Johan Björkesten

Medical research is developing an ever greater need for comprehensive high-quality data generation to realize the promises of personalized health care based on molecular biomarkers. The nucleic acid proximity-based methods proximity ligation and proximity extension assays have, with their dual reporters, shown potential to relieve the shortcomings of antibodies and their inherent cross-reactivity in multiplex protein quantification applications. The aim of the present study was to develop a robust 96-plex immunoassay based on the proximity extension assay (PEA) for improved high throughput detection of protein biomarkers. This was enabled by: (1) a modified design leading to a reduced number of pipetting steps compared to the existing PEA protocol, as well as improved intra-assay precision; (2) a new enzymatic system that uses a hyper-thermostabile enzyme, Pwo, for uniting the two probes allowing for room temperature addition of all reagents and improved the sensitivity; (3) introduction of an inter-plate control and a new normalization procedure leading to improved inter-assay precision (reproducibility). The multiplex proximity extension assay was found to perform well in complex samples, such as serum and plasma, and also in xenografted mice and resuspended dried blood spots, consuming only 1 µL sample per test. All-in-all, the development of the current multiplex technique is a step toward robust high throughput protein marker discovery and research.

Understanding the dynamics of the human proteome is crucial for developing biomarkers to be used as measurable indicators for disease severity and progression, patient stratification, and drug development. The Proximity Extension Assay (PEA) is a technology that translates protein information into actionable knowledge by linking protein-specific antibodies to DNA-encoded tags. In this report we demonstrate how we have combined the unique PEA technology with an innovative and automated sample preparation and high-throughput sequencing readout enabling parallel measurement of nearly 1500 proteins in 96 samples generating close to 150,000 data points per run. This advancement will have a major impact on the discovery of new biomarkers for disease prediction and prognosis and contribute to the development of the rapidly evolving fields of wellness monitoring and precision medicine.

An important motivation for the construction of biobanks is to discover biomarkers that identify diseases at early, potentially curable stages. This will require biobanks from large numbers of individuals, preferably sampled repeatedly, where the samples are collected and stored under conditions that preserve potential biomarkers. Dried blood samples are attractive for biobanking because of the ease and low cost of collection and storage. Here we have investigated their suitability for protein measurements. Ninety-two proteins with relevance for oncology were analyzed using multiplex proximity extension assays (PEA) in dried blood spots collected on paper and stored for up to 30 years at either +4 °C or −24 °C.Our main findings were that (1) the act of drying only slightly influenced detection of blood proteins (average correlation of 0.970), and in a reproducible manner (correlation of 0.999), (2) detection of some proteins was not significantly affected by storage over the full range of three decades (34 and 76% of the analyzed proteins at +4 °C and −24 °C, respectively), whereas levels of others decreased slowly during storage with half-lives in the range of 10 to 50 years, and (3) detectability of proteins was less affected in dried samples stored at −24 °C compared with at +4 °C, as the median protein abundance had decreased to 80 and 93% of starting levels after 10 years of storage at +4 °C or −24 °C, respectively. The results of our study are encouraging as they suggest an inexpensive means to collect large numbers of blood samples, even by the donors themselves, and to transport, and store biobanked samples as spots of whole blood dried on paper. Combined with emerging means to measure hundreds or thousands of protein, such biobanks could prove of great medical value by greatly enhancing discovery as well as routine analysis of blood biomarkers. An important motivation for the construction of biobanks is to discover biomarkers that identify diseases at early, potentially curable stages. This will require biobanks from large numbers of individuals, preferably sampled repeatedly, where the samples are collected and stored under conditions that preserve potential biomarkers. Dried blood samples are attractive for biobanking because of the ease and low cost of collection and storage. Here we have investigated their suitability for protein measurements. Ninety-two proteins with relevance for oncology were analyzed using multiplex proximity extension assays (PEA) in dried blood spots collected on paper and stored for up to 30 years at either +4 °C or −24 °C. Our main findings were that (1) the act of drying only slightly influenced detection of blood proteins (average correlation of 0.970), and in a reproducible manner (correlation of 0.999), (2) detection of some proteins was not significantly affected by storage over the full range of three decades (34 and 