Search Results

The CDC DENV-1-4 Real-Time RT-PCR Assay is intended for use on an Applied Biosystems (ABI) 7500 Fast Dx Real-Time PCR Instrument:

• For the diagnosis of dengue in serum or plasma collected from patients with signs and symptoms consistent with dengue (mild or severe) during the acute phase:
• For the identification of dengue virus serotypes 1, 2, 3 or 4 from viral RNA in serum or plasma (sodium citrate) collected from human patients with dengue during the acute phase;
• To provide epidemiologic information for surveillance of circulating dengue viruses.

The CDC DENV-1-4 Real-Time RT-PCR Assay is a panel of oligonucleotide primers and dual-labeled hydrolysis (TaqMan®) probes designed to be utilized in real-time RT-PCR assays using the ABI 7500 Fast Dx Real-Time PCR instrument for detection of DENV-1, -2, -3 and -4 viral RNA in serum or plasma from human patients with dengue.

The CDC DENV-1-4 Real-Time RT-PCR Assay includes control materials as described below:

• . Positive control materials, which consist of heat-inactivated DENV-1 Haw, DENV-2 NGC, DENV-3 H87, and DENV-4 H241 viruses.
• A Human Specimen Control (HSC), consisting of noninfectious (beta ● propiolactone inactivated) cultured human cell material that provides a positive signal in the assay and demonstrates successful recovery of RNA as well as the integrity of the RNA extraction reagents.
• . Internal positive control, the human RNase P (RP) primer and probe set detects human RP and is used with each clinical specimen to indicate that adequate isolation of nucleic acid resulted from the extraction of the specimen. A positive RP assay result also ensures that common reagents and equipment are functioning properly and demonstrates the absence of inhibitory substances.

RNA extractions can be done using the Qiagen OIAamp® DSP Viral RNA Mini Kit either manually or in combination with the QIAcube Instrument (small centrifuge) and/or Roche MagNA Pure LC total nucleic acid isolation kit and Magna Pure Instrument.

The RT-PCR assay is run on the Applied Biosystems® 7500 Fast Dx Real-Time PCR instrument using Invitrogen SuperScript™ III Platinum® One-Step Quantitative RT-PCR System. The assay can be run in singleplex (each DENV serotype detected in a separate reaction) or in multiplex (four DENV serotypes are run in the same reaction). The two formats provide equal sensitivity.

Acceptance Criteria and Device Performance for CDC DENV-1-4 Real-Time RT-PCR Assay

1. Table of Acceptance Criteria and Reported Device Performance

For the CDC DENV-1-4 Real-Time RT-PCR Assay, the acceptance criteria are implicitly derived from the performance goals demonstrated in the clinical studies. These studies evaluate the assay's ability to accurately detect and serotype Dengue Virus RNA in human serum samples.

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

• Reproducibility Study:

  • For each DENV serotype (DENV-1, -2, -3, -4) and control type (moderate positive, low positive, high negative, negative), 20 samples were tested per site (total of 60 samples per condition across 3 sites).
  • Data Provenance: Cultured, quantified (pfu/mL) stocks of whole virus (strains DENV-1 Haw, DENV-2 NGC, DENV-3 H87, and DENV-4 H241) spiked into dengue-negative human serum. The study was performed at three testing sites, including the CDC Dengue Branch Lab and two external independent sites. This appears to be a prospective experimental study.

• Limit of Detection (LoD) Study:

  • For each DENV serotype, dilution, and matrix (serum/plasma), 5 replicas were tested.
  • Data Provenance: panels prepared from cultured, quantified (pfu/mL) stocks of whole virus (strains DENV-1 Haw, DENV-2 NGC, DENV-3 H87, DENV-4 H241) serially diluted into dengue-negative human serum or plasma. This is a retrospective/controlled laboratory study using spiked samples.

