Prostate cancer tumor markers

Prostatic cancer represents an important health problem worldwide, being the third most common malignancy in men older than 55 years, aft er pulmonary and colon cancer. In early stages prostatic cancer remains hard to be diagnosed at clinical or imagistical studies, the most effi cient way to detect it being based on the determination of tumor markers. This is a literature review of the largest studies conducted on the theme of prostatic tumor markers.


INTRODUCTION
In the world, in terms of frequency, the prostati c cancer is the third form of cancer in men being more than 55 years of age, aft er the pulmonary cancer and the colon cancer. In terms of incidence, this is higher in the older persons, 60% of the newly diagnosed cancers being encountered in persons of more than 70 years old (1). The Gleason staging system is used in order to improve the diagnosis and the management of the prostate cancer. It has four stages. Stage A refers to tumors detected only microscopically.
Stage B refers to tumors which can be macroscopically detected. Stage C refers to tumors characterized by an extracapsular invasion, but without clinical expression. Stage D refers to metastati c tumors. As we can see, in an early stage, the prostate cancer can't be detected only by the digital transrectal examinati on. The transrectal ultrasonography has sensiti vity and a specifi city of about 20-30% in these cases and the computer tomography and the magneti c resonance imaging have low sensiti vity. This is the reason why the use of tumor markers is so important in this type of malignancy (1).

PROSTATE SPECIFIC ANTIGEN
PSA (prostate specifi c anti gen) is a serine protease produced by the prostate gland. It has a role in the liquefacti on of the seminal coagulum. It has fi ve isoforms: A and B are enzymatically acti ve, C, D, and E are inacti ve. Only small quanti ti es of PSA are found in the blood of healthy individuals. The enzymati cally acti ve forms are bound to the serum anti proteases α-1 anti chymotrypsin (α-1 ACT) and α-2 macroglobulins (α-2 MG). The unbound inacti ve forms can be quanti fi ed in the blood by means of the free PSA immunoassays (2).
The low specifi city of PSA for the detecti on of prostate cancer is due to various benign conditi ons (nodular hyperplasia of prostate, adenocarcinoma of prostate and prostati ti s) and situati ons associated with increased PSA levels like the physical sti mulati on of the gland: biopsy, radical prostatectomy, ejaculati on, catheterizati on, transrectal ultrasonography and digital rectal examinati on (3).
The serum PSA levels found in benign prostati c hyperplasia and the early stages of prostate cancer are diffi cult to be separated by a cutoff value (2).
The serum total PSA concentrati ons are highly correlated with the tumor volume and the disease stage (1,4). According to literature data, the pati ents diagnosed with stage A prostate cancers have PSA concentrati ons between 0 and 70 ng/ ml, pati ents diagnosed with stage B prostate cancer have PSA concentrati ons between 1.3 and 237 ng/ml, pati ents with stage C have levels between 4.6 and 830 ng/ml and pati ents with stage D have levels between 10.2 and 2093 ng/ml (4).
Other parameters with a higher specifi city than PSA are: PSA density (rati o of PSA to prostate volume), PSA velocity (change of PSA over a ti me period) and the age-specifi c reference ranges (3,5). The determinati on of PSA density is made by means of the determinati on of total PSA and of the prostate volume by transrectal ultrasound. The two parameters are well correlated in the benign prostate hyperplasia, but in pati ents with prostate cancer, the total PSA levels can be much higher than we would expect on the basis of the prostate volume. The PSA velocity is determined measuring the total PSA levels at a given ti me interval. In pati ents with benign prostati c hyperplasia, there is a constant increase rate over ti me, whereas in pati ents with prostate carcinoma we can fi nd an accelerated increase rate (5).
As for the correlati on between total PSA and age, literature data show that the cut-off value of 4 ng/ml is more appropriate for the men in their 60s, while a concentrati on which is higher than 2.0 ng/ml or 3.0 ng/ml is not usually encountered in men in their 40s or 50s. It has been proposed that these pati ents be monitored every 3-6 months by means of the PSA velocity, using a cut-off of 1ng/ml per year (5).
Of the total PSA in the plasma, 65-95% is bound to some proteins and the remaining quanti ty, named free PSA and is used as an additi onal marker in order to discriminate between benign and malignant causes of increased total PSA levels (3).
The percentage of free PSA has a good specifi city in disti nguishing between a benign and a malignant cause of total PSA increase (2).
A study realized in 2000 included 179 pati ents, 129 diagnosed with nodular hyperplasia, having a mean age of 71.5 years (53-90 years) and 50 with adenocarcinoma of the prostate, having a mean age of 67.2 years (51-80 years) (2). The serum determinati ons of total and free PSA levels were done by means of a solid phase two-site chemiluminescent enzyme immunometric assay on Immulite Automated Analyzer. The results showed that a large number of pati ents having a diagnose of a benign or malignant disease had total PSA values in the 5.0-20.0 ng/ml interval.
This latest diagram shows that a small number of pati ents having early prostate cancer or benign hyperplasia have free PSA percentages in the same interval. Using the Receiver Operating Characteristi c (ROC) curves, the sensiti vity of the total PSA for the detecti on of adenocarcinoma was 74% and the sensiti vity of the percent free PSA at a cutoff of 15% was 96%. Both values were determined for the same specifi city. An increased sensiti vity in diff erenti ati ng benign from malignant tumors reduces the number of unnecessary biopsies (2). Literature data showed that from the individuals with PSA levels less than 4 ng/ml, only a small number suff er unnecessary biopsies (6).
Another study realized by Christensson et al. included 121 pati ents diagnosed with prostati c adenocarcinoma and 144 pati ents diagnosed with nodular hyperplasia. The fi rst group was characterized by a signifi cantly lower free PSA percentage than the second group. The cutoff value for the free PSA percentage in the detecti on of malignancy was 18% and the sensiti vity was 90% (7).
Establishing a cut-off of 20% for the percent of free PSA, Launderer et al found a sensiti vity of 88% for the detecti on of malignancy (8).
Lilja et al. realized a study and found that the pati ents diagnosed with prostate cancer had a mean free PSA percentage of 18%, much lower than the mean percentage of the pati ents with benign prostati c hyperplasia (9).
In their practi ce, some laboratories consider that a free PSA percentage of less than 10% is highly suggesti ve of prostati c cancer and a percentage higher than 20% reduces the probability of prostati c cancer, but doesn't eliminate it (5).
A multi -center trial realized by Catalona et al. included 773 individuals with negati ve digital rectal examinati ons and total PSA values between 4.0 and 10.0 ng/ml. Following the ultrasound guided needle biopsy of the prostate gland, 19% proved to have cancer. These pati ents had higher total PSA levels and significantly higher free PSA percentages (10).
Other literature data sustain that from the pati ents with total PSA levels in the interval of 4-25 ng/ml and with palpably normal prostates, only 25% of the prostate biopsies led to the detecti on of cancer (11).
PSA has also some precursor forms named proPSAs, which can be used for the diagnosis and management of the early prostate cancer. One of them is [-2]proPSA . It is considered to be very cancer-specifi c and has very increased concentrati ons in pati ents with prostate cancer. A high specifi city characterizes the %[-2] proPSA ([-2]proPSA/free PSA [fPSA] × 100) (12).

