**2.2 Age and life expectancy**

104 Prostate Cancer – Diagnostic and Therapeutic Advances

follow from this data. One is related to high percentage of high grade cancers in population with median age 69 and relatively low PSA values. Once upon a time, it was suggested we should not measure PSA in men above 70. Should we now turn this completely around and biopsy all men with life expectancy of 10 yeas or more at the age of 70 irrespectively of their

Free PSA is measured to calculate fraction towards total PSA value, which is expressed as percentage. In PSA ranges between 4 and 10, with % free PSA below 10%, probability for biopsy detected prostate cancer in DRE negative patients was 56% and with % free PSA above 25% probability of cancer was only 8%. In PSA range between 2.5 and 4, with % free PSA below 10%, probability of cancer was 46% and with % free PSA above 20% probability

[-2]proPSA or p2PSA is new kid on the block, which is not yet universally available or accepted. It seems it will become next widely used PSA derivative, which is also used together with free PSA value (p2PSA/freePSA) or together with total and free PSA as value, calculated as p2PSA/freePSA times square root of total PSA and proprietary named by Beckman Coulter (Brea, California, USA) as PHI – prostate health index. Cut-off values for biopsy decisions using %p2PSA (p2PSA/free PSA) and PHI are not yet universally accepted and differ with regard to free PSA and total PSA calibration method (Hybritech or WHO), but it seems PHI values above 40 or 45 indicate high and below 21 low risk of prostate cancer. PHI of more than 48.5 was reported as 43% specific at 90% sensitive for detecting prostate cancer at initial prostate biopsy (Guazzoni et al., 2011). Another study reported PHI values above 34.2 to show increased probability for high risk disease (Isharwal et al., 2011). PHI was shown to indicate development of prostate cancer years before biopsy and correlated well with grade of future prostate cancer (Bektic et al., 2010). On negative side, using p2PSA compared to free PSA using ROC curves at area of high sensitivity (if we are to

PSA density and different variants (PSA density of transitional zone etc), calculated from total PSA value and measurement of prostate volume using transrectal ultrasound, were in the past extensively evaluated. Drawback is variability of prostate volume measurements, need for this additional investigation and low sensitivity (Pepe et al., 2009). One example of cutoff value was density above 0.15 ng/ml PSA /ml prostate volume indicating higher risk of cancer. It may represent additional useful piece of information, also for example regarding prediction of disease course for patients on watchful waiting (Kotb et. al, 2011). PSA velocity was in recent years proposed as another potentially helpful tool in picking cancers or high-risk cancers. Cutoff value of 0.35 ng/ml/year was proposed when total PSA values are less than 4. This may need adjustment for race (Tang et al., 2011). Although some doubts about value of PSA velocity in biopsy decisions were posed from analysis of PCPT trial data (Vickers et al., 2011), there are suggestions a lot of clinically important cancers in younger men may be detected by regularly and meticulously following and analyzing

PSA values increase with age due to BPH. This became known early after PSA test introduction and "age-specific" PSA values were suggested. For Caucasian Americans, median, 75th and 95th percentiles of PSA were for age group 40-49 – 0.7, 1.0 and 2.1, for 50-59 age group 1.0, 1.6 and 3.6, for 60-69 age group 1.4, 2.5 and 4.3, for 70-79 decade 1.8, 3.5 and 5.8 (Morgan et al., 1996). One approach of integrating age specific PSA values in prostate biopsy decisions may be using as cutoff half of age-specific 95th percentile PSA value together with %free PSA below 18%, which was in part used in Innsbruck (Tyrol, Austria).

PSA? Or should we use other tests, like PSA isoforms derivatives?

find cancer), increase in performance may seem marginal.

sequential PSA values (Bektic et al., 2011).

of cancer was 8% (Catalona et al., 1998).