• Analytical Reactivity Study:

  • For each of the 29 DENV-1-4 isolates, triplicate samples were tested at 10³ pfu/mL and 10² pfu/mL dilutions.
  • Data Provenance: 29 DENV-1 - 4 isolates obtained from patients from different countries, cultured and quantified. This is a retrospective/controlled laboratory study using cultured clinical isolates.

• Cross-Reactivity Study:

  • Triplicate samples for each of the 12 pathogens were tested.
  • Data Provenance: Nucleic acids extracted from 12 organisms (including WNV, YFV, SLEV, CHIKV, HSV-1/2, CMV, VZV, Leptospira, Borrelia, HCV, HAV). Ten of these were spiked into human serum (confirmed negative for dengue virus); HCV and HAV were clinical serum samples with unknown concentration. This is a retrospective/controlled laboratory study using spiked samples and some clinical samples.

• Interference Study:

  • For each DENV serotype, viral dilution (1:10 above LoD, at LoD, 1:10 below LoD) and interfering substance, three replicate samples were tested.
  • Data Provenance: Cultured, quantified (pfu/mL) stocks of whole virus (strains DENV-1 Haw, DENV-2 NGC, DENV-3 H87, DENV-4 H241) diluted into normal human serum (NHS) or NHS containing specific interfering substances. This is a retrospective/controlled laboratory study using spiked samples.

• Carry-Over/Cross Contamination Study:

  • 8 high-positive and 8 high-negative replicas were tested in an alternating series for each DENV serotype. Total of 32 negative and 32 positive samples tested.
  • Data Provenance: Cultured, quantified (pfu/mL) stocks of DENV-1, -2, -3 and -4 diluted to high-positive and high-negative concentrations. This is a retrospective/controlled laboratory study using spiked samples.

• Fresh vs. Frozen Specimens Study:

  • Moderate and low positive dilutions prepared in triplicate sets, subjected to 5 freeze/thaw cycles.
  • Data Provenance: Spiked human serum samples. This is a retrospective/controlled laboratory study using spiked samples.

• Clinical Performance - Prospective Study:

  • Sample Size: 86 serum samples.
  • Data Provenance: Prospectively collected from dengue-suspected, febrile patients during the first 5 days of symptoms at 3 public health laboratories (2009-2011). Country of Origin: Not explicitly stated, but likely US (given public health labs) or US territories (e.g., Puerto Rico). This is a prospective study.

• Clinical Performance - Retrospective Study:

  • Sample Size: 371 serum samples.
  • Data Provenance: Archived CDC routine dengue surveillance specimens collected in pairs (acute/convalescent) during 2007-2011. Country of Origin: Non-United States dengue cases (n=39); Puerto Rico dengue cases (n=82); 250 negative cases from Puerto Rico. This is a retrospective study.

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

• For Analytical Studies (Reproducibility, LoD, Analytical Reactivity, Cross-Reactivity, Interference, Carry-Over, Fresh vs. Frozen): The ground truth was established by preparing samples with known concentrations of specific DENV serotypes or other pathogens, often quantified by plaque-forming units (pfu/mL). Therefore, no human experts were explicitly used or required to establish the ground truth for these in vitro analytical studies. The accuracy relies on the lab's ability to precisely prepare and quantify these samples.

• For Clinical Performance - Prospective Study:

  • Ground Truth Method: Bi-directional sequencing of the DENV E gene (1485 bp). Genetic sequencing is considered a highly definitive molecular method for pathogen identification and sub-typing.
  • Number of Experts/Qualifications: The document does not specify the number or qualifications of experts involved in performing or interpreting the sequencing results. However, molecular sequencing is typically performed and interpreted by trained laboratory professionals, often with expertise in molecular diagnostics, virology, and bioinformatics.

