Test Page for Joe

From QIBA Wiki
Revision as of 21:57, 24 June 2013 by Jkoudelik (Talk | contribs) (1C Performance Protocol)

Jump to: navigation, search

Scratch Page for CT Volumetry

Trial Implementation Process-DRAFT PAGE

Julie's Test Page

  • This site will be used to test wiki features and posting options only.
*FDA 1C zip files

QIBA vCT Profile v1.1 edit site

  • Quantitative-MRI"

Subcommittee Members

  • Timothy Turkington, PhD, Subcommittee Chair <timothy.turkington@duke.edu>
  • Ronald Boellaard, PhD <r.boellaard@vumc.nl>
  • Patricia E. Cole, PhD, MD <p.cole@imagepace.com>
  • Eric Perlman, MD <perlman@radpharm.com>

Patricia C. Cole, PhD, MD

Cancer Imaging Program (NCI)

NBIA at CBIIT Image Collection CIP Wiki with background information and metadata associated with collections hosted by NBIA

1C Performance Protocol

UCLA Pilot Data

  • Test report documents
  • Images (noise and resolution slices)

Duke Pilot Data

  • Test report documents
  • Images (noise and resolution slices)

FDA Pilot Data

  • Test report documents
  • Images (noise and resolution slices)

Johns Hopkins Pilot Data

  • Test report documents
  • Images (noise and resolution slices)

Univ Maryland Pilot Data

  • Test report documents
  • Images (noise and resolution slices)


Dr Tucker's New Test Page


  • Are we ready? A time for measurement of physiological parameters of the lung using multidetector row CT scans. Hatabu H. Acad Radiol. 2009 Mar; 16(3):249. Academic Radiology
  • Helical multidetector row quantitative computed tomography (QCT) precision. Bligh M, Bidaut L, White RA, Murphy WA Jr, Stevens DM, Cody DD. Acad Radiol. 2009 Feb; 16(2):150-9. Academic Radiology
  • Comparison of the accuracy of CT volume calculated by circumscription to prolate ellipsoid volume (bidimensional measurement multiplied by coronal long axis). Rkein AM, Harrigal C, Friedman AC, Persky D, Krupinski E. Acad Radiol. 2009 Feb; 16(2):181-6. Academic Radiology

QIBA COPD/Asthma Committee

QIBA FDG-PET/CT SUV Subcommittee Site

National Biomedical Imaging Archive Databases

NBIA Log-in Page

  • From the above NBIA link, it is possible to log-in or register as a NBIA user
  • Once logged in, click on Simple Search
  • Filter as follows:
    • Collection(s):RIDER
    • Available on NBIA
    • From (mm/dd/yyyy): 07/17/2009 to 07/17/2009
    • Submit
    • Follow prompts to 32 MSK Coffee Break CT Image Sets

moc-up page

moc Technical Committee Page

Quantitative Imaging Biomarker Alliance (QIBA) for Volumetric CT Image Analysis:

Subgroup 1C – Assessing Impact of Instrumental Variability on Volumetrics November 10, 2008


