Cop+Ed+fisse 2006 Archivio Italiano di Urologia e Andrologia 2022; 94, 118 ORIGINAL PAPER No conflict of interest declared. prevalence in the region (4). Prostate cancer is expected to become the most common cancer in males in 2020 (4). A general goal of this study was to estimate the level at which PSA value and prostate volume value are indicative for a biopsy procedure in the Lebanese population. The primary objective was to evaluate the diagnostic strength of prostate specific antigen density (PSAD) versus PSA level in the Lebanese men in correlation with biopsy outcomes to avoid unnecessary prostate biopsy. The secondary objectives of the study were: 1) to identi- fy age-related cutoffs which may be used in the clinical practice for the diagnosis of prostate cancer, 2) to identi- fy a cutoff for the PSA level which may be used in the clinical practice for the diagnosis of the prostate cancer, 3) to identify a cutoff for the PSAD level which may be used in the clinical practice for the diagnosis of the prostate cancer. METHODS Study design and patient population This study was a retrospective chart review, conducted in Bahman hospital, including patients who were screened for prostate cancer and underwent prostate biopsy. All patients were admitted to Bahman hospital during the last 15 years, between January 2006 and December 2019. Patients were selected according to predefined inclusion and exclusion criteria (Table 1) and PSA testing was pri- marily used to screen for prostate cancer. Accordingly, patients were chosen, and data was collected and submit- ted for statistical calculation and further analysis. The protocol was reviewed and granted written study approval from the research committee in the Lebanese University, and approval from the ethical committee of the hospital. The study was conducted in accordance with the US Code of Federal Regulation 45-CFR-46.107, 21-CFR- 56.107, Good Clinical Practice ICH Section 3 and the prin- ciples laid down by the 18th World Medical Assembly (Helsinki, 1964) and all applicable amendments. All partic- ipants had a designated code. Records will be stored, and none can access the sheets except the researchers. The sample size was estimated on the assumption of an incidence of prostate cancer in the Lebanese males of Objective: Being the second most common cancer in men, prostate cancer detection relies on laboratory tests, imaging, and surgical procedures, although biopsy remains the mainstay in diagnosis of prostate cancer. No clear cut-off of prostate specific antigen density (PSAD) for suspecting prostate cancer has been established in the Lebanese population. Our primary objective was to evaluate the diagnostic strength of the PSAD value versus total prostate specific antigen (tPSA) level in the Lebanese men in correlation with biopsy outcome to avoid unnecessary prostate biopsy. Methods: A retrospective study of 347 patients with history of prostate biopsy done for cancer suspicion included tPSA, prostate volume, and prostate density values and results of prostate biopsy. Data was collected from Bahman hospital and statistical analysis of the mean values of tPSA, prostate volume and PSAD in different age groups was done. Significance of the results was tested using. Results: On average, patients with negative biopsies were younger and they had lower tPSA levels, lower PSAD values and larger prostate volume compared to patients with positive biopsies. A PSAD cutoff of 0.185 ng/ml2 revealed the highest predictive strength for prostate cancer (6 times risk) compared with other parameters. These findings were mainly referred to patients with PSA > 10 ng/ml. Conclusions: A multifactorial approach must be conducted including all parameters in order to decide upon the need for prostate biopsy. PSAD proved to be a good marker in favor or against a prostate biopsy with a cut-off of 0.185 ng/ml2, espe- cially in patients with tPSA level higher to 10 ng/ml. A multi- center study was recommended for better and more reliable results and more precise cut-offs. KEY WORDS: PSAD; Prostate biopsy; PSA; Age; Prostate volume. Submitted 22 September 2021; Accepted 15 October 2021 INTRODUCTION