Melengestrol acetate and megestrol acetate are prostatic tumor inhibiting agents G. M. PADILLA,~ R. C. YACULLO, J. J. PADILLA, B. PAYNE,AND V. PETROW Department of Cell Biology, Duke University Medical Center, Durham, NC, 27710 U.S.A. Received March 12. 1990
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
PADILLA, G. M., YACULLO, R. C., PADILLA, J. J., PAYNE, B., and PETROW, V. 1990. Melengestrol acetate and megestrol acetate are prostatic tumor inhibiting agents. Biochem. Cell Biol. 68: 1181-1 188. We had previously reported that 6-methylene progesterone, an inhibitor of 5a-reductase, the enzyme which converts testosterone to dihydrotestosterone, markedly inhibited growth of the androgen-dependent Dunning R3327-H rat prostatic tumors. We now find that the progesterone derivatives melengestrol acetate (MGA) and megestrol acetate (MA) inhibit both the androgen-dependent (Dunning R3327-H) and the androgen-independent (Dunning R3327-AT3) prostatic tumors. Growth of the AT3 tumors was suppressed by -53% after 9 days of daily S.C. injections with MGA at 10 mg/kg body weight. MGA also caused a 54% weight reduction of the ventral prostate and a 53% reduction of the seminal vesicles. Adrenal weights were reduced by 42%. A 24-day oral treatment with MGA (at - 15-17 mg/(kg -day)) inhibited AT3 tumor growth by 59% and caused a weight reduction in the following tissues: prostate (46%), seminal vesicles (19%), testes (12%), and adrenals (52%). Under the same protocol, MA inhibited AT3 tumor growth by 32% and reduced the weight of the ventral prostate by 49% and the weight of the adrenals by 18070, but had no effect on the seminal vesicles and testes. The extent of the MGA-induced prostatic regression was accompanied by cytological changes similar to those effected by 6-methylene progesterone, i.e., shrinking of the acinar epithelium. The AT3 tumors in MGA-treated rats displayed a limited degree of apoptosis. Atrophy of the adrenal cortex and lowered plasma levels of corticosterone and dihydroepiandrosterone were also observed. A therapeutic role for MGA and MA against androgenindependent prostatic neoplasms in man is forecast by these observations. Key words: prostatic cancer, androgen independent, melengestrol acetate, growth inhibition, steroids.
PADILLA, G. M., YACULLO, R. C., PADILLA, J. J., PAYNE, B., et PETROW, V. 1990. Melengestrol acetate and megestrol acetate are prostatic tumor inhibiting agents. Biochem. Cell Biol. 68 : 1181-1188. Nous avons deja montre que la 6-methylkne progesterone, un inhibiteur de la 5a-reductase, enzyme qui transforme la testosterone en dihydrotestostbone, inhibe de facon marquee chez le rat la croissance des tumeurs prostatiques Dunning R3327-H dependantes des androgenes. Nous trouvons maintenant que le melengestrol acetate (MGA) et le megestrol acetate (MA), deux derives de la progestbone, inhibent a la fois les tumeurs prostatiques (Dunning R3327-H) dependantes des androgenes et les tumeurs (Dunning R3327-AT3) independantes des androgenes. A p r b 9 jours d'injections sous-cutanees de MGA a 10 mg/kg de poids corporel, la croissance des tumeurs AT3 est reduite d'environ 53%. Le MGA rkduit egalement de 54% le poids de la prostate ventrale et de 53% celui des vesicules seminales. Le poids des surrenales est aussi diminue de 42%. Un traitement oral de 24 jours avec le MGA (d'environ 15 a 17 mg/(kg.jour)) inhibe de 59% la croissance de la tumeur AT3 et rkduit le poids des tissus suivants : prostate (46%), vesicules seminales (19%), testicules (12%) et surrknales (52%). Dans les m&mesconditions, le MA inhibe de 32% la croissance de la tumeur AT3, rtduit de 49% le poids de la prostate ventrale et de 18% le poids des surrtnales, rnais il n'exerce aucun effet sur les vksicules seminales et les testicules. L'importance de la regression prostatique induite par le MGA s'accompagne de changements cytologiques semblables a ceux effectues par la 6-mkthylene progesterone : retrkcissement de I'epithClium acinaire. Chez les rats trait& avec le MGA, les tumeurs AT3 manifestent un degre limit6 d'apoptose. Nous observons tgalement une atrophie du cortex surrknalien et de faibles taux plasmatiques de corticosterone et de dihydroepiandrosterone. Ces observations nous permettent de prkdiie un rBle thbapeutique pour le MGA et le MA dans les nbplasmes prostatiques independants des androgenes chez l'homme. Mots clgs : cancer prostatique, dependance aux androgenes, mklengestrol acetate, inhibition de la croissance, sttroides. [Traduit par la revue]
Introduction Castration as a treatment of prostatic cancer was introduced by Huggins et al. (1941) on the premise that the testicular androgen, testosterone, was the main trophic hormone of the prostate and its derived neoplasms. Alone or with administration of estrogen, castration has stood the test of time as a therapeutic modality and still represents the main standby for management of the advanced disease. In 1962 MGA (Fig. 1) was discovered by Kirk et al. (1962) as part of a program directed towards new orally active contraceptive progestogens. MGA is an analogue both of medroxyprogesteroneacetate and of MA (Fig. 2), but differs
ABBREVIATIONS: MGA, melengestrol acetate (17-acetoxy-6MA, megestrol methyl-16-methylenepregna-4.6-diene-3,40-dione; acetate; DHT, dihydrotestosterone; DHEA-S, dihydroepiandrosterone sulfate; LD,,, mean lethal dose. ' ~ u t h o rto whom all correspondence should be addressed. Printed in Canada / Imprim6 au Canada
from them principally by possessing significant antiinflammatory and immunosuppressive activities (Duncan et al. 1964). MA has been successfully used as a palliative in advanced prostatic cancer in combination with diethylstilbestrol, thereby achieving a reversible chemical castration (Geller et al. 1978). In an early clinical trial, MGA proved to be effective in radium-resistant endometrial carcinoma. It was well tolerated by women on chronic administration. It was subsequently tried against a series of experimental tumors in the rat, including squamous cell carcinoma of the prostate, with varying degrees of tumor inhibition (Petrow 1976). Against this background, it was decided to examine the effects of MGA on the Dunning R3327-H (androgen dependent) and Dunning R3327-AT3 (androgen independent) transplantable rat prostate tumors (Isaacs et al. 1978). We find that MGA (and MA to a lesser extent) inhibits growth of both Dunning tumor lines and induces regression of the
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
BIOCHEM. CELL BIOL. VOL. 68, 1990
FIG. 1. Structural formula of MGA.
Dose, mg/kg body weight
FIG. 3. Dose-response of adult male CCFl rats to daily s.c. MGA injections, at the doses shown in the abscissa, for 26 days. CCFl rats (110-120 g) were implanted with Dunning R3327-H tumors and injected with MGA or carrier (see Materials and methods). Adrenal and tumour weights are shown on the left ordinate and the prostate weights on the right ordinate. Standard errors of the means are indicated (n = 5 rats per group; two tumors per rat). TABLE1. Dose-response of Dunning R3327-AT3 prostatic tumors to MGA injections (10 day study)
o ~s
FIG. 2. Structural formula of MA.
adrenals, prostate, and seminal vesicles. Portions of this work were reported earlier (Padilla,et al. 1986, 1988). Materials and methods Animals and tumor implantation Male Copenhagen x Fisher F1 rats (CFFl) bearing Dunning R3327-H tumors were obtained from Dr. Norman Altman of the Department of Comparative Pathology, School of Medicine. University of Miami, Miami, FL. Nontumorous CFFl rats (Charles River) were implanted in our laboratory with cryopreserved Dunning R3327-H and R3327-AT3 tumor cells. Cultured AT3 tumor cells were kindly provided by Dr. D. Mickey (Medical School, University of North Carolina, Chapel Hill, NC). AT3 cells were grown on RPMI medium supplemented with 10% fetal bovine serum or CPR-5 (Sigma, St. Louis, MO), insulin (1 IU/mL), DHT (5 mg/mL), retinol (0.1 mg/mL), amphotericin B (12.5 pg/mL), and gentamycin (25 pg/mL). Prior to tumor implantation and treatment with MGA or MA, rats were acclimatized for 2 weeks and allowed to reach a body weight of 100-150 g. Retired breeder rats (250-300 g) were also employed. Rats were kept in hanging cages at constant temperature, under a 12 h light : 12 h dark cycle, fed, and watered ad libitum. Saline solution (0.9% NaCl) was the sole source of drinking water for MGA-treated rats to prevent weight loss due to electrolyte depletion. Rats were implanted with tumors in the flanks either ( a ) via a s.c. injection of 0.2-0.5 mL of trypsin- or collagenase-dissociated tumor cells (-2 x lo6 cells/ injection, at >85% viability, trypan blue exclusion) or (b) by inserting 2- to 3-mm3 tumor pieces into a small s.c. pocket. All surgical operations were performed on rats anesthetized with xylazine-ketamine (3:4 v/v) i.p. injection (0.1 mL/100 g body weight) or methoxyflurane inhalation (Metofane, Pitman-Moore, Inc., Washington Crossing, NJ). Rats were sacrificed by C 0 2 asphyxiation prior to excision of tumors and removal of organs and tissue samples.
