Computer-Aided Diagnosis Tool for Lung Nodules

(ARCADES Trial)

Recruiting at 2 trial locations

Overseen By Anil Vachani, MD, MSCE

Age: 18+

Sex: Any

Trial Phase: Academic

Sponsor: Abramson Cancer Center at Penn Medicine

Disqualifiers: Lung cancer, Active cancer, others

No Placebo Group All trial participants will receive the active study treatment (no placebo)

Approved in 2 Jurisdictions This treatment is already approved in other countries

Compensation: Varies

Number of Visits: Unknown

300 Participants Needed

Trial Summary

What is the purpose of this trial?

This is a pragmatic clinical trial that will study the effect of a radiomics-based computer-aided diagnosis (CAD) tool on clinicians' management of pulmonary nodules (PNs) compared to usual care. Adults aged 35-89 years with 8-30mm PNs evaluated at Penn Medicine PN clinics will undergo 1:1 randomization to one of two groups, defined by the PN malignancy risk stratification strategy used by evaluating clinicians: 1) usual care or 2) usual care + use of a radiomics-based CAD tool.

Do I need to stop my current medications for the trial?

The trial information does not specify whether you need to stop taking your current medications. It's best to discuss this with the trial coordinators or your doctor.

What data supports the effectiveness of the treatment Optellum Virtual Nodule Clinic for lung nodules?

Research shows that virtual nodule clinics, like the Optellum Virtual Nodule Clinic, provide high-quality care and help patients follow medical guidelines, leading to patient satisfaction and cost savings. Additionally, computer-aided diagnosis tools have been shown to help doctors identify more lung nodules and assess their risk, which can guide personalized patient care. ^{1 2 3 4 5}

Is the Computer-Aided Diagnosis Tool for Lung Nodules safe for humans?

The research articles provided do not contain specific safety data for the Computer-Aided Diagnosis Tool for Lung Nodules or its related names. They focus on lung nodule identification and management, but do not address safety concerns directly. ^{2 6 7 8 9}

How does the computer-aided diagnosis tool for lung nodules differ from other treatments?

The computer-aided diagnosis (CAD) tool for lung nodules is unique because it uses advanced computer systems to automatically detect and assess lung nodules on chest CT scans, helping radiologists identify more nodules and evaluate changes over time. This tool enhances diagnostic accuracy and workflow efficiency, unlike traditional methods that rely solely on manual evaluation by radiologists. ^{3 4 5 9 10}

Research Team

Roger Y. Kim, MD, MSCE

Principal Investigator

University of Pennsylvania

Eligibility Criteria

Adults aged 35-89 with newly discovered solid pulmonary nodules (8-30mm) on CT scans, scheduled for evaluation at a pulmonary nodule clinic. Participants must have CT imaging compatible with the Optellum Virtual Nodule Clinic software available by their first clinic visit.

Inclusion Criteria

Chest CT imaging compatible with Optellum Virtual Nodule Clinic software

Scheduled to be evaluated at a pulmonary nodule clinic

I am between 35 and 89 years old.

Exclusion Criteria

I have had a low-dose CT scan of my chest.

I have had lung cancer in the past.

I have an implant in my chest that makes it hard to see my phrenic nerve.

Timeline

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Randomization and Initial Assessment

Participants undergo 1:1 randomization to either usual care or CAD-based risk stratification, followed by initial assessment of pulmonary nodules

1-2 weeks

Follow-up

Participants are monitored for safety and effectiveness of the management strategy, including imaging surveillance and biopsy outcomes

12 months

Treatment Details

Interventions

• Optellum Virtual Nodule Clinic

Trial Overview This trial is testing the impact of a radiomics-based computer tool called Optellum Virtual Nodule Clinic on doctor's decisions in managing lung nodules, compared to standard care without this tool. Patients are randomly placed into one of these two groups.

Participant Groups

Group I: Clinician assessment + CAD-based risk stratification Experimental Treatment 1 Intervention

In the experimental arm, evaluating clinicians will receive a Lung Cancer Prediction report from an artificial intelligence radiomics-based computer-aided diagnosis tool for risk stratification of pulmonary nodules.

Group II: Usual care (clinician assessment) Active Control 1 Intervention

In the usual care arm, clinicians will evaluate individuals with indeterminate pulmonary nodules as part of routine clinical care. No specific guidance regarding pulmonary nodule risk stratification will provided to evaluating clinicians.

Who Is Running the Clinical Trial?

Abramson Cancer Center at Penn Medicine

Lead Sponsor

Trials

425

Recruited

464,000+

Findings from Research

Indeterminate pulmonary nodules are common in asymptomatic individuals, and their management is often inconsistent due to varying guidelines and adherence issues.

Establishing a dedicated pulmonary nodule clinic can provide standardized, evidence-based care, utilizing multidisciplinary teams and advanced technology to improve patient outcomes and streamline the management process.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Systematic approach to the management of the newly found nodule on screening computed tomography: role of dedicated pulmonary nodule clinics. Korst, RJ.[2018]

The increasing use of chest high-resolution computed tomography has led to more frequent detection of lung adenocarcinomas, but there is a need for better diagnostic tools to assess the risk and manage these varying lesions effectively.

The computer-aided nodule assessment and risk yield software is a validated, automated tool that can noninvasively stratify the risk of adenocarcinoma lung nodules, correlating well with histology and patient outcomes, which may improve individualized patient management.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Computer-Aided Nodule Assessment and Risk Yield Risk Management of Adenocarcinoma: The Future of Imaging? Foley, F., Rajagopalan, S., Raghunath, SM., et al.[2019]

The Virtual Lung Nodule Clinic (VLNC) evaluated 365 patients in its first year, primarily focusing on those at high risk for lung cancer, with 43.8% discharged after their initial assessment.