76% of the analyzed proteins at +4 °C and −24 °C, respectively), whereas levels of others decreased slowly during storage with half-lives in the range of 10 to 50 years, and (3) detectability of proteins was less affected in dried samples stored at −24 °C compared with at +4 °C, as the median protein abundance had decreased to 80 and 93% of starting levels after 10 years of storage at +4 °C or −24 °C, respectively. The results of our study are encouraging as they suggest an inexpensive means to collect large numbers of blood samples, even by the donors themselves, and to transport, and store biobanked samples as spots of whole blood dried on paper. Combined with emerging means to measure hundreds or thousands of protein, such biobanks could prove of great medical value by greatly enhancing discovery as well as routine analysis of blood biomarkers. It may be said that every individual is a medical experiment that potentially could help define biomarkers to diagnose diseases by combining analysis of biological samples with information about medical conditions and environmental factors. These analyses will require samples from very large numbers of individuals, but it is also important to secure multiple samples from each individual. The availability of consecutive samples increases the chance that samples will be available from a time point optimal for diagnosis, before a disease becomes clinically manifest, and it also allows trends to be monitored over time where each individual serves as his or her own control. Blood is an ideal matrix for molecular analysis because of its accessibility and the fact that disease processes anywhere in the body are potentially reflected in altered levels of molecules released into circulation. Traditional blood sampling by venipuncture and centrifugation to obtain plasma requires trained personnel and laboratory equipment, and the samples are typically stored as tubes frozen at −80 °C, taking up considerable space in energy-consuming freezers. Dried blood spots (DBS), 1The abbreviations used are: DBS, Dried blood spot; ADAM 8, Disintegrin and metalloproteinase domain-containing protein 8; ANOVA, Analysis of variance; DLL1, Delta-like protein 1; DPS, Dried plasma spot; EDTA, Ethylenediaminetetraacetic acid; GPNMB, Transmembrane glycoprotein NMB; GWAS, Genome wide association studies; HGF, Hepatocyte growth factor; LOD, Limit of detection; NPX, Normalized protein expression; PEA, Proximity extension assay; RT, Room temperature; S100A4, Protein S100-A4; S100A11, Protein S100-A11; TCL1A, T-cell leukaemia/lymphoma protein 1A; TXLNA, Alpha-taxilin; ULSAM, Uppsala longitudinal study of adult men; WIF1, Wnt inhibitory factor.1The abbreviations used are: DBS, Dried blood spot; ADAM 8, Disintegrin and metalloproteinase domain-containing protein 8; ANOVA, Analysis of variance; DLL1, Delta-like protein 1; DPS, Dried plasma spot; EDTA, Ethylenediaminetetraacetic acid; GPNMB, Transmembrane glycoprotein NMB; GWAS, Genome wide association studies; HGF, Hepatocyte growth factor; LOD, Limit of detection; NPX, Normalized protein expression; PEA, Proximity extension assay; RT, Room temperature; S100A4, Protein S100-A4; S100A11, Protein S100-A11; TCL1A, T-cell leukaemia/lymphoma protein 1A; TXLNA, Alpha-taxilin; ULSAM, Uppsala longitudinal study of adult men; WIF1, Wnt inhibitory factor. offer several advantages over conventional blood sampling. These advantages include convenient collection during routine blood sampling or even via a finger prick by a lancet, permitting home sampling, and DBS can be sent to a lab or a biobank by regular mail, and be stored without taking up much space (1.Grüner N. Stambouli O. Ross R.S. Dried blood spots - preparing and processing for use in immunoassays and in molecular techniques.J. Vis. Exp. JoVE. 2015; (doi:10.3791/52619)Crossref PubMed Scopus (90) Google Scholar, 2.Wilhelm A.J. den Burger J.C. Swart E.L. Therapeutic drug monitoring by dried blood spot: progress to and PubMed Scopus Google Scholar, The of dried blood spots for to PubMed Scopus Google Scholar, a can dried blood spots as a for biomarkers into PubMed Scopus Google These can significantly the cost of blood sampling and and for of DBS is the of collected in it that protein levels can be from DBS samples longitudinal of protein biomarkers in DBS using an PubMed Scopus Google Scholar, A.J. of in Dried Blood 2015; PubMed Scopus Google Scholar, Dried blood of proteins with and PubMed Scopus Google protein abundance can be from a from a DBS using the proximity extension (PEA) The of dried blood spots for to PubMed Scopus Google Scholar, and PubMed Scopus Google This of several proteins from a levels of have with of the by using an DBS A.J. of in Dried Blood 2015; PubMed Scopus Google Scholar, Dried blood of proteins with and PubMed Scopus Google An increases of some proteins in DBS compared with conventional plasma and PubMed Scopus Google of the of proteins from blood in the levels of range range collection and and the wide range may to of protein levels in DBS samples compared with the conventional plasma study of the of in DBS of levels of or during years of storage at +4 °C of in dried blood spots during years of PubMed Scopus Google the of our information is available the of proteins in The of