Evaluation of the Lung Nodule Sensitivity of Stationary Chest Tomosynthesis in Patients With Known Lung Nodules

Evaluation of the Lung Nodule Sensitivity of Stationary Chest Tomosynthesis in Patients With Known Lung Nodules

Evaluation of the Lung Nodule Sensitivity of Stationary Chest Tomosynthesis Versus Chest Radiographs in Patients With Known Lung Nodules

Purpose: The proposed research, if successfully implemented, will result in a low-dose, low-cost, and highly effective method for screening of lung cancer, leading to reduction of the lung cancer mortality rate. Using the stationary Digital Chest Tomosynthesis (s-DCT) system the imaging dose for a full tomosynthesis scan is expected to be only 10% of that from a low-dose computed tomography (CT). The targeted imaging time of 2s is 1/5 to 1/3 of that from a current commercial digital chest tomosynthesis (DCT) system at the same imaging dose. Beyond lung cancer screening, the low-dose and sensitive 3D lung imaging modality will likely to find applications in areas such as monitoring of pediatric cystic fibrosis patients where reduction of imaging dose is critical. Participants: One hundred (100) patients undergoing a clinical non-contrast CT for their lung nodules will be asked to have this procedure (scan) within 2 weeks (+/- 1 week) of their clinical CT and chest radiograph, with no intervening procedures or therapies. Procedures (methods): The purpose and endpoint of this study is to compare the sensitivity of the s-DCT system to the conventional chest radiographs with non-contrast chest CT as the reference standard. One hundred (100) patients undergoing a clinical non-contrast CT for their lung nodules will be asked to have this procedure (scan) within 2 weeks (+/- 1 week) of their clinical CT and chest radiograph, with no intervening procedures or therapies.

This is a one arm study of 100 patients with one or more lung nodules based on standard imaging (chest CT) at UNC Hospitals that consent to undergo an experimental s-DCT to characterize the number and extent of lung nodules. Most patients will also have undergone a standard of care (SOC) chest radiograph (CR). The sensitivity of the CR and of the experimental s-DCT will be estimated separately, using the CT as the gold-standard. The sensitivities of the CR and the s-DCT will then be compared. In the event that CRs are not included in the SOC imaging or are unusable (outside of acceptable window (i.e. do not accommodate s-DCT within 2 weeks) or poor quality), the scout image acquired by CT will be used in place of a CR. The scout image is a preliminary radiograph that is taken prior to performing a definitive CT. Working in collaboration with the University of North Carolina (UNC) Multi-Disciplinary Thoracic Oncology (MTOP) group, the investigators plan to inform patients that they may be contacted to participate in this study to evaluate the sensitivity of the s-DCT system, using the SOC CR and chest CT for comparison. CRs are included to provide a basis for comparing system sensitivity, representing a minimum standard. Once a patient has been referred, or the investigators have collaborated with the treating clinic regarding the eligibility of a subject, the patient will be approached by a coordinator from Radiology to assess interest in participation. The coordinator will either go to the treating clinic, or will call the patient at home, after he/she have been apprised of the study by his/her treating physician/nurse. If the patient is interested in participation, he/she will be consented either then (in their treatment clinic) or when he/she arrives to have his/her s-DCT, but prior to any study procedures. Review of the consent will take place in the privacy of an exam room, or when possible, a sample consent form will be sent to the patient via email prior to arriving for the scan to allow for ample review. Once the patient has consented, women of child bearing potential (WCBP) will be given a urine pregnancy test in order to ensure that they are not pregnant. If a urine pregnancy test shows a result positive for pregnancy, the patient will be excluded from the study per the exclusion criteria because the investigators cannot, in good conscience, expose a fetus to unnecessary radiation exposure. If the urine pregnancy test shows that the patient is not pregnant, she may participate in the study. The study scan, s-DCT, will be performed within two weeks of his/her clinical evaluations by chest CT and CR. There cannot be any intervening therapies or procedures (i.e. biopsy or excision of lesions) done in between the SOC imaging and the s-DCT. In the event that CRs are not included in the SOC imaging or are unusable (outside of acceptable window or poor quality), the scout image acquired by CT will be used in place of a CR. Images will be acquired by a trained radiologist technologist. The technologist will comfortably position the patient in the supine position on the imaging table. Then, all patients will have a breath held s-DCT scan in an anterior-posterior direction. Images will be reconstructed off-line and transferred for review on conventional picture archival and storage (PACS) workstations. Images will be reviewed by radiologists. A secondary objective of this study is to determine the specificity of the s-DCT system in determining malignancy. In order to assess this objective each patient will be followed for up to 1 year following study imaging. If a known nodule is biopsied, the malignancy will be determined as indicated by the pathology report. If the nodule is followed but not biopsied within 1 year, the nodule will be considered to be not malignant for the purposes of this research study. However, some instances may occur in which nodules are not biopsied and are considered malignant. In these instances, subsequent CTs and clinical reports will be used to determine malignancy during this 1 year follow-up period. A reader study will be performed after all patients have been accrued. Each of the 3 scans per patient will be de-identified. As reading of all tests will take place after the patient has undergone clinical decision making (treatment versus following), the results of this study will not affect patient care. A total of 5 radiologist readers, with at least two years of CT experience will serve as readers. Each reader will separately interpret the CR and s-DCT study for all patients. For each lesion, readers will measure a cranial caudal (CC) diameter, to evaluate the system's ability to reproduce CT size measurements. Lesion identification and averaged measurement of the CC diameter by Thoracic Radiology subspecialists measured on conventional CT with clinical interpretation will serve as reference standard for the study. Each reader will separately interpret CR and s-DCT images in random order at two different reading sessions with a one month wash-out period in between reads. The radiologists will identify lesions, measure the CC size of the lesion and rate their confidence in deciding a lesion is present based on radiographic features evaluation, specifically the identification of lesions.

