Brifbh Journalof Urobgy (199% 65,504-508 0 1990 British Journal of Urology

0007 I 3 3 I /90/0065 050416 10.00

Pituitary Adrenal and Gonadal Endocrine Suppression for the Primary Treatment of Prostate Cancer

GORDON WILLIAMS, R. ASOPA, P. D. ABEL and C. SMITH

Department of Surgery, Hammersmith Hospital and Royal Postgraduate Medical School, London

Summary-Sixty new patients with advanced local or metastatic prostatic cancer were randomised to receive either bilateral orchiectomy, orchiectomy plus dexamethasone 0.5 mg in the morning and 0.3 mg at night, or orchiectomy plus cyproterone acetate 100 mg 3 times per day. All surviving patients have been followed up for a minimum of 2 years. An improvement in both objective and subjective responses in patients receiving dexamethasone compared with those receiving cyproterone acetate suggests a possible additional role for the pituitary in the control of prostatic tumour growth. A larger study with longer follow-up may be indicated.

Endocrine therapy for the treatment of advanced prostatic cancer has been available for over 40 years. Despite many improvements in such therapy, there is still at best a 60% objective response rate with no convincing evidence of any increase in survival for those who present with metastatic disease.

Most therapies have been aimed at suppression of the testicular production of testosterone using oestrogens or orchiectomy and more recently by the use of LHRH analogues (Allen e ta l . , 1983). Additional androgen suppression aimed at blocking the action of androgens produced by the adrenals and by peripheral conversion of precursor steroids has been attempted using a combination of an LHRH analogue or orchiectomy with an anti- androgen (Labrie et al., 1985). Such attempts have been associated with an increased early response rate and at 3 years an improvement in the median time to disease progression compared with single conventional therapy (Garnick, personal commu- nication).

It has been clearly shown that pituitary ablation, either by surgery or some form of radiotherapy, produces significant subjective improvement, par- ticularly pain relief, in the majority of patients with

Accepted for publication 27 September 1989

carcinoma of the prostate in whom the disease has relapsed after conventional endocrine ablation (Fitzpatrick er al., 1980). The mechanism of this response is unclear and may be due to the suppression of some as yet unidentified hormonal peptides. In addition, mechanisms of contro: of adrenal androgen secretion are not fully understood (Anderson, 1980). There may be other pituitary factors which affect androgen production, particu- larly in the pre-pubertal period (Harding, 1979), and in some conditions there is a disassociation between adrenal androgen and cortisone secretion (Grumbach et al., 1978; Parker and Odell, 1979).

To examine whether endocrine suppression aimed at the pituitary and gonadal axis has any significant additive value compared with androgen suppression aimed solely at the testes or testes, adrenals and peripherally produced androgens, a prospective randomised study was carried out to compare orchiectomy versus orchiectomy plus dexamethasone, orchiectomy plus cyproterone ace- tate or orchiectomy plus ketoconazole in patients with locally advanced or metastatic prostatic cancer.

Following the manufacturer’s advice as to a significant incidence of septicaemia following high dose ketoconazole therapy, this arm of the study was withdrawn after 16 months and these patients are reported in a separate study. The remaining 3

504

ENDOCRINE THERAPY FOR PROSTATE CANCER 505

groups of patients have now completed 2 years’ follow-up and are the subject of this report.

Patients and Methods Sixty new patients with advanced local or metas- tatic prostate cancer were randomised using ran- dom number tables to receive either bilateral orchiectomy, orchiectomy plus dexamethasone 0.5mg in the morning and 0.3mg at night, or orchiectomy plus cyproterone acetate 100 mg 3 times per day. Pre-operative staging and follow-up examinations were carried out at regular intervals and included clinical examination, digital rectal examination, serum acid phosphatase (total and tartrate labile) and radio-isotope bone scans. In addition, the following serum endocrine studies were performed at each visit : testosterone, dihydro- testosterone, androstenedione, progesterone, corti- sol and luteinising hormone. Patients were evaluated objectively at 1 and 2 years using British Prostate Group criteria (Chisholm and Beynon, 1982).

