MAJOR ONGOING INITIATIVES

Program for the Assessment of Clinical Cancer Tests

http://cdp.cancer.gov/scientificPrograms/pacct.htm

Contact:

Barbara A. Conley, M.D.
301-496-8639, conleyba@mail.nih.gov

Many decisions relating to cancer patient management depend on information derived from clinical laboratory tests. Significant research and development are involved in producing a test that is reli­able enough for routine clinical use. CDP launched the Program for the Assessment of Clinical Cancer Tests (PACCT) in 2000 to develop a process for moving the advances in new technologies and new understand­ing of cancer biology more efficiently and effectively into clinical practice.

Biomarker Development Process

A primary goal of PACCT is to develop more informative laboratory tools to help maximize the impact of cancer treatments. The program focuses on issues and barriers affecting the development of improved tests for cancer diagnosis, prognosis, and prediction of response to therapy. PACCT is not a grants program and has depended primarily on leveraging other NCI-supported activities to accomplish goals identified by the PACCT Strategy Group. The strategy group is comprised of scientists, drawn from academia, industry, the Food and Drug Administration (FDA) and NCI, with expertise in clinical oncology, pathology, basic cancer biology, diagnostics technology and assay development, clinical trials methodology, and statistics.

The strategy group establishes working groups to address critical diagnostic issues in specific tumors as needed. The Breast Cancer Working Group’s efforts led to the Trial Assigning IndividuaLized Options for Treatment (TAILORx), which is assessing the utility of a molecular signature to identify women with early stage breast cancer who can be treated with only hormonal therapy after primary surgery because their risk of recurrence is low. The trial will address whether the signature also identifies those patients who will benefit from the addition of chemotherapy.

The Colon Cancer Working Group focused on assay standardization and validation issues with a goal to validate tests to determine whether it is possible to identify a subgroup of patients with stage II colon cancer at sufficiently high risk to benefit from adjuvant chemotherapy. As part of these activities, PACCT initiated a study to determine the intra- and inter-laboratory reproducibility of assays to determine loss of heterozygosity at chromosome 18, since this assay is being used to stratify patients in a colon cancer clinical trial being led by the Eastern Cooperative Oncology Group (ECOG). The performance of the assay was being evaluated in collaboration with the investigators running the assay for the trial and pathologists from two other cooperative groups. The results validated the analytical performance of the assay as it is performed in the ECOG trial and the results were presented at an American Society of Clinical Oncology Annual Meeting.

Another subcommittee of the PACCT Strategy Group developed criteria for prioritizing studies of essential clinical assays that will be funded through the Coordinating Center for Clinical Trials (CCCT). These criteria were reviewed and approved by NCI’s Clinical Trials Advisory Committee (CTAC).

A subcommittee was also formed to develop a document specifying standards of evidence required to support the inclusion of an assay in a phase 3 clinical trial. The data justifying the use of the clinical assay must be included in the clinical trial protocol. The document outlining the categories of information to be required was approved by CTAC.

A working group of PACCT organized an NCI/FDA/Industry workshop to consider strategies, challenges, and barriers to the co-development of targeted therapies and predictive assays necessary to optimize the use of the therapies. Three case studies were used to illustrate the challenges and barriers: HER2/trastuzumab, epidermal growth factor receptor (EGFR)/EGFR inhibitors, and a current co-development project presented by industry. A commentary based on discussions at the workshop and including a set of recommendations was published in the Journal of the National Cancer Institute. The recommendations addressed the information needed to help make decisions about whether and how to develop the drugs and predictive assays and how best to generate that information.

Taube SE, Clark GM, Dancey JE, McShane LM, Sigman CC, Gutman SI. A perspective on challenges and issues in biomarker development and drug and biomarker codevelopment. J Natl Cancer Inst 2009:101;1453-63. http://www.ncbi.nlm.nih.gov/pubmed/19855077

The PACCT Strategy Group addressed the problems inherent in the qRT-PCR assay that is used to monitor molecular recurrence in Chronic Myelogenous Leukemia (CML). CML is the paradigm for molecular oncology not only because it is the first malignancy to be treated successfully by a rationally designed tyrosine kinase inhibitor (imatinib) but also because molecular recurrence may be assayed by a peripheral blood test. Over 150 laboratories were performing this test in the United States but were using laboratory-developed assays. Proficiency testing by the College of American Pathologists (CAP) indicated that there was a much as a 3-log variation in results among laboratories participating in CAP surveys.

