Effect of Early Treatment of Prostate CancerWith the 5a-Reductase Inhibitor Turosteride in

Dunning R3327 Prostatic Carcinoma in Rats

Tiziana Zaccheo,* Donata Giudici, and Enrico di Salle

Laboratory of Experimental Endocrinology, Department of Oncology Research,Pharmacia and Upjohn, Nerviano, Italy

BACKGROUND. Turosteride, a selective 5a-reductase inhibitor, was reported to be effectivein inhibiting the growth of established tumors in the Dunning R3327 rat prostatic carcinomamodel. We evaluated the preventive effect of turosteride when administered during thelatency period in this prostatic tumor model.METHODS. Turosteride was given orally, 6 days a week for 10–15 weeks, starting at differenttimes: 1) 5 weeks after tumor implantation, when tumors were not yet palpable, or 2) 1 dayafter tumor implantation. In each experiment, one group of animals was castrated on the firsttreatment day.RESULTS. When treatment started 5 weeks after tumor implantation, neither turosteride (at50 and 200 mg/kg/day) nor castration reduced tumor incidence (91–100%). Tumor growthwas reduced in groups treated with the highest dose of turosteride and in castrated rats.When treatment started 1 day after tumor implantation, castration resulted in a 62% tumorincidence compared to 100% in controls, while turosteride at 200 mg/kg/day was not effec-tive in reducing tumor incidence. However, as in the previous experiment, the compound washighly effective in reducing tumor growth.CONCLUSIONS. The antitumor activity profile of turosteride seems not to be related to thetiming of treatment. Given either 5 weeks or 1 day after tumor implantation, the compounddid not reduce tumor take, while it maintained effective tumor growth-inhibiting activity inboth cases. Prostate 35:237–242, 1998. © 1998 Wiley-Liss, Inc.

KEY WORDS: FCE 26073; tumor prevention; experimental model

INTRODUCTION

Androgens are necessary for the growth, mainte-nance, and functional activity of the prostate gland.Androgens are also involved in prostatic cancer de-velopment, and modification of hormonal stimulationmay be useful in preventing or inhibiting tumorgrowth [1]. In normal and malignant prostatic cells,testosterone is irreversibly metabolized via the 5a-reductase enzyme to 5a-dihydrotestosterone (DHT),which appears to be the major intracellular androgen.In fact, the DHT concentration in these tissues ishigher than that of testosterone, and the androgen re-ceptor has a higher affinity for DHT than for testos-terone [2–4]. Therefore, reduction of DHT synthesisthrough inhibition of 5a-reductase activity can pro-

vide an interesting approach to androgen deprivationfor prostatic tumor control.

Turosteride (FCE 26073; 1-[4-methyl-3-oxo-4-aza-5a-androstane-17b-carbonyl]-1,3-diisopropylurea) is anew, potent, and selective 5a-reductase inhibitor [5].When tested on human and rat prostatic 5a-reductases, the compound showed a similar degree ofpotency, with IC50 values of 55 and 53 nM, respec-tively [6]. Turosteride was found to be more effectiveon the type 2 than type 1 isoform of the human re-

*Correspondence to: Dr. Tiziana Zaccheo, Department of OncologyResearch, Pharmacia and Upjohn, Viale Pasteur 10, 20014 Nerviano,Italy. E-mail: tiziana.zaccheo@eu.pnu.comReceived 18 September 1996; Accepted 5 November 1997

The Prostate 35:237–242 (1998)

© 1998 Wiley-Liss, Inc.

combinant enzyme [7]. The compound showed nonoteworthy binding to receptors for androgens, estro-gens, progesterone, glucocorticoids, and mineralocor-ticoids [8]. Twenty-day oral administration of turoste-ride to adult male rats at doses of 3, 10, or 30 mg/kg/day reduced prostatic DHT content by 61%, 74%, and78%, respectively, without any change in serum lu-teinizing hormone (LH) and serum and prostatic tes-tosterone (T) [8]. In addition, the compound was re-ported to be effective in inhibiting the growth of es-tablished tumors in the R3327 Dunning rat prostaticcarcinoma model: at an oral dose of 200 mg/kg/day,given for 9 weeks, the compound reduced tumorgrowth by 45% and ventral prostate weight by 60%[9]. No changes in serum LH and T levels were ob-served in these animals, thus confirming that turoste-ride is devoid of any antiandrogenic and antigonado-tropic effects.

