The Tempus xR IVD assay is a qualitative next generation sequencing-based in vitro diagnostic device that uses targeted high throughput hybridization-based capture technology for detection of rearrangements in two genes, using RNA isolated from formalin-fixed paraffin embedded (FFPE) tumor tissue specimens from patients with solid malignant neoplasms.

Information provided by xR IVD is intended to be used by qualified health care professionals in accordance with professional guidelines in oncology for patients with previously diagnosed solid malignant neoplasms. Results from xR IVD are not intended to be prescriptive or conclusive for labeled use of any specific therapeutic product.

xR IVD is a next generation sequencing (NGS)-based assay for the detection of alterations from RNA that has been extracted from routinely obtained FFPE tumor samples. Extracted RNA undergoes conversion to double stranded cDNA and library construction, followed by hybridization-based capture using a whole-exome targeting probe set with supplemental custom Tempus-designed probes. Using the Illumina® NovaSeq 6000 platform qualified by Tempus, hybrid-capture–selected libraries are sequenced, targeting > 6 million unique deduplicated reads. Sequencing data is processed and analyzed by a bioinformatics pipeline to detect gene rearrangements, including rearrangements in BRAF and RET.

Alterations are classified for purposes of reporting on the clinical report as Level 2 or Level 3 alterations in accordance with the FDA Fact Sheet describing the CDRH's Approach to Tumor Profiling for Next Generation Sequencing Tests and as follows:

• Level 2: Genomic Findings with Evidence of Clinical Significance
• Level 3: Genomic Findings with Potential Clinical Significance

xR IVD is intended to be performed with the following key components, each qualified and controlled by Tempus under its Quality Management System (QMS):

• Reagents
• Specimen Collection Box
• Software
• Sequencing Instrumentation

1. Reagents

All reagents used with respect to the operation of xR IVD are qualified by Tempus.

2. Test Kit Contents

xR IVD includes a specimen collection and shipping box (the Specimen Box). The Specimen Box contains the following components:

• Informational Brochure with Specimen Requirements
• Collection Box Sleeve
• Collection Box Tray
• Seal Sticker
• ISO Label

3. Software

The proprietary xR IVD bioinformatics pipeline comprises data analysis software necessary for the xR IVD assay (software version is displayed on the xR IVD clinical report). The software is used with sequence data generated from NovaSeq 6000 instruments qualified by Tempus. Data generated from the pipeline is saved to a cloud infrastructure.

4. Instrument

xR IVD uses the Illumina NovaSeq 6000 Sequencer, a high throughput sequencing system employing sequencing-by-synthesis chemistry. The xR IVD device is intended to be performed with serial number-controlled instruments. All instruments are qualified by Tempus utilizing the Tempus Quality Management System (QMS).

5. Sample preparation

FFPE (Formalin Fixed Paraffin Embedded) tumor specimens are received either as unstained tissue sections on slides or as an FFPE block using materials supplied in the Specimen Box and prepared following standard pathology practices. Preparation and review of a Hematoxylin and Eosin (H&E) slide is performed prior to initiation of the xR IVD assay. H&E stained slides are reviewed by a board-certified pathologist to ensure that adequate tissue, tumor content and sufficient nucleated cells are present to satisfy minimum tumor content (tumor purity).

Specifically, the minimum recommended tumor purity for detection of alterations by xR IVD is 20%, with macrodissection required for specimens with tumor purity lower than 20%. The recommended tumor size and minimum tumor content needed for testing are shown in Table 1, below.

Table 1: Tumor Volume and Minimum Tumor Content

*These requirements are based on the specimen's tumor content

6. RNA extraction

Nucleic acids are extracted from tissue specimens using a magnetic bead-based automated methodology followed by DNAse treatment. The remaining RNA is assessed for quantity and quality (sizing) at RNA QC1, which is a quality check (QC) to ensure adequate RNA extraction. The minimum amount of RNA required to perform the test is 50 ng. RNA is fragmented using heat and magnesium, with variable parameters, to yield similar sized fragments from RNA inputs with different starting size distributions.

7. Library preparation

Strand-specific RNA library preparation is performed by synthesizing the first-strand cDNA using a reverse transcriptase (RT) enzyme followed by second-strand synthesis using a DNA polymerase to create double stranded cDNA. Adapters are ligated to the cDNA and the adapter-ligated libraries are cleaned using a magnetic bead-based method. The libraries are amplified with high fidelity, low-bias PCR using primers complementary to adapter sequences. Amplified libraries are subjected to a 1X magnetic bead based clean-up to eliminate unused primers, and quantity is assessed (QC2) to ensure that pre-captured libraries were successfully prepared. Each amplified sample library contains a minimum of 150 ng of cDNA to proceed to hybridization.

8. Hybrid capture

After library preparation and amplification, the adapter-ligated library targets are captured by hybridization, clean-up of hybridized targets is performed, and unbound fragments are washed away. The captured targets are enriched by PCR amplification followed by a magnetic bead-based clean-up to remove primer dimers and residual reagents. To reduce non-specific binding of untargeted regions, human COT DNA and blockers are included in the hybridization step. Each post-capture library pool must satisfy a minimum calculated molarity (≥2.7 nM) to proceed to sequencing (QC3). The molarity is used to load the appropriate concentration of library pools onto sequencing flow cells.

