Indian Journal of Urology
: 2000  |  Volume : 17  |  Issue : 1  |  Page : 6--12

The significance of early changes in serum alkaline-phosphatase levels following endocrine treatment in patients with advanced prostate cancer

Aditya Pareek 
 Pontefract General Infirmary, Pontefract WF8 IPL, England, U.K

Correspondence Address:
Aditya Pareek
Department of Urology, Suite 136, MRC 1st Floor, Bombay Hospital Institute of Medical Sciences, New Marine Lines, Mumbai - 400 020, India


Objectives: The prognostic significance and the fre­quency of occurrence of serum alkaline-phosphatase fare in patients with advanced prostate cancer commenced on hormonal treatment is controversial. Present study was carried out to clarify its prognostic value and the frequency of occurrence in relation to progression and survival in these patients. Methods: Case notes of 168 (101 metastatic and 67 lo­cally advanced) patients with advanced carcinoma of the prostate diagnosed between July 1984 through December 1995 were analysed retrospectively. All cases were fol­lowed until death or until March 1997 when the cases were censored. Serum alkaline-phosphatase values ob­tained immediately prior to and within 6 weeks of initia­ting the hormonal therapy were analysed in relation to progression free and overall survival. Results: Of the total of 168 evaluable cases, 101(60.1 %) had metastatic and 67 (39.9%) locally advanced tumours. Mean pre-treatment serum alkaline-phosphatase levels were higher (350IU/1) in the group which exhibited fare phenomenon post-treatment than the other which didn«SQ»t (180 IU/1). Overall, flare activity in SAP levels was ob­served in 59 (35.1 %) of the total cases. Also, a greater proportion of the cases with metastatic disease had SAP flare than the other who had locally advanced disease (52.4% verses 8.9%). Flare in SAP levels were associated with decreased progression free and overall survival (p=0.001; p=0.002); greater the rise in SAP levels post­treatment, worse the prognosis. Conclusions: A significant proportion of patients with advanced carcinoma of the prostate have flare in SAP le­vels soon after initiating the treatment. We confirm that the flare activity in SAP levels is a negative prognostic indicator and it is possible to identify early treatment fail­ures by frequent measurements of SAP levels during first 6 weeks of commencing the therapy. The significance of SAP fare and androgen resistance is discussed.

How to cite this article:
Pareek A. The significance of early changes in serum alkaline-phosphatase levels following endocrine treatment in patients with advanced prostate cancer.Indian J Urol 2000;17:6-12

How to cite this URL:
Pareek A. The significance of early changes in serum alkaline-phosphatase levels following endocrine treatment in patients with advanced prostate cancer. Indian J Urol [serial online] 2000 [cited 2021 Jan 24 ];17:6-12
Available from:

Full Text


The incidence of carcinoma of the prostate (CAP) is reported to be increasing along with age-specific mortal­ity from this disease [1],[2] which, together with an expanding ageing population make it an important health and socio­economic problem. It is the leading cause of cancer re­lated deaths (35.000-40,000 deaths per year) in USA where 103,000 new cases were diagnosed in 1989 and 140,000 in 1992. [3],[4] In European Community it causes 35,000 deaths annually. [5] In the UK, about 11,000 new cases of CAP are diagnosed and 8,000-9,000 of them die each year. [6],[7]

Increasing awareness of the malignant potential, [8] and from the late eighties, the widespread availability and use of the PSA are now responsible for an early diagnosis of this malignancy. Nevertheless, a significant proportion of the patients still present when the tumour is already well advanced locally or systemically. [9],[10] Primary mode of treat­ment in advanced disease has remained some form of an­drogen deprivation since early forties. [11] However response to the first-order hormonal manipulation is unpredictable and varies widely from as low as 30% to as high as 80%. [12],[13],[14],[15],[16],[17] Thus, a significant proportion of these tumours are pri­marily resistant to endocrine treatment. This has been explained by environmental selection model described by Isaac and Coffey in 1981. [9] Even those which respond fa­vourably initially, the majority of these usualy relapse in 18-36 months. [14] This has been explained by mutations [18] and amplification [19] in the androgen receptor gene. Once relapsed, most of these patients die within a few months. [20],[21],[22],[23],[24] Thus, it is highly desirable to identify as early as possi­ble those patients who derive no benefit or who are soon likely to become refractory to this form of treatment. The serum prostate specific antigen (PSA), serum prostatic acid phosphatase (PAP) and serum alkaline-phosphatase (SAP) are most widely used biochemical tumour markers. How­ever, the sensitivity, specificity and cost and convenience of measurement of these markers varies significantly.

