This news release touts a high-definition scan called multiparametric magnetic resonance imaging (MS-MRI), shown in studies to increase the accuracy of prostate cancer detection. The scan is used to identify suspicious lesions in the prostate that can be biopsied for cancer. A study at the Center for Cancer Research (National Cancer Institute, NCI) in Bethesda, Md., showed that targeted MRI-guided biopsies found more cases of high-risk prostate cancer while detecting fewer low-risk cancers compared with standard biopsies.
The news release presents this data but does not mention cost, associated risks, and the fact that more research is needed on whether targeted biopsies ultimately lead to fewer deaths. The news release hypes the procedure as having a “more than 90 percent success rate” without explaining what that means. Moreover, it stokes fear by selectively describing cancer cases discovered by MRI and stating that without advanced imaging cancer may not be found until it’s “too late.” Reading this, one might conclude that all men with an elevated prostate-specific antigen (PSA) level would do well to hop on the next plane to Richmond.
The American Cancer Society estimates that 26,120 men will die this year from prostate cancer, one of the most common cancers in men and the second most deadly, after lung cancer. The current standard for detecting prostate cancer, called the transrectal ultrasound-guided (TRUS) biopsy, has been faulted for diagnosing too many low-risk cancers, leading to unnecessary treatment, and for missing aggressive tumors that can occur in the anterior portion of the prostate. MRI-guided biopsies have been gaining traction as a more accurate alternative, albeit a pricier one.
There are many challenges with screening for prostate cancer. The PSA test that is used as an initial screening test can lead to both false-negative and false-positive results. This release highlights examples of an elevated PSA test where it was difficult to identify a cancer. Not mentioned in the release is the fact that PSA tests can be elevated and the patient really doesn’t have prostate cancer. The higher the PSA test result, the more likely that it is due to cancer. Though a level of 4 is often used, lower levels have been shown to be associated with prostate cancer. The focus here is on levels above 4. Between 4 and 10, many patients will not have cancer or the cancer may be of a less dangerous type that may not require active treatment. Generally, when the level is above 10, cancer is more likely and more likely to be a problem. It has long been known that for patients with elevated PSA tests with negative results on standard “blind” biopsies may still harbor a tumor. This has led to repeat biopsies as described in the release. For some, the cancer may appear to be in the anterior portion of the prostate — an area that is hard to biopsy using standard techniques. MRI is becoming an increasingly common test when the PSA is elevated and the biopsy is normal. But there is no standard for what level of PSA at which an MRI should be performed. Some would argue that the level of 5.3 that one patient had wouldn’t justify such a search, whereas very high levels would justify the effort and expense. But the radiologist described in this release would go even further stating that guided biopsies with MRI should become the norm.
The key thing for readers to know is that neither focused MRIs (after negative standard biopsies) or initial screening with an MRI have led to better outcomes. By “better outcomes” we mean less metastatic prostate cancer and fewer deaths from prostate cancer. Without such data, we can’t be certain whether the additional cost of such evaluations is justified.
MRIs cost significantly more than ultrasounds, but that fact isn’t mentioned. According to a story in the Wall Street Journal, an MRI can cost from $600 to more than $2,000, although Medicare and many private insurers typically cover some or all of that cost. Cost is all the more significant if, as VCU’s director of oncologic and prostate imaging speculates in the news release, MRIs are eventually used like mammograms, with men getting baseline MRIs followed by repeated periodic scans if warranted by bloodwork and family history.
Some mention of cost, relative cost or cost-effectiveness was warranted.
The news release offers some solid data along with confusing statements. It says MRIs and targeted biopsies have “provided relief” for patients who have not been able to get an accurate diagnosis, citing data from the 2015 NCI study that showed targeted MRI biopsies detected 17 percent fewer low-risk tumors than standard biopsies. It says MRI “is 30 percent more accurate” in diagnosing high-risk, aggressive cancers than a standard biopsy, a misstatement. The study actually says MRI-guided biopsies “diagnosed 30 percent more high-risk cancers.”
The news release tells readers that VCU performs about 1,200 MP-MRI-guided biopsies annually, which is useful information. But it also claims a “more than 90 percent success rate” without ever explaining what that means or where the number comes from. It even quotes a patient who says he compared “success rates” at various medical facilities. Where did the patient get his figures, and what do they mean? We aren’t told.
Ideally one would like to read the sensitivity and specificity of the test — the parameters that make up a test’s accuracy. Sensitivity is the test’s ability to identify those who have the disease, and specificity refers to the test’s ability to exclude those who do not have the disease.
