Abstract
Humanised recombinant antibodies specific to the SARS-CoV-2 Spike protein were calibrated against the NISTmAb standard human antibody to produce a fully quantitative antibody assay. The assay allows comparative studies between patient cohorts to be performed from which common properties may be derived. Two cohorts comparing patient vaccine response to AstraZeneca ChAdOx1-S (AZ, 35 patients) and Pfizer/BioNTech BNT162b2 (Pfizer, 25 patients) shows close association of the 31st percentile of the AZ distribution (2.90 ± 1.10 mg/L) and the 7th percentile of the Pfizer distribution (1.11 ± 1.10 mg/L) corresponding to the efficacy of the vaccines at preventing infection. The AZ IgG response distribution varies from 0.6 mg/L–25.4 mg/L with an average (mode) of 3.3 ± 1.0 mg/L; the Pfizer response distribution varies from 0.6 mg/L to 33.1 mg/L with a mode of 3.7 ± 1.0 mg/L. A third patient cohort looked at the recovery of 195 SARS-CoV-2 RT-PCR-positive patient samples and 200 pre-pandemic patient samples. A fourth patient cohort reviewed the NIBSC Anti-SARS-CoV-2 Verification Panel. The diagnostic cut-off for RT-PCR-positive patient samples was 1.34 ± 1.10 mg/L and the NIBSC panel separated seropositive and seronegative samples at 1.90 ± 1.10 mg/L. The mean value of the two prevention and two recovery thresholds is 1.8 mg/L with 95% confidence limits of 0.2–3.4 mg/L. In recovery and, critically, infection prevention, an antibody concentration threshold estimate of 3.4 mg/L appears mechanistically important. An antibody immunity threshold predicting a mucosal concentration preventing SARS-CoV-2 colonisation of the nasopharyngeal cavity is discussed.
Introduction
Vaccination programmes against SARS-CoV-2 have brought the ongoing COVID-19 pandemic under significantly greater control. In many countries, however, population vaccine immunity is waning1, with vaccine-induced antibody and T cell half-lives reported at a median of 200 days2 and at 60 – 200 days for natural antibodies following infection3. The half-life range implies immunity waning is clearly a personalised characteristic suggesting a population immunity spectrum leading to a range of infection virulence from asymptomatic infection to viremia requiring hospitalisation. The vaccines have different efficacies at preventing infection. The Pfizer vaccine has an efficacy of 93% for reducing symptomatic cases4 14 days or longer following the second dose. Therefore, the lower 7th percentile of the distribution of antibody vaccine response represents a potential immunity threshold. Similarly, for AZ with an equivalent efficacy of 69% – 74% in a real-world setting5, the lower 31st percentile potentially corresponds to a similarly protective antibody level. Although not investigated here, the same could be suggested for the Moderna vaccine, which shows 95.2% efficacy (95% confidence limits 91.2 % – 97.4 %)6. There, the 5th percentile of the antibody response spectrum potentially represents an immunity threshold. Identification and application of an immunity threshold will require standardisation of immunity assays.
The differing efficacies of the vaccines at preventing hospitalisation vs. preventing infection suggest different mechanisms of immune response. Many developed countries have embarked on vaccination programmes and achieved high rates of vaccination in their populations7. Booster programmes have been implemented, based not on individual immunological need but rather on perceived risk category. An immunity test would produce a new tool in the management of the ‘vaccine shield’ based on personal vulnerability, enabling a strategy of prevention of infection and the consequent reduction in the prevalence of the virus. None have thus far been established quantitatively.
Investigations of antibodies as a correlate of immunity to SARS-CoV-2 have been mixed; some report unsuccessful correlations8 others report robust correlation9. Similar investigations correlating antibody levels and efficacy have been explored for 18 vaccines previously10,11, with some successes. Elsewhere, potential correlates of protection have been investigated across the complex immune response triggered by infection and other vaccines12, most critically B cells, T cells and antibody levels. One interesting example is the bacterium H. influenzae, where prevention of nasopharyngeal colonisation was achieved with specific IgG levels of 5 mg/L or higher in the plasma13. This example is directly relevant to SARS-CoV-2 infection because of the shared characteristics of transmission (airborne droplets) and site of entry (mucosal membranes). The importance of the nasal mucosa suggests a differential role for antibodies over cells: the total T cell density is ∼ 650 mm-2 (with a factor of 5 variation)14, whereas the total IgG15 is of order 11.4 – 110 (median 40.8) µg mL-1. These observations and some SARS Cov-2 studies in primates16 suggest the mucosal immunity is critical at preventing SARS-CoV-2 infection both in the nose and the lower respiratory tract.