Risk of prostate cancer significantly increases with age. Age is significant independent predictor of high risk prostate cancer. Although we need age as a criterion (we do not start to think about prostate cancer before age of 40 or 35), it is clear age could not be the only factor which would preclude decisions about PSA measurement, digital rectal exam and biopsy. Age is only one parameter in estimation of life expectancy. Life expectancy, not age, is crucial factor in decisions regarding prostate cancer screening interventions. Age does increase probability of cancer and cancer related death and therefore higher age – for example 75 years or more (with additional criterion of more than 10 years life expectancy) means strong indication in-favor of PSA screening and prostate biopsy. It is true that not all men at age 75 or more have more than 10 years of life expectancy, but significant number have and for them, prostate cancer screening is most useful and fruit bearing (much higher yield compared to younger men). This thinking is in strong contrast to past belief (and even now supported by some outdated recommendations of non-urological organizations), when PSA testing was not recommended generally for all older men.

#### **2.3 Digital rectal exam**

Digital rectal exam remains important part of prostate evaluation (Gosselaar et al., 2008). Although still on occasion declined by a patient and despite hopes PSA or any other method would completely replace it (Schroder et al., 1998), one still finds from time to time a case, where PSA is low and DRE is suspicious. For such a patient, next year, when PSA has increased, may be (and cases were confirmed it was) too late. Although after many suspicious DRE's biopsy comes negative, we are more and more aware of the fact that prostate cancer may progress without PSA increase. DRE can estimate sphincter tone, in a way prepare patient for biopsy, if needed, on occasion register some other, non-urological pathology and appreciate prostate size, which may give different dimension compared to ultrasound impression. DRE is necessary for sampling prostate cells in post-massage urine – at present for PCA3 only, but in the future maybe also for other markers. DRE is necessary for clinical staging of prostate cancer. DRE is limited, it can not palpate whole gland, but at present it is here to stay (Yossepowitch, 2008).

#### **2.4 PCA3**

Comparison of mRNA expression patterns of prostate cancer and benign tissues identified significantly different expression of non-coding mRNA sequence, first called DD3 (Bussemakers et al., 1999) and later renamed PCA3. Technology (a variant of quantitative nucleic acid amplification test) was developed to identify very minute amounts of PCA3 mRNA from prostate cells shed in prostate urethra after prostatic massage and washed out immediately in first voided urine portion. Its estimated amount (number of copies) is

Future of Prostate Biopsy: Who Will Get It and How? 107

It is accepted increased risk due to family history should result in earlier start of screening (at age 40 or 35). Some risk calculators do use family history information to modify risk estimation (increased risk for presence of cancer in first degree relatives, this means father, brother or son), which may influence biopsy decision, if based on calculated risk. But apart from this, it does not tell us anything – PSA and other cut-offs for decision about biopsy are the same as in men without family history. Also procedures for biopsy, treatment and also treatment outcomes are the same between familial, hereditary and sporadic prostate cancer. There is no known 100% risk situations, which would provoke preventive removal of

Recent research focused also on risk of prostate cancer in men whose family members have other forms of cancer. Men with first degree relative with breast cancer, who developed breast cancer at the age 35 or earlier, are at significantly increased risk of prostate cancer (5 to 18 times in case of BRCA mutation) (Dite et al., 2010). Other cancers in family may also

Race is used in many prostate cancer risk calculators – African American race increases risk compared to Caucasians, Hispanic or Asian have decreased risk. But, the same as for family history, apart from earlier start of screening or a little different risk estimates from

Previous negative biopsy decreases risk of positive subsequent biopsy. There are, however, two findings, which differ: PIN (=HGPIN – high grade prostatic intraepithelial neoplasia, low grade PIN was abandoned) and ASAP (atypical small acinar proliferation). HGPIN is most probable precursor of prostate cancer. There are structurally benign ducts and acini with abnormal secretory cells with prominent nucleoli (Schoenfield et al., 2007). HGPIN is a histological entity in itself, it is not cancer. ASAP indicates finding of small focus of structure in a 2-dimensional view, which is not large enough to satisfy criteria for cancer. It may represent only specific cut of top of benign acinus or PIN. One may define different characteristics under ASAP category, like adenosis, intraductal hyperplasia... which have in common that pathologist can not say for certain, whether they represent cancer or not. As ASAP may represent cancer, is was identified as predictor of positive repeat biopsy. Identification of multiple HGPIN in a biopsy specimen was also identified as predictor of positive repeat biopsy (Akhavan et al., 2007). Therefore, ASAP and HGPIN require consideration about repeat biopsies. ASAP may require immediate scheduling for repeat biopsy. HGPIN requires careful follow up and in most cases other criteria (PSA increase)

