Development and validation of a multiplex real-time PCR assay for detection and quantification of Streptococcus pneumoniae in pediatric respiratory samples

ABSTRACT Streptococcus pneumoniae (Spn) is a bacterial pathogen that causes a range of disease manifestations in children, from acute otitis media to pneumonia, septicemia, and meningitis. Primary Spn laboratory diagnostic identification methods include culture, antigen testing, single-plex real-time PCR, and syndromic PCR panels. However, each method lacks sensitivity, specificity, and/or cost efficiency. We developed and validated a quantitative, multiplex PCR assay using three Spn genomic targets (lytA, piaB, and SP2020) for improved sensitivity and specificity to detect Spn in pleural fluid (PF), bronchoalveolar lavage (BAL), tracheal aspirate (TA), and upper respiratory (UR, research only) samples. Validation testing included analytical sensitivity (limit of detection), specimen storage, analytical specificity (cross-reactivity), and accuracy studies. Limit of detection is 500 genome copies/mL in lower respiratory samples and 100 copies/mL in upper respiratory specimens, with a quantification range of 1,000 to 10,000,000 copies/mL. Specimens can be stored frozen at least 60 days and Spn DNA is stable through three freeze-thaw cycles. No cross-reactivity was observed against 20 closely related microorganisms and/or microorganisms that can be detected in similar sample types, including Streptococcus pseudopneumoniae. In testing of residual clinical specimens, Spn was detected in 5 of 23 (21.7%) PF, 2 of 19 (10.5%) BAL, 1 of 20 (5.0%) TA, and 44 of 178 (24.7%) UR residual specimens. For accuracy studies, 98 specimens were tested and overall percent agreement with a qualitative, lytA-based comparator assay was 96.9% across all sample types. This multiplex, quantitative PCR assay is a sensitive and specific method for Spn detection in pediatric respiratory samples. IMPORTANCE Streptococcus pneumoniae (Spn) is the world’s leading cause of lower respiratory tract infection morbidity and mortality in children. However, current clinical microbiological methods have disadvantages. Spn can be difficult to grow in laboratory conditions if a patient is pre-treated, and Spn antigen testing has unclear clinical utility in children. Syndromic panel testing is less cost-effective than targeted PCR if clinical suspicion is high for a single pathogen. Also, such testing entails a full, expensive validation for each panel target if used for multiple respiratory sources. Therefore, better diagnostic modalities are needed. Our study validates a multiplex PCR assay with three genomic targets for semi-quantitative and quantitative Spn molecular detection from lower respiratory sources for clinical testing and from upper respiratory sources for research investigation.

