A novel algorithm for diagnosis of invasive pulmonary aspergillosis based on pentraxin 3 gene polymorphisms and its adjusted value among autoimmune diseases patients

Background: Invasive pulmonary aspergillosis (IPA) is a rapidly progressive and fatal disease for those with autoimmune diseases. The performance of existing diagnostic tools is unsatisfactory, and a novel algorithm based on pentraxin 3 (PTX3) gene polymorphisms with adjusted PTX3 and galactomannan (GM) values is urgently needed. Methods: Levels of PTX3 and GM were measured in the bronchoalveolar lavage fluid (BALF) and blood samples of patients who had autoimmune diseases with IPA between June 2017 and June 2021. Urea dilution was applied internally to correct the real BALF PTX3 and GM values. Three single-nucleotide polymorphisms (SNPs; rs1840680, rs2305619, and rs3816527) in the PTX3 gene were detected by polymerase chain reaction direct sequencing, and their associations with IPA were evaluated. Receiver operating characteristic (ROC) curves based on different variables were generated to determine the best algorithm for IPA diagnosis. Results: This study enrolled 50 patients with IPA and 100 without IPA in the control groups (comprising 50 patients with Aspergillus airway colonization and 50 patients with bacterial pneumonia). The levels of adjusted BALF PTX3 and GM were higher in the IPA group than in the control groups (P<0.05, respectively). For diagnosing IPA, the best adjusted cutoff value for PTX in BALF was 14.5 ng/mL and the best adjusted cutoff value for GM in BALF was 2.5 optical density index (ODI). The SNP rs1840680 AA homozygote was associated with a higher risk of IPA [odds ratio (OR) 18.86, 95% confidence interval (CI): 7.96-44.69; P<0.01], while no genotypic distribution differences were observed for the other 2 SNPs (rs2305619 and rs3816527). Six algorithms were established based on PTX3 gene polymorphisms. The algorithm consisting of PTX3 gene polymorphisms with adjusted BALF PTX3 and BALF GM values demonstrated the best diagnostic performance (sensitivity 90.03%; specificity 97.09%; area under the curve 0.94). Conclusions: It was revealed that our new algorithm based on PTX3 gene polymorphisms combined with adjusted BALF GM and BALF PTX3 values performed well in diagnosing IPA.