The strategy group suggested further study to gain more data and to see if harmonization of the assay could be achieved and provide more consistent results. The BCR-ABL Working Group was formed and met in 2008 in Bethesda, MD, with attendees representing commercial reference laboratories, two Cooperative Oncology Group Reference labs, Comprehensive Cancer Centers, the FDA, the NY Public Health Labs that oversee CLIA certification, CDP, and CTEP. The group recommended that variation in baseline and major molecular response (MMR) values among labs be assessed, log reduction and traceability be retained, and a non-plasmid RNA calibrator be introduced for daily use. A study with two phases was designed and has been completed. In phase 1, clinical samples from both newly diagnosed and treated CML patients were analyzed along with RNA from a CML cell line diluted into a non-CML line. The data indicate that there was at least a 2- log difference in the values of BCR-ABL gene transcripts measured either at the time of diagnosis or after a major molecular response had been obtained. Phase 2 tested the postulate that use of a RNA calibrator combined with the use of a common assay would reduce the variability in estimated transcript number among the laboratories to less than a logarithm. The result was that variability in the estimation of BCR-ABL transcript number among the participating laboratories was within a logarithm when both the calibrator and the common assay were used. Results have been presented at a national meeting, and next steps will be considered with the strategy group.

Over the past several years, the PACCT Strategy Group has identified a series of barriers to the efficient development of clinically useful assays. Earlier, CDP had advertised a Request for Information (RFI) to identify specific needs. Results from the RFI indicated the need for specific resources, including specimens, technical assistance, support for statistical expertise, and assay standards as well as educational materials available to help inform the research community about issues related to clinical assay development. To respond to these identified needs and provide the necessary resources to the research community, CDP developed a plan for a program envisaged initially as a program that would assess and optimize assays for integral markers in late phase trials. The American Reinvestment and Recovery Act (ARRA) enabled funding the initiation of this program as well as creation of a laboratory to characterize tumors at the genetic level. These two efforts are now termed the Clinical Assay Development Program (CADP) and the Molecular Characterization (MoCha)-Clinical Assay Development Laboratory. The first pilot project is an extension of ongoing work in the Center for Cancer Research and the Strategic Partnering to Evaluate Cancer Signatures (SPECS) (http://cancerdiagnosis.nci.nih.gov/scientificPrograms/ specs.htm) program and will focus on the conversion of discovery-based molecular profiling of Diffuse Large B Cell Lymphomas into useful clinical assays.

CADP now accepts applications from assay developers for further validation work (see http://cadp.cancer.gov). It will provide resources (not a grant) for this purpose and includes a network of CLIA-certified laboratories (Clinical Assay Development Network–CADN) with the capabilities to optimize and validate clinical assay performance. Once validated, the assay is returned to the submitter. This will require access to tumor specimens since the performance characteristics of an assay must be assessed in the same types of specimens as the ones on which the assays will be performed in clinical practice. CDP is establishing a system that will be able to provide sets of appropriate specimens as needed to facilitate evaluation of an assay’s analytical performance and initial assessment of clinical utility. These sets of specimens will come from community settings (formalin-fixed, paraffin-embedded) and will be associated with clinical and outcome data.

To assist in this process a pilot study was undertaken to determine whether software developed by a group at Harvard as part of an earlier NCI initiative can identify cases that meet defined assay development needs. The software was designed to interact with existing medical records at the participating institutions, to strip the records of protected health information (PHI), and to use natural language processing to collect specified data related to treatment and outcomes. Sufficient progress has been made during the pilot study to allow preliminary implementation of the system.

As part of the CADP, an Assay Development Workshop was held in Baltimore, MD, in 2009 in concert with an American Association of Pharmaceutical Scientists-FDA meeting on biomarkers to discuss needs for the CADP with stakeholders in academia, commercial reference laboratories, FDA, and representatives several pharmaceutical companies.  The discussion quickly focused on Intellectual Property (IP) issues as a major problem for assay development. This prompted the FDA to collaborate with CDP to explore management of IP issues. A second workshop held at NCI in late 2009 focused on different models of management of IP. The discussions resulted in proposed changes to the licensing agreements that CTEP negotiates with its sponsors and is incorporated into intellectual property negotiations by the CADP.

Taube SE, Clark GM, Dancey JE, McShane LM, Sigman CC, Gutman SI. A perspective on challenges and issues in biomarker development and drug and biomarker codevelopment. J Natl Cancer Inst 2009:101;1453-63. http://www.ncbi.nlm.nih.gov/pubmed/19855077

Jessup JM, Dobbin K, Hamilton S, Thibodeau S, Redston M, Taube S, Wang Z, Benedetti J. Program for the Assessment of Clinical Cancer Tests (PACCT) 18q LOH Team. Interlaboratory assay reproducibility study for loss of heterozygosity on chromosome 18 (18q LOH) in colon cancer. J Clin Oncol 2009: 27(15s) (suppl; abstr 4052).