PURPOSE: Urologists are often referred patients who initially present with an extremely high serum prostate specific antigen (PSA) level. Despite a presumptive diagnosis of prostate cancer, many of these men undergo biopsy to obtain a tissue diagnosis before treatment with androgen ablative therapy. We examined a data base of men undergoing prostate biopsy to determine the accuracy of high PSA levels (greater than or equal to 20 ng./ml.) in predicting prostate cancer. MATERIALS AND METHODS: We reviewed the records of 1,250 consecutive patients undergoing transrectal ultrasound guided prostate biopsy at 1 institution. From this data base we identified all patients with PSA greater than or equal to 20 ng./ml. at the time of prostate biopsy. The accuracy of PSA in predicting cancer was determined by calculating positive predictive values for PSA ranges and PSA cutoffs. RESULTS: We identified 187 men (15%) presenting with PSA greater than or equal to 20 ng./ml. Of these 187 men 157 (84.0%) were diagnosed with prostate cancer on initial biopsy. Due to a negative initial biopsy, yet a high suspicion of cancer, 12 (6.4%) patients underwent at least 1 repeat biopsy. Of these 12 men 6 (50%) were diagnosed with cancer on repeat biopsy. Overall, 163 of the 187 men (87.2%) were diagnosed with prostate cancer by biopsy. Stratified by PSA ranges, positive predictive values were 73.6% for 20 to 29.9, 90.3% for 30 to 39.9, 93.8% for 40 to 49.9, 100% for 50 to 99.9, 95% for 100 to 199.9 and 100% for greater than or equal to 200 ng./ml. Using PSA cutoffs positive predictive values were 95.7% for PSA greater than or equal to 30, 97.6% for PSA greater than or equal to 40 and 98.5% for PSA greater than or equal to 50 ng./ml. CONCLUSIONS: Serum PSA, when increased above 50 ng./ml., is 98.5% accurate in predicting the presence of prostate cancer on tissue biopsy. Nonetheless, since transrectal prostate biopsy has a low complication rate and is relatively well tolerated, we recommend continuing to biopsy most patients with high PSA levels. However, carefully selected elderly patients on chronic anticoagulation, with severe co-morbidities or presenting with spinal cord compression may not require biopsy before androgen ablative therapy since PSA is highly accurate in diagnosing prostate cancer at levels greater than 50 ng./ml.
PURPOSE: Urologists are often referred patients who initially present with an extremely high serum prostate specific antigen (PSA) level. Despite a presumptive diagnosis of prostate cancer, many of these men undergo biopsy to obtain a tissue diagnosis before treatment with androgen ablative therapy. We examined a data base of men undergoing prostate biopsy to determine the accuracy of high PSA levels (greater than or equal to 20 ng./ml.) in predicting prostate cancer. MATERIALS AND METHODS: We reviewed the records of 1,250 consecutive patients undergoing transrectal ultrasound guided prostate biopsy at 1 institution. From this data base we identified all patients with PSA greater than or equal to 20 ng./ml. at the time of prostate biopsy. The accuracy of PSA in predicting cancer was determined by calculating positive predictive values for PSA ranges and PSA cutoffs. RESULTS: We identified 187 men (15%) presenting with PSA greater than or equal to 20 ng./ml. Of these 187 men 157 (84.0%) were diagnosed with prostate cancer on initial biopsy. Due to a negative initial biopsy, yet a high suspicion of cancer, 12 (6.4%) patients underwent at least 1 repeat biopsy. Of these 12 men 6 (50%) were diagnosed with cancer on repeat biopsy. Overall, 163 of the 187 men (87.2%) were diagnosed with prostate cancer by biopsy. Stratified by PSA ranges, positive predictive values were 73.6% for 20 to 29.9, 90.3% for 30 to 39.9, 93.8% for 40 to 49.9, 100% for 50 to 99.9, 95% for 100 to 199.9 and 100% for greater than or equal to 200 ng./ml. Using PSA cutoffs positive predictive values were 95.7% for PSA greater than or equal to 30, 97.6% for PSA greater than or equal to 40 and 98.5% for PSA greater than or equal to 50 ng./ml. CONCLUSIONS: Serum PSA, when increased above 50 ng./ml., is 98.5% accurate in predicting the presence of prostate cancer on tissue biopsy. Nonetheless, since transrectal prostate biopsy has a low complication rate and is relatively well tolerated, we recommend continuing to biopsy most patients with high PSA levels. However, carefully selected elderly patients on chronic anticoagulation, with severe co-morbidities or presenting with spinal cord compression may not require biopsy before androgen ablative therapy since PSA is highly accurate in diagnosing prostate cancer at levels greater than 50 ng./ml.
Authors: W S A Smellie; J Forth; S Sundar; E Kalu; C A M McNulty; E Sherriff; I D Watson; C Croucher; T M Reynolds; P J Carey Journal: J Clin Pathol Date: 2006-05-19 Impact factor: 3.411
Authors: Pierre I Karakiewicz; Serge Benayoun; Michael W Kattan; Paul Perrotte; Luc Valiquette; Peter T Scardino; Ilias Cagiannos; Hans Heinzer; Simon Tanguay; Armen G Aprikian; Hartwig Huland; Markus Graefen Journal: J Urol Date: 2005-06 Impact factor: 7.450
Authors: Jeremy Yc Teoh; Steffi Kk Yuen; James Hl Tsu; Charles Kw Wong; Brian Sh Ho; Ada Tl Ng; Wai-Kit Ma; Kwan-Lun Ho; Ming-Kwong Yiu Journal: Asian J Androl Date: 2015 Sep-Oct Impact factor: 3.285