A fully quantitative antibody measurement calibrated against a standard material is required to compare measurements from different samples and results derived from different studies. Standard materials take time to develop; those currently available for SARS-CoV-2 serology include WHO reference materials and NIBSC samples which define an arbitrary unit to a pooled serum sample, typically 1000 U/mL There is also a NIST standard human antibody17 specific to the respiratory syncytial virus protein F (RSVF)18, which can be used to calibrate a biophotonic assay platform and define mass purity. The mass standard can be used to calibrate a humanised antibody raised to the SARS-CoV-2 Spike (S) protein, which in turn can be used to calibrate the patient antibody response quantitatively.
In this paper, we investigate antibody immunity thresholds in different patient cohorts using a fully quantitative immunoassay for SARS-CoV-2 S antibodies, standardised against the NIST reference human antibody. The test is used to profile the anti-S IgG serum response in patients receiving the AZ and Pfizer vaccines, as well as in SARS-CoV-2 RT-PCR-positive (PCR+) convalescent patients and negative samples collected pre-pandemic. The test is further used to measure the separation between positive and negative NIBSC standard samples. Two immunity thresholds are derived from the vaccine efficacy at preventing infection 31% and 7% from the AZ and Pfizer vaccine response curves. A further two immunity thresholds are made from recovery cohorts. The four threshold estimates are then averaged to suggest a composite immunity threshold which is discussed in the context of mucosal immunity. Further, three patient booster responses are presented and discussed, with personalised management of immunity proposed as a strategy to maintain a population vaccine shield.
Methods and Materials
Methods
Biophotonic Multiplexed Immuno-kinetic assay
The core technology is an immuno-kinetic assay using a biophotonic detection event and has been described in detail elsewhere in a SARS-CoV-2 antibody sensing application19. Briefly, an array of gold nanoparticles is illuminated in a total-internal-reflection configuration exciting a localised plasmon in the particles scattering the light. The intensity of scattered light depends on the mass in the plasmon field (formally permittivity) so a greater mass of proteins leads to greater intensity in scattered light from the array surface. This signal is captured as a video image in real time. The multiplexed nanoparticle array is functionalised with capture molecules to give analytic specificity to the target analyte. The array of the Attomarker COVID-19 Antibody Immunity Test consists of five tests: total IgG captured by Protein A/G (PAG), polyclonal goat antibodies to measure C-reactive protein (CRP), and the SARS-CoV-2 proteins Nucleocapsid (N), Spike (S) and Receptor Binding Domain (RBD). In addition, a recombinant human serum albumin (HSA) channel acts as a control to adjust for temperature at the array, non-specific binding and light intensity variation. A 20 μL serum sample is diluted 100-fold, centrifuged and then fat-filtered, before being flowed over the array. The brightness change in the scattered light intensity is integrated over 120s during a capture step, which is followed by a wash step to remove unbound protein and low affinity antibodies, followed by a quantitative detection step using anti-analyte antibodies (integrating for 90s), completing a sandwich assay to deliver results in 7 minutes.
The response of the biophotonic array is calibrated using the international NIST standard which is a high-purity, monomeric, recombinant humanized IgG1ĸ with a known sequence20 specific to the respiratory syncytial virus protein F (RSVF)18. The purity of the antibody is measured using the PAG assay on the array, which binds antibodies specifically via the Fc region. The Antibody Immunity Test can then be calibrated in a two-step process:
Calibration of the site density of the PAG channel using a 7-point calibration curve derived from the NISTmAb; and
Determination of the concentration of a humanised mouse antibody specific to the SARS-CoV-2 Spike protein S2 subdomain (Figure S1).
The Spike antigen site density is then calibrated using three concentrations using the humanised mouse antibody to span the concentration range in the samples.
