Comparison of seven hepatitis B virus (HBV) nucleic acid testing assays in selected samples with discrepant HBV marker results from United States blood donors.

BACKGROUND

Sensitive triplex nucleic acid tests (NATs) are implemented for blood donation screening worldwide. Assays have variable ability to detect low-level hepatitis B virus (HBV) DNA. At borderline DNA detection levels, where Poisson distribution impacts results, distinguishing true-positive from false-positive results is challenging. Algorithms are needed to confirm such low-level HBV DNA-positive samples.

STUDY DESIGN AND

METHODS

A total of 135 blood donor samples reactive by one or more HBV markers that provided discrepant results were tested undiluted with four commercial NATs: Ultrio, Ultrio Plus, MPX, and a quantitative assay (SuperQuant). To further explore discrepancies, three additional in-house NATs including real-time polymerase chain reaction (PCR) and nested PCR and sequencing were performed.

RESULTS

The numbers reactive of these 135 "difficult" samples by four commercial NATs were as follows: 39 of 107 (36%) with SuperQuant, 40 (30%) with Ultrio, 100 (74%) with Ultrio Plus, and 102 (76%) with MPX. Of the seven NATs, 109 (81%) samples were reactive by at least two assays and thus considered confirmed positive of which 67 (50%) generated a sequence. Ultrio Plus and MPX performed similarly as above (80%-85% detected of 109 and 81%-90% of 67, respectively). Older (median, 49 years), HBV core antibody-reactive donors carried predominantly Genotype A (58%) with high-frequency amino acid substitutions in the major hydrophilic region of the S-protein. Younger (median, 24 years) hepatitis B surface antigen-positive donors carried wild-type strains predominantly Genotype B (32%) and E (24%), the latter in an apparent cluster.

CONCLUSIONS

Highly sensitive NATs require new confirmatory algorithms as presented optimally using different genomic regions or sequence generation. The introduction of immigration-related HBV genotypes may impact HBV epidemiology in the United States.