Main Page

Imaging Clinical Trials:: Board of Contributors
Imaging clinical trialists face complexities not shared by other clinicians conducting clinical experiments.
Though much has been written on the subject of ClinicalTrials, this Wiki serves as a resource for those imaging professionals who need more than mere familiarity with the design and conduct of clinical trials. Its content has been refined from an annual mentoring program conducted by a distinguished Board of Contributors by RSNA and National Cancer Institute at NIH for young faculty who intend a career as a professional imaging clinical trialist.
Consult User's Guide for information on using the wiki software.----

Concepts and FAQ's addressed in this course

How does a screening trial differ from a therapeutic trial?

How does a randomized control trial (RCT) differ from an observational trial?

Must the control arm of a therapeutic RCT always be a placebo?

What distinguishes Phase 1, 2, 3 and 4 trials?

What are ‘Phase 0’ trials and what role would imaging have in them?

What data do you need to determine a statistically valid study size?

When does a trial need a Data Safety Monitoring (DSM) board?

What functions do interim and futility analyses serve?

What are the definitions of type 1 and type 2 errors?

How do you constitute an imaging central reading review panel and how does it function?

What is the regulatory definition (e.g. FDA) of a “surrogate’ biomarker?

How do “prognostic” biomarkers differ from ‘predictive’ markers?

What happens if an ‘adverse event’ occurs to a study subject?

Trials: their types

• Phase 1 - less than 30 subjects - purpose to determine therapy dose and delivery. Commonly pharmacokinetics and/or dose esclation to maximum tolerated dose (MTD)
• Phase 2 - 100 or so subjects - to determine 'efficacy'
• Phase 3 - large numbers of subjects - to determine effect relative to standard therapy. Often designed for regulatory approval of therapy
• Phase 4 - large subject numbers - to determine ' efficaciousness' after therapy is in widespread use
• Phase 0 (a new category often using nuclear imaging or MRI) - less than a dozen subjects - to determine if a 'targeted therapy' reaches the intended disease tissue. Microdose (<100 µg) with no therapeutic intent.

Screening Studies

• Definition: Screening can be defined as the systematic testing of individuals who are asymptomatic with respect to some target disease. The purpose of screening is to prevent, interrupt, or delay the development of advanced disease in those with a pre-clinical form of the target disease through early detection and treatment.
• Screening vs Diagnosis: Screening involves asymptomatic non-patients performed on a population with a low prevalence of disease. Moreover, the test not intended to be fully diagnostic
• Screening risks the biases of early detection These include: Lead time bias, Length bias, and Overdiagnosis bias. Overdiagnosis implies diagnosis of a condition that would not have become clinically significant had it not been detected

Diagnostic Technology Assessment Trials

Diagnostic tests are conducted to determine whether cancer is present, where it is located and its stage. Some diagnostic trials compare two or more techniques to diagnose cancer and determine whether one technique is superior. Genetic tests are being evaluated as diagnostic tools to further classify cancers so as to direct therapies or improve treatments for people with specific genetic changes.

Basic Protocol Considerations

Clinical trials must follow strict scientific guidelines. These guidelines should clearly define the study’s design and who will be able to participate in the study (patient eligibility). Every trial has principal investigator(s). The principal investigator prepares a plan for the study (a protocol) which describes the clinical trial in complete detail, including imaging acquisition and analysis procedures. The protocol explains what the trial will do, how the study will be carried out, and why each part of the study is necessary. It includes information on:

• The importance and rationale that justifies the study
• A statistically justified cohort size (for each arm) or clear explanation for an exploratory trial of the information obtained will allow for the design of statistically robust trials
• Eligibility criteria (requirements may involve type of cancer, general health, age, prior therapy)
• An easily understood 'schema' depicted as a flow-chart
• The intervention or therapy being tested (e.g. drug dosage, etc.)
• What tests participants will have and how often (a ‘patient calendar’)
• What information will be gathered about the participants and how their privacy will be protected
• The participants options if they choose to withdraw
• A readable consent form (at an eighth-grade readability level)
• The endpoints (primary and secondary) and correlative studies
• Methods of analysis
• Regulatory requirements and reporting obligations

Biases and Their Control

What is Bias? Any systematic deviation (i.e. not random deviation) of observations or study results from the true clinical state. Here are some of the common biases in clinical radiology studies and how to control them in your study:

• Referral Bias: A kind of selection bias where the composition of the sample does not represent the full spectrum of patient and disease characteristics because only patients referred to the imaging test by their treating physician are included in the sample. Control: Design the study to recruit patients when they present, not after they have been referred. Alternatively, one can report the accuracy of the imaging test stratified on the patient and disease characteristics in your sample. Reference: Ransohoff and Feinstein (1)
• Verification Bias: Patients with positive (or negative) test results are referred for the gold standard more often than patients with negative (positive) test results; then sensitivity and specificity are estimated only on the subset of patients who underwent the gold standard. Control: Design the study so that the test results aren’t used to decide who undergoes the gold standard. Alternatively, there are statistical methods to correct for this bias. Reference: Ransohoff and Feinstein (1)
• Imperfect Gold Standard Bias: The gold or reference standard used in the study is not 100% accurate. Control: Design the study so that a true gold standard is used. Alternatively, there are statistical methods to correct for this bias. Reference: Valenstein (2)
• Uninterpretable Test Result Bias: Patients with technically unacceptable test results are excluded from the assessment of the test. Control: Report the frequency of uninterpretable test results. Repeat the test if possible. Reference: Begg et al (3)
• Location Bias: For tests designed to locate lesions, sensitivity is estimated without requiring that the radiologist correctly locate the lesion. Control: Define sensitivity such that a true positive requires both correctly locating the lesion and diagnosing the lesion as positive.
• Test-Review Bias: The imaging test is conducted and/or interpreted with knowledge of the results from the gold standard. Control: Mask the radiologist interpreting the test to the results of the gold standard.
• Diagnostic-Review Bias: The gold standard/reference test is conducted and/or interpreted with knowledge of the results from the imaging test under evaluation. Control: Mask the radiologist interpreting the gold standard to the results of the imaging test being evaluated.
• Reading Order Bias: When comparing two tests A and B, the radiologist interpreting test B recalls the results from test A. Control: Randomize the order of the tests such that the radiologist interprets ½ of cases with test A first and ½ of cases with test B first. Then, after a wash-out period, the radiologist interprets the cases with the test not used previously.

• 1. Ransohoff DJ and Feinstein AR. Problems of spectrum and bias in evaluating the efficacy of diagnostic tests. NEJM 1978; 299: 926-930.
• 2. Valenstein PN. Evaluating diagnostic tests with imperfect standards. Am J Clin Pathol. 1990; 93: 252-258.
• 3. Begg CB, Greenes RA, and Iglewicz B. The influence of uninterpretability on the assessment of diagnostic tests. J Chronic Dis. 1986: 39: 575-584.
• A good book reference is: Zhou XH, Obuchowski NA, and McClish DK. Statistical Methods in Diagnostic Medicine. Wiley 2002.

Imaging as a Measure of Therapeutic Response

Suppose you were asked to devise an imaging acquisition protocol and analysis scheme for a new drug therapeutic trial that might be submitted to FDA. Can you come up with a list of the key considerations you might want to go through?

• If you'd like a Socratic list of 20 questions you ought to consider,

try this link -- Instructive use-case problems for clinical trial imagers

Informed Consent and Ethics

Informed consent is an essential part of the ethical conduct of clinical trials.For detailed discussion of how it impacts imaging clinical trials, follow the link below titled 'Informed Consent'

Other Wikipedia threads are available that deal with informed consent which are available as 'out-links' http://en.wikipedia.org/wiki/Informed_consent from which you can get a more extensive background on such important documents as the Tuskegee experiment (http://en.wikipedia.org/wiki/Tuskegee_Syphilis_Study), Belmont report (http://en.wikipedia.org/wiki/Belmont_Report) and Nuremberg Code (http://en.wikipedia.org/wiki/Nuremberg_Code). But with your focus on imaging you might wish to limit yourself to the particulars of informed consent primarily applicable to 'imaging' via the link

Informed Consent

Monitoring of Clinical Trials: IRB, DSM, CTEP

Review Committees

Most clinical trials are subject to different kinds of review designed to protect participants. Large societally significant or complex clinical trials may have their own Data Safety Monitoring board (DSM or DSMB). Clinical trials sponsored by NCI — whether funded by a grant, run by a Cooperative Group, or through a Cancer Center — are reviewed by different types of panels, including experts who review the scientific and technical merit of the proposed research (the NCI uses both external Steering Committees and its own Clinical Tials Evaluation Program - CTEP). Many other clinical trial sponsors, such as pharmaceutical companies, also seek expert advice on the scientific and technical merit of their trial protocols.

In addition, all federally-funded clinical trials and all trials with investigational drugs or devices must be reviewed by local or 'central' groups called Institutional Review Boards (IRBs). All trials with investigational drugs or some investigational devices must be reviewed by the appropriate FDA division, except for highly limited radiopharmaceutical trials that are reviewed by Radioactive Drugs Research Committees (RDRC), which are under FDA oversight. Note that a marketed drug or device can be investigational and must apply to the FDA for Investigational New Drug or Device status (IND or IDE).