Most of the funding and resources were provided by the Department of Laboratory Medicine and Pathology at Children's Hospital Colorado.A portion of the funding and resources was provided by Pfizer (New York, NY) for a separate, collaborative project, of which the data and conclusions are not included in this manuscript.S.A.J. has received support from DiaSorin Molecular LLC, Karius Inc, and bioMérieux for conference attendance.DiaSorin Molecular LLC and Abbott Molecular have partnered with the Children's Hospital CO Clinical Microbiology Laboratory and have provided reagents and supplies.M.B. received support from Pfizer for conference attendance.Pfizer has partnered with the Children's Hospital CO Clinical Microbiology Laboratory and provided reagents and supplies.S.R.D. received grant support from Pfizer and BioFire Diagnostics and serves as a consultant for Karius and BioFire Diagnostics.K.H. receives research support from Pfizer.
of disease states including acute otitis media, pneumonia, bacteremia, and meningitis (1)(2)(3).As of 2016, Spn was the leading cause of lower respiratory tract infection (LRTI) morbidity and mortality worldwide (1,2,4).When compared with other common pathogens associated with LRTI (Haemophilus influenzae, influenza virus, and respiratory syncytial virus), Spn was found to cause more deaths than these three other pathogens combined (4).
A necessary prerequisite for invasive pneumococcal disease is Spn upper respiratory colonization (1).Rates of nasopharyngeal (NP) carriage of Spn are highest in children aged 2 to 3 years, reaching near or above 50%, although exact colonization rates vary by geographic location, community vaccination status, and socioeconomic factors (1,5).Children less than 5 years have the highest Spn colonization rates and have the highest Spn-associated morbidity and mortality compared to other age groups (4).
Accurate Spn laboratory identification from lower respiratory sources during confirmed or suspected LRTI can aid in clinical management and initiation of appropriate antimicrobial therapy.However, Spn detection in upper respiratory sources could identify Spn colonization instead of diagnosing true infection and could lead to unneeded antimicrobial therapy.
Current commercial and research Spn identification methods include standard microbiology culture techniques, antigen testing, real-time PCR, and syndromic multiplex PCR panels.Each of these methods has certain disadvantages.The viability of Spn to grow in culture deteriorates in standard laboratory storage conditions due to quorum sensing regulated-autolysis and is frequently hindered by antibiotic pretreat ment (6)(7)(8).If culture growth occurs, phenotypic identification techniques, including optochin susceptibility and bile solubility, are characteristics shared by other members of the Streptococcus mitis group, and some Spn isolates may exhibit optochin resistance or bile insolubility (9).Antigen testing, most commonly performed on urine specimens, has unclear clinical utility in diagnosing pneumonia in children due to relatively high Spn colonization rates (10).Commercial multiplex syndromic PCR panels with capability to detect multiple pathogens are valuable tools for rapid diagnosis, but often lack cost effectiveness, particularly if suspicion is high for a specific pathogen and are often restricted to only the sample types that have received U.S. Food and Drug Administration approval.Validation of additional specimen types by clinical laboratories can be lengthy and expensive since each panel target requires its own validation per sample type.
Distinction between Spn and closely related species, particularly Streptococcus pseudopneumoniae, by single-plex real-time PCR is challenging due to the high degree of genetic transfer between and within Spn isolates, closely related bacterial species, and co-colonizing organisms of the upper respiratory tract (9,11).Spn gene targets commonly used in PCR include lytA-which has been identified in several Streptococcus mitis isolates-and piaB, which is not always encoded by Spn isolates (9,(12)(13)(14)(15).Recent work to identify genomic markers universally and exclusively found within Spn isolates identified SP2020, a genomic region encoding a putative transcriptional regulator (14,15).
Given the high degree of genetic diversity within Spn isolates and the potential for any given single gene target to be found in a non-Spn species, we designed a multiplex real-time PCR which targets three Spn genomic regions (lytA, piaB, and SP2020) for improved sensitivity and specificity to detect Spn in respiratory samples.In this study, we evaluated the performance of this assay in lower respiratory specimens including pleural fluid (PF), bronchoalveolar lavage (BAL), and tracheal aspirate (TA) samples for clinical purposes, as well as in upper respiratory specimens collected in saline, skim milk tryptone glucose glycerol [STGG, media type used for Spn recovery (16)], or viral/ universal transport media (VTM/UTM) for research purposes.

Center and laboratory description
This study was conducted in the Clinical Microbiology Laboratory at Children's Hospital Colorado (CHCO).CHCO is a quaternary academic pediatric hospital with greater than 500 beds serving the Denver metropolitan area, the state of Colorado, and seven surrounding states.Residual specimens used in this study were samples previously collected, tested, and reported for clinical and research purposes.Study investigators conducted retrospective chart review from the electronic health record.Approval was obtained from the Colorado Multiple Institutional Review Board (COMIRB # 21-4052).

Reference and clinical material
The reference strain used in this study was Streptococcus pneumoniae serotype 19F (ATCC 49619).Quantified whole organism Spn (strain 19F, reference number 0801439) used for analytical sensitivity (limit of detection, LOD) studies and construction of standard curve was obtained from ZeptoMetrix (Buffalo, NY, USA).The titer of this material was determined by culture techniques; thus, stock concentration reported by the manufac turer is in colony forming units per milliliter (CFU/mL).However, a more appropriate unit for reporting concentration by nucleic acid amplification assays is genome copies per mL (copies/mL).Therefore, we relied on the hypothesized equivalency that 1 CFU = 1 genome and assumed all specimens used for LOD and standard curve construction were equivalent to clinical and contrived specimens used for analytical testing.
Clinical Spn isolates used in this study were isolated in the CHCO Clinical Microbi ology Laboratory from TA, BAL, cerebrospinal fluid (CSF), and blood cultures.Isolates grown from TA, CSF, or BAL fluid were identified as Spn by Gram stain, catalase testing, hemolysis characterization, and optochin sensitivity.Isolates grown in blood culture were identified as Spn by the BioFire FilmArray Blood Culture Identification Panel (BCID) or Blood Culture Identification Panel 2 (BCID2, BioFire Diagnostics, Salt Lake City, UT, USA).Viridans group streptococci clinical isolates used for specificity testing were identified by Gram stain, hemolysis characterization, optochin sensitivity, and MALDI-TOF MS.Clinical samples used for analytical sensitivity, positivity rate in residual clinical specimens, and accuracy studies for lower respiratory sample types were randomly selected residual PFs, BALs, and TAs previously collected for routine clinical testing.Residual clinical specimens were tested by Spn PCR initially to determine negativity prior to use in analytical sensitivity and other validation studies.Residual clinical samples used for positivity rate and accuracy studies from upper respiratory sample types were residual NP swab samples in sterile normal saline (0.9% NaCl) or UTM/VTM (Copan, Murrieta, CA, USA), previously tested on the BioFire Respiratory Pathogen Panel 2.1 test (RPP, BioFire Diagnostics, Salt Lake City, UT, USA).STGG was prepared in-house and composed of skim milk (2% wt/vol), tryptic soy broth (3% wt/vol), glucose (0.5% wt/vol), and glycerol (10% vol/vol) and sterilized by autoclave. Non