a of investigated were during of storage at °C to °C analyzed by of storage over A.J. of in Dried Blood 2015; PubMed Scopus Google Here we have investigated on of proteins using by the drying as well as of storage. a multiplex analysis of proteins to we compared results for dried samples of whole blood or a of the proteins in investigated DBS from stored for up to 30 years either at +4 °C or at −24 °C, and we compared protein for plasma samples stored at °C for of from adult Our results that drying only slightly but influenced of protein detectability over time the protein and the storage the of of proteins stored as blood was from a blood at Uppsala DBS samples were by blood a collection The of blood were to from the of the and the paper to be taking not to the paper with the The DBS samples were to for at of the blood were at +4 °C. The of the blood were for at and plasma was and Dried plasma spots were in the as the in were from the DBS and with a from an of the sampling was in every to The abundance of proteins to oncology was from from DBS and and in of blood and plasma Protein for of by the assays in the was from in The of each protein in the was of were in with and dried samples were by of proteins at from to to Protein The of the spots was on the of the The protein spots were dried at for before three in were from the of the spots for The of plasma in was from the the of the and the that the of proteins in the is the samples the analyzed was the of the of of samples from to were used to for paper with were analyzed to potential on protein by the paper DBS samples stored for and years from each were collected from the for at the in DBS were collected from and analyzed a from sampling at +4 °C before DBS were collected in the of in to for the that were collected in The paper used for sampling have not the from was of the of paper used and was and was used and is paper are to have very the storage conditions of the biobank have over and the biobank was at the biobank was to a with a of +4 °C and less the samples were to a with the to the of the was but in was to the and the biobank under The DBS are stored on of each in in DBS samples stored for and years from each were collected from the for at in DBS from were analyzed a from sampling at °C before DBS were collected in the of in to for the that were collected in of was by and The biobank at −24 °C The samples from the and biobanks were to we not even the of the samples from stored for and years from with of and years, were collected from the biobank in The sampled were not the the time The samples had stored at °C. It is not some of the samples had and plasma samples from from each of the and years were from the of at the Uppsala These samples were used to for the of and or were also used as a to and by much the protein levels had in the samples stored for of at °C. The DBS used in study were from the of the the of protein DBS, were collected from the and from DBS, from a finger prick from were collected and analyzed from DBS we the abundance of of proteins in DBS, and in and as protein by and PubMed Scopus Google of to measure levels of the The of with by of such the can be by a The of for each is used as a measure of numbers of as by using the The was using an extension to for and using a to for The are Normalized Protein an value on a to protein samples were used to define the of detection as protein the by three its This was for each protein in each for and of can be at the the of drying on protein detectability using blood and plasma samples, we the the analysis of a of biobank samples, samples for of DBS, and the was measure proteins of relevance for oncology and they include assays and are and to measure proteins in of the assays have to convenient detection of proteins in DBS with only to the and PubMed Scopus Google of plasma to the a DBS is with the is the as for samples but is not to the paper in the of the well after the is to the from the paper are with the the used for the DBS samples stored at −24 °C samples to the control. These were over the storage the for samples it was that the levels from samples from the The samples were and the were used for analyses were in on to can be at the were from to to for the paper DBS were by from the paper and the from the DBS were to for and of or dried samples of blood or plasma for proteins were with the protein were used to from of individual protein levels in blood compared with in the of for DBS samples stored for under conditions an was of samples with each were with of storage time on individual proteins were by with years of storage as and the protein abundance levels as The of were by an analysis of were for the proteins that were significantly decreased after for multiple value in samples stored at +4 °C and −24 °C. The half-lives were from the of the as the time to by of for individual proteins the were to protein in samples stored for and 30 years were by the protein in samples stored for years to storage. of of of proteins were with the of protein DBS samples, were collected from the or of DBS from the individual were for proteins were by the the of of the sampled from of in the protein were by taking into the individual and the or in that were from The of were by The levels of