CA2198119A1 - Methods of inhibiting endometrial cancer - Google Patents

Abstract

A method of inhibiting endometrial cancer comprising administering to a human in need thereof an effective amount of a compound having formula (I) wherein R1 and R3 are independently hydrogen, -CH3, (a), or (b), wherein Ar is optionally substituted phenyl; R2 is selected from the group consisting of pyrrolidine, hexamethyleneimino, and piperidino; or a pharmaceutically acceptable salt of solvate thereof.

Description

W096105833 ~ ~ ~ PCT1US95110651 METHODS OF INHIBITING ENDOMETRIAh CANCER
The uterine lining (endometrium) is composed of tissue, blood vessels, and -glands that grow when stimulated by the hormone estrogen. In women with normal menstrual cycles, hormonal--fluctuations triggerthe growth and shedding of the~endometrium each month. If conception occurs, the endometrium nourishes the developing embryo.

Most cases of endometrial carcinoma are associated with a precursor lesion termed endometrial hyperplasia.~~ The classification of endometrial hyperplasia is based on the presence or absence of cytologic atypia, the-presence of dysplasia, and the degree-of complexity of-the architectural pattern. Cytologic atypia is the most predictive criterion for the likelihood of progression-to carcinoma.

In simple or cystic hyperplasia with cytologic atypia present there is about an 8~ chance of progression to cancer. With complex or adenomatous hyperplasia with cytologic atypia-present, there is 29$ chance. When no cytologic atypia is present, the progression rate is 1$ for simple and 3~ for complex hyperplasia.

With continuously elevated estrogen levels, the endometrium remains in its growth phase at all time, in some cases Leading to. an overabundance of endometrial tissue or endometrial hyperplasia. Overgrowth of the endometrium is often a benign condition, but it can also be a precursor-of-endometrial cancer. Because of this risk, doctors urge women to avoid long-term unopposed estrogen therapy, which-can-cause endometrial overgrowth if the lining is not continually shed, and to seek prompt treatment for conditions that cause excessive estrogen ' production.(The-use of progesteronein-hormone replacement therapy causes breakdown bleeding and shedding of endometrial build up).

W096105833 . , -F, , PCTIUS95110651 _2_ Doctors base their treatment decisions on several factors. First, they examine the cells obtained in the biopsy or D&C. If the cells are normal but simply over , abundant, future development of cancer is less likely than if the cells are atypical, displaying-enlarged nuclei-and , other unusual features: In some cases, a D&C will show that cancer-has already developed.
Endometrial cancer is the most common ' gynecologic pathology and-the fourth most common malignancy in women, after breast, colorectal, and lung caricer. --°
Approximately 30,000-40,000 new cases of endometrial cancer are diagnosed each year. 4dhile it is the most common -pathology, most patients present in theearly stage.
Endometrial cancer affects mainly postmenopausal women, as the average age at diagnosis is 58 years, and fewer than 5~ of cases occur prior to age 40. The incidence of endometrial cancer'is higher among women with a history of breast, endometrial, or ovarian malignancies, and also-zn women that belong to a high socioeconomic , status.
The most significant risk factors for endometrial cancer are bbesity and the presence of estrogen unopposed by progesterone.
The inaccuracy in clinical staging of endometrial carcinoma impedes optimal therapy and analysis of treatment results. Unless-m~tastatic-or-systemic disease is identified, the initial approach for all _ medically fit patients is currently a total abdominal hysterectomy/bilateral salpingo=oophorectomy.
Adjunctive therapy, if needed; can be planned, depending- on whether the surgical-pathologic findings ' ' indicate intrauterine only or extrauterine disease. The patient may receive external beam radiation to the pelvis ' if pelvic nodes are positive and of external beam radiation to the para-aortic fields if those nodes are positive.
Patients with other sites--of extrauterine disease may VUO 96105833 ~ ~ ' ~. ~~ 1. ~ C~ pGT%fIS95110651 _3_ , require whole abdominal irradiation. -Some patients may need systemic therapy in addition to radiation therapy, . depending on sites-of spread.

Patients with Stage II disease are at higher risk for having extrauterine disease and recurrence. If the cervix-is of normal size and grossly normal, one appraach is an extrafasdial TAH/BSO with complete surgical staging followed-,by postoperative irradiation. With gross cervical involvement, two options are available. The first is whole pelvic irradiation followed by one intracavitary implant, which is then followed by a TAHIBSO and para-aortic lymph-node--sampling. The second option-is a radical hysterectomy. BSO, and pelvic and para-aortic lymphadenectomy with irradiation tailored to the surgical findings, if necessary.