PCR protocol
Nucleic acid, including internal positive control DNA (Exogenous Internal Positive Control Reagents, ThermoFisher, Waltham, MA, USA), was isolated on Qiagen EZ1 Advanced XL platform using the DNA Tissue Kit (Qiagen, Hilden, Germany) and subjected to real-time PCR analysis on an ABI 7500 real-time PCR system (ThermoFisher, Waltham, MA, USA).PCR reaction components included PerfeCTa qPCR Tough Mix, Low ROX (Quantabio, Beverly, MA, USA), forward primer, reverse primer, and probe for each of the three Spn targets at final reaction concentration of 200 nM, internal positive control primers and probe mix (ThermoFisher, Waltham, MA, USA), and molecular grade water for a 25-µL reaction volume.Primer and probe sequences are listed in Table 1.Real-time PCR cycling conditions were 95°C × 10 min, followed by 45 cycles of 95°C × 15 s and 60°C × 1 min.

Interpretation criteria
A C T <40 was considered positive amplification for each Spn target (lytA, piaB, and SP2020).A sample was then considered Spn-positive if at least 2 of the 3 targets were detected with C T <40.A sample was considered negative if 0-1 of the 3 Spn targets were detected with C T <40, and the internal positive control C T value was <35.A sample was considered invalid if there was no amplification of any Spn targets, and the internal positive control C T value was ≥35.

Accuracy
The comparator assay used for accuracy studies is a lytA-targeting qualitative real-time PCR for Spn molecular detection (17,19).Quantification of positive samples sent for accuracy studies ranged from less than 1,000 copies/mL to greater than 10,000,000 copies/mL.Samples sent included both clinical and contrived specimens in lower respiratory matrices (PF) and upper respiratory matrices (saline and STGG).A summary of sample type distribution is presented in Table 2.

Analytical sensitivity
To determine analytical sensitivity, we prepared two independent dilution series of Spn in each matrix (PF, BAL, TA, saline, STGG, and VTM), ranging from 10 7 to 10 1 copies/mL.At least 10 replicates at each concentration were tested over two consecutive days.To establish interpretation criteria, we evaluated the relationship between each of three targets over a range of concentrations from 10 7 through 10 2 copies/mL (Fig. 1A through  C). piaB had the lowest efficiency of the three targets, but there was a high degree of correlation (R 2 >0.9) between the C T values of each of the three targets when compared pairwise across the range of concentrations (Fig. 1A through C).At a concentration of 100 copies/mL in saline and STGG, we observed consistent amplification of lytA and SP2020 (positive in 24 of 24 replicates, 100%) with inconsistent piaB amplification (positive in 13 of 24 replicates, 54.2%), which confirms that this target is the least sensitive of the three.We therefore established positivity criteria to be amplification of at least two targets before 40 cycles.Using these positivity criteria, we found 100% Spn detection in all sample types at 1,000 copies/mL.For lower respiratory matrices (PF, BAL, and TA), Spn was detected in  100% of replicates at 500 copies/mL, but we observed less reliable detection at 100 copies/mL (<90%).For upper respiratory matrices (saline, VTM, and STGG), we observed 100% detection at 100 copies/mL but minimal detection at 10 copies/mL (<10%).Therefore, we established the LOD to be 500 copies/mL in lower respiratory sample types and 100 copies/mL in upper respiratory sample types (Table 3).Due to the reliably lower C T values of SP2020 compared to the other two targets at each tested concentration, we elected to use SP2020 C T values for construction of a standard curve to determine Spn quantification.Comparison of polynomial and linear regression revealed a loss of linearity at the highest concentration tested (1 × 10 8 copies/mL, data not shown), which led us to establish an upper limit of quantification (LOQ) at 1 × 10 7 copies/mL.We selected a lower LOQ of 1 × 10 3 copies/mL because it was the lowest concentration tested at which the standard deviation (SD) of SP2020 C T values was less than 1 C T value.Standard curves constructed in each of the matrices analyzed were consistent, with a y-intercept range from 39.9 to 41.9 and a slope range from −3.6 to −3.3 (Fig. 2).Lower respiratory matrices (PF, BAL, and TA) had consistently higher C T values across the concentration range, likely due to the presence of PCR inhibitors in these sample types.
We also assessed assay sensitivity against Spn clinical isolates.These clinical isolates were previously identified by our laboratory as Spn by standard microbiology culture techniques (n = 13) or by BCID/BCID2 (n = 9).Of these, 21/22 (95.5%) were Spn-positive by PCR.The single isolate that was identified by standard culture techniques to be Spn but was PCR-negative was sent to an independent, reference laboratory and was identified as "S.mitis-group, not S. pneumoniae." For the 21 isolates that were Spn PCR-positive, amplification of all three targets (lytA, piaB, and SP2020) was observed.There was a single isolate that was determined to not be Spn, and there was no amplification of any of the targets.