9. Sequencing

The amplified target-captured libraries are sequenced with a 2x76 read length to an average of 50 million total reads on an Illumina NovaSeq 6000 System using patterned flowcells (SP/S1, S2, or S4). Pooled sample libraries are fluorometrically quantified and normalized into a sequencing pool of up to 28 samples (SP flowcell), 56 samples (S1 flowcell), 140 samples (S2 flowcell), 336 samples (S4 flowcell) with each flowcell including 2 external controls. Partial batches are supported using a set threshold of loading capacity down to a defined percentage. Pooled sample libraries are loaded on a sequencing flow cell and sequenced.

10. Data Analysis

a. Data Management System (DMS): Sequence data is automatically processed using software that tracks sample names, sample metadata processing status from sequencing through to analysis and reporting. Reports of identified alterations are available in a web-based user interface for download. Sequencing and sample metrics are available in run and case reports, including sample and sequencing quality.

b. Demultiplexing and FASTQ Generation: Demultiplexing software generates FASTQ files containing sequence reads and quality scores for each of the samples on a sequencing run. The FASTQ formatted data files are used for subsequent processing of samples.

c. Indexing QC Check: Samples are checked for an expected yield of sequence reads identified to detect mistakes in pooling samples. Samples outside the expected range are marked as failed.

d. Read Alignment and BAM Generation: Genome alignment is performed to map sequence reads for each sample to the human reference genome (hg19). Alignments are saved as Binary Alignment Map (BAM) formatted files, which contain read placement information relative to the reference genome with quality scores. Aligned BAM files are further processed in a pipeline to identify genomic alterations.

e. Sample QC check: A sample QC check (QC4) evaluates the quality of the samples processed through the bioinformatics pipeline (sample level metrics in Table 2). Samples are evaluated for contamination by evaluating the percent of a tumor sample contaminated with foreign nucleic acid with a threshold below 5%. Sample sequencing coverage is assessed through RNA gene-ids expressed which counts all genes raw expression abundance (>12,000) and RNA GC-distribution (45-59%). The sample mapping rate (>80%), RNA strand % sense (>88%) and RNA strand % failed (≤ 10%) metrics provide confidence in the sample quality.

f. Alteration calling: A fully automated pipeline for bioinformatic analysis is used to identify gene rearrangements. The assay is validated to report specific gene rearrangements. Gene rearrangements are identified based on observations of reads supporting gene rearrangements in genomic alignments of discordantly mapped or split read pairs.

11. Controls

a. Negative control: A no template control (NTC) is processed to serve as a negative control to validate the acceptability of all the test samples processed through extraction, library preparation and hybridization and capture steps by testing for sample or reagent contamination. The NTC is not included on the sequencing run.

b. Positive control: xR IVD uses multiple external controls consisting of contrived material with synthetically derived alterations or a pool of multiple cell lines. A positive control sample containing known gene rearrangements will be included with each sequencing run. The external controls are processed from library preparation through sequencing to serve as an end to end control to demonstrate assay performance. The external controls are checked during library preparation and after sequencing. Failure of the external control to meet the pre-defined quality metrics will result in all test samples on the run being reported as Quality Control (QC) failure.

12. Result reporting

xR IVD reports oncologically relevant gene rearrangements as genomic findings with evidence of clinical significance or with potential clinical significance. Gene rearrangements are assessed as oncogenic based on required genomic regions specified in a Tempus-developed curated database. Gene rearrangements that retain the genomic region(s) required for oncogenicity are assigned a level of clinical significance consistent with FDA's Fact Sheet and reported. Gene rearrangements that do not retain the region(s) required for oncogenicity are not reported.

13. Quality metrics

Reporting takes into account the quality metrics outlined in Table 2. Quality metrics are assessed across the following categories:

• Batch-level: Metrics that are quantified per sequencing run; if the positive control fails these criteria, no results are reported for the entire batch of samples.
• Sample-level: Metrics that are quantified per sample; no device results are generated for samples failing these metrics. These metrics are also referred to as sequencing quality control (QC4).
• Analyte-level: Metrics that are quantified for individual alteration types. Alterations passing analyte-level metrics (threshold) are reported.

Table 2: Summary of xR IVD Post-Sequencing Key Quality Metrics at Batch, Sample (QC4), and Analyte Levels

N/A

Ask a specific question about this device

The ACTOnco IVD assay is an in vitro diagnostic test that uses targeted next generation sequencing of formalin-fixed. paraffin-embedded tumor tissue from patients with solid malignant neoplasms to detect genetic alterations in a broad multi gene panel. The test is intended to provide informations, small insertions and deletions, ERBB2 gene amplification, and tumor mutational burden for use by qualified health care professionals in accordance with professional guidelines, and is not conclusive or prescriptive for labeled use of any specific therapeutic product. ACTOnco IVD is a single-site assay performed at ACT Genomics.

The ACTOnco IVD assay is an in-vitro diagnostic assay intended to provide information for use by qualified health care professionals in accordance with professional guidelines, and is not conclusive or prescriptive for labeled use of any specific therapeutic product.

The assay is a custom targeted sequencing platform, utilizing amplicon-based sequencing, to detect point mutations (single nucleotide variants, or SNVs), small insertions and deletions (Indels), ERBB2 gene amplification, and tumor mutational burden (TMB) in tumor specimens. The assay uses custom DNA primers corresponding to all exons and selected introns of oncogenes, tumor suppressor genes, drug metabolism genes, and immune-related genes. Primers are synthesized by a secondary manufacturer (Thermo Fisher Scientific). An overlapping amplicon approach is utilized in which tiled primers are designed to generate multiple overlapping amplicons of the same region to avoid allele dropout. In total, the primers target approximately 1.8Mb of the human genome. Genomic DNA is extracted from FFPE tissue samples.