Different nomograms were developed, which can predict risk of positive biopsy. They take into account risk for age, family history, race, PSA and free PSA, digital rectal exam and some also for body mass index and even new decision aids, for example PCA3 and PHI. Nomograms have introduced new dimension in patient-physician pre-biopsy discussion. Need for individual's risk adjustment is illustrated by cardiology example. Do general practitioners and cardiologists decide whether one needs aspirin or statin on a single total cholesterol measurement (for example with result of 5.3 mmol/L, just above normal)? No. They measure HDL, LDL cholesterol, they ask for age, gender, blood pressure, smoking history. They consult nomograms and then decide whether statin drug is necessary (and reimbursed) or not. But urologists were dependent on a single PSA value? This has changed

prostate, as is known for breast cancer, although people sometimes do ask for this.

calculators, there are no other practical consequences of this fact at present.

cause repeat biopsy sooner or later (for example in 1-2 years time).

**2.6 Risk based strategies** 

increase risk of prostate cancer (Izmirlian et al., 2011).

normalized to estimated number of PSA mRNA copies in the same sample and multiplied by 1000, resulting in PCA3 score. Test is known as Progensa PCA3 Assay (Gen-Probe Inc, San Diego, USA). Higher PCA3 score values indicate greater risk of presence of prostate cancer. Values from approximately 4 to 125 can be obtained, 35 being most often suggested cut-off, which, according to package insert has 53% sensitivity and 74% specificity for detection of prostate cancer on subsequent biopsy. Other cut-off opinions are continuously evaluated (Auprich et al., 2011). Although PCA3 mRNA theoretically could be measured in urine also with standard molecular-medicine techniques, which first isolate mRNA from cells in the sample and then perform RT-PCR, is this, due to very small amounts of PCA3 mRNA, very difficult and inconsistent, therefore dedicated patented method from Gen-Probe which uses direct amplification and avoids isolation step, has become standard.

PCA3 use was suggested and marketed as help in biopsy decisions in previously negative prostate biopsies, in men with other risk factors (family history) and normal PSA values, in men with big prostates or other prostatic conditions where it is unclear whether high PSA is driven by cancer or other factor and in men on watchful waiting where high PCA3 score may indicate higher burden of disease.

Use of consecutive PCA3 tests for follow up of biopsy candidates or men on watchful waiting was questioned due to high variability of results, if test is repeated in the same individual with low risk disease (Shikanov et al., 2011).

Pre-biopsy PCA3 test results were evaluated together with PSA in relation to tumor characteristics in radical prostatectomy specimens (Vlaeminck-Guillem et al., 2011). It was found PCA3 score did not correlate with PSA and prostate volume. PCA3 score was also not related to Gleason score and pT stage. PCA3 did correlate with total tumor volume and index lesion volume. Further, multifocality was significantly correlated with PCA3. PCA3 expression can be identified on malignant and benign prostate tissue samples, only quantity varies (Quiles et al., 2011).

PCA3 test is, with many mentioned drawbacks and despite its high price, widely marketed and used in USA and EU. It seems sometimes may help in biopsy related decisions, but its role at present is still evolving.

#### **2.5 Family history, race, previous biopsy result**

Theoretically one may differentiate between familial (unpredictable clustering of disease in a family, slightly increased risk for offspring) and hereditary cancer (strong clustering pattern consistent with passage of a susceptibility gene via Mendelian inheritance) (Potter & Partin, 2000). Criteria for hereditary prostate cancer would be three or more first-degree relatives had prostate cancer, three successive generations had prostate cancer and where two relatives were affected at or before age 55. Despite identification of such families, their investigation has not resulted in such successful discoveries as in colon, ovary and breast cancers. In those classic hereditary cancers presence of a certain identifiable mutation with high probability predicts development of cancer in all affected offspring, which has resulted in formal screening programs for affected families and preventive surgery at early age. This is at present not known in prostate cancer.