Analysis revealed that many discharged patients had previously received poor management, with 66.9% not adhering to follow-up guidelines, indicating a need for improved adherence to care protocols.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Multidisciplinary virtual management of pulmonary nodules. Polanco, D., González, J., Gracia-Lavedan, E., et al.[2022]

An automated method combining diagnostic codes and natural language processing successfully identified 7112 patients with lung nodules, demonstrating high sensitivity (96%) and specificity (86%) compared to clinician review.

Among the identified patients, 14% were later diagnosed with lung cancer, highlighting the importance of this automated identification method in managing lung nodules in community settings.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Automated identification of patients with pulmonary nodules in an integrated health system using administrative health plan data, radiology reports, and natural language processing. Danforth, KN., Early, MI., Ngan, S., et al.[2022]

Computer-aided diagnosis (CAD) systems significantly enhance the ability of radiologists to detect lung nodules, improving diagnostic performance in clinical settings.

CAD can also assess nodule malignancy and treatment response through volumetry, while providing objective analysis of nodule morphology, which streamlines the workflow for follow-up studies.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

A computer-aided diagnosis for evaluating lung nodules on chest CT: the current status and perspective. Goo, JM.[2022]

The International Lung Screening Trial (ILST) will assess the effectiveness of the PLCOm2012 model compared to the U.S. Preventive Services Task Force (USPSTF) criteria in selecting participants for lung cancer screening, involving 4,500 participants over a minimum of 5 years.

The trial will also evaluate the efficiency of nodule management using the PanCan probability calculator against the Lung-RADS protocol, aiming to improve lung cancer detection rates and management strategies.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Protocol and Rationale for the International Lung Screening Trial. Lim, KP., Marshall, H., Tammemägi, M., et al.[2021]

A new computer program can quickly and flexibly simulate lung nodules in digital chest radiographs, improving the efficiency of lung nodule detection studies.

Psychophysical and statistical studies demonstrated that radiologists and nonradiologists could not visually distinguish between real and simulated lung nodules, confirming the realism and validity of the simulations.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Computer-simulated lung nodules in digital chest radiographs for detection studies. Yocky, DA., Seeley, GW., Ovitt, TW., et al.[2019]

Incidental pulmonary nodules are commonly found in CT scans, with rates as high as 48% in CT angiograms, highlighting the need for systematic management to prevent missed diagnoses of lung cancer.

The establishment of an incidental pulmonary nodule program improved the identification and follow-up of these nodules, ensuring better communication between radiologists, emergency department providers, and patients, which is crucial as the number of detected nodules is expected to rise with increased chest CT imaging.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Creating an Incidental Pulmonary Nodule Safety-Net Program. Van Gerpen, R.[2021]

An automated tool for identifying lung nodules was developed and validated using Electronic Healthcare Records from 14,586 patients, achieving high sensitivity (93%) and specificity (99%) for nodule identification, which can streamline healthcare service evaluations.

The tool demonstrated excellent inter-observer agreement (0.94 internally, 0.98 externally) and accurately predicted concerning nodule features with 94% accuracy, indicating its potential for improving the efficiency of lung nodule management in clinical settings.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Development of a Structured Query Language and Natural Language Processing Algorithm to Identify Lung Nodules in a Cancer Centre. Hunter, B., Reis, S., Campbell, D., et al.[2021]

The automated computer system successfully identified 86% of nodules in chest CT scans, detecting 318 out of 370 nodules across 16 studies involving eight oncology patients.

The system's assessment of nodule size changes closely matched that of a thoracic radiologist, with a high correlation coefficient of 0.932, indicating its potential reliability for monitoring nodule progression.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Chest CT: automated nodule detection and assessment of change over time--preliminary experience. Ko, JP., Betke, M.[2016]

References

1. United States pubmed.ncbi.nlm.nih.gov

Systematic approach to the management of the newly found nodule on screening computed tomography: role of dedicated pulmonary nodule clinics. [2018]

2. United States pubmed.ncbi.nlm.nih.gov

Computer-Aided Nodule Assessment and Risk Yield Risk Management of Adenocarcinoma: The Future of Imaging? [2019]

3. Spain pubmed.ncbi.nlm.nih.gov

Multidisciplinary virtual management of pulmonary nodules. [2022]

4. United States pubmed.ncbi.nlm.nih.gov

Automated identification of patients with pulmonary nodules in an integrated health system using administrative health plan data, radiology reports, and natural language processing. [2022]

5. Korea (South) pubmed.ncbi.nlm.nih.gov

A computer-aided diagnosis for evaluating lung nodules on chest CT: the current status and perspective. [2022]

6. United States pubmed.ncbi.nlm.nih.gov

Protocol and Rationale for the International Lung Screening Trial. [2021]

7. United States pubmed.ncbi.nlm.nih.gov

Computer-simulated lung nodules in digital chest radiographs for detection studies. [2019]

8. United States pubmed.ncbi.nlm.nih.gov

Creating an Incidental Pulmonary Nodule Safety-Net Program. [2021]

9. Switzerland pubmed.ncbi.nlm.nih.gov

Development of a Structured Query Language and Natural Language Processing Algorithm to Identify Lung Nodules in a Cancer Centre. [2021]

10. United States pubmed.ncbi.nlm.nih.gov

Chest CT: automated nodule detection and assessment of change over time--preliminary experience. [2016]