Serum PSA nadir of 3-6 months and the rate of decline in PSA values over a few weeks are reported to identify the group of the patients who are at risk of treatment fail­ure. [25],[26],[27],[28],[29] However, the serum PSA measurements are rela­tively expensive and no more sensitive or specific than SAP especially its bone isoenzyme form [30],[31] in monitor­ing the metastatic disease. In fact, the value of serum PSA in monitoring disease progression has been seriously questioned. [31],[32],[33],[34]

The serum alkaline phosphatases are non-specific phosphomonoesterases and exist in multiple isoforms, being derived from various tissues including osteoblasts, hepatocytes, intestinal cells and placenta. [35] SAP levels are easily measurable, relatively inexpensive and readily avai­lable in most laboratories. Several reports have confirmed the value of SAP as an independent prognostic indicator in advanced carcinoma of the prostate. [13],[30],[31],[36],[37] Also, changes in serum levels of alkaline phosphatase and other bone markers following endocrine treatment of metastatic CAP have been described and correlated with clinical re­sponse.

Increase followed by decline in SAP levels with hor­monal therapy of metastatic CAP is known since the days of Huggins and Hodges. [38] This so called flare phenome­non in SAP levels, also seen as apparent deterioration of metastatic lesions in bone scans is not unique to prostate cancer. It is known to occur in metastatic breast cancer as well where, as in prostate cancer, it is believed to repre­sent successful therapy. [39] In fact, most of the early reports have suggested that the flare phenomeon, although un­common, signifies healing of the osseous metastases following treatment. [39],[40],[41]

On the contrary, a recent and relatively large study by Pelger et al [25] has reported that the flare in SAP activity is a frequent event and the magnitude of flare correlates with survival negatively. If the SAP flare has any significance with respect to treatment in advanced prostate cancer, it could help monitoring these patients and planning the need, if any, of second line or alternative therapeutic strategies at an early stage. It is even more important to identify early treatment failures considering the concept of primary androgen resistance. [9] In an attempt to resolve these con­troversies, and in an attempt to identify value of SAP flare in early treatment failures, a retrospective analysis of the patients with advanced carcinoma of the prostate was car­ried out at this institution.

 Materials and Methods

Case notes of the patients who presented with CAP bet­ween July 1984 through December 1995 were analysed retrospectively. All were followed up until death or until February 1997 when the cases were considered censored. Cases were considered eligible for the study based on the following criteria:

1. Metastatic or locally advanced disease was demon­strated by imaging (bone scans, skeletal X-rays, TRUS, abdominal US/CT).

2. Cases were treatment-naive at entry.

3. Serum alkaline-phosphatase values were available immediately prior to and at least one value within the first six weeks after initiation of the treatment. Also at least two more values were available in the subsequent three to six months.

SAP levels were measured in subsequent follow-ups as well, usually three to six monthly.

In addition, all patients had haemoglobin, ESR, serum testosterone, total and prostatic acid-phosphatase, gamma­GT, bilirubin and creatinine measured at first and on fol­low-up visits. Since PSA measurements became available only in late eighties, not all patients had this marker meas­ured.

4. Clinical and imaging parameters used in monitoring were:

(a) Bone pains categorised into 3 grades:

(i) mild, (ii) moderate, (iii) severe.

(b) Amount and nature of analgesia.

(c) Performance status (WHO criteria).

(d) Imaging bone scans, skeletal X-Rays, abdominal US/ CT which were done at entry and at follow-up intervals usually 3 to 12 monthly.

Disease was considered progressive when at least two criteria (clinical and/or imaging) showed deterioration independent of biochemical markers.

On the basis of the SAP levels post treatment, the cases were grouped, as previously described, [25] as follows:

I. No flare, i.e., patients in whom SAP levels remained within laboratory reference range during first six weeks after initiating the treatment and following 3 months.

II. Cases with a rise in SAP level above normal but un­der 1½ times the basal value within the first 6 weeks.

III. Cases with a rise in SAP levels above normal be­tween 1½ to 2 times the basal value within the first 6 weeks.

IV. Cases with a rise in SAP levels over twice the basal values.

Based on these criteria, 168 cases were found to be evaluable.

Survival analysis and comparison between different subgroups was done using actuarial (life table) method and Wilcoxon (Gehan) test of statistics with a statistical software package on a personal computer.


