A test of merging MRI scans of a patient’s prostate with ultrasound-guided biopsies shows that in some cases the probes hit the target more often. But while this story paints a picture of a dramatic step forward, it fails to tell readers that four out of five core samples still came up empty and that the MRI technique failed to find any cancer in many men who were shown to have cancer using the standard biopsy procedure.
The story uses the phrase “watchful waiting” to describe men who have elevated PSA values, but negative biopsies. The term actually applies to men with diagnosed cancers who are “watched” for signs of cancer progression–and then receive treatment only to relieve symptoms. The comment that a biopsy is the only way to look at “the tumor cells and decide how dangerous the cancer is” is confusing. For men with an elevated PSA, the biopsy is the only way to determine whether they HAVE cancer. For men with low-risk cancers who are undergoing active surveillance (such as the subjects in the study), the biopsy is a helpful test (along with PSA) to determine whether the cancer is becoming more dangerous. The US Preventive Services Task Force advised against routine screening because they concluded that the net benefit (prostate cancer deaths prevented) was outweighed by the harms (overdiagnosis, treatment complications).
We do applaud the story for at least including some important context about fundamental doubts about the overall value of prostate cancer testing and treatment.
Stories about experimental techniques need to be clear about what researchers actually tested and whether the specific results can be said to show that a new technique is really “better” in terms of giving men and their doctors solid information that helps them make health care decisions.
Standard biopsies have a false negative rate of up to 20%. However, cancers found after repeated biopsies (following an abnormal PSA test)–the patient in the story is an unfortunate exception–are often non-aggressive. The value of finding microscopic cancers earlier through this new technology is probably limited–the recent PIVOT trial showed that men with screen-detected cancers do not benefit from treatment, at least through 10 years. This imaging technology, though hard to discern from the story, is probably most useful for men with active surveillance. These men have low-risk cancers and are in a monitoring protocol that involves repeating biopsies every 1 to 2 years. If the biopsies show cancer becoming more aggressive, the men will be offered active treatment. The technology, if it more accurately demonstrates that the cancer is becoming aggressive, would lead to earlier treatment that could improve outcomes. However, there is currently no convincing evidence supporting that hypothesis.
The story is confusing because it muddles screening and active surveillance. The patient is used as an example of why the new imaging technology is beneficial–it found cancer that had previously been missed during biopsies (incidentally, biopsies performed under general anesthesia are not standard; for most men biopsy is an office procedure. When biopsy is performed under anesthesia the number of prostate samples is about 10-fold higher than with standard). The fact that the diagnosis was made when the PSA was 18–and had already spread to the seminal vesicles–implies that the cancer was aggressive and probably would have been found by standard biopsies at that time. Finding the cancer at this relatively advanced stage also means that surgery alone cannot cure the cancer–the patient will need additional treatment. The actual benefit for cancer diagnosis would be if the technology identified aggressive cancers before they had progressed–which was not the case with this patient. The other problematic issue is that 106 of the patients were under active surveillance–they already were known to have cancer–and were being monitored because they would attempt curative therapy (radiation or surgery) if their cancer was progressing. Therefore, the statistics on cancer detection are meaningless–finding cancer in a man known to have cancer is not a benefit. What would be important is if the technology could guide biopsies to find areas of cancer that showed progression–indicating that treatment is necessary. The article does not address this issue.
There is no discussion of how much the MRI scans cost or how they might affect the overall cost of testing or treatment.
We welcome the space devoted to discussing questions and uncertainties about the effectiveness of prostate cancer screening and treatment, but the headline promising “better prostate cancer testing” and the way the story casts this experimental approach as solving shortcomings of standard techniques go far beyond what this trial actually demonstrated. The story also implies that the featured patient is likely to have a better treatment result because of the experimental MRI technique, but that is neither known nor was it part of this study.
Readers weren’t told that four out of five of the MRI-guided probes came up empty. That yield was an improvement over systemic (what the story calls “blind”) probes, but nowhere near the nearly unerring aim implied by the lead paragraphs. The story would have put the findings in better context if it had included cautionary statements from editorials published alongside the research article, for instance that the potential clinical uses of this MRI technique remain “speculative” and that “significant hurdles remain to broad adoption.”