1. Charge and scope of QIBA CT Volumetrics, Subgroup 1C, characterizing instrumental variability.

Conduct a designed experiment to analyze CT scans of a phantom imaged under a range of conditions to assess the impact of scanners on the performance relative to the requirements posed by the use of the measurements in clinically relevant scenarios. Use radiologist markup as another volume measure to be assessed along side of automated measurements. Use “physical” volume measures as reference “truth”. 2. Questions to be answered Measure the precision and bias of CT volume measurements on phantom data, as applied to two profiles: (1) the smaller nodules as typical in Stage I neoadjuvant or early diagnostic settings, and (2) larger tumors as typical in Stage IV that form the mainstay of most drug development trials. [Note that stages II and III would draw from this work.] 3. Instrument recruitment/selection A key contribution of this work is to address multi-center and multi-vendor variability, so it would be desireable to have broad manufacturer participation as well as a range of centers FDA has scanned the Shiba Soku phantom with a Siemens 64 and a Philips 16. 4. Experimental design The primary basis for the activity is drawn from a systems analysis of the sources of variability in volumetric CT represented in a matrix . Based on this analysis, sources of variability that are relevant to each considered profile would constitute “factors”, and levels for these factors would be derived to sample the appropriate range as might exist in practical real-world settings. The experimental design is organized so as to optimize the use of resources (time, equipment, organizational attention etc.) of stakeholders to satisfy necessary precursor questions needed to set the profile details. This essentially entails the use of fractional factorial designs that restrict the number of treatment combinations for factor subsets needed to characterize the variability as would be seen in applying the profile. Designs are constructed that limit the number of treatment combinations within subsets of factors. Such designs have been applied in a number of settings, and rigorous statistical work-up is pursued. For example, to provide a comprehensive analysis evaluating all of the sources of variability using a full factorial would exhaust the stakeholders, but it is possible to analytically determine what treatment combinations are actually necessary with respect to the specific clinical conditions as well as to understand in what ranges experiments need to be performed. Others have cited the benefits of such design efficiencies while retaining a comprehensive nature to the results. This subgroup effort documents and pursues the design for the stated profiles as well as establishing methodology that could be applied more broadly for other profiles in CT and/or other quantitative imaging settings. Examples of scanner Settings – which parameters are critical to control for constancy between scanners include slice thicknesses, non-ovelapping slices, mAs, kVp, FoV/voxel spacing. Fuller listing of scanner dependent parameters that may be studied. a. Potential differences in imaging protocol 1. reconstructed slice thickness and spacing/overlap 2. collimation 3. mAs 4. pitch 5. reconstruction kernel 6. kVp 7. Reconstructed Field of View 8. With larger detector row arrays providing extended z-axis coverage, sequential scans may be performed, and lesions may would span - or be at the interface of - several acquisitions FOVs

b. Potential differences between imaging hardware MDCT 1. number of detector rows 2. number and size of detector elements in each row 3. number of sources (e.g. dual source CT) 4. x-ray spectrum/beam filtration (e.g softer vs. harder x-ray) 5. software version: acquisition, correction or reconstruction

5. Phantoms for DICOM imaging - propose Image Set 1: FDA-type phantom with 3 to 20 mm phantom nodules. Phantom QA – NIST Pocket Phantom? Should QIBA acquire a phantom? 6. Algorithms: - At least one algorithm - Measured precision and bias - Intra and inter-reader variability - Precision across repeat scans.

7. Identification of required resources: Existing FDA image data sets acquired on multiple CT scanners are . Readers Other resources needed: algorithm, automatic format for study parameters.

8. Next steps Recruit/identify readers and reading tasks. Recruit participation from multiple scanners. Specify experimental conditions – source of phantom, scan parameters, Specify data interface for automated algorithm/CAD tool iinput. Proposal: adopt specification from CT Volumentrics 1B. 9. Notional complexity estimate: 5 scanners, 2 slice thicknesses, 4 nodule sizes/shapes, 3 repeat scans = 15 scans. x 2 reconstructions x 4 nodules = 120 nodule volumes to analyze.

ACRIN 6678 CT Acquisition Parameters

DICOM Tag# Parameter GE Ultra GE LS 16 GE VCT(64) Philips Brilliance 16 Philips Brilliance 64 Siemens Sensation 16 Siemens Sensation 40 Siemens Sensation 64 Toshiba Aquillion

8-slice/0.5 sec.

16-slice/0.5 sec. 64-slice/0.5 sec. 16-slice/0.5 sec. 64-slice/0.5 sec. 16 X 0.75 40 X 0.6 (beam collimation 20 X 0.6) 64 X 0.6 (beam collimation 32 X 0.6) 16-slice/0.5 sec.
0018,0050 Nominal Reconstructed Slice Width 1-1.5mm 1-1.5mm 1-1.5mm 1-1.5mm
0020,1041 Reconstruction Interval 0-20% overlap 0-20% overlap 0-20% overlap 0-20% overlap
0028,0030 Voxel Size 0.55-0.75mm 0.55-0.75mm 0.55-0.75mm 0.55-0.75mm
Motion/Breathing Artifact None None None None
Intravenous Contrast Media None None None None