Another statement implies that VCU’s procedures are superior to “most” hospitals because they have a single professional reading results and performing biopsies. There’s no data to back this up.
The news release includes two glowing anecdotes involving patients whose cancers were promptly detected at VCU and safely treated. Were there other patients who traveled to VCU only to come away with an inconclusive finding, a negative finding, or a fatal tumor?
The news release does not mention possible side effects of biopsies such as serious infections, pain, and bleeding. There’s also no mention of the harms that can come from a false-positive test that can lead to more tests, mental anguish and unnecessary procedures.
Also, it does not mention that MRIs may be inappropriate for patients with kidney disease or implanted devices such as pacemakers, stents and inner ear implants, due to the strong magnetic field used.
The news release does not mention limitations of the 2015 NCI study it referenced, which said more research is needed to determine whether targeted biopsies actually lead to meaningful clinical improvements such as preventing recurrence of disease and stemming prostate cancer mortality. Further, authors said most of the 1003 patients in the study had one or more more previous biopsies, and the sample consisted of patients referred to a single institution, which could have introduced selection bias.
To be specific, one would need to follow and compare patients treated with conventional care and those that are followed using MRI. One would look at diagnoses, but what we’d really want to see is less diagnosis of metastatic disease and lower prostate cancer related death. We know that such studies are challenging because prostate cancer is a slow growing tumor and that there are effective treatments even for advanced prostate cancer. So it could take years of follow-up to show a real benefit.
While prostate cancer is serious, this news release seems calculated to instill fear. It omits some important facts, such as that PSA levels often fluctuate and can be elevated for reasons that have nothing to do with prostate cancer. In fact, only about 25 percent of men who have a biopsy due to an elevated PSA level actually have prostate cancer, according to cancer.gov.
Relating the story of a patient whose cancer was diagnosed with an MRI after standard biopsies were inconclusive, the news release says other patients who continue to “hit a brick wall with negative results from TRUS biopsies yet persistently high PSA levels may not be as fortunate.” A clinician is quoted saying: “If you do not have this kind of advanced imaging, sometimes a patient runs into a situation when a tumor can’t be found and treated because it’s too late — it has already metastasized. He ends up diagnosed with cancer only after the tumor has spread to his bone or lymph nodes years later.”
The statement, “When caught early, prostate cancer is highly treatable,” is also troubling. In fact, prostate cancer does not always need to be treated, and treatment can result in long-term sexual side effects as well as urinary and bowel problems.
To avoid disease mongering, the release could have clarified that an elevated PSA test does not always identify cancer. Men can have elevated levels without cancer. Some of those with an elevated PSA who undergo an MRI will have a negative test — just like a negative biopsy.
The release doesn’t name a funding source, but it’s not based on one study. It focuses heavily on selected outcomes and patient anecdotes from a single treatment center, the VCU Massey Cancer Center.
There are no apparent undisclosed conflicts of interest or funding sources related to VCU. However, two authors of the NCI’s Center for Cancer Research study (briefly referenced in the release) had reported holding a patent related to the MRI biopsy platform, and a third reported holding multiple patents in the field. That research was supported by the National Institutes of Health, the National Cancer Institute, the Center for Cancer Research, and Center for Interventional Oncology. The NIH, Philips Healthcare, and devicemaker Invivo Corp. have a cooperative research and development agreement.
The news release offers comparisons between MRIs and the current standard of ultrasound-guided biopsies.
One option not mentioned is foregoing a PSA test. Indeed, that is the recommendation of the U.S. Preventive Services Task Force.
It’s also worth noting that researchers are investigating ways to improve PSA tests to better distinguish between potentially lethal cancerous tumors and benign conditions, according to the NCI. Better PSA tests could result in fewer invasive and costly biopsies.
The news release states that “VCU Medical Center is one of the major teaching hospitals in the country doing advanced multiparametric MRI.” That appears to be accurate.
It would have helped readers not in the VCU vicinity if the release had included more information about where tests could be done. Is it generally available at major teaching hospitals? If not, how many have it?
The news release does not overstate the novelty of this procedure.
We address examples of unjustifiable language in the quality of evidence and disease-mongering sections. There are other troublesome statements, such as that a urologist at another institution “was ready to throw in the towel” after a patient’s three inconclusive biopsies, which makes that clinician sound either uncaring or incompetent.
The statement: “Other patients who continue to hit a brick wall with negative results from TRUS biopsies yet persistently high PSA levels may not be as fortunate.” This implies that not trying hard enough may prove fatal for patients — something that is not true in terms of available evidence.