By Turna Ray
Several studies presented at the San Antonio Breast Cancer Symposium last week investigated whether CYP2D6 testing can help doctors decide whether to treat breast cancer patients with tamoxifen, but the body of evidence did not resolve questions about the role of CYP2D6 genotypes or myriad other factors contributing to variability in tamoxifen metabolism.
Tamoxifen is a hormonal treatment that has been shown to reduce the risk of breast cancer recurrence. CYP2D6 enzymes convert tamoxifen to the metabolite endoxifen, which suppresses the development of cancer cells by blocking the estrogen receptors that drive tumor growth. Some studies have shown that patients with certain CYP2D6 mutations that make them "poor" or "intermediate" metabolizers of tamoxifen experience a higher breast cancer recurrence rate while on tamoxifen than "normal" metabolizers of the drug. Meanwhile, other published studies have shown no association between CYP2D6 genotypes and patient outcomes when treated with tamoxifen.
As far as the US Food and Drug Administration and payors are concerned, the clinical evidence supporting the utility of pharmacogenetic testing ahead of treating patients with tamoxifen is far from convincing. The FDA has been mulling whether to update tamoxifen's label with information on CYP2D6 testing since 2006 (PGx Reporter 11/15/06).
Also, earlier this year, the Centers for Medicare & Medicaid Services' Medicare Coverage and Evidence Development Advisory Committee discussed whether there was sufficient evidence on CYP2D6 testing for tamoxifen to warrant coverage in the Medicare population. Based on an independent review of the published literature, MEDCAC found that 11 of 13 studies did not report an association between CYP2D6 status and disease recurrence. Furthermore, most of the reviewed studies were retrospective in design, while only two studies out of 13 were "repurposed randomized-controlled trials." (PGx Reporter 02/03/10).
At the breast cancer conference held in San Antonio, Texas, last week, Matthew Goetz, assistant professor of oncology at the Mayo Clinic, reviewed several studies on tamoxifen metabolism and CYP2D6 testing. He pointed out that since 2003, 14 published studies have found that CYP2D6 genotyping to administer tamoxifen improves outcomes, while 15 studies have shown there is no association between genetic testing and improvement in a patient's likelihood of breast cancer recurrence.
"There is marked heterogeneity with regard to the studies that are out there," Goetz told PGx Reporter this week.
In Goetz's view, the knowledge gaps regarding tamoxifen metabolism variability center around the specific interplay between endoxifen concentrations, differing features of tumor subtypes, and CYP2D6 and other genotypes. He noted that retrospective analysis in this case is unlikely to lead to conclusive answers since it wouldn't account for patients who didn't take the drug for the recommended five years, or those who simultaneously took other CYP2D6 inhibitors or over-the-counter drugs that might have interfered with tamoxifen's efficacy.
"What we don't know is how important is this reduction in CYP2D6 activity [and] what are the other factors that contribute to the variability in endoxifen concentrations," Goetz said. "Probably the only way we're going to sort this out is by doing prospective studies."
The data presented at SABCS, while informative, still didn't yield conclusive answers about the impact of CYP2D6 gene variations on the variability of tamoxifen response.
For example, one study, conducted by the International Breast Cancer Study Group and the Breast International Group 1-98 Collaborative Group, conducted CYP2D6 genotyping in tumors of patients who participated in the double-blind randomized trial BIG 1-98 from 1998 to 2000, to see if testing was predictive of disease outcome. The original BIG 1-98 trial enrolled more than 8,000 patients and compared patient outcomes when they were treated with the aromatase inhibitor letrozole, tamoxifen, and sequences of letrozole and tamoxifen as adjuvant endocrine therapy in post-menopausal women with endocrine-responsive breast cancer.
At SABCS, the researchers presented interim results after genotyping formalin-fixed, paraffin-embedded tumor samples from 2,000 patients. Going into the study, the authors hypothesized that certain CYP2D6 variants would be associated with shorter breast cancer-free intervals on tamoxifen monotherapy, and that a similar relationship would not be observed with letrozole monotherapy as a control. The researchers are also exploring whether there is a link between CYP2D6 genotypes and sequential therapies.
The preliminary results, as reviewed by Goetz in his presentation, show no association between CYP2D6 *4 genotypes or CYP2D6 haplotypes with breast cancer recurrence, either in the tamoxifen or the letrozole arm. The researchers plan to genotype another 2,000 patients before announcing the final results.
Another SABCS abstract, by researchers from Baylor College, Indiana University, and the University of Michigan, presented data from a large community-based cohort study that found evidence contrary to the hypothesis that low CYP2D6 confers worse outcomes when treated with tamoxifen.
The researchers in this study assessed whether there are associations between clinical outcomes and CYP2D6 variants and variants in UGT2B7. While CYP2D6 enzymes help convert tamoxifen to its active metabolite endoxifen, UGT2B7 is an enzyme that eliminates endoxifen. The study enrolled 500 patients with ER-positive, early breast cancer treated with adjuvant tamoxifen and another 500 who did not receive any adjuvant treatment. Study participants were genotyped and received a score between 0, for no CYP2D6 alleles, and 2, indicating high activity of CYP2D6 alleles. These patients were also grouped according to UGT2B7 status, based on the presence of the *2 allele.