The study scan, s-DCT, will be performed within two weeks of his/her clinical evaluations by chest CT and CR. There cannot be any intervening therapies or procedures (i.e. biopsy or excision of lesions) done in between the SOC imaging and the s-DCT. In the event that CRs are not included in the SOC imaging or are unusable (outside of acceptable window or poor quality), the scout image acquired by CT will be used in place of a CR. Images will be acquired by a trained radiologist technologist. The technologist will comfortably position the patient in the supine position on the imaging table. Then, all patients will have a breath held s-DCT scan in an anterior-posterior direction.

Sensitivity for our study is defined as the ability of s-DCT to detect a lung lesion known to exist based on non-contrast CT (the gold standard) for any sized lung nodule.

Specificity for our study is defined as the ability of s-DCT to correctly identify the absence of a lung nodule as confirmed by CT (gold standard).

Specificity for this objective is defined as the ability of the s-DCT system to identify a lesion as non-malignant over the course of 1 year post s-DCT. The gold standard will vary depending on the clinical situation as follows: In the subset of who undergo a biopsy of a suspicious lesion, the pathology report will be the gold standard. If a biopsy is not performed, but CT scan is performed within the 1 year timeframe, and a diagnosis of malignancy based on this nodule is made, the CT scan will be the gold standard. If a biopsy or CT scan is not performed, but the clinical records indicate the lesion is malignant (within the one year timeframe), this will be the gold standard.

To evaluate radiologists confidence in evaluating lung nodules of all sizes comparing s-DCT to CT. Readers will be scored using a 7 point Likert scale from -3 to +3. Negative scores in favor of CT and positive scores more in favor of s-DCT. Readers rated images based on shape/morphology, calcifications, and architectural distortion

Inclusion Criteria: ≥18 years of age Known lung lesion(s) based on SOC non-contrast CT Negative urine pregnancy test in women of child-bearing potential (WCBP) within 1 week prior to s-DCT Undergone a non-contrast chest CT in a time frame that will accommodate experimental imaging (s-DCT) within 2 weeks Institutional Review Board (IRB) written informed consent obtained and signed Exclusion Criteria: Unable to provide consent Pregnant or lactating BMI > 33 (Patients who may not fit on a 35 x 35 detector) Planned procedures or therapies in between SOC scans and study scan on s-DCT, e.g., biopsy or excision of lung lesion

Deidentified individual data that supports the results will be shared beginning 9 to 36 months following publication provided the investigator who proposes to use the data has approval from an Institutional Review Board (IRB), Independent Ethics Committee (IEC), or Research Ethics Board (REB), as applicable, and executes a data use/sharing agreement with UNC.

The investigator who proposes to use the data has approval from an IRB, IEC, or REB, as applicable, and an executed data use/sharing agreement with UNC.