In this study we investigated the tumor-preventiveeffect of turosteride, administered in the latency periodof the Dunning tumor. This tumor takes at least 6weeks to become a palpable subcutaneous nodule af-ter transplantation. Turosteride treatment started 5weeks after tumor implantation in the first experimentand on the day of tumor implantation in the second.Tumor incidence and tumor growth were recorded, aswell as the effect of the compound on endocrine organweight.

MATERIALS AND METHODS

Animals

Male Copenhagen rats, weighing approximately200 g, were supplied by Harlan Nossan Srl (Correz-

zana, Italy). Animals were housed in temperature-controlled rooms (22 ± 2°C) on a circadian rhythm of12 hr of light (6 AM–6 PM) and 12 hr of darkness.

Prostatic Tumor Model

The Dunning R3327 prostatic carcinoma was kindlyprovided by Dr. H. Altman (Papanicolaou Cancer Re-search Center, University of Miami, Miami, FL) andmaintained by serial transplantation in Copenhagenrats. The tumor was passed by harvesting fresh tumor,dicing it with sterile scissors in sterile 0.9% NaCl so-lution, and aseptically implanting a single tumor frag-ment (approximately 3–4 mm in size) subcutaneouslyin the flank of the recipient animal under mild diethyl-ether anesthesia.

Treatment and Experimental Setup

Turosteride (synthesized in the Pharmacia and Up-john Research Laboratories, Nerviano, Italy) was sus-pended in 0.5% Methocel (methylcellulose 400, DowChemicals, Midland, MI) containing 0.4% Tween-80(Merck, Darmstadt, Germany) and administered orallyin a volume of 5 ml/kg, 6 days a week for 10–15weeks. Castration was performed on the first treat-ment day under diethyl-ether anesthesia.

In the first experiment, treatment started 5 weeksafter tumor implantation. Rats were divided into fourgroups (n = 23): controls (treated orally with the ve-hicle), turosteride 50 mg/kg/day, turosteride 200mg/kg/day, and castration. In the second experiment,treatment started the day after tumor implantationand animals were divided into three groups(n = 24–30): controls (treated orally with the vehicle),turosteride 200 mg/kg/day, and castration.

Tumor incidence was recorded, and after the tumorbecame palpable, tumor growth was followed by mea-suring the two perpendicular diameters with calipers.Tumor weight was calculated according to the for-mula

d2 × D/2,

where d is the minimum and D is the maximum di-ameter [10]. Only animals with a tumor were includedin calculation of the tumor growth curve. The areaunder the tumor weight vs. time curve (AUC) wascalculated by the linear trapezoidal method up to 15weeks. Animals were sacrificed at the end of the 15thweek (4 hr after the last dose), and the ventral prostateand seminal vesicles were removed and weighed.

RESULTS

Effect on Tumor Incidence and Growth

To determine if turosteride could prevent tumorgrowth, animals were inoculated with the tumor and

Fig. 1. Effect of 10-week treatment with turosteride on tumorincidence in the Dunning prostatic tumor model in rats (23 animalsper group). Turosteride was given orally, 6 days a week, starting 5weeks after tumor transplantation. Castration was performed onthe first treatment day.

238 Zaccheo et al.

left untreated for 5 weeks. At that time, tumors werenot yet palpable and animals were randomized intofour groups: a control group, two turosteride-treatedgroups (50 and 200 mg/kg/day p.o.), and a castratedgroup. Figure 1 and Table I show tumor incidenceduring the 10-week treatment period. The percentageof animals with tumors reached 91–100% at the end ofthe treatment period, and no difference between con-trol, turosteride-treated, and castrated groups was ob-served. Despite the similar tumor incidence, tumorgrowth was reduced in the group treated with thehighest dose of turosteride and in castrated rats (Fig.2). The AUC values (g × week) are reported in Table I.The Dunning R3327 tumor grew progressively in thecontrol group (AUC, mean ± SE = 28.4 ± 4.4), whereasin castrated animals tumor growth was markedly de-creased (AUC = 2.8 ± 0.6). Turosteride at the oral doseof 50 mg/kg/day was not effective in inhibiting tumor

growth (AUC = 23.1 ± 3.5), while at the dose of 200mg/kg/day the compound significantly decreased tu-mor growth (AUC = 13.5 ± 2.3) by 53%. A furtherexperiment was carried out to examine the effect ofturosteride treatment starting 1 day after tumor im-plantation. Animals were inoculated with the Dun-ning tumor and then randomly allocated to the fol-lowing experimental groups: control, turosteride (200mg/kg/day p.o.), or castration. Treatment lasted 15weeks. As shown in Figure 3 and Table II, all controlrats showed tumors by week 12; turosteride did not re-duce tumor incidence, while castration, performed 1day after tumor implantation, resulted in a tumor inci-dence of only 62%. However, as in the previous expe-riment, turosteride was effective in reducing tumorgrowth; castration caused a marked decrease of tumorgrowth in those animals that developed tumors (Fig. 4).The AUC values reported in Table II show a signifi-

TABLE I. Effect of 10-Week Treatment With Turosteride, Starting 5 Weeks After Tumor Implantation, on DunningProstatic Tumor Growth in Rats

Treatmentgroup

Dose mg/kg/dayp.o.