Sequence libraries are prepared through a multiplex polymerase chain reaction (PCR) amplification step to enrich target sequences. Target sequences are tagged with index oligonucleotide to identify individual sample and adaptor oligonucleotide to anchor the amplicon to complimentary oligonucleotides embedded on the surface of the sequencing bead. Target sequences on the sequencing beads are amplified using emulsion PCR before sequencing. Multiple barcoded sequence libraries (from different patients) are pooled and then sequenced on a Thermo Fisher Ion GeneStudio™ S5 Prime System. Sequence reads are then aligned to the reference human genome. By comparing the identity of bases from the tumor DNA and the reference human genome, variant alterations are identified in the tumor.

The assay system includes a sequencing instrument, reagents (DNA extraction, library preparation and sequencing), software (operation of the sequencing instrument and variant calling), and standard operating procedures (SOPs) for the use of the system. ACT Genomics takes the responsibilities in monitoring the instrument; reagents and consumable materials which will be used in the assay process.

Multiple software components will be used in the assay. The NGS raw read analysis will be done using Thermo Fisher software. Variant calling for SNVs, insertions and deletions will be done using Thermo Fisher software. Mutation and variant annotation will be done using software from ACT Genomics, the Cunningham Lab and Golden Helix software. ERBB2 gene amplification will be done using software from Boeva Lab. Tumor Purity and Zygosity will be done using software from Halgamuge Lab (Kaushalya Amarasinghe). Calculations for tumor mutational burden will be done using ACT Genomics software.

I will analyze the provided text to extract information about the acceptance criteria and the study proving the device meets these criteria. I will then structure this information according to your requested format.

However, based on the provided text, it appears to be a 510(k) summary for a medical device (ACTOnco IVD, a next-generation sequencing-based tumor profiling test). The document describes performance testing (precision/reproducibility, analytical sensitivity/limit of detection, analytical specificity/interference, cross-contamination, DNA input, DNA extraction) and method comparison data.

Crucially, this document
does not contain a direct table of acceptance criteria nor explicit statements about "acceptance criteria" met by the device.

It also does not describe a "Multi Reader Multi Case (MRMC) comparative effectiveness study" or information about "human readers improve with AI vs without AI assistance" as it is a diagnostic test and not an AI-assisted diagnostic imaging device/software for human readers.

Therefore, I will provide the acceptance criteria based on the performance observed in the various analytical validation studies, as these implicitly define the performance considered acceptable for the device. I will also make clear when certain information is not present in the provided text.

Here is the information structured according to your request, extracted from the provided 510(k) summary:

Device: ACTOnco IVD - Next generation sequencing based tumor profiling test

1. Table of Acceptance Criteria (Inferred from Performance Studies) and Reported Device Performance:

Since explicit acceptance criteria are not called out as a separate table, I will infer them from the "Performance Testing" sections, assuming that the reported performance metrics were considered acceptable for the device's clearance.

2. Sample Size and Data Provenance for Test Set:

• Precision/Reproducibility: 20 unique samples (12 single clinical FFPE, 8 pooled DNA from multiple FFPE blocks). Samples covered 10 cancer types.
• Analytical Sensitivity (LoD): 10 different cancer specimens for SNVs/INDELs. For TMB LoD, 5 FFPE tumor specimens. For ERBB2 LoD, 1 amplified sample.
• Analytical Specificity / Interference: 6 FFPE specimens for spike-in (skin, breast, colorectal, endometrial, lung, kidney cancer). 469 specimens for compare-to-similar (necrotic tissue) covering 21 cancer types.
• Cross Contamination: Positive and negative process control samples (specific number not given, but refers to "12 positive control samples alternating with 12 negative control samples within 96-well plate").
• DNA Input: 10 FFPE samples representing 8 cancer types.
• DNA Extraction: Retrospective review of 1526 specimens.
• Method Comparison (SNV/Indels): 438 FFPE samples spanning 21 cancer types.
• Method Comparison (ERBB2 Amplification): 129 FFPE samples spanning 21 cancer types.
• Method Comparison (TMB): 45 FFPE clinical samples covering 14 cancer types.

Data Provenance: The document explicitly states that the samples are "clinical samples (FFPE)" and refer to "various cancer types". It implies that these are real patient samples.
The country of origin is not explicitly stated, but the submitter (ACT GENOMICS Co., LTD) is located in Taipei City, Taiwan. The submission correspondent (K2 Regulatory Consulting, LLC) is in Burlingame, CA, USA. Given that this is an FDA 510(k) submission, the data would typically be representative of the U.S. population or a general population if not geographically specific. The retrospective review of 1526 specimens for DNA extraction likely refers to historical data from ACT Genomics' laboratory.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications:

This information is not explicitly stated in the provided document. For a diagnostic test like this, ground truth is typically established by:

• Comparator Assays: The document states "a well-validated NGS based assay" for SNV/Indels, "HER2 Dual ISH DNA Probe Cocktail test (DISH)" for ERBB2, and "whole exome sequencing (WES)" for TMB. These comparator methods ARE the ground truth.
• Pathology Reports: The presence of FFPE tumor tissue as the specimen type implies pathological confirmation of cancer. Tumor purity is also mentioned, which would involve pathological assessment.

There is no mention of human experts providing consensus reads or establishing ground truth in the way a radiological AI would need.

4. Adjudication Method for the Test Set:

This information is not applicable in the context of this diagnostic test validation, as the ground truth is established by well-defined comparator assays (NGS, DISH, WES) and internal laboratory procedures. There is no indication of multiple human readers or adjudication processes for the results of the assay itself or its comparators in the manner described for imaging studies.