MRI is generally a safe technology and the researchers did not report any specific problems caused by the experimental technique they used, but the story failed to address the potential harms from tests that provide imperfect or useless information. The researchers specifically cautioned that “some significant tumors may be missed.” So a test promoted as being “better” could leave patients with a false sense of security. The story also did not tell readers about infections, bleeding or other problems that may occur with any biopsy procedure.
With the increasing number of biopsies being performed (from 6 to 12 and now even more), the risks of complications have increased. Hospitalization rates for serious infections can be up to 4% and some studies have found post-biopsy mortality rates as high as 0.3%. The bigger issue is that the article implies that the imaging technology is valuable for finding cancers when standard biopsies are negative. However, if the cancers are that hard to find, then they are probably at the indolent end of the spectrum. The article does note the 1 million+ over-diagnosed cases and the ACS warnings, but doesn’t make the connection that using this technology for screening could exacerbate the problem.
The story presents the MRI targeting technique as better than standard biopsies, without clearly noting that this study looked only at very narrow measures of test performance… and did not look at whether or not the MRI technique improves the success of treatment or the ultimate health outcomes for patients.
The researchers reported that the biopsy cores guided by the MRI technique were three times as likely to contain cancer as those obtained with the standard approach guided only by ultrasound, but still only 21% of the MRI guided biopsy cores contained cancer; that is, four out of five came up empty. Also, of the 84 men found to have prostate cancer, 69 had cancer cells in cores done using the standard technique. While, there were 15 men who had cancer cells in MRI-guided biopsy samples only, there were 31 who had cancer in standard-technique biopsy samples only.
The details show a much more complex and limited result than the story reports.
Finding tumors in 53% of 171 men is not a meaningful statistic because it doesn’t account for the fact that 106 were under active surveillance–they were already known to have cancer. The story does not report how the new imaging technology compares to the standard techniques. The value of the new technology would be in identifying more aggressive cancers–but the story only cites the author as stating that tumors appearing dangerous on MRI were also dangerous on biopsy–and provides no quantitative data.
Mixed bag here.
The story included some important caveats about the overall value of prostate cancer testing and treatment. It appropriately describes the types of patients who might be candidates for this sort of testing.
Room for improvement:
The story really doesn’t explain the concept of active surveillance–just saying that men were under observation for slow-growing tumors. The comment that men with early-stage cancers can choose from a range of active treatments (e.g., surgery, radiation), completely misses the point that active surveillance is a recommended strategy for the low-risk cancers (and indeed was the treatment strategy being employed by the men “under observation”). Not all men need to be treated.
Because that important point wasn’t emphasized or made clear, we give an unsatisfactory score here.
It appears that the only interview done for this story was with a patient. The researcher comments are merely copied from the journal article. There are no comments from independent sources that relate to this specific study. There is no discussion of potential conflicts of interest, though it should be noted that the research was funded by a National Cancer Institute grant and foundation support.
The story presents this MRI technique as better than standard prostate biopsies in a way that overstates the actual results of the study.
Describing standard biopsies as “blind” is misleading. Biopsies will be targeted towards abnormalities felt during a digital rectal examination or abnormal areas seen on ultrasound–this is not guessing. Only in the absence of these findings would the biopsies be “blind” — which actually means that urologists systematically obtain samples from the prostate.
The story does not discuss availability of this MRI technique. It fails to alert readers that the researchers and authors of editorial comments in the journal warn about the high level of specific training needed. One editorial writer said several steps have to be done just right for this MRI technique to work and that, “There are potential errors in each of these steps.”
The story does not give readers an accurate sense of how many hurdles need to be cleared before this sort of MRI technique could be offered in general clinical practice.
One of the editorial writers noted that his clinic has been testing a similar technique since 2005. The story does not tell readers that this latest study is part of a much larger research enterprise that has been underway for some time.
The quote taken from the research article and the interview with a patient indicate this story does not merely rely on a news release.