X-ray Tube Current x Exposure Time

Exposure Exposure X-ray Tube Current x Exposure Time
Scanner dependant mAs (Regular-Large) 135-220 95-245 95-245 120-310 100-260 120-310 100-260 100-260 120-310
0018,0060 KVP 120 120 120 120
0018,1210 Reconstruction Algorithm STD B B30 FC10

Wiki Markup Test

Column 1 Column 2 Column 3

Clinician’s Perspective: Developing Quantitative Response Metrics for Lung Cancer

James L. Mulshine, MD

Lung cancer is the leading cause of cancer death in the United States and the second leading cause of death overall in our society {Jemal, 2008 #2}. Mortality outcomes have improved only modestly over the last thirty years {Wakelee, 2006 #1}. For these reasons, there is an intense focus by pharmaceutical companies to develop better treatments for lung cancer. A number of challenges exist for the pharmaceutical industry for improving lung cancer drugs. Major issues include the cost and time duration of the clinical trials required to establich the utility of a drug so that it can be formally approved by federal regulatory agencies {Mayburd, 2008 #3}.

The Volumetric CT of the RSNA Quantitative Imaging Biomarkers Alliance (QIBA) has proposed that the use of quantitative spiral CT may increase the analytical power per subject enrolled in clinical trials in such a way that (1) the number of total subjects enrolled in an arm of a clinical trial can be reduced or (2) that the length of time that an individual needs to be followed to reliably establish drug response can be shortened.

Clinical Trial Endpoints

With the transition from conventional cytotoxic chemotherapy to new molecularly targeted therapeutics, a new trend has emerged for drug response endpoints. With certain molecularly targeted drugs, there may not be any appreciable shrinkage of tumor size after drug exposure. Rather, the cancer stops growing resulting in what is called stable disease (SD). In some instances, molecularly targeted drug therapy can result in disease stabilization lasting for years. In these cases, the success of the drug may be best measured with the clinical trial endpoint such as freedom from progression-free survival (PFS). With PFS, the interval is measured from the date of study entry until the date of disease progression. There are many nuances in interpreting the significance of this endpoint compared to other trial endpoints such as overall survival (OS). Overall survival is measured from the date of study entry to the time of patient death.

Measurement Accuracy

The focus of this discussion is to determine the accuracy of the measurement of these various parameters (more precise word here than parameters?) with imaging tools in clinical trials. Measurement can be quite variable related to how disease progression occurs as well as how the imaging studies are performed. Depending on the study setting, volumetric imaging may or may not be informative. Lung cancer begins with a cancer arising within the cells of the airway of the lung. Localized cancer is called a primary lung cancer. Cancers are generally lethal to their hosts due to a predilection of spreading to other parts of the body. The first place a lung cancer typically spreads as it advances from localized disease is to the neighboring lymph nodal structures of the lung. This is called regional metastatic spread. In contrast, in the most advanced stage of cancer, the cancer metastasizes to a distant site such as the brain or liver. This is called a distant site of metastatic spread. In clinical trials, the discovery of a new site of metastatic dissemination is the basis for declaring failure of the efficacy of a new drug. This is a qualitative not quantitative determination. In virtually all lung cancer clinical trials, there are situations when either a quantitative or a qualitative endpoint may be relevant, but it is likely that quantitative endpoints will be most frequently informative in trials involving early stage lung cancer.

With advanced disease, there is a tendency to have disease progression more frequently with a distant metastatic site rather than spread due to extension from the primary tumor {Wakelee, 2006 #1}. These patterns of disease progression impact how a clinical trial is designed to measure drug response. There is also a specific example of brochioalveolar carcinoma, which tends to spread extensively within the lung but seldom to distant sites {Gandara, 2006 #31}.

Cancer Staging

The extent of lung cancer dissemination is defined at the time of initial diagnosis of a patient in a process called staging. The schema (TNM Classification of Malignant Tumors?) for staging lung cancer has been recently updated so that it more accurately clusters patients who benefit from particular therapeutic interventions with predictable outcomes {Goldstraw, 2007 #7}.

