|Year : 2004 | Volume
| Issue : 2 | Page : 138-143
Measurement of serum PSA in benign and malignant enlargements of prostate in Indian population: Relevance of PSAD in intermediate range PSA
Madhu S Agarwal1, Sumit Sinha1, Sanjay Juyal2, AK Gupta3
1 Department of Surgery, SN Medical College, Agra, India
2 Department of Surgery, DDU Hospital, New Delhi, India
3 Department of Medicine, SN Medical College, Agra, India
Madhu S Agarwal
4/18c, Bagh Farzana, Civil Lines, Agra - 282 002
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Objectives : PSA has revolutionized the diagnosis and management of men with prostate cancer. However as PSA is organ-specific but not cancer-specific, its usefulness is somewhat limited due to high false-positive and false-negative results. This study assesses the PSA levels in benign and malignant enlargements of prostate and evaluates the advantages of PSA density (PSAD) over PSA in early diagnosis of carcinoma of the prostate.
Methods : This study was conducted on 184 cases of prostatic enlargement presenting with symptoms of outflow obstruction. One hundred sixty of these patients had BPH and 24 were histologically proven carcinoma. Serum PSA and prostate volume using TRUS was measured in all patients.
Results : In BPH, there was a statistically significant correlation of PSA with age and volume of the gland. In carcinoma of the prostate there was no correlation of a age and volume with PSA, but there was moderate correlation of PSA with clinical stage and significant correlation with Gleason scoring.
PSAD was better correlated with the clinical stage and Gleason scoring as compared to PSA. PSAD demonstrated better specificity as compared to PSA in diagnosis of carcinoma, especially in the intermediate range (4-10 ng/ml). The construction of receiver operating characteristic (ROC) curves demonstrated the most appropriate cut-off value of PSAD to be 0.3 (at which level the sensitivity and specificity were 75% and 84.4% respectively), a level higher than those reported in American, European and Japanese literature.
Conclusions: In this study, in northern Indian population, baseline values of PSA and PSAD in BPH patients were found to be higher than those reported in western literature. PSAD demonstrated better specificity as compared to PSA in diagnosis of carcinoma, especially in the intermediate range (4-10 ng/ml).
Keywords: Benign prostate hyperplasia, carcinoma prostate, PSA, PSA density.
|How to cite this article:|
Agarwal MS, Sinha S, Juyal S, Gupta A K. Measurement of serum PSA in benign and malignant enlargements of prostate in Indian population: Relevance of PSAD in intermediate range PSA. Indian J Urol 2004;20:138-43
|How to cite this URL:|
Agarwal MS, Sinha S, Juyal S, Gupta A K. Measurement of serum PSA in benign and malignant enlargements of prostate in Indian population: Relevance of PSAD in intermediate range PSA. Indian J Urol [serial online] 2004 [cited 2019 Jun 18];20:138-43. Available from: http://www.indianjurol.com/text.asp?2004/20/2/138/21530
| Introduction|| |
Prostate specific antigen (PSA) is a serine protease, elaborated almost exclusively by epithelial cells lining the acini and ducts of prostate. Once produced, it is secreted into the prostatic ductal system and is present in high concentrations in seminal plasma in which it serves the purpose of liquefying the seminal coagulum. It gains access to general circulation by seeping through disruptedphysiological barrier in diseases affecting the prostate gland.
PSA is organ-specific but not cancer-specific, resulting in limitation of its ability to differentiate carcinoma of the prostate from a number of benign abnormalities that can produce elevated PSA. Additionally, PSA is not increased in all patients with carcinoma of the prostate. Thus, though PSA can play an important role in early detection and screening of carcinoma of the prostate, its usefulness is limited by false-positive and false-negative results. To improve the ability of PSA to detect organ-confined carcinoma of the prostate, and reliably differentiate between carcinoma of the prostate and BPH, several new concepts have emerged in recent years, including PSA density, PSA velocity, age-specific PSA and free PSA.
