Stenman U H, Hakama M, Knekt P, Aromaa A, Teppo L, Leinonen J
Department of Clinical Chemistry, Helsinki University Hospital, Finland.
Lancet. 1994 Dec 10;344(8937):1594-8. doi: 10.1016/s0140-6736(94)90405-7.
Prostate cancer can be detected at an early, potentially curable stage by screening based on digital rectal examination and serum prostate specific antigen (PSA). The value of screening appears doubtful, based on high 10-year survival rates in selected cases of early prostate cancer, but this follow-up time may be insufficient. By linking the information on 21172 men who took part in a screening examination in Finland, 1968-73, with data from the Finnish Cancer Registry, 44 cases of prostate cancer diagnosed up to 1980 were identified. Serum samples from cancer cases and from 74 controls matched for age and time of sampling were assayed for PSA and its complex with alpha 1-antichymotrypsin (PSA-ACT). With a cut-off for PSA of 2.5 micrograms/L giving 92% specificity, 95% of the cancers developing within the first 5 years, and 52% developing in 6-10 years tested positive. As a potential screening test with a 5-year interval for men under 65, the sensitivity would be 92% and specificity 97%. The ratio of PSA-ACT to total PSA was lower in controls than in patients with cancer. Using this ratio, we could eliminate half of the false-positive results in the range 2.5-25 micrograms/L without loss of sensitivity. Cancer was typically diagnosed 5-10 years after PSA exceeded 2.5 micrograms/L, and the median survival after diagnosis was 3.6 years. 10-year survival after drawing the sample was 71% in cancer cases with a PSA concentration less than 4 micrograms/L and 48% in those with higher concentrations. The corresponding figures at 15 years were 53% and 27%, and at 20 years 43% and 18%, respectively. These results suggest it is advisable to confine screening for prostate cancer to men with a life expectancy of clearly more than 10 years--ie, younger men, who have the greatest chance to benefit from early detection.
通过基于直肠指检和血清前列腺特异性抗原（PSA）的筛查，前列腺癌能够在早期、可能治愈的阶段被检测出来。基于早期前列腺癌特定病例的高10年生存率，筛查的价值似乎存疑，但这段随访时间可能并不充分。通过将1968 - 1973年在芬兰参加筛查检查的21172名男性的信息与芬兰癌症登记处的数据相联系，确定了截至1980年诊断出的44例前列腺癌病例。对癌症病例以及74名年龄和采样时间匹配的对照的血清样本进行PSA及其与α1 - 抗糜蛋白酶复合物（PSA - ACT）的检测。PSA临界值设定为2.5微克/升时，特异性为92%，在前5年发生的癌症中有95%检测呈阳性，在6 - 10年发生的癌症中有52%检测呈阳性。作为一项针对65岁以下男性、间隔5年的潜在筛查测试，敏感性将为92%，特异性为97%。对照中PSA - ACT与总PSA的比值低于癌症患者。使用该比值，我们可以在不损失敏感性的情况下消除2.5 - 25微克/升范围内一半的假阳性结果。癌症通常在PSA超过2.5微克/升后的5 - 10年被诊断出来，诊断后的中位生存期为3.6年。采样后，PSA浓度低于4微克/升的癌症病例10年生存率为71%，浓度较高的病例为48%。15年时相应的数字分别为53%和27%，20年时分别为43%和18%。这些结果表明，将前列腺癌筛查限于预期寿命明显超过10年的男性是明智的——即年轻男性，他们最有可能从早期检测中受益。
