With all the time and money that has been channelled into breast cancer research, with largely positive results, there is one gaping hole that is apparently not going to be filled anytime soon. That is, the issue of whether or not women (particularly pre-menopausal women) with estrogen/progesterone-responsive cancer will receive any benefit from chemotherapy.
A typical breast cancer patient will probably never know about this gaping hole in our knowledge, because if she has any positive lymph nodes, or a tumor that is larger than 1 centimeter, her oncologist will almost certainly tell her that she must receive chemotherapy. If she is relatively young, her oncologist will also almost certainly push on her the most extreme chemotherapy regimen possible.
The trouble is, no one, including oncologists, really knows if chemotherapy makes a useful impact on hormone-positive breast cancer.
How is this possible, with all the research that has been conducted? The answer lies in the examination of studies on the interaction between hormone therapy and chemotherapy. Here is what we do know:
First, the biggest impact on future disease-free survival for hormone-positive breast cancer is made by hormone therapy. The various options include taking tamoxifen, an estrogen blocker; taking an aromatase inhibitor, which reduces estrogen production in post-menopausal women only; ovarian removal, irradiation, or chemical suppression (because ovaries are the primary source of estrogen in pre-menopausal women); or a combination of these.
Second, while hormone-negative breast cancer is clearly responsive to first-generation chemotherapy – that is, something similar to the long-time standard for breast cancer, cyclophosphamide + methotrexate + fluorouracil (CMF) – hormone-positive cancer is not at all clearly so. In hormone-responsive cancer patients, when CMF treatment is compared to ovarian supression, overall 10-year survival rates are similar. A recent study shows this even for women at higher risk for recurrence, i.e. with large tumor size and/or positive lymph nodes (Ejlertsen et al., 2006).
In fact, for CMF patients, it is possible that the benefit received is mostly, if not entirely, due to the common occurrence of chemotherapy-induced amenorrhea, which is the technical way of saying that chemotherapy usually shuts down women’s ovaries (in many cases, permanently). So, those receiving ovarian ablation and CMF are actually receiving their therapeutic benefit via the same mechanism – except for those receiving CMF, there are a lot more unpleasant side-effects.
What we do not know is whether second- and third-generation chemotherapy regimens (using anthracyclines and/or taxanes) do have a truly cytotoxic effect, or also a mainly hormonal benefit. There are simply not yet data comparing these regimens to straight hormonal therapy. The data that do exist show a marginal benefit to adding chemotherapy to hormonal therapy for higher-risk hormone-positive patients – meaning, statistically, undergoing these regimens adds only a few percentage points to your likelihood of remaining free of cancer for the next ten years. But it is certainly possible that as with CMF therapy, the statistical benefit received from later-generation chemotherapy is derived primarily from suppression of ovarian function. And the side-effects in this case are not only significantly more unpleasant than with CMF, but include a small probability of long-term heart damage (with anthracyclines) or nerve damage (with taxanes).
Why do we not have the data to allow women to make a truly informed decision about chemotherapy? There are multiple reasons.
First, the use of anthracyclines and taxanes is recent enough that there hasn’t been the time to conduct a lot of studies, as has been done with CMF, which has been around for decades. On the one hand, cancer patients today are lucky that there has been so much recent progress in chemotherapies that are, statistically, clearly more effective than earlier generations. On the other hand, we know a lot less about how beneficial these treatments will be for an individual patient.
Second, there is an attitude clearly prevalent among both doctors and their patients that given lack of definitive data, the default is to do “all we can” to kill the cancer. Ironically, this probably leads to less demand for definitive research on what the true benefit of treatment actually is. There is a definite sense that it is almost more important to feel as though one is doing all one possibly can, than to actually do what the science says makes the most sense. The fact that the science for many brutal treatments is weak or marginal has not affected our cultural impression that to treat cancer, one must suffer. There are always researchers looking for ways to make chemotherapy more directed, and therefore less toxic to non-cancer cells in our body, but will patients trust it to work if they ever do?
The idea that stronger medicine must be better is highlighted by a multitude of research that has been asking the question, “does ovarian suppression add a treatment value to women undergoing chemotherapy?” The researchers are, in most cases, looking at the benefit of ovarian suppression as a component of chemotherapy, which it is assumed the patient should do. Given the difference we already know in the relative benefits of hormone therapy vs. chemotherapy, this is simply the backwards way to ask the question. But researchers are biased toward assuming the more modern treatment of chemotherapy, rather than hundred-year-old hormone therapy, must be better.
