After the recommendation against massive prostate cancer screening by PSA many studies have shown an increase in the diagnosis of high grade, locally advanced and metastatic prostate cancer (22–24). Our results showed a significant impact of screening policies, with a 45% decrease in the total number of biopsies performed per year and a significant increase in positive biopsy rates between 2012 and 2015. Our results confirm the findings of other groups that found a significant decrease in the median number of biopsies (25, 26) and a 29% increase in positive biopsy rate (26). Contrary to what we expected, our study was not able to show a significant increase in PSA at initial presentation, which was a constant in our examination of the literature (23, 28–30). Although we are not able to give a definite explanation to this finding, we believe it might occur because of different derivation criteria of the associated centers, different level of compliance of the indication not to perform PSA screening and the rising utilization of prebiopsy MRI in the studied period.
We observed a significant increase in local tumoral aggressiveness, mostly because of an increase of the clinical staging cT2-4, and an increase in the ISUP 4 and 5. In their study, Banjeri et al. reported similar findings, with higher clinical stage (cT2b, p = 0.003; cT2c-3a, p = 0.027) and with D´Amico high risk scores (p = 0.036) after the USPTF recommendation (29). We analyzed the histological aggressiveness using the ISUP grading system. Several authors reported their results using the Gleason score and found a significant increase in grade (26, 30, 31). This increase in the diagnostic Gleason score has been also confirmed in radical prostatectomy final pathology (32).
We identified a significant raise in metastatic prostate cancer at the time of initial diagnosis, which in our opinion is the most important negative consequence of the implementation of non-screening recommendations. Our data supports the results described by other authors showing increased incidence of metastatic prostate cancer at time of diagnosis. Bernstein et al., analyzing the SEER database reported that in men ≥ 75 years old, the diagnosis of distant metastases increased in 2012 compared to 2011 (IR 1.13, 95% CI 1.02–1.24, p < 0.05) (33). Using the same database, Hu et al. confirmed this increase between 2010 and 2013 either in men < 75 years (2.7%; 95%CI, 2.5%-2.9% vs 4.0%; 95%CI, 3.8%-4.2%) and > 75 years (6.6%; 95%CI, 6.2%-7.0% vs 12.0%; 95%CI, 11.2%-12.7%) (34). In a population-based data review from 18 SEER registries, Dalela et al. noted that the incidence of metastatic prostate cancer increased significantly between years 2009 and 2013 at a rate of 3.1% per year (P < 0.05) (35). Interestingly, Weiner et al found that the increase in the annual incidence of metastatic prostate cancer was higher among men aged 55–69 years (36). This is especially bothersome, because this group is the most likely to benefit from definitive treatment. We failed to confirm a significant increase in pelvic lymph node metastasis, although we observed a rising trend such as the result obtained by Blair et al. It is worth noting that these results contrast with those reported by Bernstein et al., who analyzing the SEER database showed a significant increase on pelvic lymph node metastasis between 2004 and 2014 (from 54.1 to 79.5 per million men (IR 1.47, 95% CI 1.33–1.62, p < 0.01) (33).
Updated results from de ERSPC trial confirm the risk reduction of developing metastasis (HR: 0.70; 95%CI, 0.60–0.82; p = 0.001) and PCa mortality, with lower number of men needed to be invited for screening and to diagnose to prevent a prostate cancer death (570 and 18 respectively) (37, 38). It is important to mention that in a predictive model, discontinuation of screening eliminates all overdiagnoses but more than doubles metastatic cases at presentation and increase 13–20% prostate cancer deaths (39).
Although our study gives important information about the impact of the 2012 recommendation on PSA screening in our population, we are aware of its limitations. In addition to it retrospective design and inherent biases, as a tertiary center with different associated centers we don’t know exactly the derivation criteria used, the level of penetration of the recommendation and the exact time of adoption in primary health centers. We also think that the increasing use of prebiopsy MRI could have biased our results.