This study assesses the role of PSA and PSAD in the diagnosis of benign and malignant enlargements of the prostate, their correlation with the size of gland, clinical stage and histological grading of neoplasia; and evaluates the advantages of PSAD over PSA in early diagnosis of carcinoma of the prostate.
| Patients and Methods|| |
From December 1998 to December 2002, one hundred eighty four patients attending surgical outpatient clinic of SN Medical College, Agra with symptoms of bladder outflow obstruction and prostate enlargement were studied. Out of 184 patients, 160 had BPH and 24 had histologically proven carcinoma of the prostate. Thirty nine cases as controls were selected from the healthy population, in the age range 40-80 years.
Patients were worked up with detailed history and clinical examination to rule out other causes of lower urinary tract symptoms (LUTS) and complications of benign prostatic hyperplasia (BPH). Serum PSA was measured using IRMA count, PSA assay and done in all patients before biopsy, digital rectal examination (DRE) or transrectal ultrasound (TRUS). TRUS was done using realtime ultrasound scanner equipped with a 4 MHz rectal probe with rotatory transducer at its tip. Volume of prostate was measured by 'Prolate ellipsoid formula' (Vol = 0.52 x L x W x H). PSAD was calculated by dividing serum PSA by volume of prostate.
Statistical analysis using Spearman's correlation coefficient ('rs') for nonparametric data and Pearson's correlation coefficient ('rp') for parametric data were calculated. Correlation coefficient value of 1 was taken as direct correlation, 0 as no correlation and -1 as inverse correlation. A `p' value of <0.05 was taken as statistically significant. Sensitivity and specificity for PSA and PSAD were calculated by following formula:
Receiver operating characteristic (ROC) curves were constructed to illustrate sensitivity and specificity of PSA and PSAD. The values of PSA and PSAD lying close to the left upper corner of the curve have maximum sensitivity and specificity.
| Results|| |
In the present study, PSA levels in cases of BPH were <4 ng/ml in 66.2%, 4-10 ng/ml in 28.2% and >10 ng/ml in 5.6% cases. There was statistically significant correlation between PSA and age, higher levels were found with increasing age (rs = 0.70, p<0.001) [Figure - 1]. There was a significant correlation between prostate volume and PSA in BPH (rp = 0.68, p<0.01) [Figure - 2]. In all patients with prostate volume <35 ml, PSA values <4 ng/ml, whereas in the patients with prostate volume >65 ml, mean PSA values were 7.9 ng/ml.
In control group, all patients had PSA <1.2 ng/ml.
In carcinoma of the prostate, PSA was found to be <4 ng/ml in 29.2%; 4-10 ng/ml in 16.6% and >10 ng/ml in 54.2%. There was no correlation of age and volume (rp = 0.28, p>0.05) with PSA in carcinoma of the prostate. Moderate correlation was found between clinical stage and PSA (p<0.01). Significant correlation was found between PSA and Gleason scoring (rs = 0.82, p<0.01). PSAD was found to be better correlated with clinical stage (rs = 0.68, p<0.01) and Gleason scoring (rs = 0.89, p<0.001) as compared to PSA in carcinoma of the prostate [Figure - 3].
Receiver operating characteristic (ROC) curves were constructed to illustrate sensitivity and specificity performance characteristics of PSA and PSAD [Figure - 4],[Figure - 5]. The effect of different PSA cut-off levels on the discrimination between carcinoma of the prostate and BPH were seen. At PSA cut-off value of 2 ng/ml, the sensitivity and specificity were 83% and 43% respectively and at PSA cut-off value of 6; the sensitivity and specificity were 80% and 82% respectively. Thus, for PSA, the best cut-off point was 6 ng/ml, represented as the upper right hand corner value of the ROC curve [Figure - 4].
On studying the effect of different PSAD cut-off values on discrimination between carcinoma of the prostate and BPH, it was seen that at PSAD cut-off value of 0.1, the sensitivity and specificity were 66.6% and 51.2% respectively. At PSAD cut-off value of 0.3, the sensitivity and specificity were found to be 66.6% and 91 % respectively. The best cut-off point for PSAD was found to be 0.3, represented as the upper right hand corner value of the ROC curve [Figure - 5].