No. of rats with tumor/total no. of rats

Tumor growth

AUC(0–15w) (g × week)aInhibition

(%)

Control 22/23 28.4 ± 4.4b

Turosteride 50 21/23 23.1 ± 3.5 19Turosteride 200 23/23 13.5 ± 2.3 53*Castration 21/23 2.8 ± 0.6 90*

aAUC, area under the tumor weight vs. time curve, calculated by the linear trapezoidal method up to 15 weeks.bMean ± SE (including only animals with tumors).*P < 0.01 vs. control group (Dunnett’s test).

Fig. 2. Effect of 10-week treatment with turosteride on tumorgrowth in the Dunning prostatic tumor model in rats (21–23 ani-mals per group). Turosteride was given orally, 6 days a week,starting 5 weeks after tumor transplantation. Castration was per-formed on the first treatment day.

Fig. 3. Effect of 15-week treatment with turosteride on tumorincidence in the Dunning prostatic tumor model in rats (24–30animals per group). Turosteride was given orally, 6 days a week,starting 1 day after tumor transplantation. Castration was per-formed 1 day after tumor transplantation.

Turosteride in Rat Prostatic Tumor 239

cant inhibition of tumor growth by turosteride (62%)and by castration (92%).

Effect on Organ Weight

The ventral prostate and seminal vesicles of tumor-bearing rats were removed and weighed at the 15-week sacrifice. Since turosteride treatment and castra-tion slightly reduced body weight gain, the relativeweights of these organs are reported in Figures 5 and6. Turosteride, at 50 mg/kg/day given for 10 weeks,caused a decrease of ventral prostate (57%) and semi-nal vesicle (40%) weight (Fig. 5). At the dose of 200mg/kg/day, the compound reduced prostate andseminal vesicle weights by 75% and 57%, respectively.Castration resulted in a marked reduction of prostate(92%) and seminal vesicle (79%) weight.

In the second experiment, turosteride at the dose of200 mg/kg/day, given for 15 weeks, reduced prostateweight by 79% and seminal vesicle weight by 59%.Castration caused a respective reduction of 96% and87% in the weight of these organs.

DISCUSSION

Inhibition of 5a-reductase provides a novel and se-lective approach to DHT deprivation in DHT-targettissues that avoids several side effects caused by totalandrogen deprivation through androgen receptor an-tagonists or LHRH agonists [11,12]. Specific inhibitorsof 5a-reductase may be useful in controlling patho-logical conditions dependent on DHT, such as benignprostatic hyperplasia (BPH) [13]. While DHT seems tobe the main androgen stimulating and maintaining thegrowth of BPH, it is not certain whether the androgendependence of prostate cancer is also mainly related toDHT. Experimental data on the activity of 5a-reductase inhibitors in the Dunning R3327 rat prostat-

ic tumor model have shown conflicting results. Thistumor model is well-characterized. It shares severalcharacteristics with human prostate cancer, includingslow growth rate, responsiveness to hormonaltherapy, and histology similar to that of a well-differentiated carcinoma [14]. Epristeride, a noncom-petitive inhibitor of 5a-reductase [15], has been re-ported to be inactive on the H subline of the Dunningtumor but highly effective on the G subline of thesame tumor [16]. Since the two sublines have differentlevels of intratumor 5a-reductase activity (high in theH subline, low in the G subline), it was suggested thatincomplete inhibition of the enzyme, with consequentincomplete depletion of DHT, could account for thelack of effect of 5a-reductase inhibitors in some tumormodels. Finasteride, the prototype of 5a-reductase in-hibitors, which has been made available for thetherapy of symptomatic BPH [17,18], has been re-

TABLE II. Effect of 15-Week Treatment With Turosteride, Starting 1 Day After Tumor Implantation, on DunningProstatic Tumor Growth in Rats

Treatmentgroup

Dose mg/kg/dayp.o.

