Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 1 of 19 PageID #: 992
`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 1 of 19 PageID #: 992
`
`EXHIBIT 33
`
`EXHIBIT 33
`
`
`
`
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 2 of 19 PageID #: 993
`
`verifi® Prenatal Test
`Payer dossier
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 3 of 19 PageID #: 994
`
`Table of Contents
`Executive Summary
`Issues with Prenatal Screening and Invasive Testing in the US
`
`Older Prenatal Screening Modalities
`
`Risks of Invasive Testing
`
`verifi Prenatal Test – Overview
`
`Intended Use
`
`Test Technology
`
`verifi Prenatal Test – Analytical Validity
`
`Analytical Performance
`
`Accuracy
`
`Precision
`
`verifi Prenatal Test – Clinical Validity
`
`Performance Data in Singleton Pregnancies
`
`Performance Data in Twin Pregnancies
`
`Applicability of NIPT for the General Pregnancy Population
`
`verifi Prenatal Test – Clinical and Economic Utility
`
`Clinical Utility and Economic Implications in the High-Risk
`Pregnancy Population
`
`Clinical Utility and Economic Implications in the General
`Pregnancy Population
`
`Technology Assessments for Noninvasive Prenatal Testing
`Medical Society Opinions and Recommendations
`References
`
`3
`
`4
`
`4
`
`6
`
`6
`
`6
`
`6
`
`7
`
`7
`
`8
`
`8
`
`9
`
`9
`
`10
`
`10
`
`12
`
`12
`
`13
`
`14
`
`15
`
`16
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 4 of 19 PageID #: 995
`
`Executive Summary
`An estimated 2.6 million maternal serum screens for fetal aneuploidy are performed in the United States annually.1
`Before the introduction of cell-free DNA-based noninvasive prenatal testing (NIPT), prenatal screening options had
`variable performance and screen-positive rates of around 5%.2 With a prenatal trisomy 21 incidence of approximately
`0.45%3 in the general pregnancy population, most screen positives were false positives. False positive results can
`lead to unnecessary patient anxiety and invasive diagnostic tests (amniocentesis and chorionic villus sampling [CVS]).
`These invasive tests carry risks, including procedure-related miscarriage.4
`
`Because of the high false-positive rate with prenatal serum screening and the inherent risks associated with
`amniocentesis and CVS, there was a clear, unmet clinical need for a more accurate prenatal screen for fetal
`aneuploidy. Cell-free DNA-based NIPT has addressed this need. More accurate results should lead to fewer
`confirmatory invasive procedures and potentially less patient anxiety.
`
`Growing evidence has demonstrated the benefit of NIPT as a first-tier fetal aneuploidy screen for all women,5 not just
`high-risk women, dramatically reducing the number of costly, confirmatory invasive procedures. Further, economic
`models have demonstrated that NIPT as a first-tier screen would be cost-effective in the general US pregnancy
`population at a price of around $650.6-8 Importantly, medical societies now support NIPT as a fetal aneuploidy
`screening option for all women.9-11
`
`As demonstrated in this clinical dossier, the verifi Prenatal Test can significantly improve current prenatal screening
`and diagnostic strategies based on the following key points:
`
`The high sensitivity12, 13 and specificity12, 13 enable a reduction in confirmatory invasive procedures, their sequelae
`and costs.14, 15
`
`Commercial laboratory experience with the verifi Prenatal Test demonstrates a test failure rate of around 0.1% and
`an average turnaround time (TAT) of 3 business days.12
`
`The verifi Prenatal Test is intended for use in women with a singleton or twin pregnancy, and can be performed at any
`time during pregnancy from 10 weeks’ gestation to term.
`
`The verifi® Prenatal Test was developed by, and its performance characteristics were determined by Verinata Health, Inc. (VHI) a wholly owned subsidiary of Illumina, Inc.
`The VHI laboratory is CAP-accredited and certified under the Clinical Laboratory Improvement Amendments (CLIA) as qualified to perform high complexity clinical laboratory
`testing. It has not been cleared or approved by the U.S. Food and Drug Administration.
