TUMOUR MARKERS IN PROSTATIC CANCER

Scand J Clin Lab Invest Downloaded from informahealthcare.com by Bausch & Lomb IOM on 01/13/15 For personal use only.

EDWARD H. COOPER Diagnostic Development Unit,Department of Chemical Pathology & Immunology, University of Leeds LS2 9JT. UK. Prostate cancer is now the third commonest cancer in men. Extensive clinical trials comparing acid phosphatase, alkaline phosphatase (ALKP) and prostate specific antigen WA) have shown that PSA is the most sensitive and specific of the tumour markers available for prostate cancer. Caution is needed when comparing the results from different assay methods, there is no international standard for PSA. In the management of localised disease, radical treatment can reduce the PSA levels to c0.4 ng/ml, similar results can be obtained for a varying duration in patients sensitive to androgen withdrawal. Raised levels >0.4 ng/ml after radical prostatectomy are indicative of residual disease. PSA is valuable in monitoring deferred treatment or the effects of hormone manipulation and give an indication of the prognosis and early warning of recurrence. In extensive metastatic disease the combination of PSA and ALP reflects the tumour activity. Less than 15 hot spots on the wintigram at presentation and a PSA c10 ng/ml3 to 6 months after commencing treatment is associated with prolonged survival. The role of PSA in population screening for early prostatic cancer is uncertain; early results suggest it can be used in combination with digital rectal examination and ultrasonic examination of the prostate. The effect of a PSA decision level at 4 or 10 ng/ml has a considerable influence on the pick up rate. Key words: Tumour markers, prostatic cancer, alkaline phosphatase, prostatic specific antigen

INTRODUCTION The decade has seen a marked increase of interest in prostatic cancer and its laboratory investigation. Prostatic cancer has become the most commonly newly diagnosed cancer in men. In the United States it is the cause of death of 28,000 men per year, accounting for 11 percent of cancer deaths, the third in men after lung and colon cancer (l), similar figures are found in Europe (2). During the decade a great debate has taken place to try to reach a concensus as to the optimum treatment of early disease; that is, when the cancer is contimed within the prostate capsule. Radical prostatectomy, radical radiotherapy, and deferred treatment, until there are symptoms or evidence of progression, all have their advocates (3). The spectrum of anti-androgen treatments for more advanced disease has shifted, orchidectomy still holds the centre ground whilst LH-RH agonists, cyproterone acetate and flutamide have all been shown to have beneficial effects in individual patients, on the other hand the use of oestrogens has tended to decline. No treatment has emerged as being statistically superior. Androgen suppression can only cause a remission of prostatic cancer, sometimes for several years, so that the patient may eventually die of other diseases, in other patients hormone resistance will eventually produce symptoms of bone pain, anaemia and weight loss and death from the effects of the cancer.

42

Scand J Clin Lab Invest Downloaded from informahealthcare.com by Bausch & Lomb IOM on 01/13/15 For personal use only.

average age at presentation, with signs of prostatism, bone or perineal pain, is over 70 years old, many patients are over 80, and uncommon below 60 years old. Hence, each clinical decision about patient management is influenced not only by the pathological stage of the disease and the biochemical information but most importantly by the medical, psychological and social status of the patient. This means that any rules that emerge for the interpretation of biochemical changes in prostate cancer, especially of the more sensitive tests, may have to be modified according to their clinical context. It is not surprising that with the heightened interest in the management of prostatic cancer that any tumour markers that went a long way to provide the clinician with valuable information about the disease would be adopted widely. At the outset enzymatic measurement of acid phosphatase and alkaline phosphatase (ALKP)were the well tried standard laboratory tests. Whilst ALKP was a valuable indicator of disease activity in metastatic disease involving bones, acid phosphatase lacked the sensitivity and specificity to make it useful for the management of limited disease. The introduction of immunometric assays for prostatic acid phosphatase (PAP) (4-7)led to an improvement of specificity and uniformity, but few clinical chemists were convinced that the gains were sufficient to warrant the extra cost of the immunoassay. Shortly after PAP was introduced Wang et al (8) in 1W9 described the isolation of prostatic specific antigen (PSA), and its clinical potential, as a tumour marker in prostatic cancer was soon confirmed. In a few years the use of PSA has become widespread, largely at the insistence of the urologists and oncologists. The measurement of prostatic acid phosphatase has largely become redundant so that today many laboratories and clinical services rely solely on PSA and ALKP for the msessment and monitoring of patients with prostatic cancer. As this is the present trend, this review will be restricted to these two tests, the literature contains comparisons of PSA and the enzymatic and immunological measurements of PAP, all showing the increased sensitivity of PSA compared to PAP in low stage disease (9-11). This period of evolution of the laboratory tests has been reviewed by Pontes (12) and Ming Chu (13).

PROSTATIC SPECIF'IC ANTIGEN

Purified PSA is a glycoprotein, molecular weight 33kd, that is distinct from PAP. The 240 aminoacids have been sequenced (14) and shown to be a member of the serine antiprotease family. PSA is a normal constituent of seminal fluid and plays a role in the liquefaction of the post-ejaculatory seminal gel (15). PSA is tissue specific and found in normal, hyperplastic and malignant prostatic tissue, but is not cancer specific. Several commercial assays are available for the measurement of serum PSA, the fmt was Hybritech's two site hunoradiometric assay using two monoclonal antibodies, others include a conventional competitive-inhibition type polyclonal double-antibody radioimmunoassay (Diagnostic Products, Yang Laboratories) and an even wider choice in Europe. The rapid expansion of the use of PSA has been very attractive to commercial kit manufactures but has lead to confusion as the cut-off values vary between kits and the positivity varies from kit to kit. For example the Pros-Check PSA produces values 1.51.8 times higher than the Hybritech assay (16,17), others such as the Delfia PSA and CIS (new version) give results close to the Hybritech values. Clearly an international standard for PSA is urgently needed. The Hybritech assay is the most widely used but high dose hook effects have been reported in a few sera with very high levels (18); dilution of samples giving inappropriately low results are advisable. The upper limit of normal levels for the Hybritech and Pros-Check assays are 2.9 and 2.5 ndml respectively. The half life in the serum is about 2 days (19). Cut-off values: There is uncertainty as to what value should be considered as abnormal. There is little difficulty when the normal population is a healthy group of men, such a blood donors, here the 95th percentile can be readily determined. But in the routine hospital practice the average age of a pztient with prostate cancer is over 70 years old, the incidence of benign

