Optimisation of dengue diagnostic tools in order to increase the knowledge of the pathogenesis

Abstract

Dengue fever (DF) is the most common global viral mosquito-borne infection, with 100 million estimated cases annually in tropical and subtropical areas. The dengue viruses (genus Flavivirus, Flaviviridae) occur in 4 serotypes (DENV-1 to DENV-4). Dengue diseases range from a mild febrile disease to severe hemorrhagic fever. Infection with one serotype induces a life-long immunity, but does not elicit cross-protective antibodies to the other serotypes. Re-infection with a second serotype is associated with a more severe disease and is also a risk factor for dengue hemorrhagic fever. Each year the Swedish Institute for Infectious Disease and Control receives serum samples from several hundred Swedish travellers, with a suspected acute or past DF after travelling to dengue endemic areas.

The aims of this thesis were to introduce methods suitable for the different phases of viremia and antibody development in the early and the late phases of disease, respectively and to determine the optimal methods in relation to the onset and sampling dates.

In paper I, dengue IgG immunofluorescence assay (IFA) negative acute serum samples from 57 previously defined Swedish dengue patients (1997-2002), were investigated by different PCR assays and by dengue IgM ELISA. Only samples collected until day 5 post onset were found positive by PCR: in 73% (35/48) of the samples, dengue RNA of serotypes 1, 2 or 3 was detected. The number of genomes/ml varied between 103 and 108, with a gradual decline over time. Dengue-specific IgM antibodies were found in 35% (20/57) of the samples. By a combination of the PCR assays and the IgM ELISA, a dengue diagnoses could be determined in as many as 84% (48/57) of early single samples. When the analyses were consecutively performed on the samples from 2002, 100% (13/13) of the samples were positive either by PCR or by IgM ELISA.

In paper II, a dengue micro-NT (m-NT) for detection and serotyping of neutralising antibodies was developed and evaluated. Early convalescent samples (<6 weeks), complemented by late convalescent samples (>5 years) from 20 patients, previously serotyped by PCR were included in the study. The correct serotype was determined in 80% (16/20) of the late convalescent samples, while the serotype could not be determined in 4 patients. One patient did not produce any neutralising antibodies, another patient had most probably had two dengue infections with equally high titres of neutralising antibodies against both. In two patients a significant difference between the serotypes could not be determined. We found no correlation between dengue IFA IgG titres and m-NT titres in samples collected 5-10 years post onset. We have demonstrated that the m-NT is a reliable diagnostic tool for detection and serotyping of neutralising antibodies in late convalescent serum samples of primary dengue cases.