Home ▶ Products and Services ▶ Clinomics TrioDx RT-PCR COVID-19 Test (FDA EUA)

1. Performance Evaluation (in Instructions for Use)

• Limit of Detection (Analytical Sensitivity)

The LoD is defined as the lowest concentration of SARS-CoV-2 that can be reliably detected at least 95% of the time. Three LoD studies were performed to determine the analytical sensitivity of the TrioDx: (1) a preliminary study with AccuPlex materials, (2) a confirmatory study with AccuPlex materials, and (3) a study with heatinactivated virus (BEI Resources). The materials used for this study are shown in Table 12 below. Table 12. Materials used in Limit of Detection Study
Required Material Product Name Catalog # Manufacturer Note
1 Recombinant virus with
reference RNA sequence
(SARS-CoV-2)
AccuPlex™ SARS-CoV-2 Reference
Material Kit
0505-0126 Seracare 4.23 x 103 cp/mL
2 Heat-inactivated virus SARS-Related Coronavirus 2, Isolate USA-WA1/2020 NR-52286 BEI Resources: 1.16 x 109 cp/mL
3 NP and OP swabs Nasopharyngeal/Oropharyngeal Swabs
+ 2mL CTM (Clinical Viral Transport Medium)
UTNFS-3B-2 Noble
Biosciences Inc.
N/A
(a) LoD studies with Recombinant Virus (Accuplex) The preliminary study was conducted using the AccuPlex™ SARS-CoV-2 Reference Material, which is a recombinant virus containing SARS-CoV-2 sequences, to evaluate the sensitivity of the TrioDx test. Reference material was serially diluted (2-fold) and spiked into 140 uL VTM/UTM containing negative nasopharyngeal swab specimen to create 5 different viral copy-number concentrations. Each dilution was tested in 5 replicates. The nucleic acids were purified using each of the following:

(1) QIAamp Viral RNA Mini Kit (manual)

(2) Maxwell RSC Viral Total Nucleic Acid Kit with the Maxwell RSC 48 instrument

(3) MagMAX Viral/Pathogen II Nucleic Acid Isolation Kit with KingFisher Flex with 96 Deepwell Head Instrument.

Tables 13, 14, and 15 show the detection rate at 5 tentative LoD concentrations for each extraction method.

