Women who experienced menopausal symptoms were less likely to adhere to preventive treatment with tamoxifen for breast cancer than those who did not experience menopausal symptoms, according to a U.K. study presented at the 2016 San Antonio Breast Cancer Symposium (Abstract S5-03).
Preventive therapy for breast cancer is recommended for women who have an increased risk of developing the disease. Currently, approximately 10 million women in the U.S. have an increased risk of breast cancer and, therefore, could benefit from preventive therapy, such as tamoxifen. However, for a variety of reasons, only a small number of women start preventive therapy and even fewer continue on therapy long enough to receive the intended benefit.
Discontinuing the drug early is often attributed to the side effects, so the study researchers aimed to determine the effect of menopausal symptoms, which are common side effects of tamoxifen, on long-term adherence. The menopausal symptoms in this study were defined as nausea and vomiting, headaches, hot flashes, and gynecological symptoms, which included irregular bleeding, vaginal dryness, and vaginal discharge. The women in this study were those who had enrolled in the International Breast Cancer Intervention Study I (IBIS-1). These women did not have a history breast of cancer but were at a heightened risk for developing the disease.
IBIS-1 Study Results
IBIS-1 was a placebo-controlled, randomized trial designed to determine whether 5 years of preventive therapy with tamoxifen could prevent breast cancer. The trial reported that 5 years of tamoxifen therapy reduced the risk of breast cancer by at least 30 percent and the effect appears to last for 20 years. A total of 3,987 women in the U.K. were analyzed, of which 2,000 were assigned to placebo and 1,987 to tamoxifen, a commonly used preventive therapy for breast cancer.
“Overall, we showed that 66.8 percent of women were adherent for 4.5 years within this trial. There was significant difference between the trial arms, such that 71.5 percent of women in the placebo arm were adherent to the therapy compared to the 62.1 percent in the tamoxifen arm,” said Samuel G. Smith, PhD, lead study researcher and Cancer Research UK postdoctoral fellow and university academic fellow at the University of Leeds in the U.K. “This difference was statistically significant.”
Also, the overall mean time on the study was 4.1 years. For the placebo arm, the mean time of the study was 4.3 months and 3.9 months for the tamoxifen arm; this difference was statistically significant. The overall rate of symptoms across both groups was 5 percent for nausea and vomiting, 7 percent for headaches, 31.5 percent for hot flashes, and 20.9 percent for gynecological symptoms. Also, dropout rates were the fastest in the first 12 months for both groups but more so for the tamoxifen-treated group.
Nausea and vomiting and headaches were both associated with adherence, and women who experienced these symptoms were less likely to adhere to treatment for both arms. Hot flashes were not associated with adherence and did not affect adherence for either treatment arm. Gynecological symptoms were not associated with adherence, but women in the tamoxifen arm who had gynecological symptoms were less likely to adhere than those in the placebo arm.
“There was an association between some menopausal symptoms and adherence, and these effects were largely similar between trial arms, suggesting women may be attributing naturally occurring symptoms to the therapy they are taking,” Smith said. “This is suggestive of the nocebo effect.”
The nocebo effect, slightly different from the placebo effect, is a phenomenon in which a person's belief that a treatment will have negative effect leads to the person having a more negative effect than he or she otherwise would. Unlike the placebo effect, the nocebo effect can occur in the treatment and placebo group.
“Is that common in cancer and in breast cancer? Absolutely,” Ann H. Partridge, MD, MPH, medical oncologist and Director of the Adult Survivorship Program at Dana-Farber Cancer Institute, told Oncology Times about the nocebo effect.
In fact, this phenomenon was also seen in the IBIS-2 study, which looked at preventive therapy with anastrozole, an aromatase inhibitor that can have side effects such as aches and pains in bones.
“The IBIS-2 trial showed that although women taking anastrozole do have an increased risk of experiencing symptoms like aches and pains in bones, women in the placebo group also reported quite a high prevalence of those particular symptoms,” noted Smith.
A common example of the nocebo effect is anticipatory nausea, in which a woman feels nauseous before she even receives chemotherapy. This negative expectation, Partridge explained, results in a patient feeling “worse than a person who didn't expect horrible symptoms.”
Regarding how to approach a patient who may be mistakenly attributing a symptom to a drug, Partridge offered clinicians some advice.
“The answer isn't to say it's definitely not the drug to an individual woman,” Partridge said. The reason why, she explained, is the patient can become upset and “shut you out” as a physician because she thinks you're not listening to her. She recommended pursuing a different approach to treatment, such as taking a temporary break from the drug, and not necessarily focusing on how to attribute the symptoms because ultimately you want to help the patient benefit.
To manage negative expectations patients may have, Partridge noted clinicians can educate patients about the realistic harms and benefits of a therapy before the therapy begins.
“Though we had this finding about the symptoms and their relationship with adherence, I think a really telling message is that only just over two-thirds of women are actually adherent to their medication for the full 5 years,” said Smith. “So there's a third of women that start taking preventive therapy as part of a clinical trial who may not be experiencing the full benefit of the prevention, and there's considerable scope for intervention there once we also address the uptake issue as well.”
In particular, during his presentation, Smith suggested intervening early during treatment because one of the findings of the study was that that dropout rates were fastest within the first 12 months. He added, “Cognitive behavioral therapy has shown to be effective in the adjuvant setting with regard to managing menopausal symptoms. It would be interesting to test such an approach in the primary prevention setting.”
During the press conference, the moderator, C. Kent Osborne, MD, Professor and Director of the Dan L. Duncan Comprehensive Cancer Center at Baylor College of Medicine, commented, “There are things that we can do to ameliorate the symptoms for sure. In patients on tamoxifen and who have vaginal problems, that's easy to fix with vaginal estrogen.” He also said that hot flashes can be dealt with by using acupuncture, the antidepressant venlafaxine, or some other antidepressant that effectively relieves hot flashes.
Although increasing initiation of preventive therapy in addition to adherence among these women is needed, Smith noted, “We shouldn't necessarily be increasing uptake among women who have made an informed decision to say they don't want to take preventive therapy. It's the women who are either not being offered the opportunity to take preventive therapy in the first place or who are either being given misleading information or who just don't come away from a consultation knowing the harms and benefits of tamoxifen or any other preventive therapy. It's those women, I think, we could be intervening with.”
One limitation of this study, as was pointed out by Partridge during the question-and-answer portion of the presentation, is the generalizability of these results. “These are women who are not only clinical trial participants but also chose to go on a prevention trial, which we know is a select minority of our patients.”
This limitation of generalizability of results is common among clinical trials and even more so among cancer prevention trials because people who are not sick generally may not see the need or benefit of taking a medication long-term. Thus, the type of people who enroll in cancer prevention trials typically do not represent the average patient.
“That doesn't mean it's not worth looking at what he did,” Partridge later told Oncology Times regarding the limitation of generalizability of the study. “I think it's fascinating data that suggests that a lot of women are having symptoms, and interestingly, they were similar between the tamoxifen- and placebo-treated arms.” She explained the symptoms are less likely attributed to the drug than they are about the state of where these women are in their lives. “We need to intervene, educate, and support women from a history management standpoint if we want them to take drugs that may give them benefits from cancer prevention.”
You need to be logged in to save this episode to a playlist.
You must be logged in to display playlists.