Quantum-enhanced nanodiamond rapid test advances early SARS-CoV-2 antigen detection in clinical diagnostics

Nat. Commun. 16, 8778 (2025) · DOI: 10.1038/s41467-025-63066-1

License: CC BY 4.0.

Abstract

Quantum biosensors, which harness quantum effects to detect biomarkers, could address the urgent need for more sensitive rapid diagnostics. Lateral flow tests using nitrogen-vacancy centres in nanodiamond labels offer high sensitivity and robustness by controlling the spin-dependent fluorescence to remove background. This is particularly important in complex and variable clinical samples. However, to date only model systems have been studied with few clinical samples. Here we show results of a clinical evaluation of a spin-enhanced nanodiamond test for SARS-CoV-2 antigen with 103 upper respiratory tract swab samples. We find 95.1% sensitivity (Ct ≤ 30) and 100% specificity benchmarked against RT-qPCR, with no cross-reactivity to influenza A, RSV, and Rhinovirus. Modelling with patient data yields a mean of 2.0-days earlier detection compared to conventional gold-nanoparticle tests (just 0.6 days after RT-qPCR) with 2.2-fold more patients detected on the first day of symptom onset, potentially reducing the transmission risk and protecting populations.

Figures

13 panels with data across 4 figures. Each panel page shows the plot, its columns and its files; each data.csv begins with a header naming the paper, the panel, the source, the license and the provenance route.

Fig. 1

• panel (c): Panel (c), inset: continuous-wave optically detected magnetic resonance of the fluorescent nanodiamonds, fluorescence intensity (%) against microwave frequency (GHz). The energy level diagram of (c) is not hosted. data.csv
• panel (d): Panel (d): normalised excitation (green) and emission (red) spectra of the fluorescent nanodiamonds against wavelength (nm). The printed NV$^0$ and NV$^-$ labels mark the zero phonon lines on the emission curve. data.csv

Fig. 2

• panel (a): Panel (a): normalised biolayer-interferometry binding [nm] against analyte concentration [M] for six SARS-CoV-2 antibodies (AbR001, AbMM08, AbR004, AbR040, AbCR009, AbCR018): measured points and fitted binding curves. The sensor schematic is not hosted. data.csv
• panel (b): Panel (b): normalised lock-in signal (log) against recombinant nucleocapsid concentration (M) for the FND assay with modulation (blue) and 40 nm gold nanoparticles (teal): mean and SD (circles, n = 3 tests), replicate rows (crosses), exponential fit with 95% band, LoD (solid) and its 95% CI (dotted). Blanks at zero concentration are in the data but not drawn on the log axis. data.csv
• panel (c): Panel (c): mean FND signal (bars, SD whiskers, n = 3 tests; log axis) for recombinant antigens of SARS-CoV-2 (positive and negative), MERS-CoV and four human coronaviruses. The three replicate values per antigen are in the data. data.csv
• panel (d): Panel (d): normalised lock-in signal (log) against concentration (TCID$_{50}$/mL) of gamma-irradiated SARS-CoV-2 for the Omicron (orange) and wild-type (red) strains: mean and SD (circles, n = 3 tests), replicates (crosses), exponential fit with 95% band, LoD (solid) and its 95% CI (dotted). Blanks at zero concentration are in the data but not drawn on the log axis. data.csv

Fig. 3

• panel (b): Panel (b): FND test-line signal (log) against RT-qPCR copy number, $\log_{10}$(copies/mL), for 53 SARS-CoV-2 positive nasal swabs: means with SD in both directions, the Bayesian errors-in-variables regression (black) and its 95% credible interval (grey dashed), and the FND threshold (red dashed). The print colours the samples by FND and AuNP result; here they are one series. data.csv
• panel (c): Panel (c): ROC curves of the FND assay on clinical swabs (53 SARS-CoV-2 positive, 37 negative): sensitivity (%) against 100% minus specificity (%) for all samples, samples with Ct $\leq$ 25 and samples with Ct $\leq$ 30, one point per threshold, with the chance line (red dotted). data.csv
• panel (d): Panel (d): FND signal (log axis) of clinical nasal swabs: 53 SARS-CoV-2 positive, 50 SARS-CoV-2 negative, 6 RSV, 6 influenza A and 1 rhinovirus positive samples, one point per sample, black bars at the means. Points are not jittered sideways as in the print. data.csv

Fig. 4

• panel (a): Panel (a): modelled viral load ($\log_{10}$ copies/mL) against days after inoculation, from the Killingley et al. challenge-trial curve: mean (black) with 95% credible band (grey), the RT-PCR, FND and AuNP thresholds (horizontal dashed), and the mean day of detection (dashed) and the day 95% of patients are detected (solid) for FND (blue) and AuNP (green). data.csv
• panel (b-1): Panel (b), left: modelled percentage of cases detected against days after inoculation for the FND (blue) and AuNP (teal) assays, with 95% of cases marked (dashed). data.csv
• panel (b-2): Panel (b), right: FND diagnostic advantage (days earlier than AuNP) against the percentage of cases detected, with 95% of patients marked (dashed). data.csv
• panel (c): Panel (c): estimated percentage of patients detected positive by the FND (blue) and AuNP (teal) assays on days 0 to 6 since symptom onset (Frediani et al. Ct distribution). The proportions of patients with Ct $\leq$ 25, 25 to 30 and above 30, drawn as grey areas in the print, are in the data but not drawn. data.csv

Cite

Alyssa Thomas DeCruz, Benjamin S. Miller, Da Huang, Max McRobbie, Felix Donaldson, Laura E. McCoy, Ciara K. O’Sullivan, Johannes C. Botha, Eleni Nastouli, Rachel A. McKendry. Quantum-enhanced nanodiamond rapid test advances early SARS-CoV-2 antigen detection in clinical diagnostics. Nat. Commun. 16, 8778 (2025). https://doi.org/10.1038/s41467-025-63066-1

When you use hosted data, cite the original paper and give the panel's URL so a reader can find the exact values you used.

All papers · About · Statistics · Submit data · API