-
Dose (mg/kg)
Body weighta (g)
Tumor weightb (a)
Tumor volumeb (cm3)
0 2 5 10 20
92.3 k 4.1 96.9+ 5.1 82.1 k 3.5 78.8 k2.3 66.7 k 6.7'
33.64k0.28 2.88 k0.13 2.79k0.33 2.75 k 0.25 1.89 k 0.14'
3.48 k0.40 2.29k0.32 2.28 k0.50 2.16k0.48' 1.30k0.17'
NOTE:CFFl rats were used in this study. Values are means SE. 'n = 5. bn = 10. 'Significantly different from control at 95% (Scheffe F-test). +_
Administration of inhibitors Rats were weighed and given s.c. injections of MGA (Upjohn Corp., Kalamazoo, MI) or MA (Bristol-Myers Co., Evansville, IN) on a daily basis. Steroid treatment was initiated 12-16 days after implantation with the Dunning R3327-H tumors or 24 h after implantation with Dunning R3327-AT3 tumors. Both steroids were dissolved in a carrier solution of propylene glycol - ethanol (9:l V/V)and injected at volumes of 0.1 mL/100 g body weight to yield the appropriate doses given in the results. Control rats were injected with equal quantities of vehicle. MGA was also administered orally either as a suspension in 3% carboxymethyl cellulose or thoroughly mixed with dry powdered lab chow (Purina), which was moistened, formed into granules, and dried at 60°C overnight. Concentration of MGA in the food was verified by organic extraction and spectrophotometry against a standard curve. Histopathology Appropriate tissue samples of adrenals, prostates, and tumors were fixed in buffered formalin at necropsy and processed for microscopic examination by the Pathology Department. Microphotographs were taken on hematoxylin- and eosin-stained specimens.
PADILLA ET
AL.
1183
TABLE2. Effect of oral administration and s.c. injection of MGA on body weight, androgen-dependent tissues, and Dunning R3327-AT3 tumors in the rat (A) Oral administration (10 mg/kg, 21 days)' Treatment
Body weight (g)
Prostate (mg)
Adrenals (mg)
Seminal vesicles (mg)
Testes (g)
485 k 73
1.42 + 0.04
375 k 43d 23
1.35 k0.7 21
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
- -
Control (n = 5 ) MGA (n = 6) To inhibition
318k6.0
438 +44
25.2k 1.2
277 k 7.4b 13
2 6 ~ + 2 6 ~ 13.8k0.89' 39 46
(B) Subcutaneous injection (10 mg/kg, 9 days)' Treatment Control (n = 5) MGA (n = 6) To inhibition
Body weight (g)
Prostate (mg)
Adrenals (mg)
AT3 tumor (9)
121k 4.9
86+ 12
29.6 k 6.2
2.10 k 0.47
108 k 5.7d 11
40k7.1b 54
17.25 1 . 8 ~ 42
0.995 + 0 . 2 5 ~ 53
NOTE:All values are means
+_ SE. "Sprague-Dawley rats were used in this study.
bp 5 0.01.
2 '0.y';
Not slgnifrant. CCFl rats were used in this study. 'p 5 0.05.
Corticosterone and DHEA analyses The serum concentrations of corticosterone and DHEA-S were determined by a radioimmunoassay procedure available from Radioassay Systems Labs, Inc. (Carson, CA). Blood samples were removed, by cardiac puncture, immediately upon sacrifice into test tubes kept in ice. The serum was separated from the clot by centrifugation and stored at - 200Cuntil needed. Samples were analyzed in duplicate and the corticosterone and DHEA contents were calculated from appropriate standard curves.
was a dose-dependent inhibition in the growth of AT3 tumors in terms of their weight and volume. Tumor volumes were determined by measuring three tumor diameters with calipers and employing the ~~~~kformula ( V = L w H x 0.5236) (Isaacs et al. 1978). Although the treated rats were not placed on saline, there were no significant losses in body weight at MGA 20 mg/(kg'da~)-Oneway analysis of variance indicates that the weights and volumes of the AT3 tumors were significantly reduced only Statistical analyses at MGA doses in the range of 10-20 mg/kg compared with The statistical significance of the results was determined as controls. Tumor weights and volumes were not significantly necessary, using the Student's t-test or ANOVA with the Statview@ different from each other within this dose range. software package designed for the Macintosh' (Abacus Concepts, Experiments were then carried out to determine whether Inc., Berkeley, CA). MGA was orally active, using its regressive effect on androgen-target tissues as an index of in vivo activity in Results comparison with MGA given via S.C. injections. The results Figure 3 shows the results of an experiment in which rats are summarized in Table 2. Adult male Sprague-Dawley rats implanted with Dunning R3327-H tumors for -2 weeks were given daily oral doses of MGA at 10 mg/kg body were subsequently treated with MGA for a period of 26 days weight for 21 days by intubation (Table 2A). MGA caused at MGA doses in the range of 2-20 mg/kg body weight. It a 13% decrease in body weight, and the mean ventral is clear that there was a dose-dependent regression in the prostate weight was reduced by 39% and that of the adrenals weights of the prostate and adrenals. While the prostate was reduced by 46%. The seminal vesicles and testes underwent maximal involution at MGA doses > 10 mg/kg, displayed a lesser loss in their mean weight (23 and 21070, the adrenals were maximally affected by MGA at doses respectively). Table 2B shows the results of an experiment > 2 mg/kg compared with the controls. MGA had a less in which duration of MGA injections was reduced to 9 days. pronounced inhibitory effect on these tumors, a maximal Rats were implanted with AT3 tumor cells and injected daily effect being achieved at doses above 5 mg/kg. Rats treated . with MGA (10 mg/kg body weight) beginning 24 h later. with 2 mg MGA/kg experienced a 7.5% loss in body weight The results show that even though MGA injections caused compared with the untreated rats that showed a 9.2% weight a slight loss of body weight (1 1Vo), the mean weights of the gain. At higher doses MGA-treated rats lost an average of ventral prostate and adrenals were reduced by 54 and 42070, 15% of their body weight. In subsequent experiments this respectively. Note that the AT3 tumor mean weight was weight loss was alleviated by using a saline solution (0.9% reduced by -53%. In summary, the results in Table 2 NaCl w/v) as the sole source of drinking water. demonstrate that MGA had similar inhibitory effects, Table 1 summarizes the results of a 10-day study designed whether it was given orally or was injected subcutaneously. to assess the dose response of AT3 tumors to daily MGA It was thus of interest to compare the effects of MGA injections in the range of 2-20 mg/kg body weight. There and MA administered in the food to AT3 tumor-bearing
1184
BIOCHEM. CELL BIOL. VOL. 68, 1990
TABLE 3. Effect of MGA and MA on Dunning R3327-AT3 tumors, adrenals, androgen target organs, and body weight of CCFl rats
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
(24-day study)
Condition (n)
Tumorsa (g)
Adrenals (mi?)