Of the 158 evaluable cases 101 (60.1%) had metastatic disease and 67 (39.9%) had locally advanced tumours at entry. Mean age was 72 years (range: 51-95 years). Fol­low-ups ranged from 15 months to 148 months (mean: 45 months). All uncensored cases were followed up until death. Histologically, well, moderately and, poorly differ­entiated tumours comprised 62 (36.9%), 66 (39.3%) and, 40 (23.8%) of 168 cases respectively. Nature of andro­gen-suppressive treatment included bilateral orchidectomy, 36 oestrogen, [24] anti-androgens, [47] Gn RH analogues with or without anti-androgen [37] and Estramustine. [24]

Overall, pre-treatment SAP levels ranged from 30-4,000 IU/L (mean 189). Baseline SAP levels were normal in 97 (57.7%) of the 168 cases. As expected, the majority of the patients (56 out of 67 or 83.5%) with locally advanced no metastatic disease had normal baseline SAP. Mean pre­treatment SAP levels were higher (350 IU/L) in the group which exhibited flare phenomenon subsequently than those which did not (180 IU/L). Logistic regression analysis of the baseline levels and post-treatment rise in SAP levels revealed a significant relationship (P = 0.04).

Rise above the normal reference range in SAP values within 6 weeks after initiating hormonal treatment was seen in 59 (35.1 %) of 168 cases. This was sustained only in 8, in the rest, 51 cases, SAP values declined progres­sively towards baseline in the following 3 to 6 months. Increase above normal reference range in SAP less than 1½ times, between 1½ and 2 times and over 2 times the basal values as observed in 20, 19 and 20 cases respec­tively. SAP levels remained within normal reference range in 109 (64.9%) out of 168 cases after commencing the treatment. Flare reaction was seen in 6 out of 67 (8.9%) cases with locally advanced tumours and 53 out of 101 (52.4%) cases with metastatic disease.

The time to disease progression and overall survival according to the magnitude of the rise in SAP levels post­treatment are shown in [Table 1],[Table 2] and [Figure 1],[Figure 2].

Median overall survival in the groups exhibiting flare and no flare activity post-treatment were 38.4 and 54.34 months respectively. This difference is statistically sig­nificant (P = 0.003). Also there is a significant difference in the time to disease progression in these groups as well (P = 0.001). Pairwise comparison of the 4 groups [Table 1] shows the group exhibiting minimal flare (less than 1 1/2) did not differ significantly from the one which showed no flare in terms of overall survival (P = 0.5). Although the overall survival appears to be shorter with greater degrees of flare activity, it only reaches statistical significance when the group with maximum flare (greater than 2 times) is compared with the group with no flare at all (P = 0.0002).

Median months to disease progression were also shorter in the groups exhibiting flare post-treatment. Again most significant difference was seen between the group exhibit­ing no flare and the one which exhibited the greatest (greater than 2 times) flare activity (P [13],[25],[30],[31],[36],[37] of the value of SAP and the value of the flare phenomenon in monitoring the patients with ad­vanced CAP. Frequency of occurrence of this phenom­enon varied widely among various reports. It is possible that in those studies which observed this phenomenon in schintigrams done 3 months after initiation of the treat­ment , [41] the flare activity occurring earlier than 3 months may have been missed. It is because the flare activity tends to dissipate in the majority of patients within a few weeks [25] as seen in the present study as well. On the contrary, how­ever, a higher proportion of the patients with metastatic carcinoma of the breast seem to show prolonged or late (greater than 3 months) scintigrafic flare, although already raised SAP levels tend to decline before that time. [39] Whether this diference between breast and prostate can­cer is due to the nature (osteolytic vs. osteoblastic) or to the magnitude of the metastatic burden, is not known. Very high proportions of the cases (87%) showed flare activity in the study reported by Pelger et al . [25]

In the present study, the SAP flare occurred only in 59 out of 168 (35.1 %) cases. Relatively low frequency of this phenomenon in this study is possibly because of the in­clusion of a relatively higher proportion of cases with lo­cally advanced non-metastatic tumours. Fewer (16.4%) of this group of patients had raised SAP pre-treatment.

Since this study, and several others, [25],[40] show a significant relationship between pre-treatment SAP and the occur­rence of `flare' post-treatment, it is not surprising that the cases without metastatic disease contributed little in the total of cases showing `flare'. However, the flare activity occurring in those without obvious metastases (6 of 67=8.9%) may have been due to occult micrometastases not detectable by imaging.