The positive rate was determined per sample dilution based on the test interpretation, i.e., a positive result in any gene is positive. Based on these results, the tentative LoD was determined to be 0.125 copies/μL. Although each gene at this concentration, did not have 5/5 positive replicates, for each replicate one of the genes was positive and therefore the sample is positive.
Table 13. Preliminary LoD study results from RNA extraction using QIAamp Viral RNA Mini Kit (Manual Prep) - AccuPlex
Concentration 
RNA copies/uL
Assay 1 (RdRp gene) Assay 2 (E gene) Assay 3 (N gene) Using all 3 genes
Mean Ct Standard Deviation (SD) Detection Rate Mean Ct Standard Deviation (SD) Detection Rate Mean Ct Standard Deviation (SD) Detection Rate Detection Rate
2 32.6 0.2 5/5(100%) 32.5 0.2 5/5(100%) 31.7 0.2 5/5(100%) 5/5(100%)
1 33.6 0.3 5/5(100%) 33.8 0.2 5/5(100%) 33.3 0.7 5/5(100%) 5/5(100%)
0.5 34.9 0.5 5/5(100%) 34.9 0.7 5/5(100%) 34.4 0.9 5/5(100%) 5/5(100%)
0.25 36.2 0.9 5/5(100%) 36.8 1.5 4/5(80%) 36.1 1.6 4/5(80%) 5/5(100%)
0.125 36.5 0.8 5/5(100%) 37.9 1.3 3/5(60%) 36.0 0.3 5/5(100%) 5/5(100%)
Table 14. Preliminary LoD study results from RNA extraction using Maxwell RSC Viral Total Nucleic Acid Kit with Maxwell RSC 48 instrument (Auto Prep) - AccuPlex
Concentration
RNA copies/uL
Assay 1 (RdRp gene) Assay 2 (E gene) Assay 3 (N gene) Using all 3 genes
Mean Ct Standard Deviation (SD) Detection Rate Mean Ct Standard Deviation (SD) Detection Rate Mean Ct Standard Deviation (SD) Detection Rate Detection Rate
2 30.7 0.7 5/5(100%) 32.0 0.1 5/5(100%) 31.0 0.5 5/5(100%) 5/5(100%)
1 32.3 0.5 5/5(100%) 33.1 0.4 5/5(100%) 31.4 0.2 5/5(100%) 5/5(100%)
0.5 33.1 0.5 5/5(100%) 34.1 0.5 5/5(100%) 33.0 0.4 5/5(100%) 5/5(100%)
0.25 34.6 0.5 5/5(100%) 35.9 0.8 5/5(100%) 34.3 1.0 5/5(100%) 5/5(100%)
0.125 35.2 0.7 5/5(100%) 36.5 1.1 4/5(80%) 33.8 0.6 3/5(60%) 5/5(100%)
Table 15. Preliminary LoD study results from RNA extraction using MagMAX Viral/Pathogen II Nucleic Acid Isolation Kit with KingFisher Flex instrument (Auto Prep) - AccuPlex
Concentration
RNA copies/uL
Assay 1 (RdRp gene) Assay 2 (E gene) Assay 3 (N gene) Using all 3 genes
Mean Ct Standard Deviation (SD) Detection Rate Mean Ct Standard Deviation (SD) Detection Rate Mean Ct Standard Deviation (SD) Detection Rate Detection Rate
2 31.6 0.5 5/5(100%) 32.5 0.7 5/5(100%) 31.2 1.0 5/5(100%) 5/5(100%)
1 33.4 2.2 5/5(100%) 34.0 1.2 5/5(100%) 31.6 1.2 5/5(100%) 5/5(100%)
0.5 33.7 0.5 5/5(100%) 34.0 0.6 5/5(100%) 32.6 0.3 5/5(100%) 5/5(100%)
0.25 34.6 0.3 5/5(100%) 35.5 0.5 5/5(100%) 33.8 0.6 5/5(100%) 5/5(100%)
0.125 36.5 2.5 4/5(80%) 37.1 1.8 3/5(60%) 34.5 0.9 4/5(80%) 5/5(100%)
Confirmatory LoD Study with Accuplex Reference

A Confirmatory LoD study was then performed on 20 replicates for each extraction method. The LoD for each target was then determined to be the lowest concentration at which at least 19/20 replicates were detected since the FDA defines LoD as the lowest concentration at which 19/20 replicates are positive. The detection rate is shown for each extraction method individually, and the result is shown in the last column the detection rate using all three genes is provided as discussed above.

The final LoD (see Tables 16) was confirmed to be 0.125 copies/μL for all three extraction methods. At least one of the three target genes (RdRp, E or N) was successfully detected in 20 out 20 replicates (100%). Note that the E gene alone was detected at the lowest concentration in just 1 of the 20 samples for two of the extraction methods in Tables 16.
Table 16. Confirmatory LoD study results from three different RNA extraction methods - Accuplex
Concentration
RNA copies/uL
Assay 1 (RdRp gene) Assay 2 (E gene) Assay 3 (N gene) Using all 3 genes
Mean Ct Detection Rate Mean Ct Detection Rate Mean Ct Detection Rate Detection Rate
Extraction using QIAamp Viral RNA Mini Kit (Manual Preparation)
0.25 35.4 20/20(100%) 36.1 20/20(100%) 34.8 20/20(100%) 20/20(100%)
0.125 37.0 17/20(85%) 37.8 14/20(70%) 36.3 15/20(75%) 20/20(100%)
Extraction using Maxwell RSC Viral Total Nucleic Acid Kit with Maxwell RSC 48 instrument (Auto Preparation)
0.25 34.9 20/20(100%) 35.6 20/20(100%) 34.5 20/20(100%) 20/20(100%)
0.125 35.6 19/20(95%) 36.6 19/20(95%) 34.8 16/20(80%) 20/20(100%)
Extraction using MagMAX Viral/Pathogen II Nucleic Acid Isolation Kit with KingFisher Flex instrument (Auto Preparation)
0.25 35.7 19/20(95%) 36.0 20/20(100%) 34.6 20/20(100%) 20/20(100%)
0.125 37.4 16/20(80%) 36.3 18/20(90%) 35.9 15/20(75%) 20/20(100%)
Confirmatory LoD Study with Heat-inactivated Virus - Contrived Clinical Specimen

A bridging study was performed using whole heat inactivated virus (BEI Resources Product # NR-52286: SARSRelated Coronavirus 2, Isolate USA-WA1/2020) to ensure that the LoD of the assay is not significantly different with intact SARS-CoV-2 virus.