PSAD was then used to discriminate between carcinoma of the prostate and BPH in patients with PSA levels in intermediate range (4-10 ng/ml). The most appropriate PSAD cut-off point for indication for performance of prostate biopsies was again found to be 0.3 (sensitivity 75% and specificity 84.4%). Below this value only one case of carcinoma of the prostate would have not been detected in the present study.
| Discussion|| |
BPH is the most common cause of prostatic enlargement, but carcinoma is the most feared one. Carcinoma of the prostate is the second most common cause of death from malignancy in males. Hence, a reliable method for early detection is required, a job which best taken care of by PSA, clearly the most tumor specific antigen known. The results are objective, quantitative and examiner-independent, and the procedure is quite acceptable to the patient, given its noninvasive nature. But it is far from being an ideal tumor marker. Since its clinical introduction by Wang et al,  it has been unequivocally demonstrated that PSA is organ-specific but not disease-specific.
PSA, when used alone, cannot be used as an effective screening tool for carcinoma of the prostate due to its low sensitivity and specificity, especially in low and intermediate range. Large series have shown that 21-43% cancers will occur in patients with PSA in normal range (0-4 ng/ml). , In the present study, 37.5% cancers had normal PSA.
The normal PSA range used (0-4 ng/ml) does not take into account the rise of PSA production with age as the gland enlarges. Thus, age-specific ranges increase the sensitivity of PSA in the young and specificity in older males, eliminating the need for unnecessary TRUS and biopsy in many cases. Oesterling et al,  Richie et al  and several others have demonstrated that PSA increases with age. In the present study on north Indian population, the PSA range showed a statistically significant correlation with age (rs = 0.70, p<0.001). Patients with PSA levels above the age-specific reference range should be followed by sextant biopsy to exclude carcinoma.
In the present study, we found statistically significant correlation between volume and PSA levels in BPH patients. Many other authors, including Yu et al  and Caitto et al  have also found statistically significant correlation between prostate gland volume and PSA in BPH. Yu et al  found that PSA level correlated more strongly with prostate volume when age was adjusted. Caitto et al  demonstrated that methods adjusting prostate volume allow a better interpretation of PSA values and may reduce benign biopsy rates. Collins et al  studied the relationship between PSA, prostate volume and age in 472 men with benign prostate. They found a modest correlation of PSA with both age and volume. PSAD also increased with age. Age and prostate volume influenced PSA independently. Romic et al  found no correlation between the age of the patients and volume of the prostate, whereas a correlation was present between PSA and the prostate volume.
In our study, no correlation was found between prostate volume and PSA levels in carcinoma of the prostate. Kuriyama et al  also do not support any correlation of PSA with volume of prostate in carcinoma.
The concept of PSAD developed by Benson et al , postulates that malignant cells produce more PSA per gram of tissue than normal or hyperplasic cells. In 1992, Benson et al released results of two studies applying PSAD. , A total of 61 patients, 41 with carcinoma and 20 with BPH were taken. Mean PSAD was 0.581 for carcinoma of the prostate group and 0.044 for BPH group, which were statistically significant. The use of PSAD further improved the diagnostic value of PSA. PSAD increased predictability of detecting carcinoma and was found to be superior to PSA alone in detecting carcinoma in males with no palpable abnormality of prostate. However, the value of PSAD in early diagnosis of carcinoma of the prostate has been questioned by Thon et al. 
In the present study, median PSAD was 0.095 for BPH group and 0.578 for carcinoma of the prostate group, which were statistically significant. PSAD was found to be better correlated with clinical stage (rs = 0.68, p<0.01) and Gleason scoring (rs = 0.89, p<0.001) as compared to PSA in carcinoma of the prostate. PSAD was found to be more predictive for histological grade for patients having Gleason score up to 6.