Ventral prostate (mg)
Control (5) MGA (5)'
40.3 k4.5 16.6k4.3 (58.8)' 27.5k5.1 (31.8)'
62.8 k 3.9 30.4k5.9 (51.6)' 51.5k4.1 (18.2)'
636k90 345k82 (45.8)' 324k52 (49.1)'
MA (4jh
NOTE:All values are means f SE. %ach rat was implanted with two tumors. b ~ a c hsteroid was mixed with powdered lab chow to a concentration of 'Decrease in weight expressed as percent of control.
Seminal vesicles (mg)
Testes (g)
384k 25 311 k49 (19.1)' 382k58 (0.5)'
1.39k0.02 1.22k0.03 (12.2)' 1.32k0.06 (5.0)'
0.25 mg steroid/g
TABLE4. Effect of MGA treatment on serum DHEA-S and corticosterone levels in CCFl rats Treatment
DHEA-S bg/mL)
Corticosterone @g/mL)
Control (n = 5) MGA (n = 5)
2.27 k 0.13 1.44k0.94
0.82 + 0.04 0.19 k 0.01'
NOTE:Rats were injected daily with MGA (s.c.) for 9 days at 10 mg/kg. Data on adrenals, prostate and AT3 tumors is given in Table 2B. Values are means SE. "Estimated value due to low plasma levels.
*
rats. Retired breeder CCFl rats (250-300 g) were employed in this study. Rats were implanted with AT3 tumors as before and were fed powdered lab chow containing MGA or MA (0.25 mg/g food). Food consumption was monitored daily. As the rats consumed 15-20 g of lab chow per day, they ingested 15-17 mg of either steroid per kilogram body weight. Table 3 summarizes the results of this 24-day study. MGA and MA, respectively, caused a 58.9 and 31.8% tumor weight loss (or growth inhibition) compared with controls (column 2). MGA reduced the adrenal weight by 51.6%, whereas MA reduced it by 18.2% (column 3). Both steroids induced prostatic involution to the same extent (MGA = 45.8%, MA = 49.1 %, column 4). The seminal vesicles were essentially unaffected by MA (0.5% inhibition), but their weights were reduced by 19.1% by MGA (column 5). The weight of testes was reduced by 12.2% by MGA, whereas MA caused a 5% weight reduction (column 6). The last column in Table 3 shows that MGA induced a 22.6% gain in body weight, whereas MA increased it by 10.5%. The control rats showed an insignificant loss of weight (5.3%). Table 4 summarizes the results of radioimmunoassays of the serum levels of DHEA-S and corticosterone. These analyses were performed on blood samples from the rats used in the experiment in Table 2B. It is clear that MGA lowered the serum levels of both DHEA-S and corticosterone. The corticosterone levels in MGA-treated rats were reduced to such an extent that the readings were close to the lower limit of this assay. Hence, the values given in the table were estimates based on extrapolation of the standard curve slightly beyond its lower limit.
-
Body weight (g) Initial
Final
337 k 6 265k 17
319+8 325 + 17
266+ 13
294+ 16
food to yield a daily dose of
Body weight change (yo) - 5.3
+ 22 + 10.5
- 15-17 mg/kg.
Since approximately 90% of the DHEA-S is thought to originate from the zona reticularis of the adrenal cortex, measurements of DHEA-S levels may be used as indication of cortical regression and (or) loss of adrenal steroidogenesis. The cytological changes induced by MGA in the adrenals are shown in Figs. 4-7. The salient feature of these changes was a shrinkage of the adrenal cortex (Figs. 4 and 5), especially in the zona fasciculata (Figs. 6 and 7), which lost its characteristic cord-like cellular organization. The individual cells no longer appeared as distinct units, the nuclei being more closely "packed" and pycnotic. Dunning AT3 tumors, although devoid of a distinct acinar organization characteristicof the prostate gland (Isaacs et al. 1978), showed several cytological changes following MGA treatment (Figs. 8 and 9). The most prominent one was an overall shrinkage of the cells giving a crowded appearance to the nuclei (Fig. 9), some of which were surrounded by a clear halo, which may represent vacuolization of the cytoplasm and loss of perinuclear organelles (e.g., Golgi). These results also suggest an apoptotic impact by MGA on the AT3 tumor cells. We are presently examining the effects of MGA on the viability of AT3 tumor cells in culture. Preliminary results indicate that MGA has an antiproliferative effect on these cells (Padilla et al. 1988). In the prostate, MGA induced a reduction in the epithelial profile, i.e., changing it from columnar to cuboidal (cf. Figs. 10 and 11). A similar reduction in the height of this epithelium was previously obtained with 6-methylene progesterone (Marts et al. 1987). MGA did not appear to induce a marked hypertrophy in the intra-acinar stroma of this gland as with castration (Marts et al. 1987).
Discussion Complete androgen ablation through castration and treatment with diethylstilbestrol has been the primary therapeutic modality in the treatment of prostate cancer for almost half a century (Huggins et al. 1941). Undesirable side effects brought about by such interventions and failure to achieve complete remission of this neoplasm, with the recurrence of androgen-independent tumors, have prompted the search
FIG. 4. Cross-section through the adrenal cortex of control CCFl rats. Slides were stained with hematoxylin and eosin. Magnification = 10 x . FIG.5. Cross-section through adrenal cortex of CCFl rats treated with MGA for 12 days. Magnification = 10 x . FIG.6. Section of zona fasciculata of adrenal cortex of control CCFl rats. Magnification = 25 x . FIG.7. Section of zona fasciculata of adrenals from MGA-treated CCFl rats (as in Fig. 5). Portions of zona reticularis is present on the right. Magnification = 25 x .
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
PADILLA ET AL.
1185
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
1186 BIOCHEM. CELL BIOL. VOL. 68. 1990
PADILLA ET AL.
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
for alternate methods of treatment of prostate cancer in man. Recent investigations from our laboratory have revealed that in the Dunning prostatic tumor model, 6-methylene progesterone was a highly effective inhibitor of R3327-H tumor growth. We now report that MGA is also an effective inhibitor of these tumors. It should be noted that the Dunning R3327-H tumors require the presence of testosterone and (or) DHT for its growth (Isaacs et al. 1978; Petrow et al. 1984; Padilla et al. 1985; Petrow 1987). On the other hand, the Dunning R3327-AT tumors (from which the AT3 subline was derived) lack the 5a-reductase enzyme (Isaacs et al. 1978). It was, therefore, not surprising to find that the 5a-reductase inhibitor 6-methylene progesterone was largely ineffective against these tumors. The major finding in the present report was the marked in vivo inhibitory effects of MGA on the androgenindependent AT3 prostatic tumors (Tables 1-3). Megestrol acetate was also inhibitory towards these tumors, but to a lesser degree (Table 3). AT3 tumor inhibition was generally accompanied by involution of the prostate, the adrenals, and to a lesser extent the seminal vesicles. Moreover, both MGA and MA had no marked effect on the testes. As discussed below, reduction of testicular weight most likely results from inhibition of gonadotropin secretion by these progestins. Changes in the prostatic acinar epithelium induced by MGA were comparable to those previously obtained with 6-methylene progesterone (Marts et al. 1987); namely a reduction in the profile of the epithelial cells i.e., from columnar to cuboidal (Figs. 10 and 11). Such a cytological change is reminiscent of loss of secretory activity (Padilla et al. 1986). Yet involution of the ventral prostate was not as severe as that obtained by castration (Marts et al. 1987). That adrenal hormonal output is compromised by MGA was verified by a reduction in serum levels of the adrenal androgen DHEA and corticosterone (Table 4). Although we did not assay for adrenal mineralocorticoids, their absence in the serum was suggested by the loss in body weight, a condition which was alleviated by substituting saline as the sole source of drinking water to MGA-treated rats. Since the ACTH levels were not measured, we cannot rule out MGA inhibition of adrenal corticopropin production. The degree of adrenal and prostatic involution induced by MGA seemed to parallel the magnitude of AT3 tumor inhibition (40-50%, Tables 2 and 3). MA induced a lesser degree of adrenal involution, even though inhibition of AT3 tumors was comparable to that produced by MGA (Table 3, columns 2 and 3). Duncan et al. (1964) had observed a reduction in body weight, accessory sex glands, and adrenals in 26- to 28-day-old Sprague-Dawley rats given oral doses of MGA of 0.8-1.6 mg/day. In the immature male rat, orally administered MGA was found to be at least four times as potent as medroxyprogesterone acetate as a gonadotropin inhibitor (Duncan et al. 1964). Acute toxicity (LDSo)from a single dose of MGA in mice was only observed at doses larger than 2.5 g/kg body weight. (Duncan et al. 1964). Our results (Table 2), which are in general agreement with these