[ pagebreak ]
"Univariately, Kaplan-Meier estimates in conjunction with the log-rank test did not reveal any significant associations between either CYP2D6 or UGT2B7 score and time to relapse," the researchers reported. However, "in tamoxifen-treated cases, after adjustment for tumor size, nodal status and PgR, Cox regression found [an] increased CYP2D6 score, reflecting increased conversion of tamoxifen to endoxifen, was significantly associated with worse time to relapse," but this was not found in those without endocrine treatment.
Furthermore, patients with high UGT2B7 activity — and, subsequently, greater elimination of endoxifen — had better survival rates in the tamoxifen-treated group, but no effect of UGT2B7 was seen in the untreated group. Finally, researchers did not find CYP2D6 and UGT2B7 genotypes to be associated with any special tumor characteristics in either arm of the study.
"In our study, CYP2D6 and UGT2B7 genotypes associated with increased endoxifen levels were associated with worse outcome in tamoxifen-treated (but not untreated) breast cancer patients," the researchers concluded. "Thus, the data from this cohort study do not support the hypothesis that breast cancer patients with low CYP2D6 are resistant to tamoxifen."
Goetz highlighted at SABCS that the strength of the BIG 1-98 trial was that the data was derived from a large prospective adjuvant hormonal study, which defined that standard of care for early-stage breast cancer patients. However, the difficulty with looking at the impact of CYP2D6 genotypes on adjuvant tamoxifen treatment is that such analysis requires exceptionally large patient cohorts and lengthy timelines.
"If one were to look at the issue of how much endoxifen is necessary and how much is needed in various tumor subtypes, and you were going to do this in the context of an adjuvant trial … with tamoxifen, or any hormonal therapy, it would require thousands of patients to be followed for five to 10 years," Goetz noted. "Because of that, what we don't have access to are studies that collected pharmacokinetic patients prospectively."
However, this type of analysis done prospectively "will be critical to understanding the levels of endoxifen that are necessary to the metabolism of tamoxifen," he said.
Bypassing Metabolism Issues
The studies presented at SABCS did not yield straight answers to the unknowns at the heart of the tamoxifen metabolism controversy: the amount of endoxifen that is necessary to inhibit tumor growth and what other factors might affect endoxifen concentrations.
Goetz noted that different tumor subtypes may require varying endoxifen concentrations. "There is much more variability with regard to endoxifen concentrations than reflected in CYP2D6. And that's why we have such great heterogeneity," he observed. "Low endoxifen concentrations may not necessarily be detrimental in all breast cancer subtypes. There may only be some tumor subtypes where achieving higher endoxifen levels is critical"
While the research community works toward conducting the necessary studies to answer these questions, the Mayo Clinic is working on new diagnostic approaches to gauge tamoxifen variability and is also investigating new drugs to entirely bypass the metabolism issues with tamoxifen.
For example, Mayo is developing a CYP2D6 breath test that measures metabolism of oral dextromethoraphan, a cough suppressant that is converted into the active metabolite dextrophan by the CYP2D6 enzyme. "In this situation, we're developing a test to measure the metabolism of dextromethoraphan through a breath assay," Goetz said. He noted that the assay under development can identify CYP2D6 poor metabolizers and gauge CYP3A activity.
Measuring dextromethoraphan metabolism via the breath assay could be "a better indicator of endoxifen concentrations and tamoxifen variability" than measuring CYP2D6 genotype directly, he suggested. The Mayo Clinic is currently recruiting patients for a trial investigating whether the enzymes measured by this breath test can more effectively identify women who don't respond to tamoxifen.
[ pagebreak ]
Another solution is to develop a drug that circumvents tamoxifen metabolism by targeting endoxifen directly. Mayo is collaborating with the National Cancer Institute to conduct Phase I studies involving an NCI-developed drug formulation called endoxifen hydrochloride. The trials are expected to begin next year.
"The idea is to completely bypass the metabolism issues inherent in tamoxifen, not just CYP2D6 but all the different isoforms, like CYP3A, that are necessary for metabolic activation, deliver the drug directly, and achieve adequate concentrations of endoxifen," Goetz explained. "The hypothesis is that if we achieve substantial concentrations of endoxifen that we should see activity of the drug in patients who previously failed on tamoxifen," Goetz said.
"These data need to be confirmed prospectively," he urged in his SABCS presentation, noting that the controversy surrounding tamoxifen's pharmacogenetics is "unlikely to be resolved through retrospective analyses," given that CYP2D6 only partially explains the extensive variability in endoxifen pharmacokinetics.
Lastly, Goetz recommended that doctors might prescribe alternatives to tamoxifen, such as GTx's anti-estrogen hormone therapy Fareston.
