Prostate cancer is one of the leading causes of cancer mortality in men, responsible for over 250,000 deaths worldwide in 2008 according to the International Agency for Research on Cancer. In an effort to reduce mortality rates, screening initiatives have been developed to improve early detection, most notably prostate specific antigen (PSA) testing. Higher levels of PSA in the blood are linked to an increased likelihood of having prostate cancer. However, standard PSA screening has been controversial as raised PSA levels can also be indicative of prostatisis and urinary tract infections. In a recent study in BMC Medicine Andrew Vickers from the Memorial Sloan Kettering Cancer Center, USA, addresses the issue of prostate cancer overdiagnosis in relation to PSA levels and age, finding that more specific parameters should be set when determining who should undergo PSA testing.
There is currently no global consensus on who should be screened, with recommendations varying considerably, from the US Preventative Services Task Force stance against PSA screening, to the New Zealand Ministry of Health recommending all men over 50 should be tested. Two major trials that have fed into this debate include the 2009 US-based Prostate, Lung, Colorectal, and Ovarian (PLCO) trial that showed no reduction in deaths from PSA screening, and the 2012 European Randomized Study of Screening for Prostate Cancer (ERSPC) trial that showed PSA screening did reduce deaths from prostate cancer. We spoke to lead author of the PLCO trial, Gerald Andriole from Washington University, USA, and lead author of the ERSPC trial Fritz Schröder from Erasmus Medical Center, the Netherlands, about their views on this controversy and the future of PSA screening.
Why is PSA screening controversial?
GA: While PSA based screening can reduce prostate cancer specific mortality, the absolute risk reduction is relatively small and during the course of screening the probability of diagnosing a non-lethal cancer is relatively high. These ‘over-diagnosed’ cases of prostate cancer pose a dilemma for the patient and physicians; more often than not, these men undergo aggressive therapy (this has been referred to as ‘overtreatment’). Aggressive treatment of prostate cancer is costly in both human and economic terms. On balance therefore, the relative benefit of screening in terms of reducing prostate cancer specific mortality may be significantly counterbalanced by the considerably larger number of men who are overdiagnosed and overtreated.
FS: In spite of a proven effect in terms of the reduction of prostate cancer mortality in men who get screened of more than 30 percent, screening remains controversial for a number of reasons, such as the low specificity of the PSA-test resulting in too many negative tests, unnecessary anxiety and unnecessary expenses. The same is true for the very low specificity of the biopsy procedure. Only one in four men who gets a biopsy because of a positive PSA-test has prostate cancer. Furthermore, of the cancers found at biopsy about 30 percent classify for minimal disease and we would have preferred to omit the biopsy in these cases all together.
What did your PLCO trial published in 2009 show?
GA: The PLCO Cancer Screening Trial showed that after approximately 13 years of follow-up (J Natl Cancer Inst 2012, 104 (2): 125-132) of a cohort of approximately 80,000 men in the United States, there was no benefit to men undergoing screening in terms of reducing prostate cancer specific mortality. This result may have been influenced by the fact that a considerable proportion of the men who were initially randomized to the ‘usual medical care’ arm underwent PSA-based testing. So PLCO is best considered as a trial testing whether ‘organized’ annual PSA-based screening was superior to ‘opportunistic’ screening as performed in routine practice in the US. It was not a test of screening versus no screening. The PLCO Trial is continuing and further updates should become available.
What did your ERSPC trial published in 2012 show?
FS: The main message is that the PSA mortality in the total trial population was decreased by 21 percent in screened men with respect to the control group. If only men who in fact were screened and complied with the biopsy recommendations were considered, this rate amounted to 29 percent. The absolute risk reduction was the prevention of one prostate cancer death per 1000 men who underwent randomization. To prevent death from prostate cancer within the 11 year follow-up period 1055 men needed to be invited for screening, and 37 cancers would need to be detected.
Do you think targeting PSA screening to people of a certain age or with specific PSA levels could reduce the potential harms associated with screening?
GA: Yes, there is no question that some form of ‘risk stratification’ is necessary to reduce the potential harms of screening which for the most part consist of overdiagnosis and overtreatment. There are a number of possible ways of risk stratifying which men should undergo screening. One of the most promising is to test PSA levels among men at a very early age (during their 40s) and intensively screen those men with the highest PSA values (generally considered those with PSA values >1). There are potentially other ways of risk stratifying who should be screened using genetic markers and family history but these will require further evaluation. I think it is becoming clear that mass population screening of men on the basis of age alone is suboptimal because of the risks of over-diagnosis.
FS: Our study shows no effect of screening in men age 70 or older. This does not exclude the possibility of offering early detection to older men with a high life expectancy. Adjustment of PSA levels will always lead to the loss of aggressive cancers and is not a good option. No level of sensitivity can be identified which would optimize specificity. Our hope is that additional tests such as the application of multi-parametric MRI testing will allow the selective exclusion of low risk men from the biopsy procedure and thereby reduce overdiagnosis.
What are your visions for the future of PSA screening?
GA: I anticipate that the use of PSA-based screening will become more risk stratified with the use of other markers in addition to PSA, such as new serum and urinary markers. Some of these markers seem to be more cancer specific (rather than prostate specific) and also tend to identify aggressive prostate cancers better than PSA. Moreover, screening will also be improved by better prostate biopsy, likely by using MRI to localize the most suspicious area within the prostate. Finally, we are recognizing that many men with screen-detected cancer can undergo surveillance, rather than aggressive therapy. That approach, possibly coupled with focal ablation of small cancers, will significantly reduce overtreatment.
FS: The medical community will have to work on reducing overdiagnosis, the main harm of screening. After this I predict that worldwide population based screening for prostate cancer will be considered.
Questions from Claire Barnard, Senior Editor for BMC Medicine.