5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study was done, and the effect size of how much human readers improve with AI vs without AI assistance:

No, this type of study was not done. The device (ACTOnco IVD) is a Next Generation Sequencing based tumor profiling test. It is not an AI-assisted diagnostic imaging device/software designed to assist human readers. Therefore, an MRMC study assessing human reader improvement with AI assistance is not applicable to this device.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done:

Yes, the performance studies described are inherently "standalone" in the context of the algorithm's analytical performance. The studies evaluate the device (ACTOnco IVD) as an automated system that performs DNA extraction, sequencing, and variant calling. The "performance testing" sections (Precision/Reproducibility, Analytical Sensitivity, Analytical Specificity, Cross Contamination, DNA Input, Method Comparison) all describe experiments evaluating the accuracy and reliability of the device's output (variant calls, TMB scores, ERBB2 amplification status) compared to reference methods or expected outcomes, without human interpretation of raw data. The human element would primarily be in the final interpretation of the molecular findings by a qualified healthcare professional, as stated in the Indications for Use, "for use by qualified health care professionals in accordance with professional guidelines". The analytical performance itself is algorithm-driven.

7. The Type of Ground Truth Used:

The ground truth for validation was established by:

• Comparator Assays / Gold Standards:
  • For SNVs, MNVs, insertions, and deletions: "a well-validated NGS based assay."
  • For ERBB2 gene amplification: "HER2 Dual ISH DNA Probe Cocktail test (DISH)."
  • For Tumor Mutational Burden (TMB): "whole exome sequencing (WES, next generation sequencing)."
• Internal Controls and Expected Outcomes: For precision, LoD, interference, cross-contamination, and DNA input studies, the ground truth was based on known properties of the samples (e.g., specific variants, dilution levels, known interference composition, or expected wild-type status in controls).
• Pathological Assessment: Implied by the use of FFPE tumor tissue samples and assessment of tumor purity, which would rely on pathological review.

8. The Sample Size for the Training Set:

The document for the 510(k) submission does not specify the sample size for the training set used to develop or train the ACTOnco IVD device. 510(k) summaries primarily focus on the validation (test set) data and may not detail the development/training phase data.

9. How the Ground Truth for the Training Set was Established:

The document does not provide details on how the ground truth for the training set (if any specific training was done for components like variant calling algorithms after initial development) was established. Generally, for NGS panels, training data often involves publicly available genomic datasets, internal curated datasets with orthogonal confirmation (e.g., Sanger sequencing, alternative NGS platforms), and expert pathology review of tissue samples. However, this is speculative as the document is silent on this point.

Ask a specific question about this device

The NYU Langone Genome PACT assay is a qualitative in vitro diagnostic test that uses targeted next generation sequencing of formalin-fixed paraffin-embedded (FFPE) tumor tissue matched with normal specimens with solid malignant neoplasms to detect tumor gene alterations in a 607-gene panel. The test is intended to provide information on somatic mutations and small insertions and deletions) for use by qualified health care professionals in accordance with professional guidelines, and is not conclusive or prescriptive for labeled use of any specific therapeutic product. NYU Langone Genome PACT is a single-site assay performed at NYU Langone Health.

Not Found

This FDA 510(k) clearance letter acknowledges the substantial equivalence of the "NYU Langone Genome PACT (Genome Profiling of Actionable Cancer Targets)" device. However, it does not contain the detailed acceptance criteria and study results typically found in the predicate device’s 510(k) summary or the manufacturer’s submission. This document is a regulatory clearance letter, not a detailed technical report.

Therefore, I cannot provide the requested information based solely on the provided text.

To answer your request, I would need access to the actual 510(k) submission summary for K202304, which would describe the device's performance characteristics and the supporting studies.

What is available from the provided text:

• Device Name: NYU Langone Genome PACT (Genome Profiling of Actionable Cancer Targets)
• Intended Use: Qualitative in vitro diagnostic test using targeted next-generation sequencing of FFPE tumor tissue matched with normal specimens with solid malignant neoplasms to detect tumor gene alterations in a 607-gene panel. Intended to provide information on somatic mutations and small insertions/deletions for use by qualified healthcare professionals.
• Regulatory Class: Class II
• Product Code: PZM
• Regulation Number: 21 CFR 866.6080
• Regulation Name: Next generation sequencing based tumor profiling test
• Type of Use: Prescription Use

The following information remains unknown from the provided text:

1. A table of acceptance criteria and the reported device performance.
2. Sample size used for the test set and the data provenance.
3. Number of experts used to establish the ground truth for the test set and their qualifications.
4. Adjudication method for the test set.
5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, and if so, the effect size.
6. If a standalone performance (algorithm only) was done.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.).
8. The sample size for the training set.
9. How the ground truth for the training set was established.

Ask a specific question about this device

The PGDx elio™ tissue complete assay is a qualitative in vitro diagnostic device that uses targeted next generation sequencing of DNA isolated from formalin-fixed, paraffin-embedded tumor tissue from patients with solid malignant neoplasms to detect tumor gene alterations in a broad multi-gene panel.

PGDx elio tissue complete is intended to provide tumor mutation profiling information on somatic alterations (SNVs, small insertions and deletions, one amplification and four translocations), microsatellite instability (MSI) and tumor mutation burden (TMB) for use by qualified healthcare professionals in accordance with professional guidelines in oncology for previously diagnosed cancer patients, and is not conclusive or prescriptive for labeled use of any specific therapeutic product.