1187
studies, also revealed that orally administered MGA and MA for 24 days were well tolerated, as demonstrated by the increase in body weight (Table 3). The results of the present investigation demonstrate that the progesterone derivates MGA and MA inhibit the growth of androgen-independent AT3 tumors, an accepted animal model of advanced cancer in man. Recent studies show that AT3 tumors are also useful for studies on the molecular basis of their androgen independence (Quarmby et al. 1990). The results of the present investigation provide evidence that not only are MGA and MA orally active, their differential activity towards androgen-independent prostatic tumors may result from their metabolic conversion into distinct reactive species. These results also suggest that AT3 tumor inhibition may be partly due to a direct (cytotoxic) action on androgen-independent clones, as indicated by apoptotic changes in the tumor cells (Fig. 9). A significant trophic influence on AT3 tumors by the adrenal androgen DHEA is rendered unlikely by the recent experiments of Quarmby et al. (1990), which demonstrate that AT3 tumors lack androgen receptors, as determined by immunostaining procedures. AT3 tumors were found to contain very low levels of androgen receptor mRNA, compared with the prostate and the Dunning R3327-7 H tumor subline (Quarmby et al. 1990). The effects of MGA and MA on anterior pituitary gonadotropins and their combined role in the inhibition of prostatic neoplasms are somewhat more complex. In the parabiotic rat model MA was found to inhibit pituitary gonadotropin production (Suchowsky et al. 1965). MGA was likewise found to be a gonadotropin inhibitor (Duncan et al. 1964). These progestins were originally synthesized by Kirk et al. (1962) as contraceptive agents in women, but there is a paucity of studies in male animals or men. Geller et al. (1981) have shown that MA plus diethylstilbestrolproduced significant suppression of LH and FSH, but diethylstilbestrol itself played a significant role in gonadotropin suppression. MA alone was also shown by Geller et al. (1976) to suppress testosterone, LH, and FSH in man. Geller (1985) also considered the possibility that small amounts of DHT generated from adrenal androgens could provide a significant stimulus to prostate cancer cells. Geller (1985) noted that there are enzymatic activities within the prostate sufficient for conversion of adrenal androgens (androstenedione or DHEA) to DHT, thus providing a rationale for blocking the androgen output of the adrenals as well as the testis in the treatment of some forms of advanced prostatic cancers that retain some hormonal dependence. This experimental approach may not be applicable to androgen-independent prostatic neoplasms, especially if they lack androgen receptors, as is the case with Dunning AT3 rat prostatic tumors (Quarmby el al. 1990). In conclusion, the present study provides evidence, for the first time, that androgen-independent prostatic tumors are inhibited by progesterone derivatives such as MGA and MA, which possess minimal in vivo anti-androgenic activ-
-
FIG. 8. Section through R3327-AT3 tumor from control CCFl rat. Note absence of acini and rather enlarged nuclei with distinct nuclear architecture. Magnification = 40 x . FIG.9. Section through Dunning R3327-AT3 tumor from CCFl rat treated with MGA (as in Fig. 5). Note the closely packed nuclei, darker stain, and presence of a clear halo surrounding some of the nuclei (arrows). Magnification = 4 0 x . FIG.10. Section through portion of two acini of the prostate from control CCFl rats. Note the linear arrangement of the nuclei of columnar epithelial cells (arrows). Magnification = 40x. FIG. 1 1 . Section through prostate from CCFl rat treated with MGA (as in Fig. 5). The epithelial cells are now cuboidal (arrows). Magnification = 40 x .
1188
BIOCHEM. CELL BIOL. VOL.
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
ity at low doses. AS MGA has shown relatively few undesirable side effects in chronic administration to women, there is every possibility that it will prove to be equally benign in men and thus be suitable for clinical evaluation in advanced human prostate cancer. Acknowledgements We wish to acknowledge the gifts of steroids from the Upjohn Corp. and Bristol Myers Co., and the Dunning AT3 tumor cells from Dr. D. Mickey. Portions of this work were supported by grant 256 from the American Cancer Society. DUNCAN, G. W., LYSTER,S.C., HENDRIX, J. W., CLARK,J. J., and WEBSTER, H.D. 1964. Biologic effects of melengestrol acetate. Fertil. Steril. 15: 419-432. GELLER,J. 1985. Rationale for blockage of adrenal as well as testicular androgens in the treatment of advanced prostate cancer. Sem. Oncol. 12: 28-35. GELLER,J . , ALBERT,J., GELLER,S., LOPEZ,D., CANTOR,T., and YEN, S. 1976. Effect of megestrol acetate (Megace") on steroid metabolism and steroid-protein binding in the human prostate. J. Clin. Endocrinol. Metab. 43: 1000-1008. GELLER,J., ALBERT,J., and YEN, S.C. 1978. Treatment of advanced prostate cancer with megestrol acetate. Urology, 12: 537-541. GELLER,J., ALBERT,J., YEN,S.S.C.. GELLER,S., and LOZA,D. 1981. Medical castration of males with megestrol acetate and small doses of diethylstilbestrol. J. Clin. Endocrinol. Metab. 52: 576-580. HUGGINS,C., STEVENS,R.E., JR., and HODGES,C.V. 1941. Studies on prostatic cancer. 11. The effects of castration on advanced carcinoma of the prostate gland. Arch. Surg. (Chicago), 43: 209-223. ISAACS,J.T., HESTON,W.D.W.. WEISSMAN, R.M., and COFFEY, D.S. 1978. Animal models of the hormone-sensitive and
68. 1990
-insensitive prostatic adenocarcinoma. Dunning R3327H, R3327H1, and R3327AT. Cancer Res. 38: 4353-4359. KIRK, D.N., PETROW,V., and WILLIAMSON,D.M. 1962. 6-Methyl-16-methylene steroids. Br. Pat. 886,619. S.A., PADILLA, G.M., and PETROW,V. 1987. A cornparison of the effects of castration and 6-methylene progesterone, I a Sa-reductase inhibitor, on the rat ventral prostrate.~iochem. ' Cell Biol. 65: 626-634. PADILLA,G.M., PETROW,V., MARTS,S.A., and MUKHERJI,S. 1985. Approaches to prostatic cancer chemotherapy using the Dunning R3327H prostatic adenocarcinoma. Prostate (N.Y.), 6: 129-143. PADILLA,G.M., YACULLO.R.C., PADILLA,J. J., and PETROW, V. 1986. Growth inhibition of androgen independent rat prostatic tumors. J. Cell Biol. 103: 27a. PADILLA,G.M., YUNMBAM,M.K., WHITSON,G.L., and PETROW,V. 1988. In vitro inhibition of cell proliferation and protein kinase-C activity of Dunning AT3 rat prostate tumor cells. 12th Biennial Conference of The International Cell Cycle Society at St. Petersburg Beach, FL, April 21-23. PETROW, V. 1976. Pharmaceutical preparations. U.S. Pat. 3,988,447. -1987. The dihydrotestosterone (DHT) hypothesis of prostate cancer and its therapeutic implications. Prostate (N.Y.) 9: 343-361. PETROW,V., PADILLA, G.M., MUKHERJI, S., and MARTS,S.A. 1984. Endocrine dependence of prostatic cancer upon dihydrotestosterone and not upon testosterone. J. Pharm. Pharmacol. 36: 352-353. QUARMBY, V.E., BECKMAN, W.C., JR., COOKE,D.B., LUBAHN, D.B., JOSEPH,D.R., WILSON,E.M., and FRENCH,F.S. 1990. Expression and localization of androgen receptor in the R-3327 Dunning rat prostatic adenocarcinoma. Cancer Res. 50: 735-739. SUCHOWSKY,V.G.K., BALDRATTI,G., ARCARI,G., and E. 1965. Die Beeinflussung von zentralen RegulaSCRASCIA, tionsmechanismen durch Steroide. Arzneim. Forsch. 15: 437-439.