Normal ranges of prostate specific antigen (PSA) and prostate volume vary among ethnicities and communities at different geographic locations and of different socioe- conomic statuses (3). Therefore, pathological PSA and prostate volume values might as well vary between eth- nicities (3). In the Lebanese men, prostate cancer incidence is expect- ed to reach 69 cases per 100000 by 2020, the highest Predictive value of PSA density in the diagnosis of prostate cancer in lebanese men Ali Msheik 1, Mohamed Mohanna 2, Ali Mhanna 3, Ali Kanj 4, Mohamad Moussa 5, Assaad Mohanna 4 1 PGY-1 Neurosurgery, Faculty of Medical Sciences, Lebanese University, Beirut, Lebanon; 2 PGY-1 Internal Medicine, Faculty of Medical Sciences, Lebanese University, Beirut, Lebanon; 3 PGY-1 General Surgery, Faculty of Medical Sciences, Lebanese University, Beirut, Lebanon; 4 Radiology Department, Bahman Hospital, Beirut, Lebanon; 5 Chairman of General Surgery and Urology Department, Faculty of Medical Sciences, Lebanese University, Beirut, Lebanon. DOI: 10.4081/aiua.2022.1.18 Summary 19Archivio Italiano di Urologia e Andrologia 2022; 94, 1 PSA density in the diagnosis of prostate cancer 1503 new cases in 2018 and a 5-year prevalence of 3405 according to GLOBOCAN (5). Hence, we estimated a minimum sample size of 10% of the estimated prevalence that is 300 patients who must fulfill the inclusion and exclusion criteria as shown in Table 1. Data collection The researchers contacted the “Archive Department” man- ager at Bahman and set a schedule to reach the medical records and collect the data. An electronic validated database was used in the data col- lection process. The data includes the following: demographic characteris- tics (age), laboratory results (PSA ng/ml), transabdominal prostate ultrasound results (prostate volume ml) and his- tology results (Gleason score). Prostate specific antigen density was calculated by divid- ing the PSA value by the prostate volume. A Gleason Score ≥ 7 was used to define a clinically signifi- cant prostate cancer (csPCa). Statistical analysis Data was analyzed using the SPSS version 22. A descriptive analysis was done, and variables were pre- sented as per their type. The categorical variables were presented as frequency and proportions. The continuous variables were presented as frequency, mean, median and standard deviation. A binary logistic analysis was done to test the factors pre- dicting the biopsy outcome. The dependent variable was “Biopsy outcome”. The corre- lation was tested between the dependent variable and the secondary variables using the Chi-square and Fisher exact test. In addition, non-parametric tests were used as Kruskas Wallis test and Mann-Whitney test. A statistically significant correlation was set at 5% (p-value less than 0.05). RESULTS Demographic results The mean age of the patients was 66.2 (± 8.8) with a min- imum of 43 years and a maximum of 90 years. The medi- an age was 66 years. Prostate cancer and laboratory values The mean prostate volume was 59.2 (± 30.8) ml with a minimum of 12 ml and a maximum of 214 ml. The medi- an prostate volume was 53 ml. The mean PSAD was 0.56 (± 1.15) ng/ml2 with a mini- mum of 0.04 ng/ml2 and a maximum of 9.75 ng/ml2. The median prostate density was 0.18 ng/ml2 (Table 2). The mean PSA level was 24.56 (± 42.57) ng/ml with a minimum of 2 ng/ml and a maximum of 394 ng/ml. The median PSA level was 10 ng/ml (Figure 1). Histology results Histology demonstrated that 49.6% of patients had benign prostatic tissue (BPH, prosta- titis), 5.2% had low-grade prostate cancer (Gleason score = 6), and 45.2% had clinically significant prostate cancer csPCa (Gleason Score ≥ 7) (Figure 2). Table 1. Inclusion and exclusion criteria utilized in the study. Inclusion criteria Exclusion criteria PSA level ≥ 3 ng/ml Past diagnosis of prostate cancer Transabdominal prostate US result available Incomplete patient record Histologically confirmed diagnosis of csPca Table 3. Correlation between variables and biopsy outcome. Biopsy outcome N Mean Std. deviation 95% Confidence interval for mean Min-max P value Lower bound Upper bound Age Benign 172 63.38 8.30 62.13 64.63 44-87 0.000 Prostate