• For Clinical Performance - Retrospective Study:

  • Ground Truth Method: IgM anti-DENV Capture Enzyme Linked Immunosorbent Assay (MAC-ELISA) for seroconversion, validated in-house by CDC, combined with bi-directional sequencing of the DENV E gene (1485 bp) for confirmation of RT-PCR positive samples. IgM seroconversion is a standard diagnostic method for recent dengue infection.
  • Number of Experts/Qualifications: The document does not specify the number or qualifications of experts involved in performing or interpreting the MAC-ELISA or sequencing results. Similar to the prospective study, skilled laboratory professionals would perform these tests.

4. Adjudication Method for the Test Set

The document does not describe an explicit "adjudication method" in the sense of a committee of experts reconciling conflicting results for the test sets.

• For the prospective clinical study, the CDC DENV-1-4 RT-PCR Assay results were directly compared to bi-directional sequencing results. Any discrepancies were noted (e.g., one sample negative by RT-PCR but positive by sequencing).
• For the retrospective clinical study, RT-PCR results were compared to IgM anti-DENV seroconversion. For discrepant or positive RT-PCR results, bi-directional sequencing was performed for corroboration. The document highlights specific discrepancies and their resolution by sequencing.

This approach suggests a direct comparison to a reference standard (sequencing or seroconversion), with sequencing serving as a higher-level confirmation for discrepancies rather than an adjudicated consensus amongst multiple independent readers.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, What was the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance

This section is Not Applicable (N/A). The CDC DENV-1-4 Real-Time RT-PCR Assay is an in vitro molecular diagnostic test designed for automated or semi-automated analysis and detection of DENV RNA. It is not an imaging or diagnostic device that involves "human readers" interpreting results in the way an AI-assisted diagnostic tool might, for which an MRMC study would be relevant. The output (Ct values) is typically interpreted against predefined thresholds by trained laboratory personnel, rather than a subjective interpretation by a "reader." Therefore, no MRMC study or AI assistance effect size is reported.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

Yes, a standalone (algorithm only) performance was effectively performed. The CDC DENV-1-4 Real-Time RT-PCR Assay, as a molecular diagnostic test, operates autonomously once the sample preparation (RNA extraction) is complete and the reagents are loaded into the ABI 7500 Fast Dx Real-Time PCR instrument. The instrument's software detects fluorescence amplification and determines Ct values without human intervention during the detection phase. The "algorithm" here is the chemical reaction itself driven by the primers and probes, and the instrument's software for signal detection and thresholding. The results (presence/absence of DENV RNA, and serotype) are derived directly from the instrument's output. Human involvement comes in interpreting these results according to the provided Users Guide (Section M. Performance Characteristics and the Clinical Studies sections predominantly represent this standalone performance, demonstrating the assay’s inherent accuracy in detecting DENV RNA and distinguishing serotypes).

7. The Type of Ground Truth Used

• For Analytical Studies: The ground truth was primarily based on spiked samples with known concentrations of cultured, quantified viruses (pfu/mL) and other pathogens.
• For Clinical Performance - Prospective Study: Bi-directional sequencing of the DENV E gene (1485 bp) was used as the ground truth.
• For Clinical Performance - Retrospective Study: A combination of IgM anti-DENV seroconversion (using CDC MAC-ELISA) and bi-directional sequencing of the DENV E gene was used. Sequencing served to corroborate RT-PCR positive results and to resolve discrepancies from seroconversion.

8. The Sample Size for the Training Set

The document does not explicitly describe a "training set" in the context of an machine learning algorithm. This is a molecular diagnostic assay that uses defined reagents (primers and probes) rather than a machine learning model that would require a dedicated training set. The assay's parameters (e.g., thermal cycling conditions, detection thresholds within the instrument software) would have been optimized during its development and validation phase, which might involve iterative testing on various samples, but these are not explicitly termed a "training set" in the context used for AI/ML.

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

Since there is no explicitly defined "training set" for an AI/ML model for this molecular diagnostic assay, this point is Not Applicable (N/A). The assay's design (primers, probes) and operational parameters were established through standard molecular biology and assay development procedures, which inherently ensure that the chemical reactions are specific and sensitive to the target RNA.

Ask a specific question about this device