In surgical Stage III disease,-primary surgery with the use of a TAH/BSO with tumor debulking may be attempted. Extrapelvic disease, depending on the site and extent, may necessitate extended field irradiation, systemic chemotherapy, or hormone therapy. Patients with Stage III disease, by virtue of-vaginal or parametrial extension, need a thorough metastatic survey and then irradiation.

Most patients with Stage IV disease are best treated with systemic therapy, which-includes hormones or chemotherapy. Pelvic irradiationor hysterectomy is reserved for palliative control purposes-:-Patients with recurrent endometrial cancer in the pelvis may-be treated with radiotherapy.

3.0 Unfortunately, the majority of these patients also have " distant metastases as well. Isolated central recurrences in the pelvis after irradiation are'rare_ However, if this situation does occur, selected patients may be candidates for pelvic exenterative.surgery. The majority of patients with recurrent disease are treated with hormones or chemotherapy.

W0961OS833 ;' ,~ PCTIUS95/10651 C.. 1 Progestins have been_used for.decades=to treat recurrent endometrial cancer. The overall response to-:-progestins is approximately 25$, although recent trials demonstrate lower response gates, in the.range of 15 to 20~. Patients with endometrial carcinoma with progesterone-positive and-estrogen-positive receptors have a better response to endocrine therapy. Most patients-pith positive receptors respond to progestins, whereas only-15~
with negative receptors respond. Medroxyprogesterone-:=:=_.
acetate (Provera) and megestrol-acetate (Megace) are the agents most commonly used. Tamoxifen (NOlvadex) has also been used to treat patients with recurring endometrial cancer, and responses are usually seen in patients who -have previously responded to progestins.
Several cytotoxic agents have activity for -.
endometrial cancer, but responses are short-lived, and the treatment for advanced and recurrent-disease is considered palliative. The two-molt active single agents are doxorubicin and cisplatin_-- Many combinations of cytotaxic agents have been used, but the results of multiagent chemotherapy do not appear to be significantly better than those of single-agent chemotherapy.
This invention provides methods of inhibiting .
25. endometrial cancer comprising administeringto a human in need thereof an effective amount of a compound of formula I
oCHZ-Rz R
(I) wherein R1 and R3 are independently hydrogen, O O
~ ~~ a -C-(C1-C6 alkyl) or -C-Ar wherein Ar is optionally substituted phenyl; -RZ is selected from the group consisting of -pyrrolidino, hexamethyleneimino, and piperidino; and pharmaceutically acceptable salts and solvates thereof.
The current invention concerns the discovery that a select group of 2-phenyl-3-aroylbenzothiophenes (benzothiophenes), those of-formula I, are useful for inhibiting endometrial cancer:
The therapeutic and prophylactic treatments provided by this invention are practiced by administering to a human-in need thereof a dose of a compound of formula I or a pharmaceutically acceptable salt-or-solvate thereof, that is effQCtive to inhibit endometrial cancer.
The term "inhibit" includes its generally accepted meaning which includes prohibiting, preventing, resxraining, and slowing, stopping or--reversing progression, severity or a resultant symptom. As such, the present method includes both medical therapeutic and/or prophylactic administration;-as appropriate.-' Raloxifene is a-preferred compound of this invention and it is the hydrochloride salt of a compound of formula 1 wherein R1 and R3 are hydrogen and Rz is 1-piperidinyl.
Generally, at least one compound of formula I
is formulated.with common excipients, diluents or carriers, and compressed-into--tablets;-or-formulated -as-elixirs or solutions for convenientoral administration, or administered by the intramuscular or intravenous-routes.
The compounds can be administered transdermally, and may be formulated as sustained release dosage forms and the like.

