Background. There is now evidence that the advantages of screening for prostate cancer based only on prostate specific antigen (PSA) are probably offset by the health and economic disadvantages arising from over-diagnosis and over-treatment. Among novel and promising biomarkers, attention has recently been focused on the PSA isoform p2PSA, alone or in combination with free PSA (as %p2PSA), and in combination with total and free PSA as the prostate health index (phi). Methods. We performed an electronic search for original articles published in English, French, Spanish and Italian reporting studies that assessed the diagnostic performance of %p2PSA for prostate cancer screening in direct comparison with PSA, using the keywords “Prostate Cancer”, and “Measurement” or “Screening”, and “proPSA” or “p2PSA” or “[-2]proPSA”. Titles, abstracts and full texts were carefully read, and articles not matching the inclusion criteria were excluded. Heterogeneity was evaluated by the I-squared test. The pooled estimates of accuracy and the resulting area under the receiver operator characteristic curve (AUC) were calculated using a random effect model. Results. On the basis of the electronic search, 11 articles and 12 studies, including a total of 5,139 subjects (2,338 with prostate cancer) and with modest heterogeneity (Isquared, 66%) were selected. The %p2PSA pooled estimates were always greater than those of PSA, and so was the cumulative pooled AUC (0.687 versus 0.538; p<0.001). The diagnostic odds ratio was more than double for %p2PSA than for total PSA (4.0 versus 1.7). With a sensitivity of 0.95, routine p2PSA testing may result in up to 10% fewer unnecessary biopsies than the use of PSA for screening. The AUC of phi was slightly but not significantly better than that of %p2PSA (0.736 versus 0.717; p=0.21). Conclusions. These findings provide a rationale for planning further studies to assess whether %p2PSA testing may generate more favourable outcomes in terms of mortality and quality of life. Direct comparison does not permit the conclusion that the phi is globally superior to %p2PSA for detecting prostate cancer.
Prostate-specific antigen (PSA) isoform p2PSA in prostate cancer screening: systematic review of current evidence and further perspectives.
LIPPI, Giuseppe;MATTIUZZI, Camilla
2012-01-01
Abstract
Background. There is now evidence that the advantages of screening for prostate cancer based only on prostate specific antigen (PSA) are probably offset by the health and economic disadvantages arising from over-diagnosis and over-treatment. Among novel and promising biomarkers, attention has recently been focused on the PSA isoform p2PSA, alone or in combination with free PSA (as %p2PSA), and in combination with total and free PSA as the prostate health index (phi). Methods. We performed an electronic search for original articles published in English, French, Spanish and Italian reporting studies that assessed the diagnostic performance of %p2PSA for prostate cancer screening in direct comparison with PSA, using the keywords “Prostate Cancer”, and “Measurement” or “Screening”, and “proPSA” or “p2PSA” or “[-2]proPSA”. Titles, abstracts and full texts were carefully read, and articles not matching the inclusion criteria were excluded. Heterogeneity was evaluated by the I-squared test. The pooled estimates of accuracy and the resulting area under the receiver operator characteristic curve (AUC) were calculated using a random effect model. Results. On the basis of the electronic search, 11 articles and 12 studies, including a total of 5,139 subjects (2,338 with prostate cancer) and with modest heterogeneity (Isquared, 66%) were selected. The %p2PSA pooled estimates were always greater than those of PSA, and so was the cumulative pooled AUC (0.687 versus 0.538; p<0.001). The diagnostic odds ratio was more than double for %p2PSA than for total PSA (4.0 versus 1.7). With a sensitivity of 0.95, routine p2PSA testing may result in up to 10% fewer unnecessary biopsies than the use of PSA for screening. The AUC of phi was slightly but not significantly better than that of %p2PSA (0.736 versus 0.717; p=0.21). Conclusions. These findings provide a rationale for planning further studies to assess whether %p2PSA testing may generate more favourable outcomes in terms of mortality and quality of life. Direct comparison does not permit the conclusion that the phi is globally superior to %p2PSA for detecting prostate cancer.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