PGDx elio tissue complete is an in vitro diagnostic assay that uses NGS to detect turnor gene alterations in genomic DNA isolated from formalin-fixed, paraffin-embedded (FFPE) tumor tissue from a variety of tumor types, using a targeted panel (505 genes). The assay takes less than 7 days from DNA to report and provides information on single nucleotide variants (SNVs) in a range of GC content and genomic contexts, insertion/ deletions (indels), 1 amplification as well as 4 translocations. It also identifies microsatellite instability based on select mononucleotide tracts and signatures of sequence mutations. The PGDx elio tissue complete assay utilizes a ~1.3 Mb region of interest (ROI) to calculate tumor mutation burden (TMB).

The provided text describes the analytical validation studies for the PGDx elio tissue complete assay, a qualitative in vitro diagnostic device for tumor profiling.

Here's an analysis of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria & Reported Device Performance

The acceptance criteria are generally established as target performance metrics (e.g., call rates, concordance rates, %CV, false positive rates). The reported device performance is the outcome of the analytical studies.

2. Sample Size Used for the Test Set and Data Provenance

• Specificity (LoB): 2 NIST reference standards (NA24531, NA24385) with 5 replicates each. Unique test cases from normal FFPE samples (n=63 for hotspot SNVs).
• Sensitivity (LoD) - SNVs and Indels: 10 unique clinical cases with 10 replicates each across 2 kit lots (total 200 observations). Additional evaluations: 11 SNVs, 3 insertions, 5 deletions from 5 clinical FFPE specimens with 5 replicates per dilution level. Cell lines involved 3 samples with 10 replicates at 5 dilution levels (150 observations).
• Sensitivity (LoD) - ERBB2, ALK, RET, NTRK2, NTRK3, MSI: 7 clinical FFPE cases (diluted) with 10 replicates per kit lot across 2 lots for translocations/amplifications. 3 MSI-H cases with 10 replicates each.
• TMB and Tumor Purity: 8 clinical FFPE cases. Samples 1-5: 3 levels, 5 replicates each (18 total per sample). Sample 6: 5 levels, 10 replicates each (50 total). Samples 7-8: 5 levels, 5 replicates each (25 total).
• DNA Extraction: 3 FFPE specimens and 1 cell line, extracted in duplicate by 2 operators using 3 methods (48 total extraction results), processed in duplicate (96 observations).
• DNA Input: 4 unique FFPE samples in triplicate at 5 input levels (10, 25, 50, 100, 200 ng).
• Sample Carryover & Cross-contamination: 29 FFPE samples (7 positive, rest negative).
• Interfering Substances: DNA from FFPE samples tested in presence of substances (5 test cases, 8 experimental + 2 baseline conditions).
• Endogenous Interference (Necrosis/Age): 448 samples for necrosis (0-75%), 378 for age (0-253 months) from the accuracy study enrollment (521 total).
• Overall Clinical FFPE Sample Acceptance Rate: 2874 unique clinical cases from >40 tumor types.
• Accuracy: 582 samples with PGDx elio tissue complete data and orthogonal data. Specifically:
  • SNVs/Indels: 582 samples.
  • ERBB2 Amplifications (FISH comparison): 147 cases (all); 120 cases (excluding borderline).
  • ALK Translocations (FISH comparison): 71 cases. Additional in silico simulation on 10 clinical samples (410 observations).
  • RET Translocations (FISH comparison): 27 cases. 3 RET translocation-positive cell lines also tested.
  • TMB: 118 cases across 8 tumor types.
  • MSI: 283 samples across 18 tumor types.
  • Wild Type Calls: 112 specimens.
• Interlaboratory Reproducibility: 13 FFPE tissue specimens and 1 cell line (14 samples total), tested in duplicate by 2 operators on 12 sequencing runs across 3 non-consecutive days at 3 independent laboratory sites (504 total replicates).
• Lot to Lot Precision: 5 test cases in triplicate across 3 unique kit lots (45 observations).

Data Provenance: The document does not explicitly state the country of origin for the clinical samples. However, it mentions "clinical FFPE specimens," implying real-world patient data. The studies are described in a factual manner, suggesting they are retrospective analyses of collected samples for analytical validation purposes.

3. Number of Experts Used to Establish Ground Truth and Qualifications

The document does not specify the number or qualifications of experts involved in establishing the ground truth for the test set.

Instead, the ground truth for performance evaluation (accuracy) is established by:

• Orthogonal methods:
  • "2 NGS targeted panels" and "PCR" (for SNVs and Indels).
  • "ERBB2 FISH," "ALK FISH," "RET FISH" (for amplifications and translocations).
  • "matched tumor-normal whole exome sequencing results" (for TMB).
  • "MSI PCR" (for MSI).
• Validated assays/literature: For the 3 RET translocation-positive cell lines.

4. Adjudication Method for the Test Set

The document does not describe any expert adjudication process for resolving discrepancies or establishing ground truth for the test set. The ground truth is primarily based on the results from the orthogonal methods. For the in silico studies, the ground truth is derived from down-sampling "clinical samples."

5. MRMC Comparative Effectiveness Study

No multi-reader multi-case (MRMC) comparative effectiveness study was mentioned or performed. This device is a molecular diagnostic assay, not an imaging AI tool, and thus comparative effectiveness with human readers improving with AI assistance is not applicable in this context.

6. Standalone Performance

Yes, the entire document describes studies of the standalone (algorithm only, without human-in-the-loop performance) of the PGDx elio tissue complete assay. The goal is to demonstrate the analytical performance of the automated workflow, from sample preparation to data analysis and variant calling, against established ground truth methods.

