This article is more than 5 years old. Information may no longer be current.
Understanding of prostate cancer risk factors may be flawed due to study detection bias
Evidence from observational studies that indicate certain factors are associated with prostate cancer risk may be flawed due to detection bias, according to an analysis of results from two large randomized clinical trials.
Risk factors for prostate cancer are often identified in epidemiologic studies and then implemented population-wide without confirmatory studies. However, challenges of epidemiologic studies include a high prevalence of biopsy-detectable disease in aging men, and that prostate cancer is typically asymptomatic until metastatic.
“As a result, if a risk factor, even one unassociated with prostate cancer, is incorporated into clinical practice, it will be found to increase cancer risk because men with the risk factor will be more likely to undergo screening; thereafter, screen-positive men will be more likely to be recommended for and undergo biopsy,” Catherine M. Tangen, PhD, member of the public health sciences division at Fred Hutchinson Cancer Research Center, and colleagues wrote.
“Ultimately, men with the risk factor will then be more likely to be diagnosed with prostate cancer.”
Tangen and colleagues used data from the Prostate Cancer Prevention Trial (PCPT) and the Selenium and Vitamin E Cancer Prevention Trial (SELECT) to identify factors related to undergoing prostate biopsy and to assess the effect biopsy detection bias may have on the understanding of disease risk factors.
Adjusting for age, PSA levels and digital rectal examination, men who were more likely to undergo biopsy included those who:
- were younger (HR = 1.29; 95% CI, 1.24-1.36);
- were married (HR = 1.48; 95% CI, 1.4-1.56);
- had a family history of prostate cancer (HR = 1.52; 95% CI, 1.45-1.59); or
- had benign prostatic hyperplasia (HR = 1.28; 95% CI, 1.22-1.35).
In addition, men who took aspirin or statins at baseline or during the trials (HR = 1.15; 95% CI, 1.11-1.2), as well as men with some college education (HR = 1.12; 95% CI, 1.07-1.17), were more likely to undergo biopsy.
Conversely, men with a BMI of 25 or higher, with diabetes, and who were previous or current smokers were less likely to undergo biopsy.
“Thus, among men who agreed to participate in a prostate cancer prevention trial and who underwent substantial PSA screening and annual clinic visits, the decision to recommend (and accept) prostate biopsy, even after accounting for PSA/digital rectal exam history, was substantially influenced by other factors, some which could be related to prostate cancer,” the researchers wrote.
Risk factor estimates for prostate cancer varied widely among men from the SELECT cohort and PCPT cohorts 1 and 3.
For instance, a BMI of 25 to 29 was associated with a 19% increased risk for prostate cancer in the SELECT cohort (OR = 1.19), no risk in PCPT cohort 1 (OR = 1.01) and 9% reduction in PCPT cohort 2 (OR = 0.91).
This variation also existed for men with a higher baseline PSA level, because these men were more likely to undergo biopsy. The OR per unit increase in PSA level was 3.02 in the SELECT trial and 1.77 in PCPT cohort 3, representing a 69% relative difference.
The influence of family history in a first degree relative on prostate cancer risk ranged from an OR of 2.2 in the SELECT cohort to 1.41 in PCPT cohort 3.
“Because rates of screening, recommendations for biopsy, and acceptance of biopsy were likely further related to measures such as family history and PSA level, the true associations between these factors and prostate cancer are likely overestimated in the absence of study-mandated biopsies,” Tangen and colleagues wrote.
Statin use in the SELECT cohort was associated with an OR of 0.65, whereas statin use in PCPT cohort 3 demonstrated an OR of 0.99, equating to a relative difference of 34%.
Further, black men in the SELECT cohort appeared to have 20% greater odds for prostate cancer compared with black men in the PCPT cohorts.
The researchers noted the “overarching implication” of the data suggest that detection bias may seriously flaw the findings. Based on the flawed data, individuals may choose an intervention, such as aspirin, to reduce prostate cancer risk, when in actuality the intervention may have no effect on cancer risk, but may increase other risks, such as for stroke or hemorrhage.
In addition, resources may be directed to preclinical and clinical studies, although the original observation prompting additional research could be flawed.
“Better methods are needed for identifying consequential prostate cancers in order to avoid unnecessary biopsies,” Tangen and colleagues wrote. “These data provide testimony that past and current screening and biopsy practice has likely led to biased conclusions regarding prostate cancer risk factors.”
They also noted this observation can be applied to other tumors — including melanoma, thyroid and breast.
“We encourage further study of this area to develop a better understanding of the true relationship between risk factors and cancer,” researchers wrote.
In its “zeal” to identify risk factors for prostate cancer, the research community can potentially encourage the wrong patients to undergo screening, Peter C. Albertsen, MD, professor of surgery, as well as chief and program director of the division of urology at University of Connecticut Health Center, wrote in an accompanying editorial.
“Efforts to identify risk factors should focus primarily on these poorly differentiated cancers, but as Tangen and colleagues remind us, researchers must be cautious with their interpretations,” he added. “Clinical hypotheses should be confirmed by clinical trials before being widely adopted.” – by Kristie L. Kahl
Disclosure: Tangen reports no relevant financial disclosures. Please see the full study for a list of all other researchers’ relevant financial disclosures. Albertsen reports a consultant role with Blue Cross and Blue Shield Technology Assessment Panel.