cancer 175 69.03 8.42 67.77 70.28 43-90 PSA ng/ml Benign 172 11.47 11.89 9.68 13.26 2-100 0.000 Prostate cancer 175 37.42 55.93 29.07 45.76 3.5-394 Prostate volume (ml) Benign 172 64.66 34.40 59.48 69.84 13-214 0.003 Prostate cancer 175 53.92 25.92 50.06 57.79 12-175 Density (ng/ml2) Benign 172 0.23 0.30 0.18 0.27 0.04-1.98 0.000 Prostate cancer 175 0.89 1.52 0.66 1.11 0.05-9.75 Figure 1. Distribution of PSA levels. Table 2. Prostate volume and density. Prostate volume (ml) Density (ng/ml2) Mean 59.25 0.56 Median 53.00 0.18 Std. Deviation 30.84 1.15 Minimum 12.00 0.04 Maximum 214.00 9.75 Percentiles 25 39.00 0.12 50 53.00 0.18 75 72.00 0.44 Archivio Italiano di Urologia e Andrologia 2022; 94, 1 A. Msheik, M. Mohanna, A. Mhanna, A. Kanj, M. Moussa, A. Mohanna 20 Factors affecting the biopsy outcome A statistically significant correlation existed between age, PSA, prostate volume, and PSAD and the biopsy outcome (Mann-Whitney test; p < 0.05) (Table 3). The results showed that age was higher in prostate cancer patients (mean = 69.03 years) in comparison to patients with benign prostatic tissue (mean = 63.4 years) (p < 0.0001); PSA was significantly higher in prostate cancer patients (mean = 37.4 ng/ml) in comparison to patients with benign prostatic tissue (mean = 11.5 ng/ml) (p < 0.0001); prostate volume was significantly higher in patients with benign prostatic tissue (mean = 66.7 ml) in comparison to prostate cancer patients (mean = 53.9 ml) (p = 0.003) and PSAD was significantly higher in prostate cancer patients (mean = 0.89 ng/ml2) in comparison to patients with benign prostatic tissue (mean = 0.23 ng/ml2) (p < 0.0001). A binary logistic analysis was done to identify the factors predicting the biopsy out- come. The results showed that the biopsy outcome is affected by three variables: patients’ age (p = 0.000), PSA (p = 0.000), and prostate vol- ume (p = 0.000). The logistic analysis showed that the biopsy outcome is at risk times “1” to deviate to be “csPCa” when patient’s age is high, PSA level is high and prostate volume is low. Cutoff by age: factors affecting the biopsy outcome Patients were distributed into two groups according to median of age (65). The first group was aged less than 65 (148 patients) and the second group was aged 65 years and more (199 patients). A statistically significant correlation existed between the age groups and the biopsy outcome (Chi-square; p < 0.0001) (Table 4). The results showed that 70.9% of the patients aged 65 years and more, had prostate cancer and 56.4% of the patients aged less than 65 years had no prostate cancer. In the group of patients aged less than 65 years, a statis- tically significant correlation existed between PSA, and PSAD and biopsy outcome (Mann-Whitney test; p < 0.05) (Table 5). The results showed that PSA was higher in prostate cancer patients (mean = 33.1 ng/ml), and PSAD was significantly higher in prostate cancer patients (mean = 0.81 ng/ml2) comparing to patients with benign prostatic tissue. In the group of patients aged more than 65 years, a statis- tically significant correlation existed between PSA, prostate volume, and PSAD and the biopsy outcome (Mann- Whitney test; p < 0.05) (Table 5). The results showed that PSA was higher in prostate cancer patients (mean = 39.2 ng/ml), prostate volume was lower in prostate cancer patients (mean = 56.5 ng/ml), and PSAD was significantly higher in prostate cancer patients (mean = 0.92 ng/ml2) comparing to patients with benign prostatic tissue. A binary logistic analy- sis was done to identify the factors affecting the biopsy outcome in the patients aged less than 65 years. Table 4. Correlation between age and biopsy outcome. Age Total P value 40-64 years 65-90 years Biopsy outcome Benign 97 75 172 < 0.0001 56.4% 43.6% 100.0% Prostate cancer 51 124 175 29.1% 70.9% 100.0% Total 148 199 347 42.7% 57.3% 100.0% Figure 2. Distribution of the patients according to biopsy outcome. Table 5. Correlation between the variables and biopsy outcome. Age Variables Biopsy outcome N Mean Std. deviation 95% Confidence interval for mean Min-max P value Lower bound Upper bound 40-64 Years PSA ng/ml Benign 97 10.35 11.32 8.07 12.63 2-85 0.010 