From the results of previous studies, it was hypothesised that the `flare' represented healing of the osseous meta­stases and successful systemic therapy. [39],[40],[41] Data in the pre­sent study contradict this view but agree with those reported by Pelger et al [25] that the flare phenomenon in these patients indicates poor prognosis, both in terms of progression as well as survival. There are limitations in the present study including the retrospective nature and case selection because of the unavailability of all para­meters required according to the study criteria. Results in the evaluable cases (168) are quite clear, however.

Serum osteocalcin and SAP are considered to be the markers of osteoblastic activity and a rise the serum levels of these would be expected to represent a positive response to the therapy. Flare response in serum osteocalcin is seen concordant with SAP flare . [39] Patients with osteomalacia treated with vitamin-D are also known to show SAP flare. [42] However, SAP is not a true osteob­lastic marker since it is increased (along with I CTP) in all patients with osteolytic metastases compared to those with sclerotic lesions alone. [43] It is also increased in osteoblas­tic metastases, and there appears to be no correlation be­tween the metastatic mass and SAP levels. [31],[35] Thus, it is difficult to justify the value of SAP either as osteoclastic or osteoblastic marker other than that it is raised in meta­static disease. The results of the present study and of the one reported by Pelger et al [25] appear to contradict the ap­parently well-established hypothesis that the flare in SAP or other osteoblastic markers represent positive response to the treatment. It is possible that the osteoclastic activity continues in presence of osteoblastic bone scan appear­ances in the progressive disease, better measured by osteo­clastic markers such as deoxypyridinoline and pyridinoline cross-linked carboxyterminal telopeptide (I CTP). [44] We have not analysed osteoclastic or other bone markers in our patients, but it would be interesting to see if these correlate with flare phenomena. Rise in serum of these markers in parallel with the degree of flare would indicate mobilisation of bone and possibly, increased metastatic activity as reflected in rapid progression of the disease seen clinically. Since osteoclastic response is mediated by paracrine growth factors secreted by malignant cells, [45] a correlation of these and the flare phenomena may shed light on the mechanism of progression in this disease, as well as the mechanism of increase in SAP levels in vari­ous forms of metastatic disease of the skeleton.

Pathological involutional changes in the prostate called apoptosis are well known after androgen withdrawal and androgen responsive tumours may show such changes in metastases as well. [46] As the `flare' phenomenon is of nega­tive prognostic value, it is possible that it occurs only in those cancers which are primarily resistant to androgen withdrawal or selects cases in whom androgen responsive cells undergo apoptosis leaving resistant and highly malignant cells intact leading to their selective and rapid clonal growth and associated flare. Progression in meta­static disease may occur by alterations (for example loss of balance) in growth factor system such as Inter leukin-6 [47] since several growth factors are known to con­trol and balance normal cellular growth [48] and affect growth of malignant cells. [6] Also, the androgen receptor genes may undergo mutations prior to [18] or following [19] such treatment and the progressive disease is reflected in SAP flare. Al­kaline-phosphatase is involved in the hydrolysis of phos­phate esters and transphosphorylation [35] and a change in phosphorylation of the androgen receptor may be one of the mechanisms of androgen resistance, [49] SAP flare and treatment failure.

In conclusion, a significant rise in SAP levels soon af­ter commencement of treatment occurs in a significant proportion of these patients. The flare activity in SAP le­vels within 6 weeks of initiating the therapy indicates poor response to it and rapid progression of the disease with short overall survival. This group of patients may be con­sidered for second-line or alternative therapeutic regimens, especially if the magnitude of the flare is twice the basal value.

Because of the retrospective nature of the study, further prospective studies should be carried out to confirm these findings. Further studies should include an analysis of serum PSA and a comparison made as to the prognostic value and the cost of measurement with SAP. Further stu­dies may also include other bone markers especially osteo­clastic markers and growth factors to elucidate mechanisms of the flare phenomenon and its possible relationship with primary or secondary resistance to treatment.


