This story on a new version of a gene test meant to help women with small breast tumors make treatment decisions provides good context, but it may have obscured a key limitation of the study.
Women diagnosed with small breast tumors known as ductal carcinoma in situ (DCIS) often have to make treatment decisions (including whether to add radiation therapy after surgery) without being certain about how much risk of a recurrence they really face. This story reports the results of a study that compared the genetic profile of tumors to the long-term cancer recurrence rate in a group of women. The story provides multiple views and useful context. However, most readers would likely assume that treatment decisions that incorporate these test results would offer better health outcomes. This story should have clearly stated that this study was not designed to answer that important question and that follow-up trials will be need to document the health effects of using this test.
Predicting risk is one thing. Knowing which treatment is most likely to produce the best outcome for a patient with a certain risk level is a different question. People often confuse the two; so news reports about risk assessment should take special care to point out that improving the ability to predict a patient’s future does not necessarily mean that you can change that future. Eric Winer’s cautious comment was an important part of the story – “he’d like to see additional studies before using the test to advise patients.”
The story reports that the existing version of this test costs over $4,000. It puts this cost in context by reporting that one of the researchers says some women could still save money if the test results prompt them to skip radiation therapy that can cost more than $21,000 (as well as exposing them to some risks and harms).
The story comes very close to meeting this criterion. It reports that this new test put 75 percent of women with this type of breast tumor into a low-risk category and that this information could support a decision to skip radiation therapy… and that the predictions are more accurate than what clinicians usually rely on today. It also notes that for this type of tumor (ductal carcinoma in situ or DCIS), radiation therapy has been shown to reduce the risk of a recurrence in the same breast, though it does not alter survival rates, thus adding valuable context to the decision that this new test is supposed to help women make.
However, this story fails to alert readers to a central point: the study did not examine whether this test actually improves outcomes for the patients. In other words, while the researchers can say women with certain test results are less likely to have another DCIS in the same breast, further trials would be needed to determine whether or not changing treatment based on the test results improves the outcomes the women care about, including reserving radiation therapy for those women most likely to benefit.
The story does include a comment from an independent expert who says more study is needed before making changes in usual treatment recommendations; but the story would have been even better if it had plainly and prominently stated that this study did not look at whether making treatment decisions based on the results of the new test makes a difference in the long-term health outcomes of women with this sort of DCIS.
Although the overall tone of the story presents this test as a clear advance in providing treatment guidance to women with DCIS, it does spell out that 5 percent of women classified by the test as “low-risk” still developed an invasive cancer within 10 years… and that over 80 percent of women labeled as “high-risk” did not develop an invasive breast cancer within 10 years. So readers are told that even with this test, physicians cannot predict the future with certainty.
Overall, the story does a good job of describing the study and putting the results in context. But as noted above, it fails to clearly alert readers to the fact that this study did not look at the effect of changing treatment recommendations based on the test results. Many readers may believe the study accomplished more than it actually did.
Also, this study was presented at a medical meeting. We like to see stories point out that meeting presentations are not subject to the same level of peer review as articles published in leading medical journals.
The story states near the top that this test is intended for the 45,000 women in the United States who are diagnosed each year with ductal carcinoma in situ (DCIS) in one of their breasts.
The story includes two independent experts who help put the study results in context. It also identifies one source (Shak) as an employee of the company that sells an earlier version of the test. The story would have been better if it had addressed the relationship between the lead researcher and the company, but since it appears (based on other reports) that Dr. Solin did not disclose any conflicts of interest, we will give the story a satisfactory rating on this criterion.
The story lays out the conundrum facing women diagnosed with DCIS: that clinicians generally can’t predict which women will ultimately develop invasive disease and which ones will be fine even with little or no treatment. As the story reports, these women and their physicians may decide to opt for conservative treatment, but that many choose to undergo radiation therapy in hopes of reducing the risk of a recurrence. Overall, the story describes this new test as being able to narrow the uncertainty somewhat.
The story reports that an earlier version of this test is already on the market and that the company plans to make the new version available within weeks. Since it appears that updating the way this test is used does not require regulatory approval, it seems adequate to rely on the company’s statement about its marketing plans.
As mentioned above, the story describes this new test as a refinement of an existing test.
The story does not appear to rely on a news release.