Familial history of prostate cancer increases risk. In a meta-analysis brother with prostate cancer increases risk 3.3 times, father 2.2 times, two first-degree relatives 5 times (Zeegers et al., 2003).

normalized to estimated number of PSA mRNA copies in the same sample and multiplied by 1000, resulting in PCA3 score. Test is known as Progensa PCA3 Assay (Gen-Probe Inc, San Diego, USA). Higher PCA3 score values indicate greater risk of presence of prostate cancer. Values from approximately 4 to 125 can be obtained, 35 being most often suggested cut-off, which, according to package insert has 53% sensitivity and 74% specificity for detection of prostate cancer on subsequent biopsy. Other cut-off opinions are continuously evaluated (Auprich et al., 2011). Although PCA3 mRNA theoretically could be measured in urine also with standard molecular-medicine techniques, which first isolate mRNA from cells in the sample and then perform RT-PCR, is this, due to very small amounts of PCA3 mRNA, very difficult and inconsistent, therefore dedicated patented method from Gen-Probe which uses direct amplification and avoids isolation

PCA3 use was suggested and marketed as help in biopsy decisions in previously negative prostate biopsies, in men with other risk factors (family history) and normal PSA values, in men with big prostates or other prostatic conditions where it is unclear whether high PSA is driven by cancer or other factor and in men on watchful waiting where high PCA3 score

Use of consecutive PCA3 tests for follow up of biopsy candidates or men on watchful waiting was questioned due to high variability of results, if test is repeated in the same

Pre-biopsy PCA3 test results were evaluated together with PSA in relation to tumor characteristics in radical prostatectomy specimens (Vlaeminck-Guillem et al., 2011). It was found PCA3 score did not correlate with PSA and prostate volume. PCA3 score was also not related to Gleason score and pT stage. PCA3 did correlate with total tumor volume and index lesion volume. Further, multifocality was significantly correlated with PCA3. PCA3 expression can be identified on malignant and benign prostate tissue samples, only quantity

PCA3 test is, with many mentioned drawbacks and despite its high price, widely marketed and used in USA and EU. It seems sometimes may help in biopsy related decisions, but its

Theoretically one may differentiate between familial (unpredictable clustering of disease in a family, slightly increased risk for offspring) and hereditary cancer (strong clustering pattern consistent with passage of a susceptibility gene via Mendelian inheritance) (Potter & Partin, 2000). Criteria for hereditary prostate cancer would be three or more first-degree relatives had prostate cancer, three successive generations had prostate cancer and where two relatives were affected at or before age 55. Despite identification of such families, their investigation has not resulted in such successful discoveries as in colon, ovary and breast cancers. In those classic hereditary cancers presence of a certain identifiable mutation with high probability predicts development of cancer in all affected offspring, which has resulted in formal screening programs for affected families and preventive surgery at early age. This

Familial history of prostate cancer increases risk. In a meta-analysis brother with prostate cancer increases risk 3.3 times, father 2.2 times, two first-degree relatives 5 times (Zeegers et

step, has become standard.

varies (Quiles et al., 2011).

role at present is still evolving.

may indicate higher burden of disease.

individual with low risk disease (Shikanov et al., 2011).

**2.5 Family history, race, previous biopsy result** 

is at present not known in prostate cancer.

al., 2003).

It is accepted increased risk due to family history should result in earlier start of screening (at age 40 or 35). Some risk calculators do use family history information to modify risk estimation (increased risk for presence of cancer in first degree relatives, this means father, brother or son), which may influence biopsy decision, if based on calculated risk. But apart from this, it does not tell us anything – PSA and other cut-offs for decision about biopsy are the same as in men without family history. Also procedures for biopsy, treatment and also treatment outcomes are the same between familial, hereditary and sporadic prostate cancer. There is no known 100% risk situations, which would provoke preventive removal of prostate, as is known for breast cancer, although people sometimes do ask for this.

Recent research focused also on risk of prostate cancer in men whose family members have other forms of cancer. Men with first degree relative with breast cancer, who developed breast cancer at the age 35 or earlier, are at significantly increased risk of prostate cancer (5 to 18 times in case of BRCA mutation) (Dite et al., 2010). Other cancers in family may also increase risk of prostate cancer (Izmirlian et al., 2011).

Race is used in many prostate cancer risk calculators – African American race increases risk compared to Caucasians, Hispanic or Asian have decreased risk. But, the same as for family history, apart from earlier start of screening or a little different risk estimates from calculators, there are no other practical consequences of this fact at present.

