This news release touted a study that showed a blood test for a biomarker called AR-V7 can identify patients with aggressive prostate cancer who would benefit if they switched from hormone therapy to chemotherapy. The study, which observed 142 patients who received the test and subsequent treatment, was published in JAMA Oncology.
This news release from the test’s developer gave survival data and mentioned that several of its employees were study authors. It didn’t address costs or mention important caveats — such as the lack of statistical significance of that survival data or the fact that patients were not randomized. In fact, it’s unclear whether this test actually extends patients’ lives.
A blood test to inform treatment decisions for aggressive prostate cancer could save money, spare patients from ineffective and toxic treatments, and extend lives. The cost of treating prostate cancer in the U.S. has been projected to exceed $15 billion by the year 2020, up from $11.9 billion in 2010, making it one of the most expensive cancers to treat.
The $3,980 price tag for this test wasn’t mentioned. The release notes that the Centers for Medicare and Medicaid Services (CMS) are currently evaluating whether the test will be covered in the future. In other words, it is not currently covered by Medicare. We aren’t told whether other insurers cover the test.
There was also no information about the costs of the treatments involved. Hormone therapies for prostate cancer are especially pricey, running more than $10,000 a month, according to published reports.
The news release stated that “The results were that patients positive for AR-V7 who were treated with taxane-based chemotherapy had superior overall survival (OS) relative to those treated with ARSI therapy (median OS, 14.3 vs. 7.3 months). Importantly, patients negative for AR-V7 who were treated with ARSi therapy had superior OS relative to those treated with taxanes (media OS, 19.8 vs. 12.8 months).”
But in the published paper, the results state: “The median survival of patients negative for AR-V7 was 19.8 months for those treated with an ARS inhibitor and 12.8 months for those treated with a taxane (hazard ratio, 1.67; 95% CI, 1.00-2.81; P = .05) (Figure 2A). In contrast, for patients with AR-V7–positive CTCs, those receiving taxanes had longer observed median survival times relative to those treated with ARS inhibitors (14.3 vs 7.3 months; hazard ratio, 0.62; 95% CI, 0.28-1.39; P = .25)”
So, for those with a negative test, the androgent (hormone) therapy had improved survival. However, for those with a positive test, there was a trend but the sample size was very small and the p-value was >0.05 (indicating weak evidence), so the results could have been due to chance. As such, we rate this unsatisfactory since the piece provides misleading information for part of the results.
No harms are mentioned. The potential risks of blood tests are minor, although the potential of a false result could have serious consequences if it leads to a patient getting an ineffective or potentially harmful treatment.
The release does a nice job describing the study parameters. But limitations of this research weren’t mentioned. For example, the difference in survival wasn’t statistically significant and could have been due to chance. Also, patients were not randomly assigned to a treatment based on the test, so it’s not clear whether differences in their outcomes resulted from the treatments or some other factor.
Moreover, they applied a risk score after the fact to better define the test results. This after the event analysis is not clearly described.
There is no disease-mongering here. The release states that about 50,000 men a year in the U.S. will face treatment decisions for which a test like this might prove useful.
The release states that Epic Sciences “designed” the test and that different research institutions led research studies on the blood test. The release could have been much more transparent about who funded the research.
The study didn’t mention all of the researchers’ industry ties, but it did mention that some of the researchers are employed by Epic Sciences, the developer of the test.
The news release does not mention other ways to determine that hormone therapy isn’t working. There may not be any comparable alternatives.
There are ways to assess prognosis and these could be considered alternatives. The paper states: “To investigate confounding factors that might influence the decision to administer an ARS inhibitor or taxane, a risk score was developed based on a prognostic model developed from our training cohort22 and applied to the current cohort.” This suggests the authors developed a comparator to their test. As such, it would have been reasonable to comment that there may be other ways to assess risk. The advantage of this one is that it targets what treatment should be given beyond the overall risk itself.
The release said the test, Oncotype DX AR-V7 Nucleus Detect, is commercially available in the U.S. However, it didn’t clarify whether it’s FDA-approved for testing patients with prostate cancer. Because the test is not currently covered by Medicare, for those patients the test may only be available as an out-of-pocket expense.
The release called this “one of the first studies to validate that a liquid biopsy test can predict therapeutic response and demonstrate a survival benefit.”
Given the lack of statistically significant results, it was premature for this news release to state that the blood test “predicts treatment response and survival” for patients with metastatic prostate cancer. The statement that the test could be “as valuable to a patient’s outcome as a blockbuster cancer drug” also seemed exaggerated.
It would have been appropriate to note that only by studying the test as part of the treatment program itself can we be certain that it led to improved outcomes.