No. of rats with tumor/total no. of rats

Tumor growth

AUC(0–15w) (g × week)aInhibition

(%)

Control 28/28 65.3 ± 10.9b

Turosteride 200 24/24 25.1 ± 4.2 62*Castration 19/30** 5.2 ± 1.2 92*

aAUC, area under the tumor weight vs. time curve, calculated by the linear trapezoidal method up to 15 weeks.bMean ± SE (including only animals with tumors).*P < 0.01 vs. control group (Dunnett’s test).**P < 0.01 vs. control group (Pearson’s chi-square test).

Fig. 4. Effect of 15-week treatment with turosteride on tumorgrowth in the Dunning prostatic tumor model in rats (19–28 ani-mals per group). Turosteride was given orally, 6 days a week,starting 1 day after tumor transplantation. Castration was per-formed 1 day after tumor transplantation.

240 Zaccheo et al.

ported to be inactive in the Dunning prostatic carci-noma in rats up to a dose of 100 mg/kg [16,19]. Thefailure of finasteride to influence tumor growth atdoses that did reduce the size of a normal prostate washypothesized to be due to the fact that, whereas DHTis fundamental for maintaining normal prostatic func-tion, the androgenic stimulus to prostatic cancer canbe provided by either DHT or T, and T tissue level isincreased after administration of a 5a-reductase in-hibitor.

We previously showed that turosteride, given torats with established Dunning prostatic tumors, hadno effect on tumor growth at the oral dose of 50mg/kg, whereas it was found to reduce tumor growthby 45% at 200 mg/kg [9]. At variance with other 5a-reductase inhibitors, the reduction in prostatic DHTcontent caused by turosteride is not coupled with anincrease in prostatic T, which remains unchanged [6].This property of turosteride could partly contribute toits antitumor activity in this experimental model, sinceboth DHT and T were hypothesized to be important intumor growth stimulation [19]. The lack of an increaseof T cannot be ascribed to an antigonadotropic effectof turosteride [8]; it could more likely be due to a

slight inhibition of other steroidogenic enzymes, e.g.,17–20 desmolase.

In the present study we investigated whether tu-rosteride dosing early after tumor implantation couldresult in better inhibition of Dunning tumor growth orcould impair tumor take. Results show that, whengiven from either 5 weeks or 1 day after tumor im-plantation, turosteride at the oral dose of 200 mg/kgdid not reduce tumor take, but was effective in reduc-ing tumor growth by 53% and 62%, respectively.These and previous results suggest that the activity ofthe compound was not related to the time when treat-ment was started. In fact, only a slight increase in theantitumor response to turosteride was observed whenthe compound was administered starting soon aftertumor implantation (62% inhibition of tumor growth),rather than shortly before the appearance of the tu-mors (53% inhibition) or in rats with already measur-able tumors (45% inhibition) [9]. Castration, when per-formed 5 weeks after tumor implantation, did not re-duce tumor incidence, while it reduced tumor take by38% when performed 1 day after tumor implantation;in addition, the growth of tumors that developed aftercastration was markedly reduced. This again indicates

Fig. 5. Effect of 10-week treatment withturosteride on the relative organ weights ofrats bearing the Dunning prostatic carcinoma.Bars represent mean ± SE ( 23 animals pergroup). **P < 0.01 vs. controls by Dunnett’stest.

Fig. 6. Effect of 15-week treatment with tu-rosteride on the relative organ weights of ratsbearing the Dunning prostatic carcinoma. Barsrepresent mean ± SE (24–30 animals pergroup). **P < 0.01 vs. controls by Dunnett’stest.

Turosteride in Rat Prostatic Tumor 241

that androgens play an important role not only in sup-porting the growth of the Dunning tumor, but also inthe early phase of its implantation. However, onlycomplete suppression of androgen stimulation by cas-tration seems to significantly impair tumor take in thisexperimental model.

The hypothesis that decreased androgen stimula-tion of prostatic tissue leads to a lower incidence ofprostate cancer is supported by the evidence that menwith low 5a-reductase activity have a lower risk ofprostate cancer [20]. In humans, 5a-reductase inhibi-tors, by reducing the synthesis of DHT without influ-encing the circulating levels of testosterone, may pre-vent the development of prostatic cancer or inhibitsubclinical disease without the toxicity of systemic an-drogen blockade. In order to test this hypothesis, astudy with finasteride in the prevention of humanprostate cancer has already been initiated [1]. In addi-tion, experimental models of chemoprevention couldhelp us to understand the value of these compoundsin preventing tumor growth.

REFERENCES

1. Ford LG, Brawley OW, Perlman JA, Nayfield SG, Johnson KA,Kramer BS: The potential for hormonal prevention trials. Cancer1994;74:2726–2733.