The PAG assay was calibrated over an extended calibration curve of 1.56 – 50.0 nM using the NISTmAb and the Spike protein assay was linear over the range 0.6 – 20.0 mg/L (Figure S2). All Antibody Immunity Test sensor chips were calibrated with 3-point calibration using either 1.0, 5.0 and 20.0 mg/L (100-fold dilution) or 1.0, 5.0 and 10.0 mg/L (50-fold dilution) calibrant samples, each measured twice. The experiments were repeated with 100-fold and 50-fold dilutions to establish dilution invariance, with a limit of detection of 1.6 mg/L for 100-fold dilution and 0.6 mg/L for 50-fold dilution. A set of NIBSC samples was also measured to allow comparison between techniques. The threshold for the Antibody Immunity Test was calibrated using two commercial samples close to the threshold and determined by maximising the Youden Index21.
Four immunity thresholds were then calculated from five patient data sets. The antibody responses of double vaccinated patients (AZ n = 35, Pfizer n = 25) were measured and data collected after obtaining their full consent. The experimental response distributions were plotted, from which percentile concentrations for vaccine efficacy were derived based on figures reported in Phase III clinical trials of both vaccines (AZ 69%; Pfizer 93%). Further immunity thresholds were determined using the separation between positive and negative NIBSC samples and the clinical threshold for the Antibody Immunity Test. The error estimates for the four immunity thresholds were determined from the repeatability of calibration samples at similar concentrations.
Materials
Materials used throughout the course of the experiments were used as supplied by the manufacturer, without further purification. Sigma-Aldrich supplied phosphate buffered saline (PBS) in tablet form (Sigma, P4417), phosphoric acid solution (85 ± 1 wt. % in water, Sigma, 345245) and Tween 20 (Sigma, P1379). Glycine (analytical grade, G/0800/48) was provided by Fisher Scientific. Assay running and dilution buffer was PBS with 0.005 v/v % Tween 20 and the regeneration buffer was 0.1M phosphoric acid with 0.02M glycine solution in deionized water.
The recombinant Human Antibody to the Spike protein S2 subdomain was a chimeric monoclonal antibody (SinoBiological, 40590-D001, Lot HA14AP2901). The antibody was raised against the following immunogen: recombinant SARS-CoV-2 / 2019-nCoV Spike S2 ECD protein (SinoBiological, 40590-V08B). C-reactive protein (CRP)-depleted serum from BBI solutions (SF100-2). NISTmAb, Humanized IgG1κ Monoclonal Antibody from National Institute of Standards and Technology (RM8671). The NISTmAb is a recombinant humanized IgG1ĸ with a known sequence20 specific to the respiratory syncytial virus protein F (RSVF)18. The detection mixture consisted of a 200-fold dilution of IG8044 R2 from Randox in assay running buffer.
Triple Antibody Test sensor chips were printed with recombinant human serum albumin from Sigma-Aldrich (A9731), casein from Thermo Scientific (37582), anti-CRP from Biorad (1707-0189G), recombinant Protein A/G from Thermo Scientific (21186), SARS-CoV-2 Spike Protein (RBD, His tag) from GenScript (Z03479-100), SARS-CoV-2 Spike S1+S2 ECD-His Recombinant protein from SinoBiological (40589-V08B1) and SARS-CoV-2 Nucleocapsid protein from the Native Antigen Company (REC31851-100).
Patient Samples
400 commercial samples
A set of 400 serum samples was purchased from two suppliers (Biomex GmbH and AbBaltis) to derive the sensitivity, specificity and clinical threshold for the Triple Antibody Test. Negative samples: 200 pre-pandemic (pre-December 2019) human serum samples were purchased from AbBaltis. All were tested and found negative for STS, HBsAg, HIV1 Ag(or HIV PCR(NAT)), HIV1/2 antibody, HCV antibody and HCV PCR(NAT) by FDA approved tests; and positive samples: 90 samples were purchased from AbBaltis which were all from PCR+ individuals. They were also tested for IgG and IgM antibodies to SARS-CoV-2; all 90 were positive for IgM antibodies and 70 were positive for IgG antibodies. No information was provided regarding symptoms of the donors. 45% of these samples were from Female donors and 55% were from Male donors. The age of donors ranged from 19-81 years. No information on time from infection to sample collection was given.
106 samples purchased from Biomex were from PCR+ individuals. All samples were YHLO Biotech SARS-CoV-2 IgG positive and Abbott SARS-CoV-2 IgG positive. A spectrum of symptoms from the following list were detailed for each sample: fever, limb pain, muscle pain, headache, shivers, catarrh, anosmia, ache when swallowing, diarrhoea, breathing difficulties, coughing, tiredness, sinusitis, pneumonia, sickness, lymph node swelling, pressure on chest, flu-like symptoms, blood circulation problems, sweating, dizziness and hospitalisation. Time from infection to sample collection ranged from 27 – 91 days. 35% of these samples were from Female donors and 65% were from Male donors.