`
`3
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 5 of 19 PageID #: 996
`
`Issues with Prenatal Screening and Invasive Testing in the US
`Practice guidelines by the American College of Obstetricians and Gynecologists (ACOG) recommend that pregnant
`women of all ages are offered aneuploidy screening before 20 weeks’ gestation.2
`
`Aneuploidy is a term used to describe a condition where there is an abnormal number of chromosomes. The clinical
`effects of aneuploidy are significant. Most pregnancies with aneuploid fetuses do not survive to term.16 Infants with
`aneuploidies that survive to birth are generally affected by congenital birth defects and/or intellectual disability.17 An
`estimated 30–60% of all miscarriages16, 18, 19 and 1 in 30016, 18 liveborns have aneuploidy. As such, aneuploidy is the
`leading known genetic cause of miscarriage and congenital birth defects.18
`
`Current prenatal screening options are primarily used to identify trisomy 21 (Down syndrome, T21), trisomy 18
`(Edwards syndrome, T18), and trisomy 13 (Patau syndrome, T13), which are the most common aneuploidies seen
`in live births. All three conditions lead to significant birth defects and intellectual disabilities.17 The incidence of these
`trisomies increases with advancing maternal age.20
`
`• Trisomy 21 occurs on average in about 1 in 660 live births.17
`• Trisomy 18 occurs in about 1 in 3,333 live births.17
`• Trisomy 13 occurs in approximately 1 in 5,000 live births.17
`Older Prenatal Screening Modalities
`
`Before the introduction of NIPT, prenatal aneuploidy screening options included the measurement of serum
`biomarkers and ultrasound examinations. In the first trimester, measurement of particular serum biomarkers
`(pregnancy-associated plasma protein A [PAPP-A] and human chorionic gonadotropin [hCG]) and ultrasound
`for nuchal translucency (NT) can be performed (commonly referred to as the “combined screen”). In the second
`trimester, a different set of serum biomarkers (hCG, Estriol, and alpha-fetoprotein [AFP] for the “triple” screen; add
`Inhibin-A for the “quadruple” screen) is measured. In some aneuploidy screening practices, both first trimester
`and second trimester measurements are performed in an attempt to increase the overall sensitivity (referred to as
`“sequential” or “integrated” screening).
`
`These screening options have suboptimal sensitivity and specificity (Table 1). Based on approximately 4M US
`births21 and a screen rate of 62%,1 around 2.5M annual prenatal serum screens are performed. Given a 5% screen-
`positive rate, there will be approximately 124,000 positive screen results. Assuming a prenatal trisomy 21 incidence
`of approximately 0.45%3 and an 85% trisomy 21 detection rate,22 this means that about 114,514 false positives
`are reported annually. With around 50% of women with a positive screening test electing invasive testing for
`confirmation,23 this equates to around 57,257 invasive tests performed in women because of a false positive trisomy
`21 screening result. Figure 1 illustrates the landscape of the prenatal screening for trisomy 21 in the United States
`before the introduction of NIPT.
`
`Table 1: Performance characteristics of prenatal screening strategies in the first and second trimesters prior to NIPT
`
`Trimester — Test
`
`1st — Combined (serum plus NT) screen for T21
`
`1st — Combined (serum plus NT) screen for T18
`
`2nd — Quad screen for T21
`
`2nd — Triple screen for T18*
`
`* Using maternal age, (cid:1091)-hCG, and PAPP-A markers
`
`Sensitivity
`
`85%22
`
`82%24
`
`81%22
`
`77%25
`
`Specificity
`
`95%22
`
`94%24
`
`95%22
`
`99%25
`
`4
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 6 of 19 PageID #: 997
`
`Figure 1: Estimation of the US trisomy 21 annual prenatal screening landscape before the introduction of NIPT*
`
`2.5 million
`serum screens
`
`124,000
`screen positive
`
`114,514
`false positive
`
`57,257
`Dx
`
`* Estimate based on published data for screening1 and diagnostic testing23 update rates, prenatal incidence of trisomy
`21,3 and performance of prenatal screening for trisomy 21.22
`
`The number of diagnostic tests (Dx) shown indicates the number of trisomy 21 false positive cases undergoing invasive diagnostic procedures.