7. Type of Ground Truth Used

The primary type of ground truth used is orthogonal methods, specifically:

• Other NGS targeted panels: For SNVs and indels.
• PCR: For certain SNVs/indels and MSI.
• FISH (Fluorescence In-Situ Hybridization): For gene amplifications (ERBB2) and translocations (ALK, RET).
• Whole Exome Sequencing (WES): For Tumor Mutation Burden (TMB).
• Cell line characterization/literature: For known variants in control cell lines.

This is a form of reference standard ground truth, where the device's performance is compared against established, validated measurement techniques.

8. Sample Size for the Training Set

The document details analytical validation (testing) and reproducibility studies. It does not provide information about a "training set" size. As a molecular diagnostic assay, its development likely involves internal data for algorithm development and optimization, but specific training set sizes are not mentioned in this regulatory submission, which focuses on validation data.

9. How the Ground Truth for the Training Set Was Established

Since the document does not discuss a specific "training set" or its size, it also does not describe how ground truth for such a set was established. The focus is on the performance of the final, locked-down algorithm against independent test data with ground truth established by orthogonal methods.

Ask a specific question about this device

The Omics Core assay is a qualitative in vitro diagnostic test that uses targeted next generation sequencing of formalin-fixed paraffinembedded tumor tissue matched with normal specimens with solid malignant neoplasms to detect tumor gene alterations in a broad multi gene panel. The test is intended to provide informations (point mutations (point mutations and deletions) and tumor mutational burden (TMB) for use by qualified health care professionals in accordance with professional guidelines, and is not conclusive or prescriptive for labeled use of any specific therapeutic product. Omics Core is a single-site assay performed at NantHealth, Inc.

The NantHealth Omics Core assay is a custom targeted whole exome sequencing platform, utilizing solution-phase exon capture and sequencing, to report somatic alterations (point mutations, small insertions and deletions) in 468 genes and sequencing of 19,396 protein-coding genes (whole exome) to determine overall tumor mutation burden in tumor specimens. Genomic DNA is extracted from both a tumor and a patient-matched normal control sample. Sequence libraries are prepared through a series of enzymatic steps including shearing of double-stranded DNA, end repair, A-base addition, ligation of barcoded sequence adaptors, and low cycle PCR amplification. Single barcoded sequence libraries are captured using the biotinylated probes. Captured DNA fragments are then pooled and sequenced on an Illumina NovaSeq 6000 as paired-end reads. Sequence reads are then aligned to the reference human genome. Somatic alterations are identified in the tumor DNA by direct comparison to the matched normal DNA.

The Omics Core assay, a targeted next-generation sequencing test, was evaluated to determine its performance specifications, including precision, analytical sensitivity (Limit of Detection), and accuracy.

Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance:

The document doesn't explicitly list "acceptance criteria" as a separate column with specific numerical thresholds for each performance metric. Instead, it presents "Performance Specifications" as the expected or demonstrated performance of the device based on the studies. The reported device performance is then detailed under each characteristic.

Table: Omics Core Performance Specifications and Reported Device Performance

2. Sample Size Used for the Test Set and the Data Provenance:

• Precision Studies:
  • Panel-Wide Reproducibility, Per Specimen, and TMB Precision: 12 FFPE clinical tumor samples and 1 commercial cell line.
  • Well-Characterized Reference Material: 1 well-characterized sample.
• Analytical Sensitivity (LoD) Studies:
  • SNVs and Indels: 13 FFPE clinical samples.
  • TMB: 10 FFPE tumor specimens.
• Accuracy – Comparison to Orthogonal Method: 401 FFPE tumor samples.
• Accuracy – Supplemental Method Comparisons Study for Wildtype Calls: 220 positive mutations and 12,612 wildtype calls (implied as part of a larger dataset from clinical samples, likely from the accuracy comparison or other internal data).

Data Provenance: The document does not explicitly state the country of origin for the clinical samples. They are referred to as "FFPE clinical tumor samples" and "clinical specimens," suggesting they are real-world patient samples. The studies are described as "Performance Data" supporting the submission, implying they are retrospective analyses of collected samples.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and the Qualifications of Those Experts:

The document does not provide information on the number of experts or their qualifications specifically for establishing the ground truth for the test set.

For the predicate device (MSK-IMPACT), it mentions "Classification criteria were developed by MSK using the in-house OncoKB database. OncoKB undergoes periodic updates through the review of new information by a panel of experts."

For the subject device (Omics Core), it states: "Classification criteria were developed by NantHealth, Inc. NantHealth periodically updates Omics Core through the review of new information available."

However, this refers to the curation and classification of clinical evidence and mutations, not the establishment of ground truth for the analytical performance studies (precision, accuracy, LoD). The ground truth for the analytical studies appears to be based on:

• Orthogonal methods: For accuracy studies, results from an alternative, validated method are used for comparison.
• Known characteristics: For precision with a well-characterized reference material, the known variants in that material serve as ground truth.
• Internal consistency/comparison: For precision and LoD studies, the consistency and detection rates across replicates and dilutions are evaluated.

4. Adjudication Method for the Test Set:

The document does not describe an explicit adjudication method (e.g., 2+1, 3+1) for establishing ground truth for the test set. The nature of the studies (NGS analytical performance) relies on comparison to orthogonal methods or predefined characteristics of reference materials rather than expert review of images or clinical reports that might require such adjudication.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:

No, an MRMC comparative effectiveness study was not done. This device is a diagnostic test (Next Generation Sequencing based tumor profiling test), not an imaging AI device that would typically involve human "readers" and MRMC studies. The "performance data" relate to the analytical validity of the test (how accurately it detects gene alterations and TMB), not its impact on human clinical decision-making or interpretation.

