Blackwell K L, Bostwick D G, Myers R P, Zincke H, Oesterling J E
Department of Pathology, Mayo Clinic, Rochester, Minnesota.
J Urol. 1994 Jun;151(6):1565-70. doi: 10.1016/s0022-5347(17)35303-x.
The serum prostate specific antigen (PSA) level was determined in 311 men with documented prostate cancer (stages T1cN0, T2N0 and T3N0) before bilateral pelvic lymphadenectomy and radical retropubic prostatectomy. The prostates were whole mounted, serially sectioned, and examined for cancer volume, capsular perforation, seminal vesicle invasion, lymph node involvement, Gleason grade, nuclear grade and nuclear deoxyribonucleic acid content. Median serum PSA level was significantly different between cancers that were organ confined, those that had capsular perforation or seminal vesicle invasion and those with positive lymph nodes (p < 0.001). Median serum PSA level was also significantly different between tumors with Gleason scores of less than 6 and those with higher Gleason scores (p < 0.001), and between tumors with greater than 30% of poorly differentiated cancer (Gleason primary grades 4 and 5) and those with 30% or less poorly differentiated cancer (p < 0.001). Bivariate analysis revealed that the strongest correlations of serum PSA level were with cancer volume (r = 0.56), per cent of poorly differentiated cancer (r = 0.42), positive surgical margins (r = 0.39) and pathological stage (r = 0.38), for all p < 0.001. Multivariate analysis showed that cancer volume was the major contributor to serum PSA level. The derivative, PSA-cancer density (serum PSA times cancer volume divided by prostate volume), accounted for the effects of prostate volume and cancer volume on serum PSA. PSA-cancer density showed a significant correlation with pathological stage (r = 0.56), Gleason score (r = 0.53) and per cent of poorly differentiated cancer (r = 0.49, for all p < 0.001), and these correlations were significantly stronger than serum PSA level alone or PSA density (serum PSA divided by prostate volume; volume determined from tissue specimens) for all variables. These results indicate that preoperative serum PSA level has significant predictive value in determining tumor burden and pathological stage, and this predictive value is increased by accounting for cancer and gland volume with PSA-cancer density.
在311例经记录确诊为前列腺癌(T1cN0、T2N0和T3N0期)的男性患者中,在进行双侧盆腔淋巴结清扫术和耻骨后根治性前列腺切除术之前测定血清前列腺特异性抗原(PSA)水平。对前列腺进行完整包埋、连续切片,并检查癌体积、包膜穿孔、精囊侵犯、淋巴结受累情况、Gleason分级、核分级和细胞核脱氧核糖核酸含量。器官局限性癌、有包膜穿孔或精囊侵犯的癌以及淋巴结阳性的癌之间,血清PSA水平中位数有显著差异(p<0.001)。Gleason评分低于6分的肿瘤与评分较高的肿瘤之间,血清PSA水平中位数也有显著差异(p<0.001);分化差的癌(Gleason主要分级为4级和5级)占比大于30%的肿瘤与占比为30%或更低的肿瘤之间,血清PSA水平中位数也有显著差异(p<0.001)。双变量分析显示,血清PSA水平与癌体积(r=0.56)、分化差的癌占比(r=0.42)、手术切缘阳性(r=0.39)和病理分期(r=0.38)的相关性最强,所有p均<0.001。多变量分析表明,癌体积是血清PSA水平的主要影响因素。衍生指标PSA-癌密度(血清PSA乘以癌体积除以前列腺体积),解释了前列腺体积和癌体积对血清PSA的影响。PSA-癌密度与病理分期(r=0.56)、Gleason评分(r=0.53)和分化差的癌占比(r=0.49,所有p均<0.001)显著相关,对于所有变量,这些相关性显著强于单独的血清PSA水平或PSA密度(血清PSA除以前列腺体积;体积由组织标本确定)。这些结果表明,术前血清PSA水平在确定肿瘤负荷和病理分期方面具有显著的预测价值,通过PSA-癌密度考虑癌和腺体体积可提高这种预测价值。