A table of how staging relates to lung cancer drug therepy approaches, the imaging approaches used in those stages and issues relative to the image requirements is summarized in Table 1.

Table 1: Summary of IMage Processing Issues Relative to Stage of Lung Cancer

Stage  % of Cases 5-year Survival % Imaging Focus / Therapy Focus Imaging Tool Issues Thoracic Segmentation Hi-Res
I 16 49 Primary tumor / Neo and adjuvant RX sCT Small cancers surrounded by air Can be straightforward Needed
II/III 35 15.2 Primary, hilar and mediastinal lymph nodes / Combined modality sCT, PET Larger tumors and nodes abut other structures Often challenging Optional
IV 41 3 Primary/regional nodes and metastatic sites / Chemotherapy sCT, PET, Bone, Brain scans Tumor response often determined outside of the chest Often challenging Optional

For this discussion, Stage I is consider separately as it is typically treated with surgery and has the highest potential for curability. Because Stage II is relatively uncommon, Stage II and III are clustered together as their clinical management can be similar involving combinations of radiation therapy and chemotherapy with or without surgery. Stage IV is the most common form of lung cancer; its treatment typically involves only the use of drug therapy approaches. There are a number of trial types listed in the fourth column.

For example, drug treatment of Stage IV lung cancer could be in a Phase I trial, where the study endpoint is how well the drug is tolerated. Imaging is done in these trials as a secondary endpoint to determine if a drug is having a measurable impact on tumor volume reduction. Stage IV lung cancer could also involve a Phase II or Phase III study. In a Phase II trial, the goal is to determine how frequently a patient has a favorable tumor response to a particular candidate drug therapy. Since drug response is the main goal of the trial, imaging is a critical aspect of the trial design. In a Stage IV lung cancer, the disease progression could be due to new growth in the primary tumor in the lung, but this is not as common. Progressive tumor growth could also occur in the regional lymph nodes. Spiral CT imaging would typically be the imaging tool to judge for potential disease progression in either the primary tumor or in the lymphatic tissue. The development of new sites of metastatic disease is often the most common type of lung cancer progression in a Stage IV clinical trial. To assess for new sites of metastatic disease, spiral CT may be used to look for thoracic, hepatic or retroperitoneal sites of metastasis. However, PET scans are frequently used to assess for progression of metastatic disease across the entire body.

Surgery is the primary therapeutic approach to manage a localized cancer. However, in experimental settings, an additional drug treatment might be given after surgical management such as in the setting of Stage I small cell lung cancer due to the frequent occurrence of metastasis with this localized lung cancer. This additional drug therapy is called adjuvant therapy{Wakelee, 2008 #10}. There is a important new type of adjuvant trial design for Stage I lung cancer in which the additional drug is given before the surgery to remove a newly diagnosed lung cancer. This is called a neoadjuvant (window of opportunity) trial {Mulshine, 2006 #9}. This trial is interesting because it allows for the exposure of new targeted therapies to untreated cancer, so that a pharmaceutical sponsor can get feedback on the efficacy of their new compound to the unperturbed native condition of the cancer. By contrast, in conventional drug development, new drugs are typically first administered to patients with late stage lung cancer, who have already received one or more course of chemotherapy. This prior drug exposure makes it very hard to determine if a particular type of cancer is efficacious in early lung cancer.

In a neoadjuvant type of trial, the drug administration period is typically a brief typically 2-3 weeks, so that the changes in tumor volume over this short time interval may be small. For this reason, high resolution spiral CT is required to reliably determine volume change across time. A recently published manuscript from the group conducting the NELSON trial evaluated the sources of variance with their imaging and demonstrated how the strength of the image measured in Hounsefield Units unit could impact the success of an imaging metric {Stoel, 2008 #12}.

In developing quantitative imaging to evaluate therapeutic response for lung cancer drug trials, defining the context of the trial is critical to permit the relevant application of imaging. From the discussion above, it becomes more understandable why there are trade-offs in defining the most direct approach to validation of quantitative imaging. For example, there are many patients on Phase III clinical trials for Stage IV lung cancer, where typically a new treatment is compared to an existing standard treatment. However, many of these clinical trials participants will manifest their disease progression not in the lung, but rather by developing new long bone or brain metastasis. These sites are scored as qualitative endpoints related to the new occurrence of disease progression. The neoadjuvant study may typically involve trials involving quantitative intrathoracic endpoints, but these are relatively uncommon types of clinical trials.

