Recently, the Clinical Oncologists for Individualized Therapy (COFIT) drafted comments refuting ASCO recommendations on the use of Chemotherapy Sensitivity and Resistance Assays (CSRAs). The following points address the assertions raised by COFIT and explain how ASCO came to its final recommendation on the use of CSRAs.

For a copy of ASCO's recommendations, published in the Sept. 1, 2004 issue of the Journal of Clinical Oncology, link here: http://www.jco.org/cgi/reprint/22/17/3631

Study Selection Comments

-- The criteria for study selection were laid out a priori by the ASCO working group. Criteria were pre-specified, explicit, and transparent to all concerned.
The criteria were reviewed, vetted, and approved by the ASCO working group. Both the Blue Cross and Blue Shield Association (BCBSA) group and the ASCO working group were open to including any assay, apotosis or otherwise, that had undergone rigorous enough evaluation to have informative data.

-- The BCBSA reviewers explicitly sought, but did not find, evidence comparing outcomes of assay-guided therapy that used cell death assays versus empiric therapy.
Since evidence was lacking, conclusions were not possible.

-- Point of clarification: The COFIT group described that the ASCO group reviewed 1,139 published ?clinical trials.? There were nowhere near 1,139 published clinical trials of assay-guided therapy.
1,139 was the number of "hits" using very wide literature search criteria, selected to make sure the ASCO and BCBSA groups didn't miss anything. The overwhelming majority of these articles dealt with technical aspects of assay development, not clinical outcomes of using assays to manage patients.

Panel Composition Comments

-- It would be impossible to put together an ASCO working group that included members with expertise in each and every assay.
The working group included individuals with broad expertise in CSRAs as a field. The working group participants are leaders in the cancer biology area; have knowledge of all types of assays; and understand the different mechanisms that exist for cell death, such as necrosis or apoptosis.

-- Members of the ASCO working group would have liked more than anyone for these assays to be shown to be ready for ?prime time.?
Several of the investigators participating in the ASCO review spent years of their careers trying to develop truly clinically useful assays, and thus did not reach these conclusions lightly.

Conflict of Interest Comments

-- The BCBSA is not ?on record for its opposition to the use of CSRAs.?
The BCBSA does not make coverage decisions; the Plans identified by COFIT make those decisions independently of each other of the BCBSA.

-- The COFIT group members fail to acknowledge explicitly their own overt conflicts of interest as providers and marketers of assay services.

-- Regarding the alleged conflict of interest of ASCO working group members who have ?built their careers on conducting trials of empiric therapy:? Since the earliest days of chemotherapy research, oncologists and patients have recommended prospective, randomized clinical trials to define standards of care around the world.
The National Cancer Institute, patient advocacy groups, and cancer specialists from all disciplines have endorsed this process. Such trials are extraordinarily powerful and reliable for determining how best to treat cancer. The oncology community has readily welcomed the use of predictive markers (such as estrogen receptor and HER2 for breast cancer, PSA for prostate cancer, and bcr/abl for leukemia, among others) into trials for cancer therapies. The use of such prospective treatment trials is an option available to any investigators who wish to demonstrate unequivocally the value of any cancer screening or treatment method.

Analytic Method Comments

-- It is misleading to assert a measure such as predictive accuracy is sufficient to demonstrate test utility.
Tests are useful to clinicians only if the information they provide guides or changes patient management and thus improves patient outcomes. Improved outcomes may include: longer duration or greater quality of life, or decreased adverse effects of treatment. Many experts have published on these established principles for assessing diagnostic and other clinical tests. When test developers (and marketers) or other advocates request a National Coverage Decision from CMS' Medicare Coverage Advisory Committee, they are expected to provide evidence that clinical decisions based on results of their test improve patient outcomes (e.g., PET for various indications).

-- The goal of these assays should not be to determine sensitivity/specificity, but to demonstrate through prospective trials that such assays can be employed to change treatment for the better.
To date, there are insufficient data for such a demonstration.

-- The COFIT group fails to acknowledge the possibility of real harm to some patients from misguidedly directing them away from a potentially effective chemotherapy regimen based on an inadequately evaluated clinical test.

-- If the COFIT group believes that its data are compelling, then it should be easy to develop and execute a prospective trial that demonstrates the superiority of their assay over empiric therapy.
When such data exist, the ASCO working group will be in a position to revisit the question.

Jenny Heumann Senior Public Affairs Manager American Society of Clinical Oncology 1900 Duke Street, Suite 200 Alexandria, VA 22314 Phone: 703/519-1427 Fax: 703/684-8618 http://www.asco.org