`
`Clinicians have a range of screening and diagnostic tests available to offer to their patients. ACOG recommends that
`physicians provide the following information to their patients to enable them to make an informed decision about
`prenatal testing4, 9: the patient’s risk for fetal aneuploidy and other genetic diseases; the difference between screening
`and diagnostic testing; detection rates, false-positive rates, and the advantages, disadvantages, and limitations of
`each screening test; the risks and benefits of invasive diagnostic testing.
`
`Figure 2 illustrates screening and diagnostic strategies for aneuploidy detection that clinicians currently provide
`to their patients, and how NIPT can, and is, being integrated into clinical care. Of note, integrated and sequential
`screens require multiple office visits and ultrasound measurement of nuchal translucency requires specialist training.
`
`Figure 2. Prenatal screening strategies in the first and second trimesters
`
`Screening options before the introduction of NIPT
`
`Integrated Screen
`
`1st Trimester
`Serum Screen
`
`NT
`Ultrasound
`
`2nd Trimester
`Serum Screen
`
`Invasive Dx*
`(Amnio/CVS)
`
`Serum Integrated
`
`1st Trimester
`Serum Screen
`
`2nd Trimester
`Serum Screen
`
`Invasive Dx*
`(Amnio/CVS)
`
`Sequential Screen
`
`1st Trimester
`Serum Screen
`
`NT
`Ultrasound
`
`Interim
`Result
`
`2nd Trimester
`Serum Screen
`
`Invasive Dx*
`(Amnio/CVS)
`
`New screening options after the introduction of NIPT
`
`Contingent NIPT
`
`Integrated or Sequential
`Screen (1st–2nd Trimester)
`
`High-risk result –> NIPT
`(verifi Prenatal Test; ≥ 10 wks)
`
`First-tier NIPT
`
`NIPT
`(verifi Prenatal Test; ≥ 10 wks)
`
`Invasive Dx*
`(Amnio/CVS)
`
`Invasive Dx*
`(Amnio/CVS)
`
`Invasive diagnostic (Dx) testing recommended for patients with a positive screening result.
`
`Definitions: NT stands for nuchal translucency; amnio is short for amniocentesis; CVS stands for chorionic villus sampling;
`and wks is short for weeks.
`
`5
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 7 of 19 PageID #: 998
`
`Risks of Invasive Testing
`
`High false-positive rates with prenatal screening can be a concern as many (≥ 50%) of these patients will go on to
`have invasive procedures.23 These procedures carry inherent risks for adverse effects such as miscarriage, amniotic
`fluid leakage, infection, and infection transmission.26, 27 The miscarriage rate for CVS and amniocentesis ranges from
`1 in 500 to 1 in 1,000.4
`
`verifi Prenatal Test – Overview
`
`Intended Use
`
`The verifi Prenatal Test is intended for use in women with a singleton or twin pregnancy who are electing to undergo
`prenatal screening for fetal aneuploidy. Testing can be performed from 10 weeks’ gestation until term.
`
`Test Technology
`
`The verifi Prenatal Test is provided through the Illumina CLIA-certified, CAP-accredited clinical laboratory. The test
`utilizes next-generation sequencing (NGS) of circulating cell-free DNA (cfDNA) extracted from a maternal blood
`sample to screen for aneuploidy of chromosomes 21, 18, and 13, and the sex chromosomes; result reporting for the
`sex chromosomes is optional.
`
`The circulating cfDNA found in maternal plasma is a combination of cfDNA from the mother and placenta (which is
`typically representative of the fetal DNA). Around 10-15% of the cfDNA in maternal blood is from the placenta,28-30
`this percentage is commonly referred to as the “fetal fraction”. Millions of fragments from an individual patient’s
`blood sample are sequenced, aligned to a reference human genome, and analyzed for any relative over- or under-
`representation of DNA from the chromosomes of interest (indicative of aneuploidy; see Sections IV and V for
`additional information on the development of the verifi Prenatal Test). Highly sensitive NGS combined with algorithmic
`analysis can be used to detect and measure aneuploidy within this mixed sample. Based on this analysis, the sample
`receives a classification of aneuploidy status for chromosomes 21, 18, and 13, as well as the sex chromosomes (if
`requested). An overview of the verifi Prenatal Test laboratory process is shown in Figure 3.