proteins of with were for by the using the from the of in the was using and with as but with the protein abundance levels as The study was in with the of the adult samples, the study was by the at Uppsala for samples and for longitudinal samples and samples were for the The and biobank samples were the of the in with the on of and for of respectively. of samples of blood or plasma to be dried on paper is an attractive means to collect and store samples, that the results of analysis are not The to be was the on protein detectability by the of drying of samples on paper. of samples were Ethylenediaminetetraacetic plasma and a from a of These were analyzed in the dried and for were for blood or plasma dried and samples were analyzed in and dried samples, and blood and plasma samples are in each for of the proteins with the The dried samples were for a in levels for paper This only the The protein levels very well and dried samples and but correlation were of for blood and plasma and and were in the numbers of proteins in either dried or blood or plasma a of or proteins were in each from a of proteins that were in at of the were in and dried samples for whole blood and plasma only on protein detectability by the act of drying the as we blood and plasma in and as samples The whole blood samples include and blood plasma The was for each protein in the blood and is as the to the of in the is a of the protein The proteins the in blood plasma in were individual as in and The of the proteins are in were we and dried of of for proteins the and dried samples the the It is important to the of protein a DBS of a DBS are to be used for investigated the abundance of proteins in either from the or the of DBS The results correlation results for protein in samples from and of a DBS for of the analyzed proteins value were samples from the and of the DBS 10 of proteins slightly levels were in samples collected from the of the The the were very with for the 10 proteins slightly in samples from the of the DBS or in from the of the for multiple of the analyzed proteins significantly value that the of the in the DBS on protein investigated protein levels in DBS samples collected and stored for of time at either +4 °C or −24 °C. were for the proteins in the investigated samples for each time and of in in each storage a the of and the for the were from and were as were as were for to each on storage The DBS samples stored at +4 °C for years for years at by years at +4 significantly from the samples value also but with value on the samples stored at °C for years were from samples stored at −24 °C were from analysis because they a compared with the samples stored for and 10 for the analysis of of individual proteins the samples stored for and 30 years were compared with the samples stored for This was for samples stored at +4 °C and −24 °C. of individual were for each of the proteins from DBS samples stored for up to 30 years at +4 °C or −24 °C The were on storage time in years and protein as on a The of the was to identify proteins detectability was significantly affected by storage at the The of and the of the were by detectability were the proteins during storage for DBS stored at +4 °C and −24 °C proteins were not affected by storage time GPNMB, TCL1A, and levels of others decreased over time TXLNA, and significantly decreased protein levels with storage time were for of multiple and of the for samples stored at +4 °C and −24 °C, of the significantly proteins at −24 °C decreased levels also at +4 °C. The of detectability for proteins was on their of the proteins at −24 °C at +4 °C. The was for stored at +4 °C value or at −24 °C value The and of the in levels for samples stored at +4 °C and −24 °C, respectively. samples stored at +4 °C or at −24 °C of individual proteins from samples stored and 30 years were compared with levels of the proteins in samples stored for This analysis was for samples stored at +4 °C and −24 °C, in of the of for proteins were to before The results were in a trends of protein levels with storage time for samples stored at +4 °C and −24 °C 10 years of storage at +4 °C or −24 °C the median of protein abundance at 80 and of the levels in samples that had stored for were to of of protein levels in samples stored at +4 °C or −24 °C. in protein detectability for samples stored at the value were for stored for 10 and years, but was for samples stored for 30 of protein as by taking into the individual and the or for the sampled was also investigated in 50 was to of the proteins analyzed the in plasma and samples This by on protein levels is with of storage time and sampling on biobank plasma protein PubMed Scopus Google Scholar, The of and on plasma levels of 2015; PubMed Scopus Google were also for of the proteins plasma and samples for the sampled of proteins in the biobank with plasma or samples that had stored at °C were for of of the and or using the from the the and storage time were to the were on the The for detectability of the proteins was less for the samples stored at °C for the DBS of the analyzed were for multiple the for samples collected from adult and stored at °C the sampled compared with the DBS samples from and of the was by storage time for the samples and the DBS samples, respectively), our to