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
PADILLA, G. M., YACULLO, R. C., PADILLA, J. J., PAYNE, B., and PETROW, V. 1990. Melengestrol acetate and megestrol acetate are prostatic tumor inhibiting agents. Biochem. Cell Biol. 68: 1181-1 188. We had previously reported that 6-methylene progesterone, an inhibitor of 5a-reductase, the enzyme which converts testosterone to dihydrotestosterone, markedly inhibited growth of the androgen-dependent Dunning R3327-H rat prostatic tumors. We now find that the progesterone derivatives melengestrol acetate (MGA) and megestrol acetate (MA) inhibit both the androgen-dependent (Dunning R3327-H) and the androgen-independent (Dunning R3327-AT3) prostatic tumors. Growth of the AT3 tumors was suppressed by -53% after 9 days of daily S.C. injections with MGA at 10 mg/kg body weight. MGA also caused a 54% weight reduction of the ventral prostate and a 53% reduction of the seminal vesicles. Adrenal weights were reduced by 42%. A 24-day oral treatment with MGA (at - 15-17 mg/(kg -day)) inhibited AT3 tumor growth by 59% and caused a weight reduction in the following tissues: prostate (46%), seminal vesicles (19%), testes (12%), and adrenals (52%). Under the same protocol, MA inhibited AT3 tumor growth by 32% and reduced the weight of the ventral prostate by 49% and the weight of the adrenals by 18070, but had no effect on the seminal vesicles and testes. The extent of the MGA-induced prostatic regression was accompanied by cytological changes similar to those effected by 6-methylene progesterone, i.e., shrinking of the acinar epithelium. The AT3 tumors in MGA-treated rats displayed a limited degree of apoptosis. Atrophy of the adrenal cortex and lowered plasma levels of corticosterone and dihydroepiandrosterone were also observed. A therapeutic role for MGA and MA against androgenindependent prostatic neoplasms in man is forecast by these observations. Key words: prostatic cancer, androgen independent, melengestrol acetate, growth inhibition, steroids.
PADILLA, G. M., YACULLO, R. C., PADILLA, J. J., PAYNE, B., et PETROW, V. 1990. Melengestrol acetate and megestrol acetate are prostatic tumor inhibiting agents. Biochem. Cell Biol. 68 : 1181-1188. Nous avons deja montre que la 6-methylkne progesterone, un inhibiteur de la 5a-reductase, enzyme qui transforme la testosterone en dihydrotestostbone, inhibe de facon marquee chez le rat la croissance des tumeurs prostatiques Dunning R3327-H dependantes des androgenes. Nous trouvons maintenant que le melengestrol acetate (MGA) et le megestrol acetate (MA), deux derives de la progestbone, inhibent a la fois les tumeurs prostatiques (Dunning R3327-H) dependantes des androgenes et les tumeurs (Dunning R3327-AT3) independantes des androgenes. A p r b 9 jours d'injections sous-cutanees de MGA a 10 mg/kg de poids corporel, la croissance des tumeurs AT3 est reduite d'environ 53%. Le MGA rkduit egalement de 54% le poids de la prostate ventrale et de 53% celui des vesicules seminales. Le poids des surrenales est aussi diminue de 42%. Un traitement oral de 24 jours avec le MGA (d'environ 15 a 17 mg/(kg.jour)) inhibe de 59% la croissance de la tumeur AT3 et rkduit le poids des tissus suivants : prostate (46%), vesicules seminales (19%), testicules (12%) et surrknales (52%). Dans les m&mesconditions, le MA inhibe de 32% la croissance de la tumeur AT3, rtduit de 49% le poids de la prostate ventrale et de 18% le poids des surrtnales, rnais il n'exerce aucun effet sur les vksicules seminales et les testicules. L'importance de la regression prostatique induite par le MGA s'accompagne de changements cytologiques semblables a ceux effectues par la 6-mkthylene progesterone : retrkcissement de I'epithClium acinaire. Chez les rats trait& avec le MGA, les tumeurs AT3 manifestent un degre limit6 d'apoptose. Nous observons tgalement une atrophie du cortex surrknalien et de faibles taux plasmatiques de corticosterone et de dihydroepiandrosterone. Ces observations nous permettent de prkdiie un rBle thbapeutique pour le MGA et le MA dans les nbplasmes prostatiques independants des androgenes chez l'homme. Mots clgs : cancer prostatique, dependance aux androgenes, mklengestrol acetate, inhibition de la croissance, sttroides. [Traduit par la revue]
Introduction Castration as a treatment of prostatic cancer was introduced by Huggins et al. (1941) on the premise that the testicular androgen, testosterone, was the main trophic hormone of the prostate and its derived neoplasms. Alone or with administration of estrogen, castration has stood the test of time as a therapeutic modality and still represents the main standby for management of the advanced disease. In 1962 MGA (Fig. 1) was discovered by Kirk et al. (1962) as part of a program directed towards new orally active contraceptive progestogens. MGA is an analogue both of medroxyprogesteroneacetate and of MA (Fig. 2), but differs
ABBREVIATIONS: MGA, melengestrol acetate (17-acetoxy-6MA, megestrol methyl-16-methylenepregna-4.6-diene-3,40-dione; acetate; DHT, dihydrotestosterone; DHEA-S, dihydroepiandrosterone sulfate; LD,,, mean lethal dose. ' ~ u t h o rto whom all correspondence should be addressed. Printed in Canada / Imprim6 au Canada
from them principally by possessing significant antiinflammatory and immunosuppressive activities (Duncan et al. 1964). MA has been successfully used as a palliative in advanced prostatic cancer in combination with diethylstilbestrol, thereby achieving a reversible chemical castration (Geller et al. 1978). In an early clinical trial, MGA proved to be effective in radium-resistant endometrial carcinoma. It was well tolerated by women on chronic administration. It was subsequently tried against a series of experimental tumors in the rat, including squamous cell carcinoma of the prostate, with varying degrees of tumor inhibition (Petrow 1976). Against this background, it was decided to examine the effects of MGA on the Dunning R3327-H (androgen dependent) and Dunning R3327-AT3 (androgen independent) transplantable rat prostate tumors (Isaacs et al. 1978). We find that MGA (and MA to a lesser extent) inhibits growth of both Dunning tumor lines and induces regression of the
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
BIOCHEM. CELL BIOL. VOL. 68, 1990
FIG. 1. Structural formula of MGA.
Dose, mg/kg body weight
FIG. 3. Dose-response of adult male CCFl rats to daily s.c. MGA injections, at the doses shown in the abscissa, for 26 days. CCFl rats (110-120 g) were implanted with Dunning R3327-H tumors and injected with MGA or carrier (see Materials and methods). Adrenal and tumour weights are shown on the left ordinate and the prostate weights on the right ordinate. Standard errors of the means are indicated (n = 5 rats per group; two tumors per rat). TABLE1. Dose-response of Dunning R3327-AT3 prostatic tumors to MGA injections (10 day study)
o ~s
FIG. 2. Structural formula of MA.
adrenals, prostate, and seminal vesicles. Portions of this work were reported earlier (Padilla,et al. 1986, 1988). Materials and methods Animals and tumor implantation Male Copenhagen x Fisher F1 rats (CFFl) bearing Dunning R3327-H tumors were obtained from Dr. Norman Altman of the Department of Comparative Pathology, School of Medicine. University of Miami, Miami, FL. Nontumorous CFFl rats (Charles River) were implanted in our laboratory with cryopreserved Dunning R3327-H and R3327-AT3 tumor cells. Cultured AT3 tumor cells were kindly provided by Dr. D. Mickey (Medical School, University of North Carolina, Chapel Hill, NC). AT3 cells were grown on RPMI medium supplemented with 10% fetal bovine serum or CPR-5 (Sigma, St. Louis, MO), insulin (1 IU/mL), DHT (5 mg/mL), retinol (0.1 mg/mL), amphotericin B (12.5 pg/mL), and gentamycin (25 pg/mL). Prior to tumor implantation and treatment with MGA or MA, rats were acclimatized for 2 weeks and allowed to reach a body weight of 100-150 g. Retired breeder rats (250-300 g) were also employed. Rats were kept in hanging cages at constant temperature, under a 12 h light : 12 h dark cycle, fed, and watered ad libitum. Saline solution (0.9% NaCl) was the sole source of drinking water for MGA-treated rats to prevent weight loss due to electrolyte depletion. Rats were implanted with tumors in the flanks either ( a ) via a s.c. injection of 0.2-0.5 mL of trypsin- or collagenase-dissociated tumor cells (-2 x lo6 cells/ injection, at >85% viability, trypan blue exclusion) or (b) by inserting 2- to 3-mm3 tumor pieces into a small s.c. pocket. All surgical operations were performed on rats anesthetized with xylazine-ketamine (3:4 v/v) i.p. injection (0.1 mL/100 g body weight) or methoxyflurane inhalation (Metofane, Pitman-Moore, Inc., Washington Crossing, NJ). Rats were sacrificed by C 0 2 asphyxiation prior to excision of tumors and removal of organs and tissue samples.