There are other yet to be defined clinical issues that may contribute to variability including biological-based sources of variance. Imaging trials may involve drug exposures that may last for several months or more. In a trial in which chemotherapy is given with radiation therapy, the participants in these trials may lose 10% or more of their baseline body weight. Change in body composition due to weight loss represents a source of biological variation that could change aspects of the tumor quantitation across the interval of drug administration.

From consideration of these clinical realities emerges the basis for defining distinct profiles to address the particular needs of imaging that will be performed in lung cancer clinical trials. This segregation will assist device or software development vendors to define the relevant needs for their product to work in supporting the conduct of lung cancer drug development. The careful work of the American College of Radiology in developing a quality control program for its lung cancer screening trial, provides a foundation for an approach to ensuring requisite quality for drug imaging as well {Cagnon, 2006 #8}.

Further work from the QIBA process will define other source of variance in the process of clinical imaging, define the magnitude of the variance and then propose measures to standardize these processes so that robust imaging is performed in clinical trials. Ongoing research will continue to define major sources of imaging variability. Other sources of variance with imaging may be related to vendor issues such as how radiation dose is handled or how image processing functions such as with the various available kernels are managed.

Lung Cancer Profiles

To move the field forward, preliminary profiles must be proposed. However, with all of proposed profiles, there is still shared need for imaging standardization and optimization. Refinements in the quality control and related processes will dynamically impact parameters related to the proposed profile.

As discussed by Dr. Michael McNitt-Gray, ACRIN has published how they worked out acquisition parameters for lung cancer detection to support the conduct of the National Lung Screening Trial {Cagnon, 2006 #8}.

<link to ACRIN/NLST CT Technique Comparison Chart: Scanner Specific Techniques Mandated by NLST Protocol>

Our group is working through a comparable process to define and review the relevance of each of these types of parameters for response assessment. Based on the analysis of other QIBA work groups, the provisions for quality control need to be defined. In the first part of this document, the contexts associated with the different profiles were discussed.

Early Lung Cancer

For the early lung cancer profile, when assessment of the response of the primary cancer will be pivotal, images should be acquired at the greatest possible resolution. From existing neoadjuvant trials, evaluation intervals may be from 2-3 weeks.

  • In a trial from Cornell evaluating Pazopanib in a neoadjuvant, window of opportunity trial, the data from all trial participants was displayed as a Waterfall plot (Alorki et al Proc ASCO, 2008). There was not comparative group for this trial, but the response of all patients was the trial result.

Action: Validation of the magnitude of volume change under defined conditions that is considered to be a robust outcome needs to be established by the QIBA process. To support this, best practice in terms of how the measurement is obtained needs to be defined.

Regionally Advanced Lung Cancer

The discussion of this profile will be similar to the discussion of imaging issues with distant metastatic disease. One important difference is that radiation therapy to the thorax is frequently used in this setting. Radiation therapy often induces an intense inflammatory response with long term fibrosis.

Action: The QIBA group has to determine how it will manage this issue with its imaging approaches.

Distant Metastatic Lung Cancer

The ACRIN parameters have to be reviewed for their applicability to response assessment for Stage IV lung cancer. Clearly, provisions will have to be made for assessing intrathoracic sites. Additional imaging modalities beyond spiral CT will be used in this setting but the integration of these issues is beyond the scope of this work group. The critical endpoints for trials in advanced lung cancer typically are scored using RECIST criteria. Since RECIST involves semi quantitative imaging endpoints, image acquisition may need to be done with the precision comparable to the ACRIN criteria, but the images may not need to be acquired at the highest resolution. Other quality control provisions also need to be defined through the discussions of the members of the QIBA process.

Action: Define how spiral CT will be used to evaluate other thoracic, abdominal and retroperitoneal sites.