the of storage under the storage of three proteins HGF, and ADAM decreased significantly in samples and DBS stored at +4 °C, whereas proteins were significantly decreased in samples and DBS stored at −24 °C. is to association to disease by wide association The into for but the into Google Scholar, years of PubMed Scopus Google Scholar, The advantages and of analysis with a PubMed Scopus Google is a in in of for biomarkers with such as and of a of PubMed Scopus Google Scholar, and as biomarkers for 2015; PubMed Scopus Google Scholar, biomarkers for in PubMed Scopus Google Scholar, on the of biomarkers in PubMed Scopus Google biomarkers are not and over individual are preferably to that may emerging is considerable that of blood proteins information of value to diagnose disease processes anywhere in the or to and but protein have in for protein analysis is available and can be It is to for the may be where samples collected either before or after of disease discovery of blood protein biomarkers be on multiple samples collected over time for each in samples that the protein during storage. Dried spots of blood or plasma are inexpensive to collect and to but three important be their suitability as biobanks for protein (1) with the detectability of proteins (2) are protein levels and (3) the samples stored dried or The study each of but it is important to that it from the of protein detection by of in It is that results on the used or are used at or are analyzed by to protein and the is of the of protein The the by drying and storage is of relevance not only for biobanking but also for the of where samples may be collected by the by a finger prick and sent by regular to a laboratory for analysis of protein levels will the and PubMed Scopus Google Scholar, The Google Our using a of protein assays that detectability of proteins is very well but on levels by that dried samples be compared with dried samples and not to The correlation and dried blood was on whereas the correlation was for and dried may identify means to for the of the reproducible and of drying may be into in to samples to from dried The considerable protein levels in whole blood and in or dried is an of the of and in whole to the protein The abundance of proteins that not from the of blood is in blood compared with in plasma This is by the fact that the blood samples and or the of plasma compared with the plasma levels of plasma or NPX, are in plasma samples compared with the levels in whole The is not for dried samples were analyzed for dried blood and dried plasma This is a of the that plasma blood to in of plasma proteins from dried plasma and dried of proteins in DBS may analyses of the in blood samples, but to plasma samples are have to collect and plasma from of before they are dried on and may the value of biobanks of dried blood samples blood plasma for medical is it PubMed Google Scholar, for in PubMed Scopus Google Scholar, blood plasma from to PubMed Scopus Google Scholar, blood plasma using 2015; PubMed Google Scholar, and their potential in blood collection to analysis of dried plasma 2015; PubMed Scopus Google Scholar, blood plasma PubMed Scopus Google Protein a DBS is important are to be Our levels of a of the analyzed proteins in samples collected from the compared with the of DBS, but the is for multiple have in DBS and as from the drying blood samples of blood using a paper sampling Google Scholar, O. N. The of drying on the of 2015; PubMed Scopus Google The we not in the from DBS, and will have on the suitability for storage of samples for protein analysis in the dried our analysis of up to 30 years dried and samples that the of the proteins over some could be even after 30 years of but for several proteins a levels was with half-lives of some protein levels as low as 10 dried samples stored at +4 °C or at −24 °C for 30 years, and of investigated proteins were significantly The for samples stored at °C for years was of the The analysis is by the fact that the sampled have protein but the protein levels is in storage the and the biobanks a of of protein levels over and we that a storage of −24 °C protein levels in dried blood samples +4 °C 80 and 93% of median protein abundance after 10 years of storage at +4 °C or at −24 °C, it that a storage have the of the DBS offer a over samples because of the very space for their storage in freezers. the and about in protein samples stored frozen either as or dried on paper several the of the in our (1) the dried samples whole blood the frozen samples were plasma or (2) The dried samples were stored at +4 °C or −24 °C whereas the samples were stored at °C. (3) The dried samples were from but the samples were from 50 to adult and was an correlation storage time and the of the sampled It may also be that in to the dried blood samples plasma samples stored frozen to for some the as as the frozen plasma or samples were for the in protein levels by of and of by analysis of in we that proteins are in plasma and samples stored at °C as DBS stored at either +4 °C or at −24 °C. a of years only of the analyzed proteins were significantly decreased stored at °C, compared with and for samples stored dried at +4 °C or at −24 °C, respectively. An of our study is that proteins full detectability to a in