The commercial sample set was tested with the Attomarker qualitative Triple Antibody test without removing any samples. The test uses Boolean logic classifying samples as positive if Nucleocapsid, Spike or Receptor Binding Domain above a threshold. The sample showed a sensitivity of 92%, specificity of 91% and unclassified of 10%, including 1 triple false positive and two double false positives. The Boolean thresholds were calibrated in the current study quantitatively by remeasuring samples close to the cut-off.
Attomarker Clinic samples
Samples were collected from patients in Attomarker clinics, all of whom provided informed consent for their anonymised data to be used in research to aid the pandemic response. The data from tests of 63 patient samples are included in this study. 60 had received two doses of either the AstraZeneca SARS-CoV-2 vaccine or the Pfizer SARS-CoV-2 vaccine at least 14 days prior to sample collection and testing by Attomarker. Three further samples had received a third vaccination.
National Institute of Biological Standards and Control (NIBSC) SARS-CoV-2 Standard samples
The NIBSC Anti-SARS-CoV-2 Verification Panel for Serology Assays (product code: 20/B770) was purchased. This panel contains 37 samples. 23 samples are from convalescent plasma packs which are known to be Anti-SARS-CoV-2 positive; 14 samples are plasma packs known to be Anti-SARS-CoV-2 negative.
Ethics
The samples were collected with informed patient consent for use in better understanding pandemic. The use of the samples has been reviewed independently by the Biosciences Ethics Committee, University of Exeter and approved.
Results
The standardisation of the biophotonic assay to Spike S2 produces a fully quantitative estimate of the response profiles of the patients to the Pfizer and AZ vaccines, Figure 1. Threshold levels associated with the lower 7th percentile of the Pfizer distribution are plotted to correspond with the 93% efficacy at protecting against infection; similarly, the 31st percentile for AZ is shown, plotted to correspond with 69% efficacy. The commercial clinical sample threshold and the NIBSC sample threshold are also indicated in Figure 1, as are the three patients who received Pfizer booster doses, Table 3.
Demographic data for samples from double vaccinated individuals collected in the Attomarker clinic
Demographic data for samples collected and tested by Attomarker where the patients had received a Pfizer booster vaccination.
Summary of immunity thresholds for vaccine responses, including preliminary literature survey (Error bars are standard deviations from nearest calibration points)
The fully quantitative spectrum of vaccine responses, showing all four immunity thresholds plotted on each graph: (A) the AZ response spectrum; and (B) the Pfizer response spectrum. The vertical lines (solid black) are individual patients who had received a booster vaccination.
The AZ IgG response distribution ranges from 0.6 mg/L – 25.4 mg/L with a distribution mode of 3.3 ± 1.0 mg/L, whereas the Pfizer response distribution ranges from 0.6 mg/L –33.1 mg/L with a distribution mode of 3.7 ± 1.0 mg/L. These data suggest that the immune response to the vaccine is highly personal. Many of the ‘super responders’ had positive SARS-CoV-2 RT-PCR tests before vaccination, indicating the vaccine may be inducing memory of the previous infection and, consequently, an enhanced response.
The NIBSC sample responses are shown in Figure S4, along with their calibrated values. These data have been plotted against four other anti-SARS-CoV-2 serology techniques, Figure 2A-D, with correlation coefficients of 0.77, 0.72, 0.51 and 0.62, respectively. The manufacturer conversion factors are shown in Table S1.
Plots of quantified Spike IgG Antibody result in mg/L vs four external tests for the NIBSC 20/B770 sample panel. Test platform (A) measures S1/S2 IgG; platform B ELISA S1 (recombinant) IgG; platform C ELISA S1 IgA and platform D S1 IgG. Sample with NIBSC sample ID 19 has been excluded from the fit. Pink circles are NIBSC negatives samples; Blue circles are positive NIBSC samples. The point depicted as a triangle is an outlier despite several repeats and was removed from the fit.