prostate cancer 51 33.15 67.00 14.30 51.99 4-394 Prostate volume (ml) Benign 97 57.10 29.98 51.05 63.14 13-200 0.038 prostate cancer 51 47.73 14.82 43.56 51.89 22-78 Density (ng/ml2) Benign 97 0.23 0.30 0.17 0.29 0.04-1.98 0.002 prostate cancer 51 0.81 1.67 0.34 1.28 0.05-7.30 65-90 Years PSA ng/ml Benign 75 12.91 12.52 10.03 15.79 3.7-100 0.000 prostate cancer 124 39.17 50.89 30.13 48.22 3.5-300 Prostate volume (ml) Benign 75 74.45 37.36 65.85 83.04 21-214 0.000 prostate cancer 124 56.47 28.95 51.33 61.62 12-175 Density (ng/ml2) Benign 75 0.23 0.30 0.16 0.30 0.04-1.89 0.000 prostate cancer 124 0.92 1.46 0.66 1.18 0.06-9.75 21Archivio Italiano di Urologia e Andrologia 2022; 94, 1 PSA density in the diagnosis of prostate cancer The results showed that the Gleason score was affected by PSA (p = 0.012), and prostate volume (p = 0.048). The logistic analysis showed that the biopsy outcome was at risk times “1” to deviate to be “csPCa” when PSA level was high and prostate volume was low. A binary logistic analysis was done to identify the factors affecting the biopsy outcome in the patients aged more than 65 years. The results showed that the biopsy outcome was affected by two variables: PSA (p = 0.000), and prostate volume (p = 0.003). The logistic analysis showed that the biopsy outcome was at risk times “1” to deviate to be “csPCa” when: PSA level was high and prostate volume was low. Cutoff by PSA: factors affecting the biopsy outcome Patients were distributed into two groups according to median of PSA (10). The first group had a PSA level less than 10 ng/ml (168 patients) and the second group had a PSA level 10 ng/ml and more (179 patients). A statistical- ly significant correlation existed between the PSA groups and the biopsy outcome (Chi- square; p < 0.0001) (Table 6). The results showed that 63.1% of the patients, who had a PSA equal to 10 ng/ml and more, were diagnosed with prostate cancer and 63.1% of the patients who had a PSA less 10 ng/ml were not diagnosed with prostate cancer. A patient with a PSA equal to 10 ng/ml and more had a risk of 2.9 to have a prostate cancer. A binary logistic analysis was performed to predict the factors affecting the biopsy outcome in patients having a PSA level less than 10 ng/ml. The results showed that the biopsy outcome is affected by two variables: the age (p = 0.000), and the prostate volume (p = 0.049). The logistic analysis showed that the biopsy outcome is at risk times “1” to deviate to be “csPCa” when: the age is high and prostate volume is low. A binary logistic analysis was performed to predict the factors affecting the biopsy outcome in the patients hav- ing a PSA level equal to 10 ng/ml and more. The results showed that the biopsy outcome was affected by two vari- ables: age (p = 0.004), and PSAD (p = 0.000). The logis- tic analysis showed that the biopsy outcome is at risk times “1” to deviate to be “csPCa” when the age is high and at risk of “6” times” when PSAD is high. Cutoff by PSA density: factors affecting the biopsy outcome A statistically significant correlation existed between PSAD groups and the biopsy outcome (chi-square; p < 0.0001) (Table 7). The results show that 66.5% of the patients, who had a PSAD more than 0.185, were diag- nosed with prostate cancer and 65.9% of the patients who had a PSAD less than 0.185 were not diagnosed with prostate cancer. A patient with a high PSAD had a risk of 3.8 to have a prostate cancer. A statistically significant correlation existed between the PSA density groups and the biopsy outcome in each of the two age groups (Chi-square; p < 0.05) (Table 8). The results showed that 46.7% of the patients aged less than 65 years, who had a PSA equal to 10 ng/ml and more, were diagnosed with prostate cancer with an odds Ratio equal to 2.5. In addition, 77% of the patients aged 65 years and more, who had a PSA equal to 10 ng/ml and more, were diagnosed with prostate cancer with an odds Ratio equal to 4.5. A statistically significant correlation existed between the median PSAD and the biopsy outcome when the PSA level was ≥ 10 ng/ml. The risk of being diagnosed with prostate cancer was 4.2% higher (95% CI 0.263-0.691) when the PSA was