Specimen storage and precision
To evaluate acceptable specimen storage conditions, stocks of high-, medium-, and low-positive and negative samples in saline and STGG were stored at room tempera ture, 4°C, −20°C, and −70°C.High-positive samples were defined as those greater than 100,000 Spn copies/mL, medium-positive defined as Spn 1,000-20,000 copies/mL, and low-positive defined as less than Spn 1,000 copies/mL (below established LOQ).Stocks at room temperature were tested at 24 h and 48 h; stocks at 4°C were tested at 4 days, 7 days, and 14 days; stocks at −20°C were tested at 10 days, 21 days, and 30 days; and stocks at −70°C were tested at 10 days, 21 days, 30 days, and 60 days.Stocks tested at frozen temperatures underwent a freeze-thaw cycle at each tested time point.Overall SD for each gene target C T value was less than 1.5 C T (data not shown) and overall SD for quantification was within 0.5 log 10 (copies/mL) across all storage temperatures and durations (Fig. 3).However, there was a notable increase in C T values for all gene targets after a fourth freeze-thaw cycle (samples stored at −70°C for 60 days), corresponding to a reduction in calculated copies/mL (Fig. 3, darkest blue "+" symbols).These data suggest that samples should not be subjected to more than three freeze-thaw cycles to maintain quantification accuracy.
For reproducibility studies, high-, medium-, and low-positive control samples, as well as negative controls, prepared in saline or STGG, were independently extracted and analyzed by real-time PCR over 6 days by three independent technologists.Variability in C T value was the highest among low-positive samples.However, in both matrices overall SD was less than 1 C T value (data not shown) and within 0.5 log 10 (copies/mL) (Fig. 4).

Analytical specificity
We performed cross-reactivity studies against organisms genetically related to Spn, as well as organisms likely to be found in similar sample types.Tested three times each, we observed no cross-reactivity with the following streptococcal species: S. mitis, S.  salivarius group, and S. sanguinis group-were tested in triplicate and no cross-reactivity was observed.There was no amplification of any of the three targets (lytA, piaB, or SP2020) against any of the above organisms.Therefore, none of the tested specimens or isolates would have met Spn detection criteria.
Due to the high genetic similarity between Spn and Streptococcus pseudopneumo niae, we performed a more extensive cross-reactivity investigation between these two streptococcal species.We tested a range of S. pseudopneumoniae concentrations over multiple days for a total of 30 replicates from ATCC strain BAA-960.We observed limited and sporadic (not associated with concentration) single-target amplification with S. pseudopneumoniae.Late amplification of the lytA target occurred in 6 of 30 replicates (20.0%), with C T values greater than 40.Positive amplification was also observed with the SP2020 target in 1 of 30 replicates (3.3%), with a C T value less than 40 (C T = 36.4).There was no detectable amplification of the piaB target in any of the 30 replicates.Despite some, but minimal, single-target amplification (lytA, SP2020) against S. pseudopneumo niae, none of the 30 replicates would have met Spn detection criteria, thereby supporting the utility of a multiplex approach for specific Spn detection.a Specimens that were not previously tested by routine culture were excluded (n = 2 BALs, 7 TAs); specimens which resulted as "invalid" were excluded (n = 1 PF, 2 BALs, 1 TA).identified through routine culture in both Spn PCR-positive BAL specimens but not in the Spn PCR-positive TA.In each of the three lower respiratory matrices (PF, BAL, TA), we observed at least one instance of PCR inhibition, with 4 of 62 (6.5%) specimens resulted as "Invalid." The comparison between Spn PCR and culture in lower respiratory specimens is summarized in Table 4. Additionally, we tested 178 residual NP swab specimens in saline or universal transport media (UTM) that had been previously tested by RPP.Samples were collected over 6 months from patients 5 years or younger.Spn was PCR-detected in 44 of 178 specimens (24.7%).It should be noted that NP swabs are not cultured in our laboratory routinely, thus culture comparison data are not available.Positivity rates across sample types for randomly selected residual clinical specimens are summarized in Table 5.

