Study shows decrease across disease stages after USPSTF recommendation
The number of new cases fell by 12% in the first month after the task force released its draft guideline in October 2011. Incidence continued to decline by 1.8% a month relative to baseline values until the end of 2012, roughly 6 months after the USPSTF released the final guideline.
An analysis by disease stage at diagnosis showed that the number of new diagnoses of intermediate- and high-risk prostate cancer also decreased, which could have potentially ominous implications if the trend persists, Daniel A. Barocas, MD, of Vanderbilt University in Nashville, Tenn., and co-authors concluded in a report in the December issue of the Journal of Urology.
"On the good side of the state, 'If you don't look for it, you won't find it,' you won't find those low-risk tumors that don't need to be treated," Barocas told MedPage Today. "Those men, when we find those tumors, we are sort of subjecting them to the harms of treatment without an anticipated benefit."
"The bad side: If you don't look for intermediate- and high-risk cancers, you don't find those. Those are the ones that we have the potential to treat at an early stage before they spread and cause painful bony metastases and cancer death. There's little doubt that screening has improved mortality from prostate cancer. The doubt is whether the net benefit is sufficient. Are we doing so much harm by treating people who didn't need to be treated that it overrides how we've benefited the people who need treatment?"
The solution is simple, he said: "When you find low-risk disease, you don't treat it."
In the same issue of the journal, two urologic oncologists defended opposing views on whether the USPSTF recommendation has led to an increased diagnosis of advanced cases of prostate cancer.
Michael J. Barry, MD, of Massachusetts General Hospital Cancer Center in Boston, argued that no nationally representative data have emerged since 2012 to support the view that advanced prostate cancer incidence has increased. Joel B. Nelson, MD, of the University of Pittsburgh, pointed out that the incidence of advanced prostate cancer at diagnosis decreased dramatically with widespread availability of PSA testing, making an increase in advanced disease without routine screening an intuitive occurrence.
Accumulating Data
The study reported by Barocas and colleagues followed two similar report published last week in JAMA, both documenting short-term declines in the incidence of early prostate cancer since the USPSTF announced its recommendation.
Prostate cancer mortality has decreased by about 40% since the introduction of prostate cancer screening with PSA in the late 1980s. As much as 70% of the decrease might be attributed to use of screening PSA tests.
The number needed to screen and the number needed to treat to prevent one prostate cancer death are similar to those associated with mammography and fecal occult blood tests, Barocas and co-authors noted. However, prostate cancer treatment -- surgery or radiation therapy -- carries substantial morbidity. That, combined with the uncertainty surrounding the benefits of PSA-based screening, drove the USPSTF decision against routine PSA testing.
Limited data have accumulated with respect to the task force's decision on prostate cancer incidence. To address the issue, Barocas and colleagues analyzed data from the National Cancer Data Base for 2010 to 2012, capturing a year or more of data before and after the release of the USPSTF draft recommendation.
Investigators identified all men with newly diagnosed prostate cancer entered into the database during the study period. Information was collected on a variety of clinical characteristics, including disease stage, biopsy Gleason score, PSA level at diagnosis, and D'Amico risk group.
Prior to the USPSTF draft recommendation, the number of monthly prostate cancer diagnoses ranged from 9,442 to 12,021, and the number of cases increased by about 45 (0.4%) per month. Immediately after the USPSTF released the draft recommendation, the number of newly diagnosed prostate cancers decreased by 1,373, which translated into a 12.2% decreases compared with the month preceding the announcement (P<0.01).
The decline in prostate cancer incidence continued without interruption for the next year, and the number of new cases was 27.9% lower in October 2012 compared with October 2011. In contrast, the incidence of colon cancer -- for which no new clinical recommendations were issued during the same time period -- did not change.
The overall decline in prostate cancer incidence differed by stage at diagnosis. In the first month after the draft recommendation was announced, the incidence decreased by 16.9% for low-risk disease, 12.9% for intermediate-risk disease, 10.1% for high-risk disease and 2.7% for nonlocalized disease. Thereafter, the corresponding monthly changes by risk status were -2.7%, -1.9%, -1.4%, and +0.1% (P<0.01).