6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done:

The studies described primarily represent the standalone (algorithm and laboratory process only) performance of the Omics Core assay in detecting mutations and TMB. The "human-in-the-loop" aspect is implicitly through laboratory technicians performing the test and the medical director signing off on reports, but the reported performance metrics (precision, accuracy, LoD) assess the analytical capabilities of the system itself, independent of immediate human interpretive assistance on a case-by-case basis during the core analysis.

7. The Type of Ground Truth Used:

The ground truth used in the performance studies varied:

• Orthogonal Method: For accuracy studies of SNVs, indels, and TMB, the Omics Core results were compared to results obtained from an "orthogonal method." The specific method is not detailed, but it implies a different, validated technology used to confirm the presence and characteristics of mutations.
• Well-Characterized Reference Material: For precision, a "well-characterized sample" (likely with known variant profiles) was used.
• Clinical Samples: For precision and LoD, findings from "FFPE clinical tumor samples" and "commercial cell line" were used, with the "truth" being established through consistent detection across replicates or successful reliable calling at certain frequencies. This implies that for clinical samples, the ground truth was based on the consensus of repeated measurements or the expectation of what should be detected in those samples.

The document does not mention pathology re-read, expert consensus, or outcomes data as direct ground truth for these analytical performance studies.

8. The Sample Size for the Training Set:

The document does not specify the sample size for the training set. The provided performance data relates to verification and validation studies (test set), not the development or training of the assay's bioinformatics component. The "Omics Core software" is mentioned as part of the bioinformatics workflow, but details on how its underlying algorithms (for variant calling, TMB calculation) were trained are not provided.

9. How the Ground Truth for the Training Set Was Established:

As the document does not specify a training set sample size, it also does not explain how the ground truth for a training set was established. The process description focuses on the analytical performance of the finished device. It mentions "Assay optimization" and that "High analytical specificity is maintained by paired tumor/matched normal sequencing, and established during assay optimization," suggesting that internal data or processes were used during development, but without details on ground truth establishment for such activities.

Ask a specific question about this device

The MSK-IMPACT assay is a qualitative in vitro diagnostic test that uses targeted next generation sequencing of formalin-fixed paraffin-embedded tumor tissue matched with normal specimens from patients with solid malignant neoplasms to detect tumor gene alterations in a broad multi gene panel. The test is intended to provide information on somatic mutations (point mutations and small insertions and deletions) and microsatellite instability for use by qualified health care professionals in accordance with professional guidelines, and is not conclusive or prescriptive for labeled use of any specific therapeutic product. MSK-IMPACT is a single-site assay performed at Memorial Sloan Kettering Cancer Center.

A description of required equipment, software, reagents, vendors, and storage conditions were provided, and are described in the product labeling (MSK-IMPACT manual). MSK assumes responsibility for the device.

Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text for the MSK-IMPACT assay.

Device Name: MSK-IMPACT (Integrated Mutation Profiling of Actionable Cancer Targets)
Type of Test: Next generation sequencing tumor profiling test
Purpose: Qualitative in vitro diagnostic test for detecting somatic mutations (point mutations, small insertions and deletions) and microsatellite instability (MSI) in formalin-fixed paraffin-embedded (FFPE) tumor tissue matched with normal specimens from patients with solid malignant neoplasms.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are generally embedded within the "Performance" section and the "Reporting" section's "Table 3. Sample Level Quality Control Metrics." The reported performance is found throughout the "Performance" section.

2. Sample Size Used for the Test Set and Data Provenance

• Test Set Sample Sizes:

  • Precision Studies:
    • 10 samples (9 FFPE specimens, 1 commercial cell line) used for panel-wide reproducibility (Table 6).
    • Well-characterized reference standard (HapMap cell line NA20810) in 23 replicates for sequencing error rates and reproducibility.
    • 12 specimens (6 MSI-H, 6 MSS) for MSI precision.
  • Analytical Sensitivity (LoD):
    • Part 1 (Dilution Series): Patient samples (number not specified, but for 5 validation exons, implying at least 5 patients) with 5-8 serial dilutions.
    • Part 2 (Confirmation): Unspecified number of samples, providing variants for 3 deletions, 4 insertions, and 6 SNVs, each tested with 5 replicates.
    • MSI LoD: CRC specimens (number not specified, but 5 replicates run).
  • DNA Input Assessment: Unspecified number of samples (historical data from >10,000 samples mentioned in pre-analytical performance context). Table 13 presents data by DNA input amounts but not sample count for each bin.
  • Accuracy (Method Comparison):
    • 267 unique mutations in 433 FFPE tumor specimens for the main comparison (Table 14).
    • 95 specimens for the supplemental wildtype calls study.
    • 138 colorectal cancer (CRC) and 40 endometrial carcinoma (EC) specimens (training set) for MSI cutoff establishment.
    • 135 CRC patients (66 with both MSK-IMPACT and IHC) for MSI cutoff validation.
    • 119 unique non-CRC and non-EC tumor-normal pair samples for MSI comparison in other cancer types.