Given the uncertainty about the role of CYP2D6 genotypes in tamoxifen metabolism, there are business opportunities for drug developers that market competing hormone therapies. GTx, the developer of Fareston, recently sponsored a survey of 700 women with breast cancer currently taking tamoxifen to gauge their knowledge and understanding of CYP2D6 genetic testing in the context of tamoxifen therapy.
The company found that nearly 80 percent of the women surveyed would choose to have a genetic test that determines CYP2D6 genotype, but only 14 percent have been tested. This analysis was conducted during a three week period in November through Breastcancer.org.
The study also showed while more than half the participants knew that commonly used antidepressants, such as fluoxetine (Prozac), bupropion (Wellbutrin), sertraline (Zoloft), and paroxetine (Paxil) might reduce tamoxifen's efficacy, around two-thirds of those surveyed were unaware that common over-the-counter drugs containing diphenhydramine (Benadryl) or acid reducers containing cimtidine (Tagamet) might also interfere with tamoxifen metabolism.
William Irvin of the University of North Carolina, Chapel Hill, a CYP2D6 expert who helped design the survey, said in a statement that the data suggest that "patients want more discussion on the pros and cons of emerging science … [on] CYP2D6 testing, even if we don’t yet have definitive answers to all of their questions."
According to McDavid Stilwell, GTx's director of communications and financial analysis, the survey showed that women aren't very aware about the limitations of CYP2D6 testing to administer tamoxifen. He said that GTx, in its conversations with doctors, attempts to bring up some of the uncertainties related to treatment with tamoxifen when discussing the benefits of Fareston.
"We do talk to doctors about the issues of how our drug is metabolized. It's not a prodrug, so it's already active," Stilwell said. "So that takes some of the guesswork out of" administering the drug.
Fareston is indicated for the treatment of metastatic breast cancer in post-menopausal women with ER-positive or unknown tumors. GTx's website for Fareston asks in bold text in a banner, "Concerned about CYP2D6? Fareston may be the answer." The company further promotes the fact that unlike tamoxifen, Fareston does not need to be converted to an active form; there are no gene-related metabolism variability issues; and there are no drug interactions with other CYP2D6 inhibitors, such as the selective serotonin reuptake inhibitors Paxil and Prozac, or Benadryl-containing over-the-counter medicines.
"These concerns don't have to enter the conversation with Fareston," Stilwell said. "So, for women with metastatic breast cancer, who are post-menopausal, Fareston is a very good alternative." Furthermore, Stilwell noted that in post-menopausal women who have trouble tolerating aromatase inhibitors, Fareston might be a good option as an alternative to tamoxifen without raising concerns about metabolism and genetic testing issues.
In terms of adverse reactions, QTc prolongation has been seen in some patients undergoing treatment with Fareston, so patients considering the drug are advised to tell their doctor if they are receiving drugs that prolong the QTc interval such as quinidine, dofetilide, sotalol, amiodarone, erythromycin, and certain antipsychotics. Additionally, the drug has been shown to increase the risk of blood clots and is contraindicated in those who are at risk of getting blood clots in the legs, lungs, or eyes.
[ pagebreak ]
In Stilwell's view, the data coming out of SABCS does not settle the issue as to the exact role of CYP2D6 genotypes in tamoxifen metabolism. "The controversy remains, and prospective studies are needed," Stilwell said, agreeing with Goetz's review of the data presented at the meeting.
Goetz, after reviewing the latest data at SABCS on tamoxifen CYP2D6 testing, recommended that women be better educated about the controversy surrounding tamoxifen pharmacogenetics.
Given the evidence gaps he cited, Goetz does not recommend that doctors routinely perform CYP2D6 testing for postmenopausal women with ER-positive breast cancer to make decisions about whether to prescribe tamoxifen or aromatase inhibitors.
However, he posited that CYP2D6 testing "could be considered" in post-menopausal women who are at high risk of breast cancer who prefer tamoxifen or aromatase inhibitors. Furthermore, Goetz advised doctors to "exert caution and avoid potent CYP2D6 inhibitors in women treated with tamoxifen."
Even with these unresolved issues about tamoxifen's pharmacogeneitcs, medical practice has changed to account for the drug-drug interactions impacting tamoxifen metabolism. "It is interesting to us that doctors understand the tamoxifen metabolism issues enough that they've changed the way they practice … to move SSRI use away from CYP2D6 inhibitor drugs, and moved toward drugs like" Effexor (venlafaxine), an SNRI (serotonin and norepinephrine reuptake inhibitor) that is not thought to inhibit CYP2D6.
GTx's survey showed that more women in the study group were on Effexor and very few were taking Prozac. "That's a change that's come about in the last five years as the CYP2D6 metabolism issues have become better understood," Stilwell said. "So, concerns there have shifted the practice of medicine, but concerns regarding the genetic disposition of these women have not yet been clarified to the extent that physicians routinely test or don't test women for their CYP2D6 status."
Have topics you'd like to see covered in Pharmacogenomics Reporter? Contact the editor at tray [at] genomeweb [.] com.