The decline in prostate cancer incidence from October 2011 to October 2012 was 37.9% for low-risk prostate cancer, 28.1% for intermediate risk, 23.1% for high risk, and 1.1% for nonlocalized cancer.
"While some effects of this guideline may be beneficial in terms of decreasing harms of overdiagnosis and overtreatment, the reduction in intermediate and high-risk cancer diagnoses raises concern for delayed diagnoses of important cancers, which are associated with inferior cancer outcomes," the authors concluded. "Future research should focus on prostate cancer screening paradigms that minimize harms and maximize the potential benefits of screening as well as account for individual patient risk factors and preferences."
More Advanced Disease: No
In his contribution to the pro/con discussion of the USPSTF recommendation and advanced disease, Barry noted that the stage migration to lower-risk disease attributed to PSA screening might not reflect a true reduction in the incidence of intermediate- and high-risk disease. Cancer registries record only the stage at initial diagnosis. If a cancer subsequently becomes more advanced, the case is not reported as advanced disease.
"Therefore, if screening results in a 'stage shift' toward more localized disease, it will appear to result in lower rates of advanced disease, even if 'true' population-based rates of advanced disease are unchanged," he said.
"To my knowledge, no nationally representative data are available on the 'true' population-based rates of advanced-stage prostate cancer since the 2012 recommendations," Barry continued.
The data reported by Barocas et al. "do not allow calculation of population-based rates .... The overall number of incident diagnoses of prostate cancer decreased an estimated 12% in October 2011, with an ongoing reduction of 1.8% per month through the end of 2012. However, the number of incident cases of nonlocalized disease did not change."
Barry concluded his argument with a "thought experiment" to provide perspective for discussions about harms versus benefits. "Imagine a policy of mandatory universal prophylactic radical prostatectomy, say at age 55 years. I have little doubt implementation of such a policy would presage a dramatic decrease in the incidence of advanced disease, followed by a dramatic reduction in prostate cancer mortality. If that policy were then rescinded, the trends would eventually reverse.
"Viewed in isolation, the observed time trends in advanced disease and mortality would strongly support such a policy, despite being the ultimate overtreatment. However, when weighed against rates of surgical complications and side effects, a quite different picture would emerge. Public health policy, whether for screening or prevention, must consider both sides of the coin."
More Advanced Disease: Yes
Nelson countered with a simple statement: "The answer is obvious."
"In the population, prostate carcinogenesis is constant, progression of prostate cancer is constant, and the diagnosis of prostate cancer cannot be delayed. Allowing a disease to progress before diagnosis means that it will be more advanced compared to earlier detection. This is a central tenet supporting all disease detection strategies and not just those for prostate cancer."
In contrast to Barry's population-wide viewpoint, Nelson brought the argument down to the level of the urologist/urologic oncologist.
"We treat individual patients, not populations, and here the logic is considerably more nuanced. This is what we know. First, PSA-based screening is more likely to detect indolent disease. This is in no one's best interest but it is inevitable. Those fish should be released. Embrace active surveillance.
"Second, screening more often fails to detect the most aggressive, and ultimately lethal, prostate cancers at a time when treatment will change the natural history: Screening does not find those who need it most. Some have suggested this is a reason to abandon screening. This non sequitur is true, true and unrelated. We need to offer treatment, with all its attendant morbidity, to those who need or want to avoid the inevitable and more profound suffering associated with untreated progressive disease."
Finally, Nelson said, "a delay in diagnosis means that disease will be more advanced and our treatments will be less effective. If they are willing to accept this fate -- and older, frailer men should -- these patients can be assured that not screening will spare them, in the short term, unnecessary diagnosis and treatment. We cannot be so sanguine about the long-term prospects. Many men happily accept a permanently impaired functional state knowing theirs was not the cancer that got away."
Không có nhận xét nào:
Đăng nhận xét