more than 10 ng/ml (p < 0.0001) (Table 9). DISCUSSION In this study, the profiles of patients submitted to prostate biopsy were outlined. The variables considered were tPSA (total PSA), prostate volume, PSAD, age and prostate biop- Table 6. Correlation between the PSA and the biopsy outcome. Biopsy outcome P value OR CI (95%) Benign Prostate cancer PSA PSA 3–9.9 ng/ml 106 62 < 0.0001 2.927 1.89-4.53 63.1% 36.9% PSA 10–19.9 ng/ml 66 113 36.9% 63.1% Table 7. Correlation between the PSA density and the biopsy outcome. Biopsy outcome P value OR CI (95%) Benign Prostate cancer PSA PSAD < 0.184 114 59 < 0.0001 3.831 2.45-5.98 65.9% 34.1% PSAD > 0.185 58 115 33.5% 66.5% Table 8. Correlation between the PSA density and the biopsy outcome in terms of the age groups. Age PSAD Gleason score P value Risk CI (95%) Benign Prostate cancer 40-64 years PSAD < 0.184 65 23 0.010 2.473 1.234-4.956 73.9% 26.1% PSAD > 0.185 32 28 53.3% 46.7% 65-90 years PSAD < 0.184 49 36 < 0.0001 4.554 2.465-8.416 57.6% 42.4% PSAD > 0.185 26 87 23.0% 77.0% * Chi-Square Test. Table 9. Individual risk of benign and prostate cancer with PSA level ≥ 3 ng/ml by PSA density. PSA PSAD Biopsy outcome P value Risk CI (95%) Benign Prostate cancer Lower Upper PSA 3–9.9 ng/ml < 0.184 ng/ml2 87 82.1% 47 77.0% 0.432* 0.827 0.522 1.309 > 0.185 ng/ml2 19 17.9% 14 23.0% PSA > 10 ng/ml < 0.184 ng/ml2 27 40.9% 12 10.6% 0.000* 0.427 0.263 0.691 > 0.185 ng/ml2 39 59.1% 101 89.4% * Chi-Square Test. Archivio Italiano di Urologia e Andrologia 2022; 94, 1 A. Msheik, M. Mohanna, A. Mhanna, A. Kanj, M. Moussa, A. Mohanna 22 sy findings. The results showed that age, tPSA level, prostate volume, and PSAD were important factors to con- sider in the decision of whether to do a biopsy of prostate or not. This study showed that each of these factors has a certain median relative to which the positive predictive value (PPV) of prostate cancer at prostate biopsy differs. Age First, about age, the results showed that above two thirds of the biopsies proved evidence of high grade cancer (Gleason > 7) in patients older than 65 years, while less than half biopsies done in patients younger than 65 years diagnosed csPCa. Accordingly, advanced age added one times risk to the detection of csPca. Hence, patients older than 65 years who had urinary symptoms that advocated prostate pathol- ogy proved to be candidates for a prostate biopsy with a high PPV for prostate cancer. These findings were antici- pated in the literature. More than 65% of prostate cancer patients are expected to be above 65 years of age (6). Volume We reported a one-time risk of prostate cancer detection in association with a low prostate volume in both age groups. Of note, the mean of prostate volume was higher in BPH patients compared to prostate cancer patients in either age groups. However, we observed an increase of mean volume with age in both BPH and prostate cancer patients. This sheds light of the possibility of concomitant occurrence of BPH and of its evolution before or along with prostate cancer. Many studies in the literature confirmed that in patients with small volume prostates, PSA levels superior to 4 ng/ml and suspicious on digital rectal examination (DRE) were more likely to show pathological evidence of prostate cancer at biopsy (7, 8). A retrospective study by Camur et al. noted that prostate volume has no significant effect on upgrading in active surveillance of appropriate patients9. This result addressed the prostate volume as a single factor, conversely correlation of volume with PSA level proved that volume has a role in the of prostate can- cer in the indication to prostate biopsy. PSA Evaluation of PSA levels showed that a value of 10 ng/ml (0.38 nmol/l) represented a median that departed values similarly to what was observed for age with a median value of 65 years. In fact, 52% of the patients in this study had a PSA supe- rior to 10 ng/ml. The prevalence of prostate cancer in this PSA group was remarkably differentiable with the PSA value. Nearly, two-thirds of patients whose PSA level was below 10 ng/ml have benign biopsy outcome compared to two-thirds of patients who have proved csPca on their prostate biopsy