This story looks at two classes of drugs – aromatase inhibitors and bisphosphonates – which it says may deliver a “powerful one-two punch” when treating breast cancer. It reports on two Lancet studies that compiled data from 9 and 26 trials, respectively – what are known as meta-analyses.
The studies reported on are newsworthy and there’s certainly some interesting and useful information here. But what’s missing from the piece are key details, including the fact that these drugs can cause significant side effects that may lead some women to stop taking them. Consulting an independent expert may have shed light on this issue for readers.
We also think some of the language used in the story was misleading. The headline focuses the “one-two punch” of using the two types of drugs in combination, but there is no evidence in either meta-analysis to support combining the two drugs. Since that wasn’t an element of the research being reported on, why leap to that in a headline?
Aromatase inhibitors have long interested scientists as an alternative to tamoxifen, the standard therapy for estrogen receptor(ER)-positive breast cancer, which may increase the risks for uterine cancer, blood clots and eye problems. Standard interventions range from surgery to drug therapy and are often invasive and toxic, sometimes causing side effects that appear months or years after treatment has ended. While non-invasive therapies found to be effective and to reduce the severity of side effects are newsworthy, readers should know that aromatase inhibitors are only used in women who’ve undergone menopause — either natural menopause or induced (e.g. by chemotherapy, oophorectomy, etc). In addition, aromatase inhibitors are not new and have been used both after a breast cancer diagnosis and to prevent breast cancer for some time. Significant side effects of aromatase inhibitors and bisphosphonates lead many women to stop taking them.
The Washington Post article calls aromatase inhibitors “inexpensive,” but it doesn’t put a price tag on them. According to some reports, patient copays for taxomifen and aromatase inhibitors are around $8 – $9 for a month’s supply of each drug, which we’ll agree is inexpensive. But the cost of bisphosphonates is not addressed. We’ll give the benefit of the doubt here for the story’s nod in the direction of costs, but ideally we’d prefer to see more specifics.
The article reports risk reduction of dying from breast cancer in relative numbers, but we think benefits of an intervention should always be stated in absolute figures when possible. We believe relative risk numbers tend to inflate the impression of how much impact the drug really has. For example, the story states that bisphosphonates could reduce the risk of death by 18% over 10 years. It would be more informative to note that the 10-year risk of death was 14.7% for women who took bisphosphonates vs 18.0% for women who did not receive bisphosphonates — a 3.3% absolute reduction.
Another issue here is that aromatase inhibitors are only used in postmenopausal women — either natural menopause or induced. The story doesn’t explain this.
Harms aren’t really addressed in this report, except for the last sentence, which states the “main side effect of aromatase inhibitors is an increase in bone loss and fractures.”
But aromatase inhibitors also carry the risk of exacerbating menopausal symptoms (including sexual side effects and sleep problems) and of serious muscle and joint pain, which has caused some women to stop taking the drug in the past.
Bisphosphonates long term can cause bone fractures (paradoxical – they build bone but not always normal bone) and rarely can cause jaw osteonecrosis (“bone death”). This is more common after dental work, but can also happen sporadically and is devastating.
Since these details are missing in the report, we give it a “Not Satisfactory” rating here.
The Washington Post report does not point out that these two Lancet journal articles are meta-analyses, meaning the researchers compiled data from various trials and then combined these results into comprehensive reports. The story states that the study came from a collaboration “that collects and analyzes information about randomized clinical trials every few years” but it never specifies what that means. Meta-analyses are essentially a summation of clinical trials, being only as good as the individual studies themselves. As a result, biases in any of these trials are often carried over to the meta-analysis. For example, if a pharmaceutical company financially backed one of these trials, it’s possible that a favorable bias could be introduced in the meta-analysis.
Large, double-blind, randomized clinical trials are the gold standard and are seen as the most reliable way to obtain reproducible results. Sometimes, a randomized clinical trial has yielded a different outcome from a meta-analysis looking at the same clinical question. A 1997 New England Journal of Medicine article documents these discrepancies.
Another fact the article brings up is that a “substantial number of trial participants switched from tamoxifen to aromatase inhibitors, which affected their results.” But how might this exactly confound the results of these trials? An independent expert could have shed some light on this topic.
There is no disease mongering in this article. Although, as noted above, we think there’s some “treatment mongering” going on both the headline and closing quote from a researcher who says, “The drugs are complementary, because the main side effect of aromatase inhibitors is an increase in bone loss and fractures, while bisphosphonates reduce bone loss and fractures as well as improving survival.” These two drugs are commonly prescribed together but the study does not address their use in combination. If the story is going to focus on an issue that wasn’t even addressed in the studies being reported on, we think it’s essential to include a statement from an independent expert about that suggestion.
This is a two-source story, with quotes from two professors who had ties to the studies. The report desperately needed an independent expert, who could talk about some of the missing details – like the design and limitations of the studies, the quality of evidence and harms of the interventions.
The article compares aromatase inhibitors to tamoxifen, which in the past has been the go-to drug for treating breast cancer. More recently, many women have been on tamoxifen for a few years and then on an aromatase inhibitor in addition to other treatments.The piece could have mentioned other treatments like chemotherapy, radiation therapy and surgery, such as ovary removal in women with very high risks for breast cancer. Often all of these approaches are used together in the course of treatment.
The article describes aromatase inhibitors as “a class of inexpensive, existing generics” and bisphosphonates as drugs typically used to treat osteoporosis.
In the first paragraph, the story makes it sound as if aromatase inhibitors were just discovered by researchers: “…and despite decades of research scientists have not been able to find anything comparable — until now.”
Researchers have known about aromatase inhibitors as an alternative to tamoxifen for at least a decade, something which the story later points out while describing the finding about aromatase inhibitors reducing mortality rates as a “surprise.” That might be overstating the case, since the original study says the evidence on this point is “uncertain” — suggesting that researchers were at least aware of the likelihood that these drugs might reduce death rates. We’ll give the benefit of the doubt, since the story does, eventually, get around to describing what’s new here and why — even if that description is somewhat questionable.
The Washington Post story relies on this news release a bit too much for our liking. The article’s lead quote from Paul Workman comes right from the press statement. And while the story acknowledges the source for that quote, it doesn’t alert readers that a second quote from Richard Gray is also pulled from the same news release. We can’t find evidence of any other original reporting, so we’ll flag the story here as Not Satisfactory.