Reply to Comments: Comparison of Methods Study between a Photonic Crystal Biosensor and Certified ELISA to Measure Biomarkers of Iron Deficiency in Chronic Kidney Disease Patients.

In this brief note, we respond to the comments made by Dr [...].

Comparison of Methods Study between a Photonic Crystal Biosensor and Certified ELISA to Measure Biomarkers of Iron Deficiency in Chronic Kidney Disease Patients.

The total analytical error of a photonic crystal (PC) biosensor in the determination of ferritin and soluble transferrin receptor (sTfR) as biomarkers of iron deficiency anemia in chronic kidney disease (CKD) patients was evaluated against certified ELISAs. Antigens were extracted from sera of CKD patients using functionalized iron-oxide nanoparticles (fAb-IONs) followed by magnetic separation. Immuno-complexes were recognized by complementary detection Ab affixed to the PC biosensor surface, and their signals were followed using the BIND instrument. Quantification was conducted against actual protein standards. Total calculated error (TEcalc) was estimated based on systematic (SE) and random error (RE) and compared against total allowed error (TEa) based on established quality specifications. Both detection platforms showed adequate linearity, specificity, and sensitivity for biomarkers. Means, SD, and CV were similar between biomarkers for both detection platforms. Compared to ELISA, inherent imprecision was higher on the PC biosensor for ferritin, but not for sTfR. High SE or RE in the PC biosensor when measuring either biomarker resulted in TEcalc higher than the TEa. This did not influence the diagnostic ability of the PC biosensor to discriminate CKD patients with low iron stores. The performance of the PC biosensor is similar to certified ELISAs; however, optimization is required to reduce TEcalc.

Enhanced sandwich immunoassay using antibody-functionalized magnetic iron-oxide nanoparticles for extraction and detection of soluble transferrin receptor on a photonic crystal biosensor.

Iron deficiency anemia (IDA) has detrimental effects on individuals and societies worldwide. A standard sandwich assay (SA) for the detection of soluble transferrin receptor (sTfR), a biomarker of IDA, on a photonic crystal (PC) biosensor was established, but it was susceptible to non-specific signals from complex matrixes. In this study, iron-oxide nanoparticles (fAb-IONs) were used as magnetic immuno-probes to bind sTfR and minimize non-specific signals, while enhancing detection on the PC biosensor. This inverse sandwich assay (IA) method completely bound sTfR with low variability (<4% RSD) in buffer and allowed for its accurate and precise detection in sera (Liquichek™ control sera) on the PC biosensor using two certified ELISAs as reference methods. A linear dose-response curve was elicited at the fAb-IONs concentration in which the theoretical binding ratio (sTfR:fAb-IONs) was calculated to be <1 on the IA. The LoDs for sTfR in the SA and IA were similar (P>0.05) at 14 and 21 µg/mL, respectively. The inherent imprecision of the IA and reference ELISAs was σ(δ)=0.45 µg/mL and the mean biases for Liquichek™ 1, 2 and 3 were 0.18, 0.19 and -0.04 µg/mL, respectively. Whereas the inherent imprecision of the SA and reference ELISAs was σ(δ)=0.52 µg/mL and the biases for Liquichek™ 1, 2 and 3 were 0.66, 0.14 and -0.67 µg/mL, respectively. Thus, unlike the SA, the IA method measures sTfR with the same bias as the reference ELISAs. Combined magnetic separation and detection of nutrition biomarkers on PC biosensors represents a facile method for their accurate and reliable quantification in complex matrixes.

A photonic crystal biosensor assay for ferritin utilizing iron-oxide nanoparticles.

Iron deficiency anemia afflicts 1 in 3 individuals, mostly women and children worldwide. A novel application using iron-oxide nanoparticles (IONPs) and a photonic crystal (PC) optical biosensor as an immunodiagnostic platform for detection of serum ferritin, a biomarker for iron deficiency, is presented. Human liver ferritin (450 kDa), clinical serum controls, and three commercially available ferritin ELISA tests were used to evaluate the PC biosensor assay in terms of inter- and intra-assay variability, spike-recovery (%), limit of detection (LOD), and matrix effects on binding. For the PC biosensor, signal response from label-free, sandwich with secondary antibody (pAb), and pAb functionalized with iron-oxide nanoparticles (FpAb) assays were detected using the Biomolecular Interaction Detection (BIND) system. Bland-Altman analysis was used to evaluate agreement between expected values for ferritin in control sera and each of the detection platforms. Inter- and intra-assay variability of the PC biosensor were both <10%. Percent mean recovery (±%RSD) of ferritin from two control sera samples were 94.3% (13.1%) and 96.9% (7.6%). Use of FpAb in PC biosensor resulted in two orders of magnitude increase in sensitivity compared to label-free assay; capable of measuring serum ferritin as low as 26 ng/mL. In comparison to ELISA tests, the PC biosensor assay had the lowest bias (-1.26; 95% CI [-3.0-5.5]) and narrower limit of agreement (-11.6-9.1 ng/mL) when determining ferritin concentrations from control sera. These proof-of-concept studies support the use of IONPs to enhance detection sensitivity of PC biosensors for determination of biomarkers of nutritional status.