Discussion
SARS-CoV-2 antibody testing has not yet been standardised which makes comparison between techniques and studies difficult. There have been reports of up to 100-fold variation in standard sample measurements between laboratories22, all presented in different in-house, arbitrary units. The correlation between techniques is variable (Figure 2) and standardisation efforts are further obstructed by variation in detection targets between techniques (e.g., IgG/IgA combined vs. IgG only). The intrinsic mass sensitivity of the biophotonic assay using the NIST Standard Human Antibody is a potential method by which all techniques can now be calibrated, enabling stoichiometric and fully mechanistic analyses to be proposed and tested.
A standardised approach presenting results in mg/L is mechanistically more powerful and can be applied to both the antigens for viral load and host response, and between vaccines for different pathogens. Many sterilising vaccines prevent infection, such as for (MMR)23 leading to quantified thresholds of immunity, Table 3. The first reports for the SAR-CoV-2 vaccines similarly reported prevention of infection and the production of good antibody levels4. The two prevention thresholds and two recovery thresholds determined in the paper are also summarised in Table 3. The assumption behind the infection prevention thresholds is that the antibody level in the distribution correlate with the vaccine efficacy at preventing infection, 93rd percentile for Pfizer4 and 31st percentile for AZ5 and by extension the 95th percentile for Moderna.6 Similarly the recovery thresholds derived from the PCR(+/-) cohort and the standardised NIBSC sample cohort all appear to coincide when measured quantitatively; Table 3. The average of all four thresholds investigated in this study suggests a mean of 1.8 mg/L, with 95% confidence limits of 0.2 – 3.4 mg/L. We propose the higher 95% confidence limit as a ‘clinical fail safe’ mucosal immunity threshold.
The concept of an antibody immunity threshold as a correlate of immunity has clearly been an objective of vaccine science for some time. There are some limitations in the estimates presented here: larger sample sizes would produce better estimates of the vaccine response distributions; and a higher detection sensitivity might allow lower levels to be measured below the detection threshold of the current technique. However, the immunity threshold estimate is established at the 95% CI for all four estimates and is deliberately over-estimated. The other two estimates based on recovery diagnosis and the NIBSC serum set are clearly different estimates of the antibody cut-off and derived independently. Further, the similarity with the threshold for prevention of H. influenzae colonisation is compelling and not accidental. Whilst concentrations for prevention of bacterial vs. viral colonisations may be different, fundamentally, antibody opsonisation and neutralisation are questions of surface area coverage of the invading pathogen in the mucosa.
The prevention of infection implies a mechanism that is either a rapid antibody and T cell response to a pre-symptomatic viral load, or the prevention of cellular infection completely by opsonisation and neutralisation. The pure prevention mechanism is consistent with mucosal immunity25-27 and allows decoupling of the antibody levels from T and B cell memory responses that require infected cells to trigger the adaptive response. Further, the total T cell density in the mucosa is ∼650 mm-2, varying over a factor of five14, and would increase response once epithelial cells have been infected by inducing recruitment of further T cells from the blood. However, the kinetics of T cell immunity are too slow to prevent infection. Thus, the mucosa separates the role of antibodies from T cells in preventing infection.
It is likely that the mechanism for infection prevention could occur in the nasal mucosa or lower in the mucosa of the lungs, quantified here, allowing a stochiometric mechanism to be assessed. The SARS-CoV-2 virus has a diameter of 91 ± 11 nm (n = 179) and contains 24 ± 9 Spike protein trimers per virion28, suggesting of order 25 antibodies to prevent viral colonisation by opsonisation or neutralisation. The nasal mucosa contains a total IgG of order 11.4 – 110 (median 40.8) µg mL-1, equivalent to 4.5 – 44 (median 16.4) ×1013 IgG molecules per mL15. Assuming that the fraction of total IgG which is anti-Spike is the same in the nasal mucosa as in the serum, 3.4 mg/L corresponds to 1011 antibodies per mL. Neutralisation and opsonisation of an infection challenge of 109 viral particles29 would be possible with 1011 antibodies per mL. Whilst this still leaves the potential for very large viral loads to cause illness, serum/nasopharyngeal anti-Spike IgG concentrations above 3.4 mg/L are sufficient to achieve the COVID-19 vaccine efficacies against infection reported in the literature, with waning antibody levels accounting for the appearance of breakthrough infections. In this hypothesis, the mucosal antibody concentration is causatively protective and prevents infection.