`
`Figure 3: Cell-free DNA-based whole-genome NGS-based noninvasive prenatal testing: the verifi Prenatal Test
`
`Several methods for NIPT are currently available. The verifi Prenatal Test harnesses the power of whole-genome
`sequencing (WGS) with a highly optimized algorithm,31-34 which has been shown to be an accurate, reliable, and fast
`approach.12, 32-35 Some NIPTs use a targeted approach, sequencing only a select number of chromosomes or select
`single-nucleotide polymorphisms (SNPs).36, 37
`
`The verifi Prenatal Test is not dependent on maternal age, maternal weight, or gestational age (after 10 weeks).
`Further, unlike SNP-based tests,38, 39 verifi can be used in twin pregnancies34 and pregnancies conceived using
`an egg donor.40 Table 2 highlights some of the important differences between the verifi Prenatal Test and targeted
`NIPT approaches.
`
`6
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 8 of 19 PageID #: 999
`
`Table 2: Comparison of verifi Prenatal Test and targeted NIPT approaches
`
`verifi Prenatal Test (a WGS assay)12, 32, 34
`
`Low failure rate (< 0.5%)
`
`Captures comprehensive genomic data
`
`Not constrained by patient factors
`
`Available for singleton and twin gestations
`
`Targeted NIPTs 37, 41-43
`
`High failure rates (1–5% or greater)
`
`Analysis limited to a few chromosomes
`
`Can rely on patient factors
`
`May not offer testing for twin gestations
`
`verifi Prenatal Test – Analytical Validity
`Multiple clinical studies have established that sequencing of cfDNA from maternal blood can accurately detect fetal
`aneuploidy. Initial NIPT studies focused on the detection of the most common fetal aneuploidy, trisomy 21.44, 45 As
`trisomy 18 is the next most common autosomal fetal aneuploidy in women undergoing prenatal screening, Verinata
`Health (a subsidiary of Illumina) developed a proprietary algorithm, SAFeR™ (Selective Algorithm for Fetal Results), to
`detect trisomy 21 and trisomy 18.31 SAFeR calculates a Normalized Chromosome Value (NCV) for each chromosome
`of interest, which significantly reduces data variation caused by GC (guanine and cytosine) content, sample-to-sample
`and run-to-run variations, and other factors.
`
`To validate the algorithm, Verinata Health conducted a blinded, prospective study in collaboration with 13 clinics
`throughout the United States.31 Blood samples were collected from 1014 pregnant women who were at least 18
`years of age and were undergoing a clinically indicated CVS or amniocentesis procedure. The blood samples were
`collected prior to the invasive procedures. Of the 1,014 eligible patient samples, 119 underwent cfDNA analysis;
`53 of the 119 samples tested were from women with fetal aneuploidy. An optimized classification algorithm was
`developed from the sequencing data on 71 of the 119 samples (training set). The optimized classification algorithm
`was then evaluated on an independent test set of 48 samples. In this study, the optimized classification algorithm
`demonstrated 100% correct classification of T21 and T18.31
`
`The study concluded that “…algorithms for quantification not only minimize random and systematic
`variation between sequencing runs but also allow for effective classification of aneuploidies across
`the entire genome…”31
`
`Analytical Performance
`
`Analytical performance was assessed by testing samples with an unaffected karyotype and positive controls. The test
`process involves [1] collection of blood at clinics and shipment of the blood to the testing site, [2] isolation of plasma
`before cfDNA extraction from the plasma, [3] preparation of DNA libraries, [4] cluster generation and multiplexed
`sequencing on an Illumina NGS machine, [5] processing of sequencing results, which involves sequence alignment to
`the human genome and counting of unique tags, [6] sample classification.
`
`The performance of the test was assessed using predefined acceptance criteria and comparison of the sequencing
`results with karyotype data. A set of unaffected samples (N=611) and controls (N=84 replicates of a contrived control)
`were used for accuracy and precision determinations.
`
`From a total of 611 unaffected samples, 582 passed quality control metrics for DNA concentration (20–250 pg/μL)
`and library concentration (10 nM). The positive control was a contrived sample comprised of a mixture of sheared
`genomic DNA from three individual trisomic cell lines (representing trisomy 21, trisomy 18, and trisomy 13) sufficient
`to yield chromosome ratios representative of a trisomic fetal state for chromosomes 13, 18, and 21. The positive
`control was run in replicate in the analytical experiments described below.