Previous negative biopsy decreases risk of positive subsequent biopsy. There are, however, two findings, which differ: PIN (=HGPIN – high grade prostatic intraepithelial neoplasia, low grade PIN was abandoned) and ASAP (atypical small acinar proliferation). HGPIN is most probable precursor of prostate cancer. There are structurally benign ducts and acini with abnormal secretory cells with prominent nucleoli (Schoenfield et al., 2007). HGPIN is a histological entity in itself, it is not cancer. ASAP indicates finding of small focus of structure in a 2-dimensional view, which is not large enough to satisfy criteria for cancer. It may represent only specific cut of top of benign acinus or PIN. One may define different characteristics under ASAP category, like adenosis, intraductal hyperplasia... which have in common that pathologist can not say for certain, whether they represent cancer or not. As ASAP may represent cancer, is was identified as predictor of positive repeat biopsy. Identification of multiple HGPIN in a biopsy specimen was also identified as predictor of positive repeat biopsy (Akhavan et al., 2007). Therefore, ASAP and HGPIN require consideration about repeat biopsies. ASAP may require immediate scheduling for repeat biopsy. HGPIN requires careful follow up and in most cases other criteria (PSA increase) cause repeat biopsy sooner or later (for example in 1-2 years time).

#### **2.6 Risk based strategies**

Different nomograms were developed, which can predict risk of positive biopsy. They take into account risk for age, family history, race, PSA and free PSA, digital rectal exam and some also for body mass index and even new decision aids, for example PCA3 and PHI.

Nomograms have introduced new dimension in patient-physician pre-biopsy discussion. Need for individual's risk adjustment is illustrated by cardiology example. Do general practitioners and cardiologists decide whether one needs aspirin or statin on a single total cholesterol measurement (for example with result of 5.3 mmol/L, just above normal)? No. They measure HDL, LDL cholesterol, they ask for age, gender, blood pressure, smoking history. They consult nomograms and then decide whether statin drug is necessary (and reimbursed) or not. But urologists were dependent on a single PSA value? This has changed

Future of Prostate Biopsy: Who Will Get It and How? 109

Prostate cancer was in the past diagnosed most often using histology from transurethral resection, which was performed for local complications of prostate cancer or merely BPH. As it became clear such diagnosis is most often too late, makes subsequent radical surgical treatment more difficult and as PSA become available, biopsy had become diagnostic method of choice. At first, it was finger guided, blind. With development of transrectal ultrasound (trus – transrectal ultrasound guided prostate biopsy) and coupling it to 18G

Most prostate biopsies around the world are performed transrectally, as this naturally developed from transrectal finger guided approach, is cheaper, simpler, does not need any analgesia or anaesthesia (although it is highly recommended, see section 4.3). Transperineal approach does need at least local anaesthesia. On same occasions, like post surgical amputation of the rectum, if it is sutured, transperineal biopsy (or transgluteal) may be the only way. However, such biopsy is guided by US probe on the perineum or in the urethra or more often by CT or MRI, which is not the way transperineal biopsy is performed regularly. Regular transperineal biopsy is performed using probe in the rectum and needle is advanced through the skin above rectal opening. Transperineal approach has similarity with the way some treatments are performed (brachytherapy...). Treatment templates may be used for biopsy which may guide biopsy needle to a specific part of prostate, which is seen as potentially more effective and with better gain. Recently, stereotactic templates (even motorized and computer guided) have become available also for transrectal prostate sampling – mechanically assisted 3D TRUS guided biopsy system (Megwalu et al., 2008). Supporters of transperineal approach suggest transperineally apical and anterior parts of the prostate may be better accessed compared to transrectal approach. For those reasons, studies were designed to compare yields of transrectal versus transperineal biopsy approach in primary 12-core and saturation 20-core schemes (Hara et al., 2008; Abdollah et al., 2011). Those studies uniformly showed no difference in cancer yields. However, this has not convinced proponents of transperineal approach. In regions, where transperineal approach was used routinely and claimed to be better compared to transrectal (for example some parts of Italy), they are only lately discovering prostate biopsy with end-fire probe (Galosi et al., 2010). Due to technical development of ultrasound equipment, some manufacturers (also very respectful one), were using side-fire probe with needle "through-the-probe" technology. This was proved to be inferior (Chin et al., 2009) in comparison to end-fire biopsy approach (see below). Therefore comparison of perineal to side-fire transrectal approach is not appropriate. Combination of transrectal and transperineal approach has also been used and described, aiming at gaining best from both approaches and maximizing cancer