| References|| |
|1.||Wang MC, Valenzuela LA, Murphy GP, Chu TM. Purification of a human prostate specific antigen. Invest Urol 1979; 17: 159. [PUBMED] |
|2.||Bahnson RR, Catalona WJ. Clinical use of PSA in patients with prostate carcinoma. J Urol 1989; 142:5. |
|3.||Partin AW, Carter HB, Chan DW, Epstein JI, Oesterling JE, Rock RC. PSA in staging of localized prostate cancer. Influence of tumour differentiation, tumour volume and BPH. J Urol 1990; 143: 747-52. |
|4.||Oesterling JE, Cooner WH, Jacobsen SJ, Guess HA, Lieber MM. Influence of patient age on the serum PSA concentration. An important clinical observation. Urol Clin North Am 1993: 20: 671. |
|5.||Richie JP, Catalona WJ, Ahmann F et al. Effect of patient age on early detection of prostate cancer with serum PSA and DRE. Urology 1993; 42: 365-74. |
|6.||Yu HJ, Chiang GJ, Chin TY. Lai MK. Relationship between PSA and prostate volume. J Formos Med Assoc 1995; 94: 666-70. |
|7.||Ciatto S, Bonardi R, Mazzotta A, Santoni R. Reliability of volume or age adjusted PSA to improve diagnostic accuracy. Int Biol Markers 1995: 10: 226-8. |
|8.||Collins GN, Lee RJ, Me Kelvie GB, Rogers AC, Hehir M. Relationship between PSA, prostate volume and age in the benign prostate. Br J Urol 1993; 71: 445-50. |
|9.||Romics I, Frang D, Bodrogi I. Significance of examination of PSA and PSAD in patients with BPH and carcinoma prostate. Int Urol Nephrol 1997; 29: 449-55. [PUBMED] |
|10.||Kuriyama M, Uno H, Watanabe H, Yamanaka H, Saito Y, Shida K. Determination of reference values for total PSA, F/T and PSAD according to prostatic volume in Japanese prostate cancer patients with slightly elevated serum PSA levels. Jpn J Clin Oncol 1999; 29: 617-22. [PUBMED] [FULLTEXT]|
|11.||Benson MC, Whang IS, Olsson CA et al. The use of prostate specific antigen density (PSAD) to enhance the predictive value of intermediate levels of PSA. J Urol 1992; 147: 817. |
|12.||Benson MC, Reinkaplan, Cooler WH. PSAD: Distinguishing BPH and prostate cancer. J Urol 1992; 147: 815. |
|13.||Thon WF, Gadben F, Truss MS, Hoetmanan U. PSAD - a reliable parameter for the detection of prostate cancer. World J Urol 1996; 14: 53-8. |
|14.||Arai Y, Maeda H, Ishitoya S, Okubo K, Takashi, Okada, Aoki Y. Prospective evaluation of PSA and systematic biopsy for detecting prostate cancer in Japanese. J Urol 1997; 158: 861-4. |
|15.||Uno H, Koide T, Kuriyama M, Ban Y, Deguchi T, Kawada Y. PSAD for discriminating prostate cancer from BPH in the gray zone of PSA. Int Urol Nephrol 1998; 30: 305-10. |
|16.||Abdalla I, Ray P. Ray V, Vaida F. Srinivasan VK. Comparison of PSA and PSAD in African-American, White and Hispanic men without prostate cancer. Urology 1998; 51: 300-5. |
|17.||Tochigi T. Kawamura S, Numahata K, Tokuyama S, Kuwahara M. Horaguchi T, Satou S. Retrospective evaluation of PSA density for selection of biopsy candidates with prostate specific antigen in the gray zone. Nippon Hinyokika Gakkai Zasshi 2001; 92: 609-14. |
|18.||Henderson RJ, Eastham JA, Culkin DJ, Kattan MW, Whatley T, Mata J, Venable D, Sartor 0. PSA and PSAD: Racial differences in men without prostate cancer. J Natl Cancer Inst 1997; 89: 134-8. |
|19.||Guo Y, Sigman DB, Borkowski A, Kyprianou N. Racial differences in prostate cancer growth: apoptosis and cell proliferation in Caucasian and African-American patients. Prostate 2000; 42: 130-6. [PUBMED] [FULLTEXT]|
|20.||Oesterling JE, Kumamoto Y, Tsukamoto T, Girman CJ, Masumari N, Jacobsen SJ, Leiber MM. PSA in community based population in healthy Japanese men; lower values than for similar white men. Br J Urol 1995; 75: 347. |
[Figure - 1], [Figure - 2], [Figure - 3], [Figure - 4], [Figure - 5]