The third reason is directly related to the second. The study that needs to be done is one in which pre-menopausal hormone-positive breast cancer patients receive hormone therapy, and then are randomized to either receive or not receive chemotherapy, in order to determine whether the latter has a true added benefit apart from its ovarian suppression. Exactly this design was attempted a few years ago by the International Breast Cancer Study Group in the Premenopausal Endocrine Responsive Chemotherapy (PERCHE) trial. Unfortunately, the experiment had to be abandoned, because not enough participants could be recruited. The reason appears to be that while women do not mind having their hormone treatment randomized, as has been successfully done in other studies, they want to retain choice over whether or not to do chemotherapy (Regan et al., 2008). Simply, women prefer to follow the recommendations of their doctors when deciding on chemotherapy, which are in many cases much more aggressive than necessary, because “better safe than sorry” trumps the admission that we do not actually know if it is truly beneficial. It is therefore a chicken-and-egg problem.
Thus, although as some frustrated authors state, “[c]learly there is a group of patients with low to intermediate risk of relapse after surgery for early breast cancer for whom chemotherapy adds little or no benefit to combined endocrine therapy” (Regan et al., 2008), most patients (and their doctors) continue to prefer to believe they are all in the group that will benefit. This pervasive faith in taking action that overcomes our ability to find the truth has always been a weakness of chemotherapy. Cancer is scary, and often fatal. We don’t want to accept that there might be nothing we can do, so we do what we can, and whether or not that something actually works or not seems to become a secondary concern in many cases. Because of this oncologists themselves can reduce the emphasis on treatments that are known to have a true benefit, perhaps because they are not painful enough, or modern enough, and thus don’t convince us as strongly that we’re doing something.
But the important facts are: Tamoxifen works pretty well at preventing cancer recurrence, but is not effective for all women and for many women loses its effectiveness over time (Wegman et al., 2005). Aromatase inhibitors are better, but only work if the ovaries are gone or have unequivocally stopped estrogen production. The best therapy for preventing recurrence of hormone-responsive breast cancer is to keep cancer receptors from receiving estrogen, and the best way to do that is to eliminate all estrogen in the body as entirely as possible. The benefit of chemotherapy is at best marginal compared to the benefit of estrogen removal. And yet far more hormone-positive American women undergo chemotherapy than destroy the estrogen sources in their body, which is much easier and less costly.
Addendum, May 2009: Some of the arguments made here have been born out by new reports that anthracyclines, long pushed by oncologists on all (especially young) breast cancer patients, have a positive effect only on women whose cancer is HER-2 positive. This is an important acknowledgment because anthracyclines (such as adriamycin) can have dangerous, irreversible effects. Three out of four oncologists I consulted with suggested I take adriamycin, but it turns out in my specific case it was correct to go with my gut and decline this drug. If you are in the process of deciding on chemotherapy for breast cancer, please review this latest research with your oncologist and ask if it is relevant to your particular pathology.
Ejlertsen, B, Mouridsen, H.T., Jensen, M., Bengtsson, N., Bergh, J., Cold, S., Edlund, P., Ewertz, M., de Graaf, P.W., Kamby, C., Nielsen, D.L. 2006. Ovarian ablation compared with cyclophosphamide, methotrexate, and fluorouracil: Similar efficacy from a randomized comparison in premenopausal patients with node-positive, hormone receptor-positive breast cancer. J Clin Oncol. 24:4956-4962.
Regan, M., Pagani, O., Walley, B., Torrisi, R., Perez, E. A., Francis, P., Fleming, G. F., Price, K.N., Thurlimann, B., Maibach, R., Castiglione-Gertsch, M., Coates A.S., Goldhirsch A., and Gelber, R.D. 2008. Premenopausal Endocrine-responsive Early Breast Cancer: Who Receives Chemotherapy? Ann Oncol. 19(7):1231-1241.
Wegman, P., Vainikka, L., Stal, O., Nordenskjold, B., Skoog, L., Rutqvist, L. and Wingren, S. 2005. Genotype of metabolic enzymes and the benefit of tamoxifen in postmenopausal breast cancer patients. Breast Cancer Res. 7(3): R284-R290.