`
`7
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 9 of 19 PageID #: 1000
`
`Accuracy
`
`The accuracy of the test is based on the comparison of the NIPT results to the karyotype results. Accuracy for
`chromosomes 13, 18, and 21 was determined for the positive controls and for the samples of unaffected karyotype.
`As a means to minimize experimental variance and account for experimental biases, chromosomal counts on test
`chromosomes were normalized by comparing to counts on a set of denominator chromosomes.
`
`Accuracy for the positive control was determined from 84 replicates of a contrived control, sequenced across 13 flow
`cells. All 84 replicates were accurately classified for trisomy 13, 18, and 21 with mean NCVs of 8.07, 7.14, and 7.82,
`respectively (Table 3).
`
`Table 3: Accuracy of positive control (n = 84 replicates)
`
`Mean
`
`Min
`
`Max
`
`Ratio 13
`
`0.27591
`
`0.27435
`
`0.27783
`
`Ratio 18
`
`0.24410
`
`0.24289
`
`0.24588
`
`Ratio 21
`
`0.08807
`
`0.08757
`
`0.08864
`
`NCV 13
`
`NCV 18
`
`NCV 21
`
`8.07
`
`6.75
`
`9.73
`
`7.14
`
`5.05
`
`10.23
`
`7.82
`
`6.47
`
`9.37
`
`Of the 582 samples that went to sequencing, there were 15 individual sequencing failures, leaving 567 samples
`that passed all quality control metrics. Of the 567 analyzed samples, 555 were classified as “Unaffected” with an
`NCV of < 2.5. Two samples had a single chromosome with an NCV > 4 and were classified as “Affected” for that
`chromosome: 1 for chromosome 18 and 1 for chromosome 21. Ten samples were unclassified with NCVs between
`2.5 and 4. Accuracy for the unaffected samples was calculated to be 99.6%.
`
`Precision
`
`Precision in unaffected samples was determined from the variability of 567 individual samples of unaffected karyotype
`that were sequenced across 13 flow cells. The mean, standard deviation, median, and CV (coefficient of variation) of
`chromosome 13, 18, and 21 ratios on each flow cell were calculated. The inter-flow cell CV was also determined.
`
`A positive control was used to determine the precision for positive results. The control was tested in replicates (4–12
`replicates within a flow cell) for a total of 84 measures across 13 flow cells. The variability (CV) of chromosome ratios
`for 13, 18, and 21 were determined; T21 ratios are shown in Table 4. The variability on multiple sequencing runs
`assessed with this positive control is an indication of the precision expected for chromosomes 13, 18, and 21 with
`affected samples.
`
`Table 4: Intra- and inter-flow cell variation of ratio T21
`
`Flow Cell ID
`
`1
`
`2
`
`Mean
`
`0.085204685
`
`0.085313555
`
`SD
`
`0.0003808
`
`0.0003887
`
`Median
`
`0.0852253
`
`0.0853179
`
`0.0850902
`
`Intra-Flow Cell CV
`(%)
`
`Inter-Flow Cell CV
`(%)
`
`0.45
`
`0.46
`
`0.41
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`0.085083745
`
`0.085176931
`
`0.085281972
`
`0.085095318
`
`0.085159962
`
`0.085135233
`
`0.085205245
`
`0.085191251
`
`0.085126717
`
`0.085234313
`
`0.085263435
`
`0.0003458
`
`0.0003682
`
`0.0004553
`
`0.0004237
`
`0.0004214
`
`0.0004944
`
`0.0003955
`
`0.0003174
`
`0.0004011
`
`0.000442
`
`0.0003756
`
`0.0851671
`
`0.0852451
`
`0.0851235
`
`0.0851081
`
`0.0851664
`
`0.085156
`
`0.0852005
`
`0.0850987
`
`0.0852298
`
`0.085258
`
`0.43
`
`0.53
`
`0.50
`
`0.49
`
`0.58
`
`0.46
`
`0.37
`
`0.47
`
`0.52
`
`0.44
`
`1.69
`
`8
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 10 of 19 PageID #: 1001
`
`verifi Prenatal Test – Clinical Validity
`
`Performance Data in Singleton Pregnancies
`
`Building on the findings of the Sehnert et al study,31 the MELISSA clinical study was initiated to validate the diagnostic
`accuracy of the verifi Prenatal Test. Blood samples were collected in a prospective, blinded study from 2,882 women