-
Dose (mg/kg)
Body weighta (g)
Tumor weightb (a)
Tumor volumeb (cm3)
0 2 5 10 20
92.3 k 4.1 96.9+ 5.1 82.1 k 3.5 78.8 k2.3 66.7 k 6.7'
33.64k0.28 2.88 k0.13 2.79k0.33 2.75 k 0.25 1.89 k 0.14'
3.48 k0.40 2.29k0.32 2.28 k0.50 2.16k0.48' 1.30k0.17'
NOTE:CFFl rats were used in this study. Values are means SE. 'n = 5. bn = 10. 'Significantly different from control at 95% (Scheffe F-test). +_
Administration of inhibitors Rats were weighed and given s.c. injections of MGA (Upjohn Corp., Kalamazoo, MI) or MA (Bristol-Myers Co., Evansville, IN) on a daily basis. Steroid treatment was initiated 12-16 days after implantation with the Dunning R3327-H tumors or 24 h after implantation with Dunning R3327-AT3 tumors. Both steroids were dissolved in a carrier solution of propylene glycol - ethanol (9:l V/V)and injected at volumes of 0.1 mL/100 g body weight to yield the appropriate doses given in the results. Control rats were injected with equal quantities of vehicle. MGA was also administered orally either as a suspension in 3% carboxymethyl cellulose or thoroughly mixed with dry powdered lab chow (Purina), which was moistened, formed into granules, and dried at 60°C overnight. Concentration of MGA in the food was verified by organic extraction and spectrophotometry against a standard curve. Histopathology Appropriate tissue samples of adrenals, prostates, and tumors were fixed in buffered formalin at necropsy and processed for microscopic examination by the Pathology Department. Microphotographs were taken on hematoxylin- and eosin-stained specimens.
PADILLA ET
AL.
1183
TABLE2. Effect of oral administration and s.c. injection of MGA on body weight, androgen-dependent tissues, and Dunning R3327-AT3 tumors in the rat (A) Oral administration (10 mg/kg, 21 days)' Treatment
Body weight (g)
Prostate (mg)
Adrenals (mg)
Seminal vesicles (mg)
Testes (g)
485 k 73
1.42 + 0.04
375 k 43d 23
1.35 k0.7 21
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
- -
Control (n = 5 ) MGA (n = 6) To inhibition
318k6.0
438 +44
25.2k 1.2
277 k 7.4b 13
2 6 ~ + 2 6 ~ 13.8k0.89' 39 46
(B) Subcutaneous injection (10 mg/kg, 9 days)' Treatment Control (n = 5) MGA (n = 6) To inhibition
Body weight (g)
Prostate (mg)
Adrenals (mg)
AT3 tumor (9)
121k 4.9
86+ 12
29.6 k 6.2
2.10 k 0.47
108 k 5.7d 11
40k7.1b 54
17.25 1 . 8 ~ 42
0.995 + 0 . 2 5 ~ 53
NOTE:All values are means
+_ SE. "Sprague-Dawley rats were used in this study.
bp 5 0.01.
2 '0.y';
Not slgnifrant. CCFl rats were used in this study. 'p 5 0.05.
Corticosterone and DHEA analyses The serum concentrations of corticosterone and DHEA-S were determined by a radioimmunoassay procedure available from Radioassay Systems Labs, Inc. (Carson, CA). Blood samples were removed, by cardiac puncture, immediately upon sacrifice into test tubes kept in ice. The serum was separated from the clot by centrifugation and stored at - 200Cuntil needed. Samples were analyzed in duplicate and the corticosterone and DHEA contents were calculated from appropriate standard curves.
was a dose-dependent inhibition in the growth of AT3 tumors in terms of their weight and volume. Tumor volumes were determined by measuring three tumor diameters with calipers and employing the ~~~~kformula ( V = L w H x 0.5236) (Isaacs et al. 1978). Although the treated rats were not placed on saline, there were no significant losses in body weight at MGA 20 mg/(kg'da~)-Oneway analysis of variance indicates that the weights and volumes of the AT3 tumors were significantly reduced only Statistical analyses at MGA doses in the range of 10-20 mg/kg compared with The statistical significance of the results was determined as controls. Tumor weights and volumes were not significantly necessary, using the Student's t-test or ANOVA with the Statview@ different from each other within this dose range. software package designed for the Macintosh' (Abacus Concepts, Experiments were then carried out to determine whether Inc., Berkeley, CA). MGA was orally active, using its regressive effect on androgen-target tissues as an index of in vivo activity in Results comparison with MGA given via S.C. injections. The results Figure 3 shows the results of an experiment in which rats are summarized in Table 2. Adult male Sprague-Dawley rats implanted with Dunning R3327-H tumors for -2 weeks were given daily oral doses of MGA at 10 mg/kg body were subsequently treated with MGA for a period of 26 days weight for 21 days by intubation (Table 2A). MGA caused at MGA doses in the range of 2-20 mg/kg body weight. It a 13% decrease in body weight, and the mean ventral is clear that there was a dose-dependent regression in the prostate weight was reduced by 39% and that of the adrenals weights of the prostate and adrenals. While the prostate was reduced by 46%. The seminal vesicles and testes underwent maximal involution at MGA doses > 10 mg/kg, displayed a lesser loss in their mean weight (23 and 21070, the adrenals were maximally affected by MGA at doses respectively). Table 2B shows the results of an experiment > 2 mg/kg compared with the controls. MGA had a less in which duration of MGA injections was reduced to 9 days. pronounced inhibitory effect on these tumors, a maximal Rats were implanted with AT3 tumor cells and injected daily effect being achieved at doses above 5 mg/kg. Rats treated . with MGA (10 mg/kg body weight) beginning 24 h later. with 2 mg MGA/kg experienced a 7.5% loss in body weight The results show that even though MGA injections caused compared with the untreated rats that showed a 9.2% weight a slight loss of body weight (1 1Vo), the mean weights of the gain. At higher doses MGA-treated rats lost an average of ventral prostate and adrenals were reduced by 54 and 42070, 15% of their body weight. In subsequent experiments this respectively. Note that the AT3 tumor mean weight was weight loss was alleviated by using a saline solution (0.9% reduced by -53%. In summary, the results in Table 2 NaCl w/v) as the sole source of drinking water. demonstrate that MGA had similar inhibitory effects, Table 1 summarizes the results of a 10-day study designed whether it was given orally or was injected subcutaneously. to assess the dose response of AT3 tumors to daily MGA It was thus of interest to compare the effects of MGA injections in the range of 2-20 mg/kg body weight. There and MA administered in the food to AT3 tumor-bearing
1184
BIOCHEM. CELL BIOL. VOL. 68, 1990
TABLE 3. Effect of MGA and MA on Dunning R3327-AT3 tumors, adrenals, androgen target organs, and body weight of CCFl rats
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
(24-day study)
Condition (n)
Tumorsa (g)
Adrenals (mi?)
Ventral prostate (mg)
Control (5) MGA (5)'
40.3 k4.5 16.6k4.3 (58.8)' 27.5k5.1 (31.8)'
62.8 k 3.9 30.4k5.9 (51.6)' 51.5k4.1 (18.2)'
636k90 345k82 (45.8)' 324k52 (49.1)'
MA (4jh
NOTE:All values are means f SE. %ach rat was implanted with two tumors. b ~ a c hsteroid was mixed with powdered lab chow to a concentration of 'Decrease in weight expressed as percent of control.