1Boring CC, Squires TS, Tong T, Mongomery S. Cancer statistics, 1994. CA Cancer J Clin 1994; 44: 7-26.
2Denis L. Trends in the diagnosis and treatment of prostate cancer. Oncology in Practice 1994; 1.
3Silverberg E, Lubera J. Cancer statistics. CA 1989; 39: 3-12.
4Bouffioux Ch. Prostate cancer: To screen or not to screen? Eur Urol 1997; 31 (Suppl 1): 2-4.
5Jensen OM, Esteve J, Moller H, Renard H. Cancer in European Community and its member states. Eur J Cancer 1990; 25: 1167­-1256.
6Byrne RL, Leung H, Neal DE. Peptide growth factors in the pros­tate as mediators of stromal-epithelial interaction. Br J Urol 1996; 77: 627-633.
7Doherty AP, Christmas TJ. Diagnosis and management of urological malignancy: the prostate. Br J Hosp Med 1996; 55: 104.
8Chodak GW, Thisted RA, Gerber GS et al. Results of conservative management of clinically localised prostate cancer. N Eng J Med 1994; 330: 242-247.
9Bloom JHM. Clinical parameters for the transition to endocrione independence of prostate cancer. In: Berns PMJJ, Romijn JC, Schroder FH (eds): Mechanism of progression to hormone inde­pendent growth of breast and prostate cancer. Parthenon 1991; 1: 3-8.
10Dal Fior S, Innone T, D'Inca G et al. Incidence and natural history of metastatic adenocarcinoma of the prostate at diagnosis. Eur J Cancer 1995; 31A (Suppl 5): 464 (Abstr).
11Huggins C, Stevens RE, Hodges CV. The effect of castration on advanced cancer of the prostate gland. Arch Surg 1941; 43: 209­-223.
12Byar DP. VACURG Studies of Cancer of the prostate. Cancer 1973; 32: 1126-1130.
13Chodak GW, Vogelzang MJ, Caplan RJ, Soloway M, Smith J. In­dependent prognostic factors in patients with metastatic (D2) pros­tate cancer. JAMA 1991: 265: 618-621.
14Soloway M. The importance of Prognostic factors in advanced pros­tate cancer. Cancer 1990; 66: 1017-1021.
15Vogelzang NJ, Chodak GW, Soloway MS et al. Goserlin versus orchiectomy in the treatment of advanced prostate cancer: Final results of a randomised trial. Urol 1995; 46: 220-226.
16Robinson MRG, Smith PH, Richards B et al. The Final Analysis of the EORTC Genito-urinary Tract Cancer Co-operative Group Phase III Clinical Trial (Protocol 30805) Comparing Orchidectomy, Or­chidectomy plus Cyproterone Acetate (CPA) and low dose stil­boestrol in the management of metastatic carcinoma of the prostate. Eur Urol 1995; 28: 273-283.
17Bertagna C, De Gery A, Hucher M. Francois JP, Zanirato J. Effi­cacy of the combination of nilutamide plus orchidectomy in pa­tients with metastatic prostatic cancer: A meta-analysis of seven randomised double-blind trials (1056 patients). Br J Urol 1994; 73: 396-402.
18Wayne DT, Buchanan G, Hickey TE, Bentel JM. Mutations in the androgen receptor gene are associated with progression of human prostate cancer to androgen independence. Clin Cancer Res 1996; 2: 277-285 (Abstract).
19de Vere White R, Meyers F, Chi SG et al. Human androgen receptor expression in prostate cancer following androgen ablation. Eur Urol 1997; 31: 1-6.
20Iversen P, Rasmussen F. Estramustine versus placebo in hormone refractory carcinoma of the prostate. Danish Prostate Cancer Group Study 9002. J Urol 1995; 153 (Suppl) 41: 239.
21Yagoda A, Petrylak D. Cytotoxic Chemotherapy for Advanced Hor­mone Resistant Prostate Cancer. Cancer 1993; 71 (Suppl): 1098.
22Blumenstein B, Crawford ED, Saiers JH et al. Doxorubicin, Mito­mycin C and 5-FU in the treatment of hormone refractory adeno­carcinoma of the prostate: A Southwest Oncology Group Study. J Urol 1993: 150: 411-413.
23Veronesi A, Lo Re G, Foladore S et al. Multidrug chemotherapy in the treatment of non-elderly patients with hormone refractory prostatic carcinoma. Eur Urol 1996; 29: 434-438.
24Waxman J, Saini A. Current Status of Scientific Research and Hor­monal Treatment for Carcinoma of the Prostate. Br J. Cancer 1991; 64: 419-421.
25Pelger RCM, Lycklama GAB, Nije Holt A et al. The flare in serum alkaline phosphatase activity after orchiectomy: a valuable nega­tive prognostic index for progression free survival in prostatic car­cinoma. J Urol 1996; 156: 122-126.
