Fewer PSA screenings linked to increases in advanced prostate cancer diagnoses
Click Here to Manage Email Alerts
U.S. states with larger declines in PSA screening from 2002 to 2016 also had larger increases in incidence of metastatic prostate cancer at diagnosis, according to a presentation at Genitourinary Cancers Symposium.
“Our data strengthen the epidemiologic evidence supporting that the rise in metastatic prostate cancer at diagnosis may at least in part be explained by the decline in PSA screening, since states with larger declines in PSA screening tended to have larger increases in metastatic disease at diagnosis,” Vidit Sharma, MD, health services fellow in urologic oncology at University of California, Los Angeles, told Healio.
Sharma and colleagues had noticed an increase in metastatic prostate cancer at diagnosis in recent years and that the timing coincided with U.S. Preventive Services Task Force recommendations against PSA screening in 2008 and 2012. But the association of these factors had never been tested.
Researchers used 2002 to 2016 data from the North American Association of Central Cancer Registries to obtain age-adjusted incidences of metastatic prostate cancer at diagnosis, and they used the Behavioral Risk Factor Surveillance System to obtain PSA screening estimates for each state. They then combined these screening and metastasis data as a multipanel time series, which they analyzed using a random-effects linear regression model with random effects at the state level.
“We had to break down data by looking at the individual states because if we looked at national data, there simply weren’t enough years since the USPSTF recommendations to adequately associate changes in PSA screening with metastatic disease,” Sharma said. “For instance, we considered using an interrupted time series approach with national data, but there were just too few data points (one per year available) since the new recommendations. Thus, we had the idea to examine each individual state’s PSA screening and metastatic prostate cancer at diagnosis data points over time. We reasoned that there would be variation in changes in PSA screening over time between states, and thus we could test if this variation in PSA screening was associated with variation in metastatic disease at diagnosis between states and over time.”
In his presentation, Sharma showed those variations and the correlation between reductions in PSA testing and increased prostate cancer diagnoses.
“[Prior to 2010], there was signification variation between states in PSA screening,” Sharma said. “However, after 2010, there was a significant decrease in PSA screening across states. Similarly, [when looking at] incidence of age-adjusted metastatic prostate cancer at diagnosis per 100,000 men by state, once again, there is a significant variation between states but after 2010, the incidents rise across states.”
The percentage of men who reported ever receiving PSA screening ranged from 40.1% to 70.3%, and the age-adjusted incidence of metastatic prostate cancer at diagnosis ranged from 3.3 to 14.3 per 100,000, according to Sharma and colleagues.
Results showed the average percentage of men aged 40 years or older who underwent PSA screening for prostate cancer decreased from 61.8% in 2008 to 50.5% in 2016 and, at the same time, the average number of men diagnosed with metastatic prostate cancer (after adjusting for age) increased from 6.4 to 9 per 100,000 men (P < .001 for both).
The random-effects linear regression model showed the longitudinal reductions in PSA screening among states correlated with increased metastatic prostate cancer diagnoses (regression coefficient per 100,000 men, 14.9, 95% CI, 12.3-17.5). The variations in PSA screening explained 27% of the longitudinal variation in metastatic prostate cancer within states, Sharma and colleagues wrote.
“Broadly speaking, our study argues against a unilateral ‘never PSA screen’ approach, given the association with the decline in PSA screening and the rise in metastatic prostate cancer at diagnosis at the state level,” Sharma told Healio. “Instead, we would recommend a shared decision-making approach with men who are candidates for PSA screening to help them make an informed decision.”
That includes the approach outlined in the updated USPSTF guideline, issued in 2018, which recommended screening as an option for men aged 55 to 69 years after a discussion with their clinician about the possible harms and benefits.
“Our study highlights that methods are also needed to optimize PSA screening practices, such that the harms of overdiagnosis and overtreatment of low-risk prostate cancer are balanced with the harms of missing more aggressive disease that can metastasize,” Sharma added.
Perspective
Back to Top
Bobby Liaw, MD
We have known that incidence of prostate cancer — particularly metastatic disease — was on the rise, and there has been some concern that this had to do with the changes to USPSTF recommendations regarding screening. Although this has been on everybody’s radar, we have never been able to dig down into the data to show a statistical association to link practice changes in PSA screening with increased incidence of metastatic prostate cancer.
Consequently, this is an important public health study, because it shows that even small changes in screening practices potentially can make a big difference. When we talk about screening, we should be focused on finding people who have disease in a state where we can potentially cure them. If we find them too late and the disease is not curable, that is a failure of the screening program.
In 2018, the USPSTF backtracked a bit on its recommendations, moving away from a very strong stance of “don’t screen at all” to more of an informed decision-making model for men of a certain age range. Going forward, hopefully we will start to see a leveling out or a reduction in incidence of metastatic prostate cancer, and we’ll begin to find more cases of localized or earlier-stage disease when it is potentially curable.
Screening may only go so far in explaining the longitudinal variation in metastatic prostate cancer diagnoses. There are other factors, such as patients’ access to primary care, insurance status, and socioeconomic or educational status. Also, is there one group of people in whom metastatic prostate cancer was more commonly found? We don’t have the granular data that would help identify some of those patient subcategories, but that is important to understand from a public health policy standpoint so we can determine whether there are specific communities we should focus on and how we can help those most at risk.
Also, stage IV disease is separated into stage IVA and stage IVB. Stage IVA disease can still be considered curable with hormone therapy, radiation therapy or multimodality therapy. Whether patients with stage IVA disease were included in this definition of metastatic disease would be interesting to understand.
PSA is a very sensitive test, and one of the criticisms when it was initially rolled out was that it was picking up cases of very low-grade, slow-growing, indolent disease that we now know often can be managed with active surveillance. We need to find a much happier medium between not screening at all — which can result in a lot of cases falling through the cracks — and being too cavalier and screening everybody without thinking, because then we’re in a position where we overdiagnose patients and put them in harm’s way.
Perspective
Back to Top
Clayton S. Lau, MD
Prostate cancer screening using PSA testing and digital rectal examination has been controversial due to the possibility of overdiagnosis of clinically insignificant disease and the morbidity of treatment and psychosocial distress. Despite the European Randomized Study of Screening for Prostate Cancer trial showing an improvement of mortality and cancer–related metastasis, the USPSTF did not recommend PSA screening both in 2008 and 2012. Many physicians who care for these men have been concerned about this and the possibility of finding more men with locally advanced and metastatic disease.
Using random-effects linear regression models, the longitudinal reductions in PSA screening across states appeared associated with increased rates of de novo metastatic prostate cancer, which suggests that states with larger declines in PSA screening had larger increases in incidence of metastatic disease at diagnosis. It would be great to evaluate the differences with race/ethnicity and socioeconomic status in this cohort.
It is imperative that physicians educate patients about the benefits and potential harms of screening, so they can make an informed decision about testing. This includes compliance with timely follow-up and possibly more evaluation and testing.
Reference:
Schröder FH, et al. Lancet. 2014;doi:10.1016/S0140-6736(14)60525-0.
Collapse
Sharma V, et al. Abstract 228. Presented at: Genitourinary Cancers Symposium (virtual meeting); Feb. 11-13, 2021.
Click Here to Manage Email Alerts