OR WAIT 15 SECS
Improvements in image acquisition and read quality: A look at the value of standardization in oncology trials.
With access to a large and genetically diverse patient population, lower overall costs, and relatively fewer trials than in the United States or Europe, Asia has become an increasingly attractive market for conducting clinical trials. Oncology trials in particular are experiencing major growth in the region, and the trend is likely to continue in parallel with improving infrastructure for hospitals and other clinical facilities. For pharmaceutical companies looking to conduct trials in Asia, this means a number of new opportunities and challenges to consider.
One such area relates to the use of medical imaging, which has become a vital component in oncology trials. Imaging biomarkers are increasingly accepted as surrogate endpoints in oncology trials, making the need for improvements in image acquisition and read quality more apparent. This includes a greater emphasis on reader training and testing. Improving the quality of reads can help better stratify patients for treatment, evaluate safety, improve dose selection, and even provide early prognostic information to assess the effectiveness of a particular treatment.
As growth in clinical trials in Asia explodes in parallel with improving infrastructure for imaging facilities, the need for standardizing image acquisition and image review is increasing. Pharmaceutical companies conducting oncology clinical trials exclusively within the Asia-Pacific region, or as part of a global study, benefit from identifying a partner that can qualify imaging facilities, standardize image acquisition protocols, manage the collection and QC of imaging data sets, and coordinate the central review of images by a panel of trained radiologists.
Standardization of image acquisition requires the collection of information about the equipment (both hardware and software) that a potential site has available to scan participating study subjects. With varieties in vendor and software platforms across a multi-center study, it is critical to confirm acceptability and compliance ahead of first patient first visit. This is where an imaging CRO's expertise can be instrumental. Providing a standard image acquisition protocol to sites in advance and confirming that the sites can acquire scans following these procedures is essential to establishing a harmonized imaging data set across sites and timepoints in a longitudinal study and is where an imaging CRO can play a key role in reducing a study's size and duration.
By centrally collecting all imaging visits, a sponsor can verify that consistent techniques were followed in scan acquisition and establish a library of complete images. It is also important to confirm that imaging visits occur within the protocol defined study window. Proactive notification to sites, reminding coordinators that imaging visits are due for a particular subject is helpful to improve compliance. Any issues identified should be followed to resolution to improve the quality of the imaging visits and accuracy of independent assessments.
In addition, imaging CROs can assist in incorporating certain design features into an oncology trial to minimize bias and increase precision and accuracy of potentially subjective observer-dependent assessments. By paying special attention to blinding of patient data and the use of central (external) review committees, sponsors can increase the uniformity of efficacy endpoints based on tumor assessments. Without independent central review, bias stemming from the sponsor or study monitor's familiarity with particular patients may influence endpoint assessments. If stringent processes are put in place to reduce the potential for bias at the onset, the validity of the conclusions drawn are less likely to be questioned.
To minimize bias in review and ultimately in the analysis of study outcome, identifying an independent review committee composed of radiologists with clinical expertise in oncology is paramount to study success. Thorough training and testing procedures should be followed to qualify the reading team prior to initiating on study reads. In many studies, it is also beneficial to have subject eligibility confirmed by the independent reading panel to ensure the right patients are included in the study, avoiding costly delays in recruitment and time wasted following patients that are inappropriately included in the intent to treat population.
When defining the image review process during imaging charter development, sponsors should determine the most appropriate reading paradigm relevant to the disease process and phase of research. For example, most Phase III oncology studies require a double-read with adjudication model if using progression-free survival as the primary endpoint. By selecting a read process, and by having an understanding of the inherent variability in reads, a sponsor can identify the best ways to monitor variability over time.
Reader testing methodology should also be defined during imaging charter development. Sponsors, together with imaging CROs, should establish proper procedures to adequately monitor reader performance during the course of the trial and assess overall reader variability at the end of the trial. The testing methodology should include guidelines for the identification of test cases, assessment by blinded readers, and regular analysis and review of reader performance results.
Many different methods can be applied to assess reader performance and improve the precision and accuracy of the data set. Working with statisticians that understand imaging data is critical to employing the right test and to interpret results appropriately. If the sponsor and CRO are equipped with a valid measure of reader variability, they can investigate the causes of low-reader reproducibility and proactively identify potential problem areas. Factors such as poor image quality, insufficient reader training, unclear read criteria, and low individual reader performance are common causes, but the efficient use of statistics allows both parties to take the appropriate remediation steps to increase the trial's chances of success.
With oncology trial growth in Asia expected to continue, standardization of image acquisition and reader variability training and testing will become increasingly important. To fully maximize the value of the imaging used in these trials, increased precision and accuracy of readers is necessary for achieving reproducibility in performance and reducing variability that adversely affects sample size of trials. By employing an imaging CRO with expertise in reader variability testing and training to carefully manage read quality, sponsors can reduce variability using imaging assessments, increase the ability to discern differences in treatment groups, and thus provide greater potential for demonstrating a drug's efficacy.
Richard Taranto, Executive Vice President, and Richard Walovitch, PhD, President, WorldCare Clinical, LLC, 7 Bulfinch Place, PO Box 8908, Boston, MA, http://www.wcclinical.com.