(P-BC-25) Development of a Chagas Disease Dried Blood Spot Assay to Address Challenges in Diagnostic Confirmation Following Positive Blood Donor Screening

Background/Case Studies: Chagas disease (CD) diagnosis requires confirmation by two distinct serologic assays. While CD screening is routine for first-time blood donation in the US, diagnostic confirmation following positive donor screening remains a challenge, especially in follow-up clinical settings. We aimed to address this challenge by developing a dried blood spot (DBS) method that leverages this alternative sampling strategy for low-cost, high-feasibility confirmatory testing or re-testing upon presentation to clinical settings. Banked specimens may also be useful for clarifying discordant testing performed after clinical follow-up.

Study

Design/Methods: Whole blood for DBS was contrived using plasma samples from 88 American Red Cross (ARC) blood donors with known CD status (n = 53 seropositive; n = 35 seronegative) and fresh CD-negative packed red blood cells (RBC) at concentrations physiologic for adults. Contrived whole blood was spotted onto Whatman 903 Protein Saver cards and allowed to dry uncovered. DBS cards were held at room temperature for up to 7 weeks before overnight extraction in Enzyme-Linked ImmunoSorbent Assay (ELISA) kit-specific diluent. DBS eluate was run head-to-head with liquid plasma on three commercially available ELISA kits for chronic CD diagnosis: Hemagen Chagas’ Kit, Wiener Chagatest Recombinante v.3.0, InBios Chagas Detect Fast. Additional iterative method development was undertaken for the highest accuracy ELISA to optimize performance. Data analysis was performed in STATA 14.2. This study was approved under ARC and UCSF IRB protocols.

Results/Findings: Commercially available ELISA kits for the serologic diagnosis of chronic CD demonstrated high feasibility and variable test performance using DBS specimens, with highest overall accuracy from InBios Chagas Detect Fast ELISA. Optimization for DBS by adjustment of the dilution factor to 1:35 (kit-recommended 1:50 for liquid serum/plasma) demonstrated 98.1% sensitivity and 100% specificity compared to blood donor testing results. At an estimated blood donor population prevalence of 0.02%, this would correspond to positive predictive and negative predictive values for CD of 100.0%.

Conclusions: Our optimized assay provided high accuracy for diagnosis of CD based on DBS contrived with plasma from a well-characterized blood donor cohort. The DBS methodology would allow index whole blood or follow-up specimens to be spotted and held at room temperature, requiring minimal storage space and cost, and reflexed for additional diagnostic purposes in the case of discordant confirmatory or follow-up results. Additionally, this low-cost, low sample volume method would be highly amenable to creating repositories of CD positive donors for future characterization.