Seminal vesicles (mg)
Testes (g)
384k 25 311 k49 (19.1)' 382k58 (0.5)'
1.39k0.02 1.22k0.03 (12.2)' 1.32k0.06 (5.0)'
0.25 mg steroid/g
TABLE4. Effect of MGA treatment on serum DHEA-S and corticosterone levels in CCFl rats Treatment
DHEA-S bg/mL)
Corticosterone @g/mL)
Control (n = 5) MGA (n = 5)
2.27 k 0.13 1.44k0.94
0.82 + 0.04 0.19 k 0.01'
NOTE:Rats were injected daily with MGA (s.c.) for 9 days at 10 mg/kg. Data on adrenals, prostate and AT3 tumors is given in Table 2B. Values are means SE. "Estimated value due to low plasma levels.
*
rats. Retired breeder CCFl rats (250-300 g) were employed in this study. Rats were implanted with AT3 tumors as before and were fed powdered lab chow containing MGA or MA (0.25 mg/g food). Food consumption was monitored daily. As the rats consumed 15-20 g of lab chow per day, they ingested 15-17 mg of either steroid per kilogram body weight. Table 3 summarizes the results of this 24-day study. MGA and MA, respectively, caused a 58.9 and 31.8% tumor weight loss (or growth inhibition) compared with controls (column 2). MGA reduced the adrenal weight by 51.6%, whereas MA reduced it by 18.2% (column 3). Both steroids induced prostatic involution to the same extent (MGA = 45.8%, MA = 49.1 %, column 4). The seminal vesicles were essentially unaffected by MA (0.5% inhibition), but their weights were reduced by 19.1% by MGA (column 5). The weight of testes was reduced by 12.2% by MGA, whereas MA caused a 5% weight reduction (column 6). The last column in Table 3 shows that MGA induced a 22.6% gain in body weight, whereas MA increased it by 10.5%. The control rats showed an insignificant loss of weight (5.3%). Table 4 summarizes the results of radioimmunoassays of the serum levels of DHEA-S and corticosterone. These analyses were performed on blood samples from the rats used in the experiment in Table 2B. It is clear that MGA lowered the serum levels of both DHEA-S and corticosterone. The corticosterone levels in MGA-treated rats were reduced to such an extent that the readings were close to the lower limit of this assay. Hence, the values given in the table were estimates based on extrapolation of the standard curve slightly beyond its lower limit.
-
Body weight (g) Initial
Final
337 k 6 265k 17
319+8 325 + 17
266+ 13
294+ 16
food to yield a daily dose of
Body weight change (yo) - 5.3
+ 22 + 10.5
- 15-17 mg/kg.
Since approximately 90% of the DHEA-S is thought to originate from the zona reticularis of the adrenal cortex, measurements of DHEA-S levels may be used as indication of cortical regression and (or) loss of adrenal steroidogenesis. The cytological changes induced by MGA in the adrenals are shown in Figs. 4-7. The salient feature of these changes was a shrinkage of the adrenal cortex (Figs. 4 and 5), especially in the zona fasciculata (Figs. 6 and 7), which lost its characteristic cord-like cellular organization. The individual cells no longer appeared as distinct units, the nuclei being more closely "packed" and pycnotic. Dunning AT3 tumors, although devoid of a distinct acinar organization characteristicof the prostate gland (Isaacs et al. 1978), showed several cytological changes following MGA treatment (Figs. 8 and 9). The most prominent one was an overall shrinkage of the cells giving a crowded appearance to the nuclei (Fig. 9), some of which were surrounded by a clear halo, which may represent vacuolization of the cytoplasm and loss of perinuclear organelles (e.g., Golgi). These results also suggest an apoptotic impact by MGA on the AT3 tumor cells. We are presently examining the effects of MGA on the viability of AT3 tumor cells in culture. Preliminary results indicate that MGA has an antiproliferative effect on these cells (Padilla et al. 1988). In the prostate, MGA induced a reduction in the epithelial profile, i.e., changing it from columnar to cuboidal (cf. Figs. 10 and 11). A similar reduction in the height of this epithelium was previously obtained with 6-methylene progesterone (Marts et al. 1987). MGA did not appear to induce a marked hypertrophy in the intra-acinar stroma of this gland as with castration (Marts et al. 1987).
Discussion Complete androgen ablation through castration and treatment with diethylstilbestrol has been the primary therapeutic modality in the treatment of prostate cancer for almost half a century (Huggins et al. 1941). Undesirable side effects brought about by such interventions and failure to achieve complete remission of this neoplasm, with the recurrence of androgen-independent tumors, have prompted the search
FIG. 4. Cross-section through the adrenal cortex of control CCFl rats. Slides were stained with hematoxylin and eosin. Magnification = 10 x . FIG.5. Cross-section through adrenal cortex of CCFl rats treated with MGA for 12 days. Magnification = 10 x . FIG.6. Section of zona fasciculata of adrenal cortex of control CCFl rats. Magnification = 25 x . FIG.7. Section of zona fasciculata of adrenals from MGA-treated CCFl rats (as in Fig. 5). Portions of zona reticularis is present on the right. Magnification = 25 x .
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
PADILLA ET AL.
1185
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
1186 BIOCHEM. CELL BIOL. VOL. 68. 1990
PADILLA ET AL.
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
for alternate methods of treatment of prostate cancer in man. Recent investigations from our laboratory have revealed that in the Dunning prostatic tumor model, 6-methylene progesterone was a highly effective inhibitor of R3327-H tumor growth. We now report that MGA is also an effective inhibitor of these tumors. It should be noted that the Dunning R3327-H tumors require the presence of testosterone and (or) DHT for its growth (Isaacs et al. 1978; Petrow et al. 1984; Padilla et al. 1985; Petrow 1987). On the other hand, the Dunning R3327-AT tumors (from which the AT3 subline was derived) lack the 5a-reductase enzyme (Isaacs et al. 1978). It was, therefore, not surprising to find that the 5a-reductase inhibitor 6-methylene progesterone was largely ineffective against these tumors. The major finding in the present report was the marked in vivo inhibitory effects of MGA on the androgenindependent AT3 prostatic tumors (Tables 1-3). Megestrol acetate was also inhibitory towards these tumors, but to a lesser degree (Table 3). AT3 tumor inhibition was generally accompanied by involution of the prostate, the adrenals, and to a lesser extent the seminal vesicles. Moreover, both MGA and MA had no marked effect on the testes. As discussed below, reduction of testicular weight most likely results from inhibition of gonadotropin secretion by these progestins. Changes in the prostatic acinar epithelium induced by MGA were comparable to those previously obtained with 6-methylene progesterone (Marts et al. 1987); namely a reduction in the profile of the epithelial cells i.e., from columnar to cuboidal (Figs. 10 and 11). Such a cytological change is reminiscent of loss of secretory activity (Padilla et al. 1986). Yet involution of the ventral prostate was not as severe as that obtained by castration (Marts et al. 1987). That adrenal hormonal output is compromised by MGA was verified by a reduction in serum levels of the adrenal androgen DHEA and corticosterone (Table 4). Although we did not assay for adrenal mineralocorticoids, their absence in the serum was suggested by the loss in body weight, a condition which was alleviated by substituting saline as the sole source of drinking water to MGA-treated rats. Since the ACTH levels were not measured, we cannot rule out MGA inhibition of adrenal corticopropin production. The degree of adrenal and prostatic involution induced by MGA seemed to parallel the magnitude of AT3 tumor inhibition (40-50%, Tables 2 and 3). MA induced a lesser degree of adrenal involution, even though inhibition of AT3 tumors was comparable to that produced by MGA (Table 3, columns 2 and 3). Duncan et al. (1964) had observed a reduction in body weight, accessory sex glands, and adrenals in 26- to 28-day-old Sprague-Dawley rats given oral doses of MGA of 0.8-1.6 mg/day. In the immature male rat, orally administered MGA was found to be at least four times as potent as medroxyprogesterone acetate as a gonadotropin inhibitor (Duncan et al. 1964). Acute toxicity (LDSo)from a single dose of MGA in mice was only observed at doses larger than 2.5 g/kg body weight. (Duncan et al. 1964). Our results (Table 2), which are in general agreement with these