with a PSA level above 10 ng/ml. These values were concordant with the findings reported by Schmid et al. (10), whereas according to Park et al. (11) and Kobayashi et al. (12), there is no significant different detection rate of cancer and pathological findings between the group with tPSA 2-4 ng/ml and 4-10 ng/ml. In the present study patients were divided in only two groups according to PSA (4-9.9 ng/ml and > 10 ng/ml). Nevertheless, a tPSA level higher to the 10 ng/ml median added a risk of 1 times to the detection of csPca, similar- ly to the age factor. Of note, the mean PSA level increased in either BPH or prostate cancer patients as they grow old similarly to the increase of prostate volume observed between the two age groups (19-28% relative increase in mean prostate volume versus 18-29% increase in mean PSA levels). The joint increase of both age and PSA in relation to the outcome of prostate biopsy demonstrated that none of the two factors could be a major predictive value by itself. A PSA value higher to the cutoff (10 ng/ml) was predic- tive of a high risk of prostate cancer if the patient’s age was higher of the age cutoff (65 years old). In fact, for a PSA value above 10 ng/ml, no more than 50% of prostate biopsies demonstrated a prostate cancer unless the age was superior to 65 years. DRE, TRUS and tPSA level are commonly used methods of screening for prostate cancer. The detection of any abnormality in the prostate volume through DRE or TRUS, and the detection of a higher than age-related tPSA level are usually followed by an ultrasound or MRI-guid- ed biopsy of the prostate to rule out prostate cancer. On note, tPSA was initially utilized as a post-operative laboratory test for recurrence detection. Its implementa- tion as a screening method has lowered morbidity associ- ated with prostate biopsies and the number of unneces- sary biopsies, and allowed earlier detection of csPca up to 81% as compared to DRE alone (13). PSAD Benson et al. in 1992, introduced the concept of PSAD, to correct PSA value by prostate volume to differentiate patient with high volume benign disease from those with prostate cancer (14). However, many authors questioned this concept, because the utilization of PSAD with a cut- off of 0.15 ng/ml2 showed a sensitivity of only 60% (14). The diagnostic efficacy of PSA density has been thor- oughly discussed in relation to its stratification for each PSA level interval showing that for tPSA levels higher than 10ng/ml, a high prostate density indicated a 6 times risk of csPca detection on a prostate biopsy. This finding defined prostate density as an extremely important tool for the indication of a biopsy for this tPSA level interval. When the tPSA level is higher to 10 ng/ml, the risk of diagnosing a prostate cancer for a prostate den- sity higher than 0.185 ng/ml2 was 4.2% (95% CI 0.263- 0.691) higher. On the contrary, when PSA level is below 10 ng/ml, a high prostate density value proved to be unreliable using the 0.185 ng/m2 cut-off. Conversely, in the PSA interval with higher incidence of prostate cancer, the risk of prostate cancer at biopsy dropped when PSAD is below the 0.184 ng/ml2 cut-off. In addition, the joint evaluation of age and prostate den- sity showed that higher values of both was predictive of a higher risk of prostate cancer detection on a prostate biopsy (77% of the patients whose PSAD and age were superior to the considered cutoffs had prostate cancer compared to only 26% when both parameters were infe- rior to the considered cutoffs). The results from this study confirmed previous reports on the value of PSAD in the biopsy indication. Jue et al. demonstrated that PSAD is better in predicting prostate 23Archivio Italiano di Urologia e Andrologia 2022; 94, 1 PSA density in the diagnosis of prostate cancer cancer versus the use of PSA level or prostate volume alone (15). Similarly, Stephan et al. showed the PSA den- sity to perform better than the tPSA level for patients whose PSA level ranged between 2 and 20 ng/ml (16). Van Iersel et al., similarly to many other authors, con- cluded that the PSAD cutoff to distinguish between prostate cancer and BPH could be 0.15 ng/ml2 