Statistical methods Patients were categorised at 2 years as responders, stable, progressing, dead from progressive disease or dead from other disease. Their response was compared between the 3 treatment groups using the Kruskal-Wallis one-way analysis of variance by ranks (adjusted for ties), using the statistical software package Minitab. The analysis was carried out both omitting the patients dead from other disease, and with these patients grouped with the patients dead from progressing disease. The 3 pairwise comparisons between groups were made by comparing the average ranks of the 3 groups, following the procedure given in Siege1 and Castel- Ian (1988) except that the adjustment for multiple comparisons they give (which in this case would effectively mean multiplying each P value by 3) was not used. All comparisons were made by two-tailed tests, with the level of significance being taken as 5%. Statistical analysis of the endocrine data was carried out using the unpaired Student’s t test. Table 1 Details of Patients

Details of patients are shown in Table 1 .

Results

Endocrine studies Serum testosterone was at all times lower in the group receiving dexamethasone but these differ- ences were not significant (Fig. 1). There were no differences in serum dihydrotestosterone between the 3 groups of patients. The reduction in serum androstenedione in the orchiectomy plus dexameth- asone group compared with the orchiectomy group was significant (Pc0.001) at 12 and 24 months (Fig. 2). In the group receiving cyproterone acetate the reduction in androstenedione at 24 months compared with the orchiectomy group was signifi- cant (P<O.O2). The reduction in serum cortisol in the group receiving dexamethasone was significant compared with the orchiectomy group at 12 and 24 months (Pc0.001) (Fig. 3). Although serum pro- gesterone was reduced in the group receiving dexamethasone compared with the other 2 groups, these differences were not significant. The reduction in serum LH in the group receiving cyproterone acetate was significant at 12 months (P<O.OOl) compared with the other 2 groups but not significant at 24 months (Fig. 4).

04 1 I I I I 3 I .21 I 3 0 0 12 ?s I 0 21 24

month

Fig. 1 represents the standard error of the mean.

Serum testosterone in the 3 treatment groups. Each bar

Tumour grade No. of Age range Age Elevated acid patients (years) (mean) I 2 3 Metastatic phosphatase

Orchiectomy 24 48-84 72 5 4 15 16 12

Orchiectomy + dexamethasone 16 62-86 73 3 5 8 14 10 Orchiectomy + cyproterone acetate 20 54-84 14 6 4 10 16 13

506 BRITISH JOURNAL OF UROLOGY

t ORPII. + + m u -+m

l n O D t h

Fig. 2 Serum androstenedione in the 3 treatment groups. Each bar represents the standard error of the mean.

o + , . . . . I . . . I o .za 1 a I m i z 10 in a1 a4

month

Fig. 3 Serum cortisol in the 3 treatment groups. Each bar represents the standard error of the mean.

4a

3a

c El0

10

0

Fig.4 Serum LH in the 3 treatmentgroups. Each bar represents the standard error of the mean.

Objective response The response to treatment at 1 and 2 years is shown in Tables 2 and 3. At 1 year, 2 patients had died from carcinoma of the prostate in the orchiectomy group, 1 at 2 months and the other at 9 months. By 2 years, an additional 2 patients had died of prostatic cancer, both at 18 months. No patient in the orchiectomy group died from intercurrent disease. In the first year, 2 patients in the cyproter- one acetate group died from prostate cancer, 1 at 2 and 1 at 9 months. In addition, 2 patients died, 1 from a myocardial infarct and congestive cardiac failure at 4 weeks and the other from respiratory failure at 6 months. In the second year, a further 3 patients died from prostate cancer at 16, 19 and 20 months and 1 patient died at 17 months from congestive cardiac failure. No patient died from prostate cancer at 1 year in the dexamethasone group but 1 died from a ruptured aortic aneurysm at 3 months. In the second year, 2 patients died from prostate cancer at 15 months and 1 from the myocardial infarct at 18 months. At 2 years, when patients dead from other diseases are omitted, the dexamethasone group performed significantly bet- ter than the cyproterone group ( P = 0.0088). Neither of the other 2 comparisons was significant, though the one between the orchiectomy and the orchiec- tomy plus dexamethosone groups was nearly so (P=0.076). When patients dead from progressing disease are included with those dead from other disease, orchiectomy plus dexamethosone gave a significantly better performance than orchiectomy plus cyproterone (P=0.017). Neither of the other 2 comparisons was significant.