43

Scand J Clin Lab Invest Downloaded from informahealthcare.com by Bausch & Lomb IOM on 01/13/15 For personal use only.

prostatic hyperplasia (BPH) is much higher at this age than in blood donors. For this reason a series of decision values for PSA have been suggested. In practice when a patient is seen for the fust time a relatively high cut-off may be used as a warning of the likelihood of cancer, say 10 ng/ml, as the clinical information is more refmed by ultrasonic and biopsy; a lower level may be adopted to give information about the probable stage. Once treatment has been initiated then the decision cut-off may be below 0.5 ng/ml following radical prostatectomy. Transient elevation of PSA for one or two days can follow transurethralprostatectomy, the effects of rectal examination on serum PSA values are less certain.

CLINICAL APPLICATION Benign prostatic hyperplasia: The levels of PSA in BPH reported from the USA tend to be lower than those reported in European series, this is probably a reflection of the difference in organisation of medical practice. In the UK Fen0 et al (20) found that 32.5% of 40 men with BPH had a PSA >10 ng/ml, and Armitage et al(21) found 11/91 (12%) with BPH admitted for elective surgery and 16/30 (53%) with acute retention had a PSA >lhg/ml, Allhoff et al (10) found 10% of 696 patients with BPH had a PSA >10 ndrnl. By contrast, series from the USA, for example Hudson et al (22) and Lange et al (23), reported 5/168 (3%) and 8/352 (2%) respectively with levels >10 nglml. It is now well recognised that there is a correlation between the size of the BPH and the level of serum PSA, so that high values, often >10 ng/ml, will be found in patients with large benign prostate glands. Untreated prostatic cancer: The most accurate correlationbetween pre-operative PSA level and stage in non-metastatic disease is from the series reported on radical prostatectomy as operation provides accurate assessment of the local spread of the tumour and its involvement of regional lymph nodes, the values observed by Stamey et al(19). using the Pros-Check assay are shown in Table I. It can be seen there is overlap between stages with the mean rising with increasing stage. Lange et al(25) found in 100 patients undergoing radical prostatectomy, even at levels of 10 ng/ml, the positive and negative predictive values (78 and 61 percent respectively) to predict extracapsular disease were not sufficient to make the test useful alone for staging. Gram for gram, the average prostatic cancer produces at least 10 times the amount of PSA that is produced by BPH tissue (19); an important consideration when comparing the gland volumes estimated by ultrasound and the PSA levels. high values in patients with near normal sized glands are highly suggestive of cancer. With increasing tumour burden the level of PSA tends to rise so that a level of >4Ong/ml has been suggested to be suspicious of extracapsulardisease (26). The levels of PSA in patients with metastatic disease involving the skeleton are over a very wide range from near normal to many thousands of ng/ml. The lowest values tend to be in asymptomatic patients with a single hot spot discovered by scintigraphy. A PSA decision level of 50-100 ndml is often taken as an indication of probable bone metastases in a patient presenting for the first time and an indication for a bone scan. Deferred treatment of prostate cancer: In a study of 88 patients with localised disease who were followed for 6-194 months, (median 34 months) from diagnosis without hormone manipulation, evidence of disease progression occurred in 15 patients. A stratificationof the first PSA level in the progressors and non-progressors is shown in Table 11. In 78 patients the rates of change of PSA could be calculated, the higher the initial PSA the greater the rate of change; for example, patients presenting with a level 20 nglml it was 2.42 (-2.97 +16.1) ng/ml/month. The rates of increase in those who progressed was significantly higher than in patients who did not (21). This suggests that the initial PSA level can provide prognostic information in patients

44

who are to be managed by deferred treatment and the risk of progression within three years is increased in patients with a PSA >20ng/ml.

Table L PSA in 230 untreated patients with prostate cancer Cancer stage

No Pts

PSA (nghnl)

Scand J Clin Lab Invest Downloaded from informahealthcare.com by Bausch & Lomb IOM on 01/13/15 For personal use only.

Mean + SD*

A1

13

3.1 + 0.7

0.0 - 8.6

A2

14

12.1 + 5.1

0.6 - 7.0

B1

43

12.3 + 4.1

1.6 - 39.8

B2

47

25.0+ 4.9

2.3 - 230

B3

21

39.7 + 11.2

1.3 - 237

C

36

102 + 28.4

4.6 - 830

D1

21

100 + 23.6

10.2 - 454

D2

35

562

+

29.7 - 2093

103.7

From Stamey and Kabalin (24) *Pros-Chek PSA assay

Table II. Stratificationof initial PSA level in 88 patients with untreated prostate cancer

Non-progressors

Progressors

10.1 -20

20.1 - 50

>50

Tumour markers in prostatic cancer.

Prostate cancer is now the third commonest cancer in men. Extensive clinical trials comparing acid phosphatase, alkaline phosphatase (ALKP) and prosta...
660KB Sizes 0 Downloads 0 Views