The BEI Resources NR-52286 Reference Material was serially diluted and spiked into 140uL VTM containing negative nasopharyngeal swab specimen (item 2 in Table 12). All specimens were isolated using 50-μL elution buffer. The nucleic acids were purified using the QIAamp Viral RNA Mini Kit. Five replicates were performed at each of the following: 1/3X LoD, 1X LoD, and 3X LoD, where the 1X LoD is taken as based on the observed 0.125 copies/uL.

The results are summarized in Table 17. 5/5 replicates were detected at the 1X LoD of 0.125 copies/uL. Consistent with the results with the Accuplex reference material, there was no detection at 1/3X LoD with the BEI reference material. The E gene alone was detected in one of the five replicates at 1X LoD.
Table 17. Confirmatory LoD study results using BEI reference. RNA extraction using QIAamp Viral RNA Mini Kit
LoD copies/uL RdRp gene E gene N gene Using all three genes
Mean Ct SD Rate Mean Ct SD Rate Mean Ct SD Rate Rate
3X 0.375 35.2 0.5 5/5 34.5 0.9 5/5 33.8 0.3 5/5 5/5
1X 0.125 36.9 1.0 4/5 36.4 0.7 5/5 35.2 1.3 3/5 5/5
1/3X 0.042 Undetermined - 0/5 37.9 - 1/5 Undetermined - 0/5 1/5
Final Reported LoD- summary of LoD for all three extraction methods

The final reported LoD is summarized below in Table 18. The LoD based on the test algorithm is in the bottom row of the table and shows the LoD is the same, 0.125 copies/μL, for all three extraction methods.
Table 18. Final reported LoD for each target for three different extraction methods
Gene QIAamp Viral RNA Mini Kit (Manual Prep) Maxwell RSC Viral Total Nucleic Acid Kit with Maxwell RSC 48 instrument (Auto Prep) MagMAX Viral/Pathogen II Nucleic Acid Isolation Kit with KingFisher Flex instrument (Auto Prep)
RdRp gene 0.25 copies/μL 0.125 copies/μL 0.25 copies/μL
E gene 0.25 copies/μL 0.125 copies/μL 0.25 copies/μL
N gene 0.25 copies/μL 0.25 copies/μL 0.25 copies/μL
Final reported LoD using all 3 genes 0.125 copies/μL 0.125 copies/μL 0.125 copies/μL

• Inclusivity (Analytical Sensitivity)

The inclusivity of the TrioDx was evaluated in silico by comparing the primer/probe sequences separately with each of the 18,927 Severe acute respiratory syndrome coronavirus 2 sequences published in NCBI and GISAID databases. A total of 9,467 publicly available SARS-CoV-2 sequences accessed via NCBI Virus SARS-CoV-2 data hub, and 9,460 sequences from GISAID (out of 64,286 sequences available as of July 14th, 2020) that are different from the NCBI database. Each was separately used for alignment with the TrioDx assay. The SARS-CoV-2 primers and probes for RdRp, E and N gene targets showed 100% homology with reference sequence (NC_045512.2). The inclusivity % calculated are summarized in Table 19. Table 19. Inclusivity test results
Database Target Genes % Homology
Forward Primer number of not 100% homology case Probe number of not 100% homology case Reverse Primer number of not 100% homology case
NCBI (n=9467) RdRp gene 99.92 % 8 99.89 % (CASE 1) 10 99.98 % 2
E gene 99.96 % 4 99.85 % 14 99.94 % 6
N gene 99.73 % 26 99.66 % 32 99.82 % 17
GISAID (n=9460) RdRp gene 99.86 % 13 99.57 % 41 99.98 % 2
E gene 99.92 % 8 99.76 % 23 99.88 % 11
N gene 99.75 % 24 99.53 % (CASE 2) 44 99.81 % 18
There was a total of 297 strains that did not demonstrate 100% homology. However, while the primers and probes in Table 19 do not all show 100% homology there was no case in which all of the RdRp, E, and N sequences showed less than 100% homology. Specifically:

• 289 strains contained single base pair mismatches in one of the primers and probes that would not impact amplification of the target due to its position. Note in general, one sequence mismatch does not critically impact primer amplification unless such a mismatch occurs in 3', the binding efficiency of the primer may decrease; however, such a case was not found.

