Should Oncologists Test for CYP2D6 Before Prescribing Tamoxifen?

Zosia Chustecka

February 09, 2010

February 9, 2009 — In clinical practice, there is a "hesitation" about testing for CYP2D6 before prescribing tamoxifen to reduce the risk for breast cancer recurrence, observes an editorial published online February 1 in the Journal of Clinical Oncology.

Tamoxifen is widely used to prevent breast cancer recurrence in women with hormone-positive disease. However, it is not the drug itself that exerts this beneficial effect, but an active metabolite, and there is a genetically determined variation in the ability to metabolize tamoxifen. This variation occurs on the CYP2D6 gene, and testing for an inherited deficiency in this gene identifies women who are poor metabolizers of tamoxifen and who appear to derive less benefit from the protection that the drug offers against breast cancer recurrence.

Commercial tests for CYP2D6 genotyping are available, and some experts have argued that they should be used in clinical practice, particularly in the case of postmenopausal women, where there is an alternative therapy to tamoxifen that can be offered (i.e., aromatase inhibitors).

But the editorial states that "routine use should await more reliable evidence from well-designed studies."

Currently, there remains an "uncertaincy" about such testing because it is based on "incomplete biological evidence and inconsistent epidemiologic evidence," write editorialists Timothy Lash, MD, PhD, and Carol Rosenberg, MD, from Boston University School of Health in Massachusetts. They also point out that there is no current consensus guideline that recommends CYP2D6 genotyping in patients prescribed tamoxifen.

One Exception — Other Drugs That Affect Metabolism

Among all of this uncertainly, however, there is one situation in which a practice recommendation can be made, the editorialists explain. They urge physicians to "to be cautious about prescribing comedication that might interact with tamoxifen via CYP2D6 inhibition." Where there are therapeutically similar drugs available, it is "prudent to prescribe medications that inhibit CYP2D6 very little," they add.

One example of this is the drugs that are often coprescribed in this patient population for depression and/or vasomotor symptoms — paroxetine and fluoxetine have been reported to interfere with CYP2D6, whereas citalopram and venlafaxine have not.

New data on the interaction between paroxetine and tamoxifen have just been published online in the British Medical Journal, and make this case more forcefully. This interaction was shown to reduce the ability of tamoxifen to protect against breast cancer recurrence to the extent that it increased the risk for death from breast cancer. As a result, clinicians have now been warned against coprescribing the 2 drugs.

Highly Variable Practice Patterns

Apart from that exception, however, the editorial emphasizes the "uncertainty" over the data that are available so far on CYP2D6.

"At present, we do not routinely perform CYP2D6 genotyping for any patient," Dr. Rosenberg told Medscape Oncology. "There are a few individualized exceptions, but they are rare and the situations can't be generalized," she added.

This is in contrast to the practice that others have reported, in particular at the Mayo Clinic in Rochester, Minnesota, where Matthew Goetz, MD, said his team has been testing for CYP2D6 for 3 years now. Dr. Goetz and colleagues were among the first to report the effect of CYP2D6 on tamoxifen efficacy, and last year published additional data on 1325 women.

"In the postmenopausal setting, where there is an alternative therapy, the role of genotyping is very compelling, and I think it's very reasonable in that situation to discuss this test," Dr. Goetz told Medscape Oncology.

The situation is different in premenopausal women, where the data are sparse and there is no proven alternative therapy, but even here, "what data there are suggest that genotyping is also important in this patient population," Dr. Goetz explained.

"The data do not suggest that the poor metabolizers do not benefit from tamoxifen, but the data suggest that they do not do nearly as well as the extensive metabolizers," he added. "I strongly believe that pharmacogenetics does play a role here."

These 2 opposing opinions are reflected in what the editorialists describe as "highly variable practice patterns for CYP2D6 genotyping." They report carrying out an informal survey among medical oncologists, and also report discovering many different views.

"Some never assay the genotype, some decide on a case-by-case basis, and a few assay frequently," the editorialists report. "Some wonder whether it is ethical to prescribe a drug that may be ineffective if an alternative exists. Most agree that interpreting the assay's results can be problematic, given the lack of definitive data."

They add that even clinicians who do carry out the test "seem to cross their fingers that the patient turns out to be a 'normal metabolizer,' so they can prescribe tamoxifen without concern."

New Study With "Dramatic" Results

The editorial was prompted by a study also published online February 1 in the Journal of Clinical Oncology, which reported that CYP2D6 testing identifies women who have a poor response to tamoxifen. The researchers, led by Kazuma Kiyotani, from the RIKEN Center for Genomic Medicine, Yokohama, Japan, reported data from a series of 282 patients, 67 of whom they had reported on previously.

These women had invasive hormone-receptor-positive breast cancer, had undergone surgery, and were prescribed tamoxifen for 5 years to reduce the risk for breast cancer recurrence.

Women who were identified by CYP2D6 testing as poor metabolizers of tamoxifen had a 9.5-fold higher rate of recurrence than those who were identified as good metabolizers, the researchers report. Those who were identified as "intermediate metabolizers" (with only 1 fully functional allele) had a 4.5-fold higher rate of recurrence.

"These results are dramatic," say the editorialists, but they add that this study shares "important limitations with earlier reports." These include selection of study participants, limited consideration of other prognostic markers, and perhaps most important, immortal person-time bias. This arises in observational studies when follow-up time is included in person-time at risk for the study outcome, even though that time precedes the last event required for entry into the study population, they explain.

In contrast, Dr. Goetz said that "this study provides additional evidence in regard to the role of CYP2D6 metabolism and clinical outcome."

"What needs to happen in our community is that we have to come up with some guidelines as to when it is and when it is not appropriate to carry out this testing," he added.

The editorialists and the researchers have disclosed no relevant financial relationships. Dr. Goetz reports that he is named, along with colleagues and the Mayo Clinic, as an inventor on nonprovisional patent applications for tamoxifen and CYP2D6; the technology is not licensed and no royalties have accrued yet. Dr. Goetz also reports having acted as a consultant for Roche and having received honoraria from Pfizer, Roche, and DNA Direct.

J Clin Oncol. Published online February 1, 2010. Abstract, Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.