Side eflects of treatment No patient in the orchiectomy group complained of unexpected side effects. In those randomised to cyproterone acetate, 1 patient who died at 16 months of prostatic cancer stopped treatment between 10 and 13 months because of lethargy. One stopped at 6 months for no apparent reason and died at 20 months from prostate cancer. One patient stopped at 12 months for no apparent reason and had progressive disease at 24 months. A further patient stopped treatment at 12 months because of t hromboembolic complications and had stable disease at 24 months. All patients receiving dexa- methasone gained weight and had an improvement in appetite and well-being. One patient who complained of gastrointestinal discomfort stopped therapy between 7 and 10 months and showed responding disease at 24 months.

ENDOCRINE THERAPY FOR PROSTATE CANCER 507

Table 2 Results at 1 Year

Dead from Group Dead from progressive no. Responders Stable Progressing other disease disease

Orchiectomy 24 8 9 5 0 2 Orchiectomy fcyproterone acetate 20 9 6 I 2 2 Orchiectomy + dexamethasone 16 10 3 2 1 0

Table 3 Results at 2 Years

Dead from Group Dead from progressive no. Responders Stable Progressing other disease disease

Orc hiectomy 22 7 6 7 0 2 Orchiectomy + cyproterone acetate I6 3 3 6 I 3 Orchiectomy + dexamethasone 15 10 I I I 2

Discussion

The majority of prostate cancers consist of a mixture of cells, some of which require the presence of androgens for their continued growth and some which do not. The suggestion by Labrie et al. (1985) that complete androgen suppression using an LH/ RH analogue or orchiectomy plus an anti-androgen should result in improvements in survival and response rates was not initially supported by the results of randomised trials carried out by other workers. However, the Leuprolide study group has now completed a 3-year follow-up of patients randomised to either single therapy with the LH/ RH analogue Leuprolide or total androgen ablation. Their results show an improved initial response rate for those randomised to total androgen ablation and also an improvement in the median time to disease progression (Garnick, personal communi- cation).

The present study is the first to report attempts to inhibit androgen production further by pituitary suppression using dexamethasone. Serum testoster- one was at all times lower in the group receiving dexamethasone compared with the other groups but this difference was not significant. However, the very marked reduction in serum androstenedi- one and serum cortisol in the dexamethasone group was highly significant at all time intervals (Figs 2 and 3).

Since the pituitary control of adrenal androgen production is not fully understood, the pituitary suppression induced by dexamethasone might also have suppressed other as yet unidentified hormone peptides.

The 3 treatment groups were very similar with regard to patients’ age, number of poorly differen- tiated tumours, patients with metastatic disease and with an elevated acid phosphatase. All have now been followed up for a minimum of 2 years. In that time 4 patients in the orchiectomy group died from progressive disease and none from intercur- rent disease; although 20 of the 24 patients are still alive, only 7 are showing evidence of disease response and 7 have progressive disease. In the group receiving cyproterone acetate, 5 died from progressive prostate cancer and 3 from cardiores- piratory failure. Of the survivors, 6 show progres- sive disease and only 3 evidence of disease response. In those receiving dexamethasone, 2 have died from prostate cancer and 2 from cardiovascular causes but of the 12 patients alive at 2 years only 1 shows progressive disease and 10 show evidence of disease response-a much higher proportion than in the other 2 groups. Statistical analysis has shown that the 3 groups did not have the same response irrespective of whether the patients who died from other diseases were omitted or included. In both analyses the group receiving orchiectomy plus dexamethasone showed responses which were significantly better than those of the orchiectomy plus cyproterone group. When patients dead from other diseases are omitted, the comparison between orchiectomy and orchiectomy plus dexamethasone was nearly significant (P= 0.076). However, when patients dead from progressive disease are included with those who died from other diseases, the P value does not approach significance.

