Combined CD25, CD64, and CD69 Biomarker in 3D-Printed Multizone Millifluidic Device for Sepsis Detection in Clinical Samples

Sepsis is a serious medical conditionthat arises from a runaway response to an infection, which triggersthe immune system to release chemicals into the bloodstream. Thisimmune response can result in widespread inflammation throughout thebody, which may cause harm to vital organs and, in more severe cases,lead to organ failure and death. Timely and accurate diagnosis ofsepsis remains a challenge in analytical diagnostics. In this work,we have developed and validated a sepsis detection device, utilizing3D printing technology, which incorporates multiple affinity separationzones. Our device requires minimal operator intervention and utilizesCD64, CD69, and CD25 as the biomarker targets for detecting sepsisin liquid biopsies. We assessed the effectiveness of our 3D-printedmultizone cell separation device by testing it on clinical samplesobtained from both septic patients (n = 35) and healthyvolunteers (n = 8) and validated its performanceaccordingly. Unlike previous devices using poly(dimethyl siloxane),the 3D-printed device had reduced nonspecific binding for anti-CD25capture, allowing this biomarker to be assayed for the first timein cell separations. Our results showed a statistically significantdifference in cell capture between septic and healthy samples (with p values of 0.0001 for CD64, CD69, and CD25), suggestingthat 3D-printed multizone cell capture is a reliable method for distinguishingsepsis. A receiver operator characteristic (ROC) analysis was performedto determine the accuracy of the captured cell counts for each antigenin detecting sepsis. The ROC area under the curve (AUC) values foron-chip detection of CD64+, CD69+, and CD25+ leukocytes were 0.96,0.92, and 0.88, respectively, indicating our diagnostic test matchesclinical outcomes. When combined for sepsis diagnosis, the AUC valuefor CD64, CD69, and CD25 was 0.99, indicating an improved diagnosticperformance due to the use of multiple biomarkers.