Maintaining antibody levels in the mucosa above 3.4 mg/L is a challenge given the waning population immunity. Three patient booster responses are shown in Figure 1 and are diverse across individuals.30 However, for public health management, the main challenge lies in how long the immunity will last. The half-life for IgG antibodies following infection varies and reported to be 60 – 200 days, with a mean of 106 days3; a further study estimated the half-life of IgG from the peak Pfizer vaccine-induced humoral response to be 21 days (95% CI 13 – 65) in initially seronegative individuals, but 53 days (95% CI 40 – 79) in initially seropositive individuals; and the estimated half-life for total antibodies was longer and ranged from 68 days (95% CI 54 – 90) to 114 days (95% CI 87 – 167) in initially seronegative and initially seropositive individuals, respectively31, with a median of 110 days. The only way to be sure about the antibody half-life of an individual is a paired antibody test 100 days apart. The lowest half-life of 21 days in combination with the immunity threshold could define an ‘edge of immunity’. A patient with a serum antibody concentration of 13 mg/L and a half-life of 100 days would have 200 immunity days before falling below the immunity threshold, falling to 120 days with a half-life of 60 days. However, the Pfizer booster patient with a serum antibody concentration of 37 mg/L, more than 9 times higher than the mode concentration (≥14 days post-first vaccine dose) prompts questions around whether a safe upper limit for vaccination exists and the risk of autoimmune diseases32.
Conclusions
Precision immunity offers the best solution for a properly managed return-to-normality in the pandemic, reducing the need for frequent antigen testing and improving vaccine supply stewardship by focussing booster campaigns on individual antibody levels and clinical vulnerability. The efficacy-threshold analysis is readily adaptable to all variants of SARS-CoV-2 and the multiplexed test can be extended for the simultaneous analysis of spike proteins from major variants of concern. The fully quantitative test allows simultaneous measurement of differential binding to variant spike proteins, providing a predicted vaccine efficacy for each. Critically, vaccine escape would be associated with a rising threshold for a new variant; if the threshold were higher than the booster response then the variant would have escaped the vaccine. Fully quantitative, personalised immunity measures could play an important role in the management of the SARS-CoV-2 pandemic, as well as for seasonal influenza and other respiratory infections.
Data Availability
All data produced in the present study are available upon reasonable request to the authors.
Funding
Exeter University Alumni, Attomarker Ltd funded PhD studentship at the University of Exeter and Attomarker Ltd funding directly.
Supplementary Data
Measurement of the concentration and monomeric purity of the chimeric anti-Spike antibody (Sinobiological, 40590-D001) relative to the NIST supplied monoclonal standard RM8671. The response shown is the area-under-curve (mili Refractive Index Units seconds mRIUs) of the Protein A/G - rHSA channels. (°) denote samples of NIST monoclonal at concentrations of 50.0, 25.0, 12.5, 6.25, 3.13 and 1.56 nM. (Δ) denote samples of anti-Spike antibody at concentrations of 50.0, 12.5, 3.13 and 1.56 nM. Vertical error bars show 1 s.d. of repeats. Horizontal error bars show 3% error as the assumed possible error from sample pipetting.
The response curve to samples of the chimeric anti-Spike antibody (Sinobiological, 40590-D001), flowed over the sensor for 120s each. The response shown is the area-under-curve of the Spike - rHSA channels. A detection mixture of 200-fold diluted Randox antiHuman IgG (IG8044) in PBS-T buffer was flowed over the sensor for 90s. Samples were made up in 100-fold diluted serum, obtained from BBI pre-pandemic (SF100-2), at concentrations of 5.33, 2.67, 1.33, 0.66, 0.33, 0.17,0.08, 0.04 nM, corresponding to pre-dilution serum antibody concentrations of 80.0, 40.0, 20.0, 10.0, 5.0, 2.5, 1.3, 0.6 mg/L. Vertical error bars show 1 s.d. of repeats. Horizontal error bars show 3% error as the assumed possible error from sample pipetting.
The calculated Spike antibody concentrations of a panel of NIBSC samples. Samples 1-23 are listed as positive; 24-37 are negative. Values were calculated using calibration samples of 40590-D001 at 1, 5, 10 and 20 mg/L. All sample measurements were single measurements, with calibrators repeated. Note that sample 12 has not been tested.
Lines of best fit and Correlation coefficients for the comparison with four platform technologies against the current fully quantitative test.
Acknowledgements
The authors would like to thank Dr Jonathan Snicker and Mr Sydney Nash for their guidance on the potential strategic and policy implications of the scientific findings.
References