The safety of orchiectomy as a primary treatment for prostate cancer has again been shown, as has

508 BRITISH JOURNAL OF UROLOGY

an incidence of cardiovascular deaths in patients receiving cyproterone acetate. The improvement in both objective and subjective responses in patients receiving dexamethasone and the absence of any significant toxicity are of interest and suggest a possible additional role for the pituitary in the control of prostatic tumour growth.

This study suggests a role for dexamethasone in the primary treatment of prostatic cancer but a much larger study with a longer follow-up is required before any definite conclusion can be reached.

Acknowledgement We acknowledge the help of Neil Alexander, Department of Medical Physics, for the statistical analysis.

References Allen, J. M., O’Shea, J. P., Mashiter, K. el a~! ( I 983). Advanced

carcinoma of the prostate. Treatment with a gonadotrophin releasing hormone agonist. Br. Med. J . , 286, 1607-1609.

Anderson, D. C. (1980). The adrenal androgen-stimulating hormone does not exist. Lancet, 2,454-456.

Chisholm, C . D. and Beynon, L. L. (1982). The response of the malignant prostate to endocrine treatment. In The Endocrinol- ogy o/ Prosrate Tumours, ed. Ghanadian R. Chapter 10, pp. 241 -262. Lancaster: MTP Press.

Fitzpatrick, J. M., Cardiner, R. A., Williams, J. P. e t d (1980). Pituitary ablation in the relief of pain in advanced prostatic carcinoma. Br. J . Urol.. 52, 301-304.

Grumbach, M. M., Richards, C. E., Conte, F. A. et d. (1978). Clinical disorders of adrenal function and puberty: an assessment of the role of the adrenal cortex in normal and abnormal puberty in man and evidence for an ACTH-like pituitary adrenal androgen stimulating hormone. In The Endocrine Function of the Human Adrenal Corre.r, ed. James, V. H. T., Serio, M., Guisti, C. and Martin, L. Pp. 583-612. London: Academic Press.

Harding, B. W. (1979). Synthesis of adrenal cortical steroids and mechanismsof ACTH effects. In Endocrinology, ed. Delstroot, L. J . Volume 2, pp. 1131-1137. New York: Grune and Stratton.

Labrie, F., Dupont, A. and Belanger, A. (1985). Complete androgen blockade for treatment of prostate cancer. In Important Adrances in Oncology, ed. Devita, V. T., Jr., Hellman, S. and Rosenberg, S. A. Chapter 10, pp. 193-217. Philadelphia : Lippincott.

Parker, L. N. and Odell, W. D. (1979). Evidence for existence of cortical androgen stimulating hormone. Am. J . Phl,.vb/., 236, E616E620.

Siegel, S. and Castellan, N. J. (1988). Non-parametric Sraristics forthe BeharioralSciencrs. Pp. 206-216. New York: McGraw- Hill.

The Authors Gordon Williams, MS, FRCS, Consultant Urologist. R. Asopa, FRCSE, Registrar in Urology. P. D. Abel, ChM, FRCS, Senior Lecturer in Urology. C. Smith, MA(Cantab), MPhil. Research Technician

Requests for reprints to: Gordon Williams, Department of Surgery, Hammersmith Hospital and Royal Postgraduate Med- ical School, Du Cane Road, London W I 2 OHS.