This news release describes findings from a retrospective study on a blood test marketed for the early detection of cancer, in this case asbestos-related malignant mesothelioma. The test screens for the presence of the protein ENOX2 which researchers say exists only on the surface of a malignant cancer cell. The MorNuCo company is “elated to share” news about this “exciting” ONCOblot cancer marker blood test. To make sure you don’t miss the “exciting” point, the word is used three times in a release that is only nine sentences long. The release falls short on data and relies on fear-mongering to engage people worried about cancer risk because of exposure to asbestos. The release flogs the myth that early detection is always a good thing, thus raising false hopes.
To give powerful cancer treatments too soon, or even unnecessarily, does enormous harm, justified only if the potential benefit is clear. Not only does this study of the ONCOblot test fail to provide any evidence that early detection could provide a benefit to people exposed to asbestos (most of whom will never get mesothelioma), the retrospective study of a few people who have been diagnosed with mesothelioma cannot even demonstrate that the test could reliably predict who is likely to develop the disease, something that will have to be tested by prospective studies. The authors of the study noted the following in their conclusions:
“As with all biomarker studies, these observations require validation in a larger, independent cohort of patients and should include prospective as well as retrospective sampling.”
The news release failed to communicate that message.
It is notable that the company that sells the ONCOblot test (which can be used for early detection of 26 different cancers) doesn’t discuss its price. Some clinics list their test cost at $850 or $1000. It is not known how often testing would need to be repeated, if it is found to be useful for people exposed to asbestos.
The release states it’s “an exciting sign of progress in the cancer detection field” because researchers detected “two mesothelioma-specific ENOX2 protein transcript variants” in the blood serum of individuals exposed to asbestos “an average of 6.2 years” before they were clinically diagnosed.
Unfortunately, that’s the only attempt at quantification found in the release. There’s no mention of accuracy or sensitivity of the test, how many people were tested, or how long the study lasted.
It’s not enough to claim that the test is useful. The sponsors should provide some data on how it improves patient outcomes, not just sound an alarm in order to get individuals to take a test. The unfortunate reality is that patients diagnosed with stage 1 disease have a median survival of 21 months and those diagnosed at stage 4 have a median survival of 12 months. It remains to be seen if early detection provides any advantage and a large clinical trial will be needed to see if this is true.
While the ONCOblot test itself involves just drawing a blood sample, the results of the test could produce life-threatening harms if it overdiagnoses people. In other words, a person told they have tested positive for these biomarkers for mesothelioma might start treatment too early or unnecessarily. Mesothelioma treatment is notably damaging. Let’s put the study into perspective: there were a total of 32 people enrolled (17 with known malignant mesothelioma and 15 without). Hardly a large enough sample size on which to make the claims made. Although as we noted previously, early detection does offer a small survival advantage, it comes at a cost. The American Cancer Society notes that chemotherapy for mesothelioma can cause hair loss, mouth sores, loss of appetite, nausea and vomiting, diarrhea, increased risk of infections, easier bruising or bleeding, fatigue and long-lasting nerve damage. Radiation treatment also can cause many of the same problems, as well as lung damage. Combining radiation and chemotherapy often makes side effects worse.
It won’t be known whether starting treatment earlier offers any benefits to people at risk for mesothelioma, but it is certain that treating more people earlier would cause more treatment-related harm.
The release fails to mention how many people were included in the study: just 17 cancer patients and 15 people who were exposed to asbestos, but had not been diagnosed with cancer.
Although the release notes near the top that the study was “retrospective” (it looked back at stored blood samples taken from people exposed to asbestos, some who had been diagnosed with mesothelioma and some without known cancer), there is no mention of the long years of work needed before anyone will be able to say the test would be useful to people who have been exposed to asbestos. The journal article notes, “As with all biomarker studies, these observations require validation in a larger, independent cohort of patients and should include prospective as well as retrospective sampling.” The release should have noted that there is an important difference between finding that people who have cancer carry a certain biomarker and what people really want to know: that the biomarker can predict who will or won’t develop cancer, and even more importantly, whether starting treatment earlier makes any difference.
The news release emphasizes that “Malignant mesothelioma is an aggressive and almost uniformly fatal tumor caused primarily by exposure to asbestos.” However, it fails to note that not only is the disease rare (fewer than 3,000 cases per year in the United States, compared to more than 1.6 million cases and half a million deaths from all types of cancer), but most people exposed to asbestos never get mesothelioma. By implying that this test might be useful for anyone exposed to asbestos, the release sets the stage for a vast sales market of people, most of whom would never have to deal with mesothelioma, regardless of testing.
The news release states that the study was done by MorNuCo, the company that developed the ONCOblot test.
Another useful flag apparent to close readers is that the only person quoted in the release is not a researcher, but the company’s vice president of business development.
The release doesn’t mention any alternatives to the sponsor’s test. Some physicians believe that repeat chest CT scans are useful but this has not been verified in a clinical trial. The other alternative is basic primary care and watchfulness.
The news release notes that the ONCOblot test is FDA approved as a “Laboratory Developed Test.” It also includes a link to a company web site with more information on the test and how to get it.
The study backs up the characterization in the release that the results represent progress. Aside from the concerns mentioned above about implying the test will benefit patients, it is reasonable to say this test could be useful to researchers planning clinical trials of early treatment of mesothelioma.
There are only nine sentences in this release, yet there is room to boast of “exciting results,” “exciting sign of progress” and an “exciting new chapter” that the company is “elated to share.” While the study results may well be exciting to some researchers who are planning clinical trials of early treatments for rare mesothelioma cancers, there is nothing in this study that is truly “exciting” to people worried about their personal risk of developing mesothelioma. The release dangles false hope in front of people exposed to asbestos by pushing the popular myth that “early detection is widely considered the corner stone of an effective strategy to reduce cancer-related deaths”.
As this online primer summarizes, the triple stumbling blocks of lead time bias, length bias and overdiagnosis stand between this study and any real understanding of whether the news may ever be exciting to people worried about the consequences of exposure to asbestos.