• Data Provenance:

  • General: The device is performed at Memorial Sloan Kettering Cancer Center (MSK), indicating the data likely originates from their patient population.
  • Retrospective/Prospective:
    • The pre-analytical performance (specimen invalid rates) used historical data from >10,000 specimens, implying a retrospective chart review.
    • The MSI validation study (CRC patients) was a retrospective-prospective chart review.
    • The clinical performance section mentions a large-scale, prospective clinical sequencing initiative using MSK-IMPACT involving >10,000 patients, whose data are publicly accessible. This cohort likely informed the broader context and understanding of the device but was not explicitly stated as the test set for the analytical validation.
    • The analytical performance studies (precision, LoD, accuracy) used clinical samples/specimens, which could be retrospective or prospectively collected for the purpose of the study. The text doesn't explicitly state for each study.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and the Qualifications of Those Experts

The text does not specify the number of experts used to establish the ground truth for the test set, nor their specific qualifications (e.g., "radiologist with 10 years of experience").

However, it does indicate:

• For the accuracy studies, results were compared to "original results obtained with the validated orthogonal methods." This implies that the ground truth was established by these validated orthogonal methods, which are presumably performed and interpreted by qualified personnel using established clinical diagnostics.
• For MSI, the MSIsensor results were compared to "a validated MSI-PCR or MMR IHC test," a "commercially available PCR assay," or a "validated IHC panel (MLH1, MSH2, MSH6 and PMS2)." Again, this suggests ground truth from established, clinical laboratory methods.
• The "Clinical Evidence Curation" section mentions that "OncoKB undergoes periodic updates through the review of new information by a panel of experts," which informs the clinical interpretation of detected mutations. This expert panel contributes to the broader clinical context of the mutations, but not directly the ground truth for the analytical test set itself.

4. Adjudication Method (e.g., 2+1, 3+1, none) for the Test Set

The text does not describe a formal adjudication method (like 2+1 or 3+1 consensus with experts) for establishing the ground truth of the test set cases. Instead, the ground truth was derived from "validated orthogonal methods."

For example, in the accuracy study, the MSK-IMPACT results were "compared to the original results obtained with the validated orthogonal methods." This indicates that the results from the comparison methods served as the reference standard, rather than requiring an additional expert adjudication process on top of those existing validated methods.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done

No, an MRMC comparative effectiveness study was not described. This document pertains to the analytical validation of a genetic sequencing assay, which inherently does not involve human readers interpreting images in a multi-reader, multi-case setup. Therefore, a comparative effectiveness study measuring human reader improvement with AI assistance (which is typical for imaging AI) is not applicable here.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

Yes, the analytical performance studies (precision, analytical sensitivity, and analytical accuracy) described are all measures of the standalone performance of the MSK-IMPACT assay, which relies on its sequencing and bioinformatics pipeline without direct human-in-the-loop diagnostic interpretation to produce the raw mutation calls.

• The "Mutation calling SNVs and Indels" section and "Summary of mutation filtering scheme" (Figure 1) describe the automated pipeline for identifying mutations.
• The "Performance" section details how characteristics like precision, LoD, and accuracy were determined for the assay itself by comparing its outputs to known or established results from other validated methods. These do not involve a human interpreting the device's output to make a diagnosis within the performance evaluation but rather assess the accuracy of the device's genomic calls directly.

7. The Type of Ground Truth Used

The primary type of ground truth used was:

• Orthogonal Methods / Comparator Assays: For the accuracy studies, the MSK-IMPACT results were compared against "original results obtained with the validated orthogonal methods." This included comparison to:
  • Validated orthogonal methods for SNVs and indels.
  • Established MSI-PCR or MMR IHC tests for Microsatellite Instability status.
• Known Reference Material: For precision, a "well characterized reference standard (HapMap cell line NA20810)" was used, with reference genotypes obtained from the 1000 Genomes database.
• Expected Values/Dilution Series: For Limit of Detection studies, serial dilutions of patient samples with "known mutations" and "expected frequencies" were used.

Therefore, the ground truth is a combination of established methods, known reference materials, and empirically derived expected values.

8. The Sample Size for the Training Set

The document explicitly mentions training data primarily in the context of the MSI cutoff:

• MSI Cutoff Training: A "training specimen dataset consisting of 138 colorectal cancer (CRC) and 40 endometrial carcinoma (EC) specimens with matched normal and having MSI status results from a validated MSI-PCR or MMR IHC test."

For the mutation calling pipeline (SNVs and indels), the text refers to:

• Optimization of thresholds: "The threshold values for the filtering criteria were established based on paired-sample mutation analysis on replicates of normal FFPE samples, and optimized to reject all false positive SNVs and almost all false positive indel calls from the reference dataset." The size of this "reference dataset" or "replicates of normal FFPE samples" used for training/optimization of filtering thresholds is not explicitly stated as a defined "training set sample size" for the SNV/indel calling. It implies an internal dataset used during development.

9. How the Ground Truth for the Training Set Was Established

For the MSI cutoff training set:

• The ground truth was established by "validated MSI-PCR or MMR IHC test" results. These are existing, established clinical diagnostic methods for determining MSI status.

For the SNV/indel pipeline optimization/threshold establishment:

• The ground truth for optimizing filtering thresholds was based on "paired-sample mutation analysis on replicates of normal FFPE samples" and "reference dataset." This suggests that the "true" status of these calls (i.e., whether they were true positives, false positives, etc.) would have been known or definitively determined through external means (e.g., highly confident calls from a different (perhaps more laborious or deeply sequenced) method, or a known characteristic of the "normal FFPE samples"). However, the specific method for establishing this ground truth for the filtering optimization is not explicitly detailed beyond being from a "reference dataset."

Ask a specific question about this device