WO 96105833 ~ 19 8119 ~~ ~~.h5".~

The compounds used in the methods of the current invention can be made according to'established procedures, such as those detailed in U:S. Patent Nos. 4,133,814, , 4,418,068; and 4,380;635 all of which are incorporated by reference herein.' In general, the process starts with a , benzo[b7thiophene having a 6-hydroxyl group and a 2-(4-hydroxyphenyl) group. The starting compound is protected, acylated, and deprotected to form the formula I compounds.
Examples of the preparation of such compounds are provided in the U.S. patents discussed above. The term "optionally substituted phenyl" includes phenyl and phenyl substituted once or twice with C1-C6 alkyl, C1-C4 alkoxy, hydroxy, vitro, chloro, fluoro, or tri(chloro or-fluoro)methyl. -The compounds used in the methods of this invention-form pharmaceutically acceptable acid and base addition salts with a wide variety of organic and inorganic acids and bases and include the physiologically acceptable salts which are often used in pharmaceutical chemistry.
Such salts are also part of this invention.-- Typical inorganic acids used to form such salts include hydrochloric, hydrobromic, hydroiodic, nitric, sulfuric phosphoric, hypophosphoric and the like. Salts derived from organic acids, such as aliphatic mono and dicarboxylic acids, phenyl substituted alkanoic acids, hydroxyalkanoic and hydroxyalkandioic acids, aromatic acids, aliphatic and aromatic sulfonic acids, may also be used. Such pharmaceutically acceptable salts thus include acetate, phenylacetate, trifluoroacetate, acrylate, ascorbate, benzoate, chlorobenzoate, dinitrobenzoate, hydroxybenzoate, methoxybenzoate, methylbenzoate, o-acetoxybenzoate, naphthalene-2-benzoate, bromide,- isobutyrate, phenylbutyrate, i~-hydroxybutyrate, butyne-1,4-dioate, hexyne-1,4-dioate, caprate, caprylate, chloride, cinnamate, ' citrate, formate, fumarate, glycollate; heptanoate, hippurate, lactate, malate, maleate, hydroxymaleate, malonate, mandelate, mesylate, nicotinate, isonicotinate;

WO96105833 '~' ~ 1 ~ g l 1 9 pCTlU895I10651 nitrate, oxalate, phthalate, teraphthalate, phosphate, monohydrogenphosphate, dihydrogenphosphate, metaphosphate, pyrophosphate, propiolate, propionate, phenylpropionate, salicylate, sebacate, succinate, suberate, sulfate, bisulfate, pyrosulfate, sulfite, bisulfite, sulfonate, benzene-sulfonate, p-bromophenylsulfonate, chlorobenzenesulfonate, ethanesulfonate, 2-hydroxyethanesulfonate, methanesulfonate, naphthalene-1-sulfonate, naphthalene-2-sulfonate, p-toluenesulfonate, xylenesulfonate~ tartarate, and the like. A preferred salt is the hydrochloride salt.
The pharmaceutically acceptable acid addition salts-are typically formed by reacting a compound of formula I-with an equimolar or excess amount of acid. The reactants are generally combined in a mutual solvent such as diethyl ether-Dr benzene. -The salt normally precipitates out of.solution within about one hour to 10 days and can be isolated by filtration or the solvent can be stripped off-by conventional-means.
Bases commonly used for formation of salts include ammonium hydroxide and alkali and alkaline earth metal hydroxides, carbonates, as well as aliphatic and primary, secondary and tertiary amines,'aliphatic diamines.
Bases especially useful in the preparation of addition salts include ammbnium hydroxide, potassium carbonate, methylamine,-diethylamine, ethylene diamine and cyclohexylamine.
The pharmaceutically acceptable salts generally have enhanced solubility characteristics compared to the compound from which they are derived, and thus are often more-amenable to formulation as liquids or emulsions.
Pharmaceutical formulations-can be prepared by procedures known in the art. For example, the compaunds can be formulated with common excipients, diluents, or carriers, and farmed into tablets, capsules, suspensions, powders, and the like. Examples of excipients, diluents, R'O 96/05833 PC'TIUS95110651 _g_ and carriers that are: suitable for-such formulations include the following: fillers and extenders such as starch, sugars, mannitol, and silicic derivatives; binding , agents such as carboxymethyl cellulose and other cellu3ose derivatives, alginates~gelatin, and polyvinyl pyrrolidone; , moisturizing agents such as glycerol; disintegrating agents such as calcium carbonate and sodium bicarbonate; agents for retarding dissolutionsuch as paraffin; resorption -accelerators such as quaternary ammonium compounds; surface active agents such as cetyl alc9hol, glycerol monostearate;
adsorptive carriers such as kaolin and bentonite; and lubricants such as talc, calcium and magnesium stearate, and solid polyethyl glycols.
The compounds can also be formulated as elixirs or solutions for convenient oral administration