Transrectal approach is recommended for routine primary and also saturation prostate biopsy. On rare occasions, where stereotactic guidance may be needed, decision on approach depends on availability of equipment. At present, equipment for transperineal stereotactic approach is more often available and more widespread, due to its use for brachytherapy (and different other new experimental approaches, such as cryotherapy or photodynamic therapy), compared to new and only limited availability of transrectal

**3. Methods of biopsy** 

needle, Pandora box of prostate cancer was opened.

**3.1 Transrectal or transperineal approach?** 

detection rate (Kawakami et al., 2007).

with risk calculators availability. Using single PSA cutoff value would be the same as claiming for a man, who runs marathon, has normal blood pressure, normal BMI, is a non-smoker, his parents lived till age 100 and has cholesterol of 5.3 mmol/L – to be at high risk for cardiac event because his cholesterol is elevated (above 5.2). Using single PSA value for biopsy decision would be the same as claiming for an overweight (BMI 30), sedentary man, heavy smoker, on three anti-hypertensive medications whose parents both died of cardiac event, because his cholesterol is 5.1 mmol/L (just normal) – not to be at risk for cardiac event. Everyone can clearly see ridiculosity, absurdity of such approach in cardiology – but we in urology use PSA as single dichotomous variable in our prostate biopsy decisions?! Risk calculators allow us to tell men with any PSA value their individual chances for biopsy outcome. Calculators report risk for high grade (Gleason 7 or more) and overall cancer probability. For example if calculator predicts 6% chance of significant prostate cancer (Gleason 7 or more) and 26% of overall cancer risk, we should tell patient his chance of being without cancer at present is approximately 74% (100-26), he has 6% chance of finding important high risk cancer and he has approximately 20% (26-6) chance biopsy would find cancer which may not be clinically important at present (but maybe in the future) and we would have to deal with this. If we explain this way, most men can and do take their own decision, whatever it is (to proceed with biopsy or continue PSA follow up).

Using risk calculators often surprises us with values one would not expect. For example use Sunnybroke Prostate risk calculator (see below) and take 75 years old healthy Caucasian male with PSA of 3.9, %free PSA 22%, IPSS 9, in a good health (for example upper 25% of his generation), no family history for prostate cancer and negative DRE. He has more than 15 years of life expectancy, 39% risk of prostate cancer and 20% risk of having at present high grade prostate cancer (Gleason 7 or more). 20% risk of high grade cancer and more than 15 years of life expectancy - would you suggest a biopsy?

At present (june 2011), one of the most modern (can take into account PCA3, p2PSA...) prebiopsy nomograms is available from UT Health Science Center, San Antonio, head Ian Thompson (http://deb.uthscsa.edu/URORiskCalc/Pages/calcs.jsp). Opponents of his calculator state it has good estimation of overall cancer risk, but underestimates high risk cancers. Large urology collection of nomograms, also many pre-biopsy nomograms including estimation of biopsy risks, are available from Pierre Karakiewicz, Cancer Prognostics and Health Outcomes Unit from University of Montreal Health Centre, Canada (www.nomograms.org). Calculators derived from European Randomized study of Screening for Prostate Cancer (Fritz Schroeder) are available at SWOP Prostate Cancer Research Fundation: http://www.prostatecancer-riskcalculator.com/via.html. Useful, simple and modern (more accurate high risk cancers estimation) pre-biopsy nomogram is available from SunnyBrook Health Sciences Centre, Toronto: http://sunnybrook.ca/ content/?page=OCC\_prostateCalc.

Different nonograms, for example taking into account 3D biopsy approach (Sakura et al., 2011) are being developed and published constantly. Also, comparisons (Oliveira et al., 2011) and evaluations (Ngo et al., 2011) are permanently performed and published. When using nomograms, one should be aware of their limitations (Vickers & Cronin, 2010).

NCCN guidelines on prostate cancer early detection not require nomogram use, follow in part risk adapted approach, cover most "before biopsy" issues, are regularly updated and modern. They are available freely after registration from www.nccn.org.