dried samples, of protein levels are the to preserve samples over either or our not a but we that of samples of levels of some of the proteins over as by storage are and it is that of protein levels at a are for samples stored dried or for protein in are and can be by to in and storage of Protein PubMed Scopus Google The of are to and of protein PubMed Scopus Google Scholar, important to protein PubMed Scopus Google Scholar, and of proteins in PubMed Scopus Google to proteins as drying have and and PubMed Scopus Google is that can be by of such as to samples by of of a protein by during Scopus Google It may prove to the of proteins in samples dried on paper by some of the processes to their The results of our study are encouraging as they suggest an inexpensive means to transport, and store biobanked samples as spots of either whole blood or plasma and dried on paper. Our main findings were that (1) the act of drying only slightly influenced detection of blood proteins with an correlation of levels in the and dried and the was reproducible with a correlation of samples of either or samples, (2) detection of some proteins was not significantly affected by storage over the full range of three This was for and 76% of the analyzed proteins in dried samples stored at +4 °C and −24 °C, levels of others decreased slowly during storage with half-lives in the range of 10 to 50 years, (3) detectability of proteins was affected in dried samples stored at +4 °C compared with at −24 °C, for the median protein abundance at 80 and 93% of starting levels after 10 years of storage at +4 °C and −24 °C, and protein levels in samples stored at °C compared with DBS stored at −24 °C. of the analyzed proteins significantly decreased stored at °C for years compared with for proteins in dried samples stored at −24 °C for 30 years, that of storage is a factor. Combined with means to measure hundreds and thousands of protein, such biobanks could prove of great medical value by discovery of biomarkers for detection of as well as for with important for for with multiplex and and for to the with

Background A number of factors regarding blood collection, handling and storage may affect sample quality. The purpose of this study was to assess the impact on plasma protein profiles by delayed centrifugation and plasma separation and multiple freeze-thaw cycles. Methods Blood samples drawn from 16 healthy individuals were collected into ethylenediaminetetraacetic acid tubes and kept either at 4 °C or 22 °C for 1-36 h prior to centrifugation. Plasma samples prepared 1 h after venipuncture were also subjected to two to eight cycles of freezing at -80 °C and thawing at 22 °C. Multiplex proximity extension assay, an antibody-based protein assay, was used to investigate the influence on plasma proteins. Results Up to 36 h delay before blood centrifugation resulted in significant increases of 16 and 40 out of 139 detectable proteins in samples kept at 4 °C or 22 °C, respectively. Some increases became noticeable after 8 h delay at 4 °C but already after 1 h at 22 °C. For samples stored at 4 °C, epidermal growth factor (EGF), NF-kappa-B essential modulator, SRC, interleukin 16 and CD6 increased the most, whereas the five most significantly increased proteins after storage at 22 °C were CD40 antigen ligand (CD40-L), EGF, platelet-derived growth factor subunit B, C-X-C motif chemokine ligand 5 and matrix metallopeptidase 1 (MMP1). Only matrix metallopeptidase 7 (MMP7) decreased significantly over time and only after storage at 22 °C. No protein levels were found to be significantly affected by up to eight freeze-thaw cycles. Conclusions Plasma should be prepared from blood after a limited precentrifugation delay at a refrigerated temperature. By contrast, the influence by several freeze-thaw cycles on detectable protein levels in plasma was negligible.

Abstract Digital PCR provides high sensitivity and unprecedented accuracy in DNA quantification, but current approaches require dedicated instrumentation and have limited opportunities for multiplexing. Here, we present an isothermal platform for digital enumeration of DNA reaction products in multiplex via standard fluorescence microscopy. Circular DNA strands, which may result from a wide range of molecular detection reactions, are captured on streptavidin-coated surfaces via hybridized biotinylated primers, followed by rolling circle amplification (RCA). The addition of 15% polyethylene glycol 4000 during RCA resulted in uniform, easily recorded reaction products. Immobilized DNA circles were visualized as RCA products with 100% efficiency, as determined by droplet digital PCR. We confirmed previous reports about the influence on RCA by sequence composition and size of RCA templates, and we developed an efficient one-step restaining procedure for sequential multiplexing using toehold-triggered DNA strand displacement. Finally, we exemplify applications of this digital readout platform by demonstrating more than three orders of magnitude improved sensitivity by digital measurement of prostate specific antigen (PSA) (detection threshold ∼100 pg/l), compared to a commercial enzyme-linked immunosorbent assay (ELISA) with analogue readout (detection threshold ∼500 ng/l), using the same antibody pair.