where a higher value is significant of a higher malignancy proba- bility (17). Although similar results were obtained in this retrospective study with a PSAD cut-off of 0.185 ng/ml2 and 0.13 ng/ml2, these values are population specific and their variance relied on a multitude of factors (18). The reliability of the prostate volume measurement dra- matically affects the significance of cut-offs used for biop- sy decisions. Two methods were implemented in the cal- culation of the prostate volume: the ellipsoid method or planimetric method. Stone et al. found that the three plane method (ellipsoid method) had a variability of 30% com- pared to the 3D-planimetric method which showed only 5% variation. Furthermore, Holmang et al. stated that the 3-plane method underestimated the volume by 20% compared to the 3D-planimetric method (19). Hence, the higher is the accuracy of the method of volume assessment, the better PSAD would assess the need for a prostate biopsy and the less unnecessary biopsies would be made. In a multi-cen- tric study of 773 patients, Catalona et al. (20) considered lowering the PSAD cut-off to 0.078 ng/ml2, because at that cut-off, 95% of tumors would be detected. In the present study, for a tPSA level between 3 and 9.9 ng/ml, when the PSAD cut-off value of 0.184 ng/ml2 was utilized, only 42.4% of csPca patients would have been diagnosed with prostate biopsy with a 57% specificity. A p-value of 0.432 for this tPSA level rendered the results less reliable. For a tPSA > 10 ng/ml, using a PSAD cut-off of 0.184 ng/ml2, the utilization of prostate biopsy was 89.4% sensitive and 40.9% specific, with a p-value of < 0.0001. Lowering the cut-off to 0.09 ng/ml2, sensitivity increased to 96% while specificity decreased to 35%. This result could be justified by the fact that 89.4% of csPca patients have a PSAD superior to 0.185 ng/ml2 compared to only 59.1% of benign patients being superior to the aforemen- tioned value. Lowering the PSAD cut-off led to a new dis- tribution of patients as shown in Table 10 and conse- quently to a change of sensitivity. It can be intuitive of the fact that most prostate cancer patients have relatively high PSAD and that lowering the PSAD cut-off recruited more patients into this category and favored an increase of the sensitivity and decrease in the specificity. Therefore, this study proved PSAD was a good predictor of the biopsy outcome for tPSA ranging between 4 to 20 ng/ml. PSAD cut-off of 0.18 ng/ml/cc could minimize the number of unnecessary biopsies. Consideration of a lower cutoff may promote better sensitivity with a slight decrease in the false negative results, which remains acceptable for a screening test. However, the acknowl- edgment of a PSAD cut-off mandates further studies for an optimal value that balances the sensitivity and false negative results of the prostate biopsy. Study limitations Every study has limitations and this study is no excep- tion. The limitations can be sorted under two titles. Population The number of the patients and the fact that they were from the same hospital limited the credibility of our results. A multicenter study with a larger population size will impart to those results more credibility, render them more reliable, help achieve more precise cut-offs, averages of minimal standard deviation, and allow generalization of the findings as representative of the whole population. Retrieval of data The missing data prevented the estimation of the positive predictive value (PPV), negative predictive value (NPV), sen- sitivity and specificity of the prostate biopsy procedure. This limited the comparison of the results versus other studies. REFERENCES 1. Rogers OC, Anthony L, Rosen DM, et al. PSA-selective activation of cytotoxic human serine proteases within the tumor microenviron- ment as a therapeutic strategy to target prostate cancer. Oncotarget. 2018; 9:22436-22450. 2. Prcic A, Begic E, Hiros M. Usefulness of total PSA value in prostate diseases diagnosis. Acta Inform Med. 2016; 24:156-61. 3. Mittal RD. Reference range of serum prostate-specific antigen lev- els in Indian men. Indian J Med Res. 2014; 140:480-1. PMID: 25488440; PMCID: PMC4277132. 