`undergoing prenatal diagnostic procedures at 60 US clinical sites.32 Chromosome classifications were made by verifi
`for each sample and compared with fetal karyotypes obtained by CVS or amniocentesis. This study demonstrated
`that the verifi Prenatal Test has high sensitivities and specificities for fetal trisomy 21, 18, and 13 (Table 5).32
`
`Table 5: verifi Prenatal Test performance in the MELISSA study32
`
`Condition
`
`Trisomy 21,
`Down syndrome
`
`Trisomy 18,
`Edwards syndrome
`
`Trisomy 13,
`Patau syndrome
`
`Monosomy X,
`Turner syndrome
`
`N
`
`493
`
`496
`
`499
`
`433
`
`Sensitivity
`
`>99.9% (89/89)
`
`95% CI
`
`95.9–100
`
`Specificity
`
`95% CI
`
`>99.9% (404/404)
`
`99.1–100.0
`
`97.2% (35/36)
`
`85.5–99.9
`
`>99.9% (460/460)
`
`99.2–100.0
`
`78.6% (11/14)
`
`49.2–99.9
`
`>99.9% (485/485)
`
`99.2–100.0
`
`93.8% (15/16)
`
`69.8–99.8
`
`99.8% (416/417)
`
`98.7–100.0
`
`Following the MELISSA study, updates to the DNA sequencing chemistry and algorithm used for verifi were made to
`improve test precision and performance. The updated test performance for the verifi Prenatal Test is shown below in
`Table 6.12,13 Further, a new classification category “Aneuploidy Suspected” was introduced for samples with borderline
`results (NCVs above the cutoff for reporting as “No Aneuploidy Detected” but lower than the cutoff for reporting as
`“Aneuploidy Detected”). While some affected cases are expected to fall in the borderline zone, samples reported
`as “Aneuploidy Suspected” are more likely than samples reported as “Aneuploidy Detected” to be unaffected
`(false positive).
`
`Table 6: verifi Prenatal Test performance in the MELISSA cohort after test updates12,13
`
`Condition
`
`Trisomy 21
`
`Trisomy 18
`
`Trisomy 13
`
`Monosomy X
`
`N
`
`500
`
`501
`
`501
`
`508
`
`Sensitivity
`
`>99.9% (90/90)
`
`97.4% (37/38)
`
`87.5% (14/16)
`
`95.0% (19/20)
`
`95% CI
`
`96.0–100.0
`
`86.2–99.9
`
`61.7–98.5
`
`75.1–99.9
`
`Specificity
`
`99.8% (409/410)
`
`99.6% (461/463)
`
`>99.9% (485/485)
`
`99.0% (483/488)
`
`95% CI
`
`98.7–100.0
`
`98.5–100.0
`
`99.2–100.0
`
`97.6–99.7
`
`Performance of the verifi Prenatal Test has since been evaluated in the clinical population submitting samples to the
`Illumina clinical laboratory.12, 35, 46, 47 Although clinical follow-up was limited, an evaluation of the first approximately
`6,000 samples submitted to the Illumina laboratory suggested that clinical performance was in line with the
`performance parameters established in validation studies.35 This is important because validation studies typically use
`later gestation samples, which can have a higher fetal cfDNA fraction, and exclude samples that can be more difficult
`to analyze (such as samples with a mosaic karyotype). In this study, the average turnaround time was 5 business
`days and the technical failure rate was 0.7%.35
`
`A larger, subsequent study of around 85,000 clinical samples submitted for the verifi Prenatal Test has now been
`published.12 During the timeframe of this study, several process improvements and analytic updates to the verifi
`Prenatal Test were implemented. These updates resulted in a reduction in the average turnaround time to 3 days, and
`a reduction in the test failure rate to 0.1%.12 Importantly, this average turnaround time12 is faster than with the targeted
`NIPT approaches that are offered by other laboratories.48 Further, the 0.1% test failure rate determined in this study12
`is significantly lower than the failure rates observed with targeted NIPT approaches (1–5% or higher).48 As in the
`
`9
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 11 of 19 PageID #: 1002
`
`earlier clinical outcome study, outcome data was limited, particularly for cases reported as No Aneuploidy Detected.