1187
studies, also revealed that orally administered MGA and MA for 24 days were well tolerated, as demonstrated by the increase in body weight (Table 3). The results of the present investigation demonstrate that the progesterone derivates MGA and MA inhibit the growth of androgen-independent AT3 tumors, an accepted animal model of advanced cancer in man. Recent studies show that AT3 tumors are also useful for studies on the molecular basis of their androgen independence (Quarmby et al. 1990). The results of the present investigation provide evidence that not only are MGA and MA orally active, their differential activity towards androgen-independent prostatic tumors may result from their metabolic conversion into distinct reactive species. These results also suggest that AT3 tumor inhibition may be partly due to a direct (cytotoxic) action on androgen-independent clones, as indicated by apoptotic changes in the tumor cells (Fig. 9). A significant trophic influence on AT3 tumors by the adrenal androgen DHEA is rendered unlikely by the recent experiments of Quarmby et al. (1990), which demonstrate that AT3 tumors lack androgen receptors, as determined by immunostaining procedures. AT3 tumors were found to contain very low levels of androgen receptor mRNA, compared with the prostate and the Dunning R3327-7 H tumor subline (Quarmby et al. 1990). The effects of MGA and MA on anterior pituitary gonadotropins and their combined role in the inhibition of prostatic neoplasms are somewhat more complex. In the parabiotic rat model MA was found to inhibit pituitary gonadotropin production (Suchowsky et al. 1965). MGA was likewise found to be a gonadotropin inhibitor (Duncan et al. 1964). These progestins were originally synthesized by Kirk et al. (1962) as contraceptive agents in women, but there is a paucity of studies in male animals or men. Geller et al. (1981) have shown that MA plus diethylstilbestrolproduced significant suppression of LH and FSH, but diethylstilbestrol itself played a significant role in gonadotropin suppression. MA alone was also shown by Geller et al. (1976) to suppress testosterone, LH, and FSH in man. Geller (1985) also considered the possibility that small amounts of DHT generated from adrenal androgens could provide a significant stimulus to prostate cancer cells. Geller (1985) noted that there are enzymatic activities within the prostate sufficient for conversion of adrenal androgens (androstenedione or DHEA) to DHT, thus providing a rationale for blocking the androgen output of the adrenals as well as the testis in the treatment of some forms of advanced prostatic cancers that retain some hormonal dependence. This experimental approach may not be applicable to androgen-independent prostatic neoplasms, especially if they lack androgen receptors, as is the case with Dunning AT3 rat prostatic tumors (Quarmby el al. 1990). In conclusion, the present study provides evidence, for the first time, that androgen-independent prostatic tumors are inhibited by progesterone derivatives such as MGA and MA, which possess minimal in vivo anti-androgenic activ-
-
FIG. 8. Section through R3327-AT3 tumor from control CCFl rat. Note absence of acini and rather enlarged nuclei with distinct nuclear architecture. Magnification = 40 x . FIG.9. Section through Dunning R3327-AT3 tumor from CCFl rat treated with MGA (as in Fig. 5). Note the closely packed nuclei, darker stain, and presence of a clear halo surrounding some of the nuclei (arrows). Magnification = 4 0 x . FIG.10. Section through portion of two acini of the prostate from control CCFl rats. Note the linear arrangement of the nuclei of columnar epithelial cells (arrows). Magnification = 40x. FIG. 1 1 . Section through prostate from CCFl rat treated with MGA (as in Fig. 5). The epithelial cells are now cuboidal (arrows). Magnification = 40 x .
1188
BIOCHEM. CELL BIOL. VOL.
Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by University of Nebraska Lincoln on 01/01/15 For personal use only.
ity at low doses. AS MGA has shown relatively few undesirable side effects in chronic administration to women, there is every possibility that it will prove to be equally benign in men and thus be suitable for clinical evaluation in advanced human prostate cancer. Acknowledgements We wish to acknowledge the gifts of steroids from the Upjohn Corp. and Bristol Myers Co., and the Dunning AT3 tumor cells from Dr. D. Mickey. Portions of this work were supported by grant 256 from the American Cancer Society. DUNCAN, G. W., LYSTER,S.C., HENDRIX, J. W., CLARK,J. J., and WEBSTER, H.D. 1964. Biologic effects of melengestrol acetate. Fertil. Steril. 15: 419-432. GELLER,J. 1985. Rationale for blockage of adrenal as well as testicular androgens in the treatment of advanced prostate cancer. Sem. Oncol. 12: 28-35. GELLER,J . , ALBERT,J., GELLER,S., LOPEZ,D., CANTOR,T., and YEN, S. 1976. Effect of megestrol acetate (Megace") on steroid metabolism and steroid-protein binding in the human prostate. J. Clin. Endocrinol. Metab. 43: 1000-1008. GELLER,J., ALBERT,J., and YEN, S.C. 1978. Treatment of advanced prostate cancer with megestrol acetate. Urology, 12: 537-541. GELLER,J., ALBERT,J., YEN,S.S.C.. GELLER,S., and LOZA,D. 1981. Medical castration of males with megestrol acetate and small doses of diethylstilbestrol. J. Clin. Endocrinol. Metab. 52: 576-580. HUGGINS,C., STEVENS,R.E., JR., and HODGES,C.V. 1941. Studies on prostatic cancer. 11. The effects of castration on advanced carcinoma of the prostate gland. Arch. Surg. (Chicago), 43: 209-223. ISAACS,J.T., HESTON,W.D.W.. WEISSMAN, R.M., and COFFEY, D.S. 1978. Animal models of the hormone-sensitive and
68. 1990
-insensitive prostatic adenocarcinoma. Dunning R3327H, R3327H1, and R3327AT. Cancer Res. 38: 4353-4359. KIRK, D.N., PETROW,V., and WILLIAMSON,D.M. 1962. 6-Methyl-16-methylene steroids. Br. Pat. 886,619. S.A., PADILLA, G.M., and PETROW,V. 1987. A cornparison of the effects of castration and 6-methylene progesterone, I a Sa-reductase inhibitor, on the rat ventral prostrate.~iochem. ' Cell Biol. 65: 626-634. PADILLA,G.M., PETROW,V., MARTS,S.A., and MUKHERJI,S. 1985. Approaches to prostatic cancer chemotherapy using the Dunning R3327H prostatic adenocarcinoma. Prostate (N.Y.), 6: 129-143. PADILLA,G.M., YACULLO.R.C., PADILLA,J. J., and PETROW, V. 1986. Growth inhibition of androgen independent rat prostatic tumors. J. Cell Biol. 103: 27a. PADILLA,G.M., YUNMBAM,M.K., WHITSON,G.L., and PETROW,V. 1988. In vitro inhibition of cell proliferation and protein kinase-C activity of Dunning AT3 rat prostate tumor cells. 12th Biennial Conference of The International Cell Cycle Society at St. Petersburg Beach, FL, April 21-23. PETROW, V. 1976. Pharmaceutical preparations. U.S. Pat. 3,988,447. -1987. The dihydrotestosterone (DHT) hypothesis of prostate cancer and its therapeutic implications. Prostate (N.Y.) 9: 343-361. PETROW,V., PADILLA, G.M., MUKHERJI, S., and MARTS,S.A. 1984. Endocrine dependence of prostatic cancer upon dihydrotestosterone and not upon testosterone. J. Pharm. Pharmacol. 36: 352-353. QUARMBY, V.E., BECKMAN, W.C., JR., COOKE,D.B., LUBAHN, D.B., JOSEPH,D.R., WILSON,E.M., and FRENCH,F.S. 1990. Expression and localization of androgen receptor in the R-3327 Dunning rat prostatic adenocarcinoma. Cancer Res. 50: 735-739. SUCHOWSKY,V.G.K., BALDRATTI,G., ARCARI,G., and E. 1965. Die Beeinflussung von zentralen RegulaSCRASCIA, tionsmechanismen durch Steroide. Arzneim. Forsch. 15: 437-439.