4. Shamseddine A, et al. Cancer trends in Lebanon: a review of inci- dence rates for the period of 2003-2008 and projections until 2018. Popul Health Metr. 2014; 12:4. 5. https://gco.iarc.fr/today/data/factsheets/populations/422-lebanon- fact-sheets.pdf 6. Instituto Brasileiro de Geografia e Estatística; Ministério do Planejamento, Orçamento e Gestão. Estimativas populacionais 1980-2010: Brasil, regiões geográficas e unidades da federação. Rio de Janeiro (Brasil): IBGE; 2010. [citado em 15 de junho de 2010]. Disponível em: http://www.ibge.gov.br 7. Nickel JC. Inflammation and benign prostatic hyperplasia. Urol Clin North Am. 2008; 35:109-15 8. Çamur E, Coskun A, Kavukoglu O, et al. Prostate volume effect on Gleason score upgrading in active surveillance appropriate patients. Arch Ital Urol Androl. 2019; 91:93. 9. Babaian RJ, Fritsche HA, Evans RB. Prostate-specific antigen and prostate gland volume: correlation and clinical application. J Clin Lab Anal. 1990; 4:135-7. 10. Bell N, et al. Canadian Task Force on Preventive Health Care. Table 10. Redistribution of patients with PSA > 10 ng/ml after lowering PSAD cut-off to 0.09 ng/ml2. Biopsy outcome PSAD Prostate cancer Benign > 0.09 ng/ml2 109 43 < 0.09 ng/ml2 4 23 Archivio Italiano di Urologia e Andrologia 2022; 94, 1 A. Msheik, M. Mohanna, A. Mhanna, A. Kanj, M. Moussa, A. Mohanna 24 Recommendations on screening for prostate cancer with the prostate- specific antigen test. CMAJ. 2014; 186:1225-34. 11. Park HK, Hong SK, Byun SS, Lee SE. Comparison of the rate of detecting prostate cancer and the pathologic characteristics of the patients with a serum PSA level in the range of 3.0 to 4.0 ng/mL and the patients with a serum PSA level in the range 4.1 to 10.0 ng/mL. Korean J Urol. 2006; 47:358-61. 12. Kobayashi T, Nishizawa K, Ogura K, et al. Detection of prostate cancer in men with prostate-specific antigen levels of 2.0 to 4.0 ng/mL equivalent to that in men with 4.1 to 10.0 ng/mL in a Japanese population. Urology. 2004; 63:727-31. 13. Gomes R, Rebello LEFS, Araújo FC, et al. A prevenção do câncer de próstata: uma revisão da literatura. Ciencia & Saude Coletiva. 2008; 13:235-46. 14. Benson MC, et al. Prostate specific antigen density: a means of distinguishing benign prostatic hypertrophy and prostate cancer. J Urol. 1992; 147:815-6. 15. Jue JS, et al. Re-examining prostate-specific antigen (PSA) densi- ty: defining the optimal PSA range and patients for using PSA densi- ty to predict prostate cancer using extended template biopsy. Urology. 2017; 105:123-128. 16. Stephan C, et al. The ratio of prostate-specific antigen (PSA) to prostate volume (PSA density) as a parameter to improve the detec- tion of prostate carcinoma in PSA values in the range of < 4 ng/mL. Cancer. 2005; 104:993-1003. 17. Van Iersel MP, Witjes WP, de la Rosette JJ, Oosterhof GO. Prostate-specific antigen density: correlation with histological diag- nosis of prostate cancer, benign prostatic hyperplasia and prostatitis. Br J Urol. 1995; 76:47-53. 18. Ediz C, Akan S, Temel MC, Yilmaz O. The importance of PSA- Density in active surveillance for prostate cancer. Arch Ital Urol Androl. 2020; 92:136. 19. Holmäng S, Lindstedt G, Mårin P, Hedelin H. Serum concentra- tion of prostate-specific antigen in relation to prostate volume in 50 healthy middle-aged men. Scand J Urol Nephrol. 1993; 27:15-20. 20. Catalona WJ, et al. Comparison of percent free PSA, PSA densi- ty, and age-specific PSA cutoffs for prostate cancer detection and staging. Urology. 2000; 56:255-60. Correspondence Ali Msheik, (Corresponding Author) newpie@mail.com PGY-1 Neurosurgery, Faculty of Medical Sciences, Lebanese University, Beirut (Lebanon) Mohamed Mohanna PGY-1 Internal Medicine, Faculty of Medical Sciences, Lebanese University, Beirut (Lebanon) Ali Mhanna PGY-1 General Surgery, Faculty of Medical Sciences, Lebanese University, Beirut (Lebanon) Ali Kanj, MD Radiologist, Bahman Hospital, Beirut (Lebanon) Mohamad Moussa, MD Urologist, Chairman of General Surgery and Urology Department, Faculty of Medical Sciences, Lebanese University, Beirut (Lebanon) Assaad Mohanna. MD Radiologist, Head of Radiology Department, Bahman Hospital, Beirut (Lebanon)