`Observed positive predictive values (PPVs), indicating the proportion of cases reported as aneuploidy detected or
`suspected that were true affected fetuses, were calculated from cases with known clinical outcomes. The observed
`PPVs ranged from 50.0% to 92.8% for trisomies 13, 18, and 21.12 These PPVs are consistent with those published
`by other NIPT clinical outcome studies.35, 48
`
`Performance Data in Twin Pregnancies
`
`In the US, the incidence of twin births is around 1 in 30, with the rate of twin births on the rise.49 Traditional serum
`screening options have lower sensitivities and specificities in twin gestations.50 Thus, the availability of an accurate
`and reliable fetal aneuploidy screen for use in twin pregnancies would be of significant value.
`
`Improvements were made to the Illumina SAFeR algorithm to enable fetal aneuploidy screening in twin pregnancies
`with the verifi Prenatal Test.34 An initial evaluation of test performance in twin pregnancies was made using maternal
`blood samples collected as part of two prospective clinical studies, MELISSA32 and CARE 5. A total of 115 twin
`pregnancy samples with known karyotypes were available for analysis. All samples were correctly classified, including
`4 (three trisomy 21, one trisomy 18) affected pregnancies and 111 unaffected pregnancies.34
`
`Test performance was next evaluated in clinical twin pregnancy samples submitted to the Illumina clinical laboratory
`for the verifi Prenatal Test.34 A total of 487 samples were evaluated, of which 479 (98.4%) received a test result;
`all cancellations were for administrative, not technical, reasons. Of these, 9 cases were reported as aneuploidy
`detected or suspected, and 470 cases were reported as no aneuploidy detected. Of the 9 aneuploidy suspected/
`detected cases, 6 were true positives (at least one twin was affected), 1 was a false positive (both twins unaffected),
`and 2 were suspected to be true positives based on ultrasound findings but confirmatory karyotypes were
`unavailable. Within the 164 cases reported as no aneuploidy detected and with known outcomes, no false negatives
`were reported.
`
`Overall, these two studies demonstrated that the verifi Prenatal Test performs well in twin pregnancies.
`
`Applicability of NIPT for the General Pregnancy Population
`
`Initial NIPT validation studies were performed using women considered high risk for fetal aneuploidies. This was
`largely because this ensured a population enriched with affected samples, which powered the studies to determine
`sensitivity. However, questions quickly arose about the performance of NIPT in low- or general-risk pregnancy
`populations.
`
`Performance of the verifi Prenatal Test in a general-risk population was determined in a prospective, blinded clinical
`study, CARE.5 This study compared the results of the verifi Prenatal Test with results of conventional prenatal
`screening in a sample of 1,914 women recruited from the general obstetrical population. Verifi detected all cases of
`aneuploidy (five trisomy 21, two trisomy 18, and one trisomy 13) within this population. Importantly, the false-positive
`rate for the verifi Prenatal Test was significantly lower than standard screening for trisomy 21 (0.3% vs 3.6%) and
`trisomy 18 (0.2% vs 0.6%). Lower false-positive rates would result in far fewer women undergoing confirmatory
`invasive procedures. Further, the PPVs were markedly higher with verifi compared with standard screening: 45.5%
`versus 4.2% for trisomy 21; 40.0% versus 8.3% for trisomy 18. Thus, for women in this cohort with a positive
`screening result for trisomy 21 and undergoing a confirmatory invasive procedure, 1 in 2 will be truly affected with
`NIPT compared with only 1 in 25 with standard screening.
`
`10
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 12 of 19 PageID #: 1003
`
`After the completion of the CARE study, updates were made to the analytics algorithm used for the verifi Prenatal
`Test that were anticipated to reduce technical causes of false positives.51 Reanalysis of the CARE cohort with this
`updated algorithm showed that over 50% (6/11) of the previous false positives5 were now correctly reported as
`“No Aneuploidy Detected”.51 This confirmed that the updated algorithm used in the Illumina laboratory for the verifi
`Prenatal Test prevents some technical causes of false positive results. A reduction in false-positive rates will result
`in higher PPVs, which is considered to be of significant value as NIPT is increasingly utilized within the general
`pregnancy population.
