Streamlining fungal diagnostics: Evaluation of a Direct-to-PCR extraction-free workflow for candida detection.

Fungal infections are an escalating health threat, and delays from conventional nucleic acid extraction hinder timely diagnosis and treatment. This study evaluated the diagnostic performance of a novel extraction-free technology, Direct-to-PCR (D2P; Scienetix, Tyler, TX, USA), for the detection of clinically significant Candida species (C. albicans, C. glabrata, C. auris, C. parapsilosis, and C. tropicalis). D2P was compared against conventional silica column-based (Qiagen) and magnetic bead-based (KingFisher) extraction methods, using microbial reference isolates, residual clinical specimens, and limit-of-detection analyses. Diagnostic sensitivity and specificity of D2P were comparable to conventional approaches, with specificity ranging from 96.77 % to 100 %. Concordance between methods was high, with Cohen's kappa coefficients (κ=0.93-1.00). Limit-of-detection analyses demonstrated strong analytical sensitivity, with excellent linearity (R²=0.924-0.999) and low replicate variability (coefficient of variation 0.2-6.3 %). Statistical comparison of cycle threshold values revealed no significant differences between methods (p > 0.05), supporting equivalent nucleic acid recovery without the need for time-intensive extraction steps. Despite these strengths, the study has limitations. The relatively small number of residual clinical specimens (n = 40) may restrict the generalizability of findings, and further validation across broader patient populations, specimen types, and clinical settings is warranted. Nevertheless, D2P offers a streamlined, rapid diagnostic workflow that reduces turnaround times and enhances accessibility, particularly in resource-limited environments. Wider adoption of extraction-free PCR platforms such as D2P could facilitate earlier detection of invasive candidiasis, improve clinical outcomes, and mitigate healthcare-associated morbidity and mortality attributable to fungal infections.