PROSTATE H EALTH INDEX
Another parameter which has the same uti lity as free PSA is the prostate h ealth index [phi; (-2) proPSA/fPSA×√PSA] (3). It has demonstrated its superiority over total PSA and free PSA in the detecti on of prostate cancer pati ents and in the predicti on of an aggressive course of the disease (12).

PCA3 -PROSTATE CANCE R GENE 3
Another test used for the determinati on of the benign or malignant character of a prostati c aff ecti on is the determinati on of the urine PCA3, a non-coding messenger RNA, which is highly sensiti ve and doesn't depend on the prostate volume, age or total PSA (3). It is used for the detecti on of low-volume disease and pathologically insignifi cant prostate cancer, but its uti lity for the detecti on of locally advanced disease and aggressive cancer is limited. The detecti on of the PCA3 overexpression may have a therapeuti c role, as it is demonstrated by the PCA3-Gene-Viro Therapy. Another uti lity of this tumor marker is in establishing the biopsy indicati ons (13).
Ultrasensiti ve PSA is a test used for monitoring pati ents aft er surgical resecti on of the prostate. Most PSA tests can determine concentrati ons higher than 0.2 ng/ml. Ultrasensiti ve PSA can detect smaller concentrati ons, even up to 0.02 ng/ml. The determinati on is very important for the detecti on of the recurrence or metastasis and several measurements can be also useful for this purpose (5).
For the future diagnosis of prostati c cancer, the detecti on of the circulati ng prostate cancer cells and the profi ling of microRNAs may be used (3).

CONCLUSIONS
Prostati c tumor markers such as total or free PSA, PSA density, PSA velocity associated with other more recent and more specifi c parameters such as prostate health index and even the associati on of microRNA profi le are extremely important in order to early detect prostate cancer and to assure a proper follow up aft er surgically treated prostati c cancer.