• 6 strains contained one mismatch in more than one primer and probe that would not impact the amplification of the target due to its position.

• 2 strains contained multi-base pair mismatches that may impact the amplification of the target.
    ○ sequence MT372483.1 (Case 1 in Table 19), a total of 4 mismatches were found in RdRp Probe target region. In this case, four base pair mismatches were confirmed in the RdRp gene probe of a total length 24 oligonucleotide sequence. While the RdRp gene may not be detected, these sequences had 100% homology with the N gene and E gene, and a sample with this sequence would consequently still be detected as positive.
    ○ sequence “Iceland/603/2020” (Case 2 in Table 19; strain name used since GenBank accession number is unavailable), a total of 2 mismatches were found in N gene Probe target region. In this case, two base pair mismatches were confirmed in the N gene probe of a total length 24 oligonucleotide sequence. While the N gene may not be detected, these sequences had 100% homology with the RdRp gene and E gene, and a sample with this sequence would consequently still be detected as positive.

Therefore, the multi target design of the test mitigates false negative results due to mismatches in an individual target. Thus, when all three SARs-CoV-2 targets were checked according to the sponsors algorithm, there were no strains that could not be detected by PCR in the 18,927 strains evaluated.

Note: The consensus sequences generated from alignment with the SARS-CoV-2 genome sequences contain ambiguity characters (W, S, M, K, R, Y-refer to IUPAC notation) to represent two positional base variations present in the region. These were not counted as mismatches, as one of the two positional variations demonstrated by the ambiguity characters always matched with the TrioDx primer/probe nucleotide sequences.

Inclusivity for new variants as of January 25, 2021

Additional inclusivity of the TrioDx was evaluated in silico by comparing the primer/probe sequences separately with each of 9,275 new variant SARS-CoV-2 sequences published in GISAID databases (through January 25th, 2021).

Three representative Emerging Variants lineages were used from: https://www.cdc.gov/coronavirus/2019-ncov/more/science-and-research/scientific-brief-emerging-variants.html. This included 31 cases in P.1 (Brazil), 8,573 cases in B.1.1.7 (United Kingdom), and 671 cases in B.1.351 (South Africa). Each new variant SARS-CoV-2 sequence was separately used for alignment with the TrioDx assay.

The SARS-CoV-2 TrioDx primers and probes for RdRp, E and N gene targets showed 100% homology with reference sequence (NC_045512.2). The inclusivity is summarized in Table 20.
Table 20. Inclusivity test results for new variants: P.1 (Brazil), B.1.1.7 (United Kingdom), B.1.351 (South Africa)
Database Target Genes Homology
Forward Primer Number of sequences in databases without 100% homology Probe Number of sequences in databases without 100% homology Reverse Primer Number of sequences in databases without 100% homology
GISAID (n=9275) RdRp gene 99.99 % 1 99.81 % 18 99.96 % 4
E gene 99.36 % 59 99.92 % 7 100 % 0
N gene 99.62 % 35 98.71 %(CASE 3) 120 99.83 % 16
P.1 (Brazil) showed 100% homology in all Primer & Probe oligonucleotide sequences.
B.1.1.7 (United Kingdom) showed 237 strains without 100% homology.
B.1.351 (South Africa) showed 20 strains without 100% homology.

Thus, there was a total of 257 strains that did not demonstrate 100% homology. However, while the primers and probes in Table 19 do not all show 100% homology, there was no new variant for which all of the RdRp, E, and N sequences showed less than 100% homology. Specifically:

• 254 strains contained single base pair mismatches in one of the primers and probes that would not impact amplification of the target due to its position. Note that, in general, one sequence mismatch does not critically impact primer amplification unless such a mismatch occurs in 3', the binding efficiency of the primer may decrease; however, such a case was not found.

• 2 strains contained one mismatch in more than one primer and probe that would not impact the amplification of the target due to its position.

• 1 strain contained multi-base pair mismatches that may impact the amplification of the target.
    ○ sequence EPI_ISL_856362 (Case 3 in Table 19), a total of 2 mismatches were found in N gene Probe target region.
In this case, four base pair mismatches were confirmed in the N gene probe of a total length 24 oligonucleotide sequence. While the N gene may not be detected, these sequences had 100% homology with the RdRp gene and E gene, and a sample with this sequence would consequently still be detected as positive.