For detailed topic information click on: Oversight Boards

Investigational Drugs and Devices

Investigational means any drug or device not currently marketed in the US in the exact form that you are using for the exact indication for which you are using it. For example, updated device software is investigational, as is investigation of an marketed drug for a disease or histology that is not explicitly in the drug labeling. For more detailed discussion follow this link -- Investigational Drugs & Devices

Nuts and Bolts of Creating a Trial Protocol

Trial protocol construction is a lot easier if you take advantage of existing templates and follow a checklist of things to consider. You should also become more familiar with common software programs that you often take for granted but haven't yet mastered and are essential for your efficiency.

Sponsorship and Economics of Imaging Trials

Organizations That Sponsor Clinical Trials

NCI, pharmaceutical companies, medical institutions, and other organizations sponsor clinical trials. NCI often partners with pharmaceutical companies to develop new agents. Regardless of sponsor, clinical trials take place at universities, large medical centers, small hospitals, and doctors’ offices. Individual physicians at cancer centers and other medical institutions can also sponsor clinical trials themselves.

For more details click: Trial Sponsorship

Why and How to Share Data

Because most clinical trials attempting to assess therapeutic efficacy are Phase I or II trials with small numbers of patients, there is a great need to assemble evidence by cross-analyzing the results of many trials thought to be similar ('meta-analysis'). Insufficient attention has been paid to be sure trial results permit that possibility therefore data-sharing begins with clear and fully defined protocols including detailed imaging parameters that can be understood by others. Case report forms (CRFs) should employ common vocabularies and common data elements (CDEs)so that results can be harmonized between trials. Cancer-applicable CRF templates, CDEs with vocabularies and Adverse Event definitions can be accessed from NCI clinical trials websites.

Images obtained during clinical trials have a special value of their own and the field greatly benefits from sharing those images in their original DICOM format with others so as to advance research in quantitative clinical image analysis. To share images one should understand the principles of de-identify and anonymize image sets.

Bibliography

Click to the Bibliography to access selected references, many of which are hot-linked, that offer more depth on topics specifically chosen to be relevant to imaging trialists.

Abbreviations and Definitions

• Aims
• Endpoints

• CR (Complete Response) after a therapy intervention implies that both 'target' and non-target tumor lesions are no longer visible
• Confirmed PR is a measure taken at the next interval (more than 6 weeks) after a declared first instance where tumor measurement meets the PR criteria (which for RECIST it is a reduction of 30%)
• DFS (Disease Free Survival) post-treatment measures the length of time before death during which there are no detectable tumor lesions
• ORR (Objective Response Rate) summarizes the percent of cases on an experimental therapy in which there is a measurable reduction in target tumor mass as depicted by imaging
• OS (Overall Survival) measures the length of time before death of cases on an experimental therapy
• PD (Progressive Disease) is defined by the time at which a case undergoing therapy is measured to have grown in dimension beyond a nadir measurement after the start of therapy. By the RECIST criteria it is defined as having reached a threshold of 20% growth in the summed diameters of measured target tumors that exceeds the nadir value reached after the inception of therapy. PD is also defined to have occured if there is the appearance of a new lesion.
• PR (Partial Response) a net measureable reduction in tumor dimensions amongst the 'target' lesions. For RECIST the threshold for PR is a 30% reduction.
• PFS (Progression Free Survival) the length of the interval for which there has neither been death nor a measureable increase in net tumor dimensions (in RECIST greater than a net 20% dimensional increase).
• RR (Response Rate) represents data that implies an reduction of the amount of tumor burden and can include either reduced symptoms or shrinkage of measured tumor size or both.
• RECIST Response Evaluation Criteria for Imaging in Solid Tumors. A solid tumor measurement method that uses the sum or 'target tumor' linear diameter measures to estimate time-sequence changes in tumor volume. Click here to see a table of the elements that define classification of RECIST Endpoints
• SD (Stable Disease) measurement intervals of 'target' lesions in which net their dimensions have neither increased by 30% nor decreased by 20%.
• TTP (Time to Progression) the time interval that has elapse before the tumor target measurements have not exceeded the threshold defined as 'PD' (in RECIST it is a net increase of 20%)
• Target Lesions tumors visible at baseline that are measured sequentially over time by the RECIST criteria http://www.eortc.org/Recist/Default.htm
• WHO A solid tumor measurement method that uses two orthogonal tumor diameter measures to estimate time-sequence changes in tumor volume

For more comprehensive lists of clinical trial glossaries and abbreviations you can out-link to: http://www.cdisc.org/glossary/CDISCGlossaryV5.pdf and http://www.cdisc.org/glossary/CDISCAcronymsv5.pdf link title

Useful (out-of-Wiki) Websites

Thorough knowledge of what's available on these other Websites is probably worth 20 added points to your 'Clinical Trialist IQ'

Funding Sources and Key Resources