Accuracy
Samples representing a range of high-, medium-, and low-positive results, as well as negative samples, were sent to an independent laboratory for accuracy studies.The comparator assay is a lytA-based, qualitative PCR for detection of Spn (17,19).Positive percent agreement among PF samples was 83.3% and negative percent agreement was 100%.The single discordant PF sample (Spn PCR negative but comparator assay positive) had an unexpectedly high C T value (37.1) for the internal positive control, indicative of PCR inhibition.The comparator lytA C T value for this specimen was 29.9 (C T positivity cutoff = 30).For saline samples, there was both 100% negative and positive percent agreement in qualitative interpretation.STGG samples also had a 100% positive agreement.Two STGG samples classified as low-positive (less than 1,000 copies/mL) by our Spn PCR were classified as negative by the comparator assay, yielding a 91.3% negative agreement.For these two specimens, lytA C T values were 32.7 and 35.9; piaB C T values were greater than 40; and SP2020 C T values were 32.2 and 34.2.Positive percent agreement across all specimen sources was 98.0%, negative percent agreement was 95.7%, and overall percent agreement was 96.9% (Table 6).
Though an imperfect comparison due to the multiplex nature of our assay, we found an acceptable degree of correlation between our lytA C T values and the lytA C T values produced by the comparator assay (PF R 2 = 0.91, saline R 2 = 0.77, STGG R 2 = 0.73) (Fig. 5).