Therefore, the multi target design of the test mitigates false negative results due to mismatches in an individual target. In summary, when all three SARs-CoV-2 targets were checked according to the algorithm, there were no strains that could not be detected by PCR in the new variant strains evaluated.

Cross-reactivity (Analytical Specificity)

The Cross-Reactivity Studies (see Table 21) have been performed by running BLAST against 32 non-target microorganisms/ closely related virus strains on 7/14/20. No homology ≥ 80% homology was found in any tested organisms, except in SARS-coronavirus (SARS-CoV).
The “no amplicon formed” designation shown in Table 21 means that there is less 80% homology in the in silico analysis and hence the amplicon will not be generated during the PCR amplification.
Table 21. List of microorganisms tested for cross-reactivity via in silico
Category Microorganisms Taxid NCBI database RdRp gene E gene N gene
Other high priority pathogens from the same genetic family Human coronavirus 229E 11137 NC_002645.1 No amplicon formed No amplicon formed No amplicon formed
Human coronavirus OC43 31631 NC_006213.1 No amplicon formed No amplicon formed No amplicon formed
Human coronavirus NL63 277944 NC_005831.2 No amplicon formed No amplicon formed No amplicon formed
Human coronavirus HKU1 290028 NC_006577.2 No amplicon formed No amplicon formed No amplicon formed
SARS-coronavirus* 694009 NC_004718.3 >80% homology found in Probe sequence >80% homology found Forward and Reverse Primers and Probe >80% homology found in Forward Primer and probe
Middle East respiratory syndrome-related coronavirus (MERS-CoV) 1335626 NC_019843.3 No amplicon formed No amplicon formed No amplicon formed
High priority organisms likely present in a respiratory specimen Human Adenovirus C1 10533 AC_000017.1 No amplicon formed No amplicon formed No amplicon formed
Human Metapneumovirus (hMPV) 162145 NC_039199.1 No amplicon formed No amplicon formed No amplicon formed
Human parainfluenza virus 1 12730 NC_003461.1 No amplicon formed No amplicon formed No amplicon formed
Human parainfluenza virus type 2 1979160 NC_003443.1 No amplicon formed No amplicon formed No amplicon formed
Human parainfluenza virus type 3 11216 NC_001796.2 No amplicon formed No amplicon formed No amplicon formed
Human Parainfluenza virus 4a 11224 NC_021928.1 No amplicon formed No amplicon formed No amplicon formed
Human orthopneumovirus (Human respiratory syncytial virus A) 11250 NC_038235.1 No amplicon formed No amplicon formed No amplicon formed
Influenza A virus (A/PR 8/1934 (H1N1)) 211044 NC_002022.1 No amplicon formed No amplicon formed No amplicon formed
Influenza virus type B 11520 NC_002204.1 No amplicon formed No amplicon formed No amplicon formed
Human Enterovirus EV68 42789 NC_038308.1 No amplicon formed No amplicon formed No amplicon formed
Human orthopnemovirus (Human respiratory syncytial virus B) 11250 NC_001781.1 No amplicon formed No amplicon formed No amplicon formed
Human rhinovirus A1 573824 NC_038311.1 No amplicon formed No amplicon formed No amplicon formed
Chlamydia pneumoniae TW-183 182082 NC_005043.1 No amplicon formed No amplicon formed No amplicon formed
Haemophilus influenzae 727 CP007805.1 No amplicon formed No amplicon formed No amplicon formed
Legionella pneumophila NCTC12273 446 NZ_LR134380.1 No amplicon formed No amplicon formed No amplicon formed
Mycobacterium tuberculosis H37Rv 83332 NC_018143.2 No amplicon formed No amplicon formed No amplicon formed
Streptococcus pneumoniae 1313 NZ_LN831051.1 No amplicon formed No amplicon formed No amplicon formed
Streptococcus pyogenes 1314 NZ_LN831034.1 No amplicon formed No amplicon formed No amplicon formed
Bordetella pertussis 18323 568706 NC_018518.1 No amplicon formed No amplicon formed No amplicon formed
Mycoplasma pneumoniae FH 722438 NZ_CP010546.1 No amplicon formed No amplicon formed No amplicon formed
Pneumocystis jirovecii 42068 NW_017264778.1 No amplicon formed No amplicon formed No amplicon formed
Candida albicans 237561 GCF_000182965.3 (8chr)(XM_715226.) No amplicon formed No amplicon formed No amplicon formed
Pseudomonas aeruginosa DSM 50071 1123015 NZ_CP012001.1 No amplicon formed No amplicon formed No amplicon formed
Staphylococcus aureus subsp. Aureus NCTC 8325 93061 NC_007795.1 No amplicon formed No amplicon formed No amplicon formed
Staphylococcus epidermidis 1282 NZ_CP035288.1 No amplicon formed No amplicon formed No amplicon formed
Streptococcus salivarius 1304 NZ_LR134274.1 No amplicon formed No amplicon formed No amplicon formed
The in silico cross reactivity study (see Table 21) yielded only one species (SARS-CoV, marked with an * in the table above) that shares more than 80% homology with multiple of the primers and probes included in the TrioDx test kit:

• RdRp gene probe

• E gene primers and probe

• N gene forward primer and probe.

The results indicate that TrioDx kit does not cross react to any of non-target microorganisms except for SARScoronavirus. Therefore, the negative samples do not show any false positives due to cross-reactivity.

Non-target microorganisms also do not interfere with the positive results of the TrioDx COVID-19 test. In the case of SARS-coronavirus no amplicons will be generated from unintended cross reactivity in RdRp and N genes, as both forward and reverse primers must match to form amplicons. The RdRp gene shows more than 80% homology with only the probe. The N gene shows more than 80% homology with forward primer and probe.

While the in silico results indicate that co-infection with SARS-CoV-1 could theoretically reduce the sensitivity of the test, this is clinical not relevant since the CDC reports that since 2004, there have not been any known cases of SARS-CoV reported anywhere in the world.

Wet testing was not performed for these reasons and the in silico analysis was deemed sufficient.

• Clinical Evaluation

The clinical performance of the TrioDx was established by testing retrospective samples using 30 natural, noncontrived positive clinical samples and 30 natural, non-contrived negative clinical samples that were procured from a reference laboratory. Commercially sourced positive clinical samples were used following the FDA's COVID-19 Test Development and Review FAQs. All 60 clinical specimen samples were Nasopharyngeal Swabs (NP) in Aptima Specimen Transport Medium or Multitrans Medium.
Nucleic acids from the 60 clinical samples were purified using QIAamp Viral RNA Mini Kit and the samples were tested using QuantStudio 6 Flex. Since the LoD was the same for both the manual and automated extraction methods, the Clinical Evaluation was not repeated using the automated extraction method.
For the comparator, all 60 clinical samples had been tested and validated as positive or negative at the Reference Laboratory on an EUA authorized assay. The corresponding results from the reference laboratory were used as a comparator with the TrioDx assay. An EUA authorized assay with high sensitivity performed at the reference laboratory on the 60 samples detects two regions of the ORF1ab gene in the same fluorescent channel. The TrioDx detects three SARS-CoV-2 targets, RdRp, E and N gene in three distinct fluorescent channels.
A second study was performed to test seven low positive clinical samples using two different EUA authorized comparator assays. These samples were considered low positive when compared to the Ct values at the mean LoD of each EUA authorized assay. The TrioDx detected all 7 low positive samples with 100% agreement with the comparator assays.
All procedures for the TrioDx clinical evaluation were conducted in accordance with Section 10 (Test Procedures) described in this Instruction for Use (IFU).
The results of the clinical evaluation study are summarized in Table 22. Both positive percent agreement (PPA) and negative percent agreement (NPA) came out to be 100%.
Table 22. Clinical evaluation results summary
n = 67 EUA Authorized Comparator Test Total
Positive Negative
TrioDx Positive 37 0 37
Negative 0 30 30
Total 37 30 67
Positive Percent Agreement: 100% (37/37) 95% CI: 90.51% - 100.00%
Negative Percent Agreement: 100% (30/30) 95% CI: 88.43% - 100.00%
Contact Information
Clinomics USA Inc.
Address: 8949 Complex Drive, Unit B, San Diego CA 92123
P: +1 (619) 261-4321
Email: markd@clinomics.com
Website: http://clinomics.com/en/main