26Arai Y, Yoshiki T, Yoshida O. Prognostic significance of PSA in endocrine treatment for prostatic cancer. J Urol 1990; 144: 1415-1419.
27Siddall JK, Hetherington JW, Cooper EH et al. Biochemical Moni­toring of Carcinoma of Prostate Treated with an LH-RH Analogue (Zoladex) Br J Urol 1986; 58: 676-682.
28Reynard JM, Peters TJ, Gillatt D. PSA and Prognosis in patients with metastatic prostate cancer - a multivariable analysis of pros­tate cancer mortality. Br J Urol 1995; 75: 507-515.
29Cooper EH, Armitage TG, Robinson MRG et al. Prostate specific antigen and prediction of prognosis in metastatic prostate cancer. Cancer 1990; 66 (Suppl 5): 1025-1028.
30Morote J, Lorente JA, Encabo G. Prostate carcinoma staging. Clini­cal utility of bone alkaline-phosphatase in addition to PSA. Cancer 1996; 78: 2374-2378.
31Wechsel HW, Petri E, Bichler KM. Skeletal alkaline-phosphatase: A marker for individual follow-up in patients with advanced pros­tate cancer. Urol Int 1997; 58: 80-83.
32Oesterling JE, Andrews PE, Suman Vj, Zincke H, Myere RP. Preoperative androgen deprivation therapy: artificial lowering of serum PSA without downstaging the tumour. J Urol 1993; 149: 779-782.
33Leo ME, Bilhartz DL, Bergstrahl EJ, Oesterling JE. PSA in hormo­nally treated stage D2 prostate cancer: Is it always an accurate indi­cator of disease status? J Urol 1991; 145: 802-806.
34Stein A, de Kernion JB, Smith RB, Dorey F, Patel H. Prostate spe­cific antigen levels after radical prostatectomy in patients with or­gan confined and locally extensive prostate cancer. J Urol 1992; 147: 942-946.
35Price C. Multiple forms of human serum alkaline-phosphatase: de­tection and quantification. Ann Clin Biochem 1993; 30: 355-372.
36Iverson P, Rasmussen F, Christensen IBJ. Serum testosterone as a prognostic factor in patients with advanced prostatic carcinoma. Scand J Urol Nephrol 1997; (Suppl) 157: 41-47.
37Urwin GH, Percival RC, Harris S et al. Generalised increase in bone resorption in carcinoma of the prostate. Br J Urol 1985; 57: 721-723.
38Huggins C, Hodges CV. Studies on prostate cancer I: The effect of castration, of oestrogen and androgen injection on serum phosphatases in metastatic carcinoma of the prostate. Cancer Res 1941; 1: 293-297.
39Coleman RE, Mashitor G, Whitaker KB et al. Bone scan flare pre­dicts successful systemic therapy for bone metastases. J Nucl Med 1988; 29: 1354-1359.
40Mackintosh J, Simes J, Raghavan D, Pearson B. Prostatic cancer with bone metastases: serum alkaline phosphatase as a predictor of response and the significance of the SAP "flare". BR J Urol 1990; 66: 88-93.
41Pollen JJ, Witztum KF, Ashburn WL. The Flare Phenomenon on Radionuclid Bone Scan in Metastatic Prostate Cancer. AIR 1984; 142: 773-776.
42Naik RB, Gosling P, Price CP. Comparative study of alkaline-phos­phatase isoenzyme, bone histology and skeletal radiography in di­alysis bone disease. BMJ 1977; 1: 307-310.
43Kylmala T, Tammela TLJ, Ristelli L. Type I collagen degradation product (ICTP) gives information about the nature of bone metastases and has prognostic value in prostate cancer. Br J Cancer 1995;71:106.
44Meads H, Koizumi M, Yoshimura K et al. Correlation between bone metabolic markers and bone scan in prostate cancer. J Urol 1997; 157: 539-543.
45Mundy GR. The hypercalcaemia of malignancy. Kidney Int 1987; 31: 142-155.
46Reuter VE. Pathological Changes in Benign and Malignant Prostatic Tissue Following Androgen Deprivation Therapy. Urology 1997; 49 (Suppl 3A): 16-22.
47Twillie DA, Eisenberger MA, Carducci MA. Interleukin-6: a can­didate mediator of human prostate cancer morbidity. Urology 1995; 45: 542-549.
48Vlodavsky I, Benezra M, Levi E, Miao HQ, Neufeld G. Growth factors control cell growth by stimulating and inhibiting effects. Eur J Cancer 1993; 29A (6): Abst 31.
49Kokontis J, Takakura K, Hay N, Shutsung L. Increased androgen receptor activity and altered c-myc expression in prostate cancer cells after long-term androgen deprivation. Cancer Res 1994; 54: 1566-1573.