DISCUSSION
Accurate and reliable Spn detection is essential for appropriate disease management in lower respiratory sources and for understanding colonization dynamics in upper respiratory sources.Despite this need, traditional Spn microbiological methods (primarily culture and antigen testing), as well as current PCR methods, lack sensitivity and specificity due to both high genetic diversity within and between Spn strains and characteristic phenotypic and genotypic Spn elements in non-Spn species (9,12,15,20).To address this need, we developed and validated a quantitative real-time PCR targeting three Spn genomic regions for improved sensitivity and specificity for detecting Spn in respiratory specimens.The multiplex approach described in this study is designed to limit the disadvantages of any single PCR target.The piaB target is the most specific target since no amplification was observed when tested against any other organism.However, the piaB target is the least sensitive of the three, with inconsistent amplification at low (100 copies/mL) Spn concentrations.The other two targets, lytA and SP2020, have higher sensitivity than piaB, but our studies and others have demonstrated that they can cross-react with other Streptococcus species (14,15).Therefore, our multiplex approach, with positive Spn detection requiring amplification of at least two targets, provides a balance between sensitivity and specificity.This assay detects as few as 500 Spn genome copies/mL in lower respiratory specimens and quantification was linear across 6 orders of magnitude.Quantification remained consistent after storage in typical laboratory conditions and through three freeze-thaw cycles.Cross-reactivity was not observed against any organism tested.
We evaluated the assay's performance in PF, BAL, and TA specimens as well as in upper respiratory specimens (NP swab samples collected in saline, UTM/VTM, or STGG).The validation of this assay in both lower and upper respiratory specimens allows its use to address clinical and research questions.In clinical contexts, accurate and timely Spn identification in lower respiratory specimens is important for initiation of appropriate antimicrobial therapy and medical management of pneumococcal disease.Using our assay, we detected Spn in lower respiratory samples from which Spn was not identified through routine laboratory testing.Due to the success of this validation study, Spn PCR from PF, BAL, and TA specimens is now available for clinical use in our laboratory.Quantification of Spn from lower respiratory specimens is reported in semi-quantitative ranges due to consistent elevation of SP2020 C T values across the quantifiable range (Fig. 2).The semi-quantitative ranges that we use for clinical reporting are: less than 1,000 copies/mL, 1,000-10,000 copies/mL, 10,001-100,000 copies/mL, 100,001-1,000,000 copies/mL, and greater than 1,000,000 copies/mL.
For research purposes, this Spn PCR may be used for understanding Spn colonization dynamics in the upper respiratory tract.Understanding these dynamics is an area of ongoing research since Spn colonization is a necessary prerequisite for Spn disease.Additionally, there are several described relationships between Spn colonization and viral respiratory disease, which have important implications for invasive disease prevention (1,5,(21)(22)(23)(24)(25)(26)(27).Application of our assay may include characterizing Spn detection and bacterial load in the upper respiratory tract and association with viral co-infection and incidence of invasive pneumococcal disease.Due to concerns regarding detection of non-pathogenic colonization resulting in unnecessary antibiotic use, our institution uses this assay only for research purposes for upper respiratory specimen types (1,2).
This study has several notable limitations.First, this study was conducted at a single institution and is therefore restricted to Spn strains available commercially (ATCC) or which are currently circulating in the Denver, CO geographic area.Additionally, this study was limited to a pediatric patient population, and there are known age-related differences in Spn serotype prevalence (28).This study was conducted during wide spread circulation of SARS-CoV-2, which has unknown implications on Spn colonization and disease dynamics (21).As with all nucleic acid amplification tests, the assay's sensitivity and specificity may be affected by genetic variation in the target organism and will be routinely evaluated against currently circulating strains.The inclusion of three Spn gene targets in this assay should aid in the early identification of such issues if single-target drop-out is observed with increasing frequency.In such a case, primerprobe sequences may be updated to maintain performance of this assay.In conclusion, this laboratory-developed test showed robustness against all studies performed.The performance of this assay is acceptable for clinical and research purposes from pediatric lower and upper respiratory sources, respectively.

FIG 1
FIG 1 Pairwise comparison of C T values of lytA (A), SP2020 (B), and piaB (C) over the range of 10 7 to 10 2 Spn copies/mL.At least 10 replicates were tested at each concentration in each matrix (PF, BAL, TA, saline, STGG, and VTM).Replicates meeting detection criteria (at least two targets with C T <40) indicated by blue circles.Replicates not meeting detection criteria indicated by red X marks.No amplification indicated as C T = 45.R 2 as indicated for each pairwise comparison calculated from replicates meeting detection criteria only.

FIG 2
FIG 2 Average SP2020 C T values for each tested concentration of standard for each matrix with indicated standard curve and R 2 value.N = at least 10 replicates at each concentration in each matrix.

FIG 3
FIG 3 Calculated Spn quantification in high-, medium-, and low-positive samples prepared in saline (A) or STGG (B).Storage duration indicated by color; storage temperature in °C indicated by shape.Each sample at each storage condition was tested in triplicate.Mean ± SD indicated for each sample.

FIG 4
FIG 4 Calculated Spn quantification in high-, medium-, and low-positive control samples prepared in saline (A) or STGG (B).Each independent run indicated by a unique color.Samples during each run were tested in triplicate.Mean ± SD indicated for each sample.

We tested 23 FIG 5
FIG 5 Comparison of lytA C T values from the comparator assay and this assay.y = x indicated by black line.

TABLE 1
Primer and probe sequences

TABLE 2
Samples sent for accuracy studies a a A total of 98 samples were sent for testing, representing both clinical and contrived specimens.Quantification range of Spn PCR positive samples was <1,000 copies/mL to >10,000,000 copies/mL.StatisticsStatistical analyses, including linear regression and coefficient of determination, were completed in R version 4.0.3.

TABLE 3
Percentage of replicates (n = at least 10) in each matrix that met Spn detection criteria (at least two targets with C T <40) a a NT indicates not tested.

TABLE 4
Comparison between Spn PCR and standard culture of residual lower respiratory specimens (n = 22 PFs, 15 BALs, and 12 TAs) a

TABLE 5
Positivity rate results for PF, BAL, and TA, and upper respiratory (NP swab) sample types

TABLE 6
Qualitative agreement between this multiplex Spn PCR and comparator, lytA-based PCR