`
`In summary, the high sensitivity and specificity for trisomies 21, 18, and 1312, 13 (Table 6) and strong performance in
`the general pregnancy population12 support that the verifi Prenatal Test can be successfully integrated into current
`prenatal screening strategies (see Figure 4). The screening strategy with the largest potential benefit to patients would
`be utilization of NIPT as a first-tier screen, as the test is availability from early pregnancy (≥10 weeks) and its use
`would result in fewer confirmatory invasive diagnostic procedures.
`
`Figure 4: Potential prenatal testing strategies.
`
`PREGNANT WOMAN
`(Singleton or twin gestation)
`
`Discuss options
`2 and 3
`
`Review of prenatal test options
`with clinician and/or
`genetic counselor
`
`OPTION 1
`Traditional screening
`(≥ 10 weeks)
`
`OPTION 2
`Noninvasive prenatal testing
`(≥ 10 weeks)
`
`OPTION 3
`Invasive Testing:
`CVS (11–13 weeks)
`Amniocentesis (> 15 weeks)
`
`Positive
`
`Negative
`
`No Aneuploidy
`Detected
`
`Aneuploidy
`Suspected
`
`Aneuploidy
`Detected
`
`Clinical follow-up based
`on clinical context
`
`Confirmation with diagnostic
`testing recommended
`
`Screening options and potential results depicted in blue; diagnostic testing option depicted in purple. Patient can
`choose/decline any, and all, prenatal testing options.
`
`11
`
`

`

`Case 1:20-cv-01644-RGA Document 1-33 Filed 12/03/20 Page 13 of 19 PageID #: 1004
`
`verifi Prenatal Test – Clinical and Economic Utility
`
`Clinical Utility and Economic Implications in the High-Risk Pregnancy Population
`
`In the prospective, blinded MELISSA study, the verifi Prenatal Test demonstrated high sensitivity and specificity for
`the detection of trisomies 21, 18, and 1332 suggesting that it could be incorporated into existing aneuploidy screening
`algorithms to reduce the number of confirmatory invasive procedures for high-risk women.
`
`To evaluate the potential impact of the verifi Prenatal Test on current prenatal screening strategies and the associated
`rate of invasive diagnostic testing, a transition state probability model of current prenatal screening and diagnostic
`strategies was developed. This model used the performance data published in the MELISSA study. Bridgehead
`International developed this model to evaluate the impact of incorporating the verifi Prenatal Test into routine high-risk
`maternal screening practice.
`
`The model took a theoretical cohort of 100,000 pregnant women at high risk for fetal aneuploidy (based on either first
`or second trimester screening) and assessed the expected clinical and cost impact of using the verifi Prenatal Test
`compared with current practice.52 In the modeled population of five million covered lives with 100,000 pregnancies
`annually, invasive diagnostic induced miscarriages are reduced from 60 to 20, a 66% reduction (Table 7).
`
` “The model demonstrates that inclusion of the verifi test into the prenatal testing paradigm for
`high-risk women will provide clear clinical benefits. The biggest benefit to women comes from a
`reduction in miscarriages due to invasive testing.”52
`
`Often, new medical technologies add significant cost to the health system. A transition state probability model is a
`method of evaluating the clinical and economic impact of incorporating new technology into current standard of care.
`The transition state probability model for the verifi Prenatal Test illustrated that incorporating it into prenatal testing
`algorithms can greatly reduce the number of costly avoidable procedures and result in an overall cost savings to the
`health system (Table 7).
`
`“The…savings are accompanied by an improved clinical experience by ruling out the need for
`clinically unnecessary invasive testing for many women.”52
`
`Table 7: The clinical and cost impact of adopting the verifi Prenatal Test52
`
`Measure
`
`Traditional Prenatal
`Testing
`
`Invasive diagnostic induced
`miscarriages
`
`60
`
`Unnecessary follow-up costs
`
`$14