars are months, months with 0 sampling are not shown; (e) Cumulative OROV cases in absolute (full line) and by gender (male dotted, female dashed) of samples from 2024 for which gender metadata was available; the inner panel shows the male gender ratio (M/N) per date (grey bars are the Wilson 95% confidence interval); (f–g) Observed (non-filled bars), background (yellow bars) and theoretical (lines) age-distributions for males (f) and females (g) of samples from 2024 for which gender metadata was available (males N = 63, female N = 70). Fitted theoretical distributions were Weibull (male: shape 2.11, scale 44.39, mean 39.15; female: shape 2.07, scale 44.22, mean 39.16; determined by fitting a wide range of possible distributions using the R-package MASS). Background distributions per gender extracted from IBGE census projection for 2024.

Figure 2

Molecular Evolution and Demographic History of OROV Segments S, M, and L. (a–c) Maximum likelihood phylogenetic trees of the three OROV segments: S (n = 376), M (n = 231) (a), L (n = 303). Tips are colour-coded according to the legend in the left corner; Inner plots indicate the effective population size (i.e. genetic diversity) of OROV infections (Log scale) over time estimated under the coalescent-based Bayesian Skygrid (BSKG) model (posterior median = solid lines, 95% HPD = pale areas) for each segment; (d–f) Regression of sequence sampling dates against root-to-tip genetic distances in a maximum likelihood phylogeny of the Brazilian 2022–2024 expansion clade (n = 254).

Figure 3

Inferred Viral Dissemination Patterns of OROV in Brazil and Globally. (a–c) Phylogeographic reconstruction of the spread of OROV (segments S, M, and L) in Brazil. Circles represent nodes of the maximum clade credibility phylogeny and are coloured according to their inferred time of occurrence. Shaded areas represent the 80% highest posterior density interval, depicting the uncertainty of the phylogeographic estimates for each node. Solid curved lines denote the links between nodes and the directionality of movement; (d) Dissemination patterns of OROV within the Americas and Europe, inferred from ancestral-state reconstructions and annotated by region. Destination countries of viral exchange routes are marked with dots, while curved lines represent movement from the country of origin to the destination in a counterclockwise direction. Dashed lines indicate probable dispersal routes inferred in the absence of genomic evidence, particularly for Cuba and Colombia, based on available travel data. Countries that have reported OROV cases but lack genomic data (Ecuador, Bolivia, and Guyana) are highlighted in light grey to provide epidemiological context.