2. Hammond GL: Endogenous steroid levels in the human pros-tate from birth to old age: A comparison of normal and diseasedtissues. J Endocrinol 1978;78:7–19.

3. Klein H, Bressel M, Kastendieck H, Voigt KD: Quantitative as-sessment of endogenous testicular and adrenal sex steroids andof steroid metabolizing enzymes in untreated human prostaticcancerous tissue. J Steroid Biochem 1988;30:119–130.

4. Liao S, Liang T, Tymoczko JL: Structural recognition in interac-tions of androgens and receptor proteins and in their associationwith nuclear components. J Steroid Biochem 1972;3:401–407.

5. di Salle E, Briatico G, Giudici D, Ornati G, Nesi M, Panzeri A:17b-acylurea derivatives of 4-azasteroids as inhibitors of testos-terone 5a-reductase. J Steroid Biochem Mol Biol 1992;41:765–768.

6. di Salle E, Giudici D, Briatico G, Ornati G, Panzeri A: Hormonaleffects of turosteride, a 5a-reductase inhibitor, in the rat. J Ste-roid Biochem Mol Biol 1993;46:549–555.

7. Iehle C, Delos S, Guirou O, Tate R, Raynaud JP, Martin PM:Human prostatic steroid 5 alpha-reductase isoforms—A com-parative study of selective inhibitors. J Steroid Biochem Mol Biol1995;54:273–279.

8. di Salle E, Briatico G, Giudici D, Ornati G, Panzeri A: Endocrineproperties of the testosterone 5a-reductase inhibitor turosteride(FCE 26073). J Steroid Biochem Mol Biol 1994;48:241–248.

9. Zaccheo T, Giudici D, di Salle E: Effect of turosteride, a 5a-reductase inhibitor, on the Dunning R3327 rat prostatic carci-noma. Prostate 1997;30:85–91.

10. Geran RI, Greenberg NH, McDonald MM, Schumacher AM,Abbott BJ: Protocol for screening chemical agents and naturalproducts against animal tumors and other biological systems.Cancer Chemother Rep 1972;3:1–103.

11. Vogelzang NJ, Kennealey GT: Recent developments in endo-crine treatment of prostate cancer. Cancer 1992;70:966–976.

12. Hsieh W-S, Simons JW: Systemic therapy of prostate cancer.New concepts from prostate cancer tumor biology. Cancer TreatRev 1993;19:229–260.

13. Metcalf BW, Levy MA, Holt DA: Inhibitors of steroid 5a-reductase in benign prostatic hyperplasia, male pattern bald-ness and acne. Trends Pharmacol Sci 1989;10:491–495.

14. Isaacs JT, Coffey DS: Adaptation versus selection as the mecha-nism responsible for the relapse of prostatic cancer to androgenablation therapy as studied in the Dunning R3327-H adenocar-cinoma. Cancer Res 1981;41:5070–5075.

15. Lamb JC, English H, Levandoski PL, Rhodes GR, Johnson RK,Isaacs JT: Prostatic involution in rats induced by a novel 5a-reductase inhibitor, SK&F 105657: Role for testosterone in theandrogenic response. Endocrinology 1992;130:685–694.

16. Lamb JC, Levy MA, Johnson RK, Isaacs JT: Response of rat andhuman prostatic cancers to the novel 5a-reductase inhibitor,SKF 105657. Prostate 1992;21:15–34.

17. Gormley GJ, Stoner E, Bruskewitz RC, Imperato-McGinley J,Walsh PC, McConnell JD, Andriole GL, Geller J, Brachen BR,Tenover JS, Vaughan ED, Pappas F, Taylor A, Binkowitz B, NgJ: The effect of finasteride in men with benign prostatic hyper-plasia. N Engl J Med 1992;327:1185–1191.

18. Stoner E: The clinical development of a 5a-reductase inhibitor,finasteride. J Steroid Biochem Mol Biol 1990;37:375–378.

19. Brooks JR, Berman C, Nguyen H, Prahalada S, Primka RL, Ras-musson GH, Slater EE: Effect of castration, DES, flutamide, andthe 5a-reductase inhibitor, MK-906, on the growth of the Dun-ning rat prostatic carcinoma, R3327. Prostate 1991;18:215–227.

20. Ross RK, Bernstein L, Lobo RA, Shimizu H, Stanczyk FZ, PikeMC, Henderson BE: 5-alpha-reductase activity and risk of pros-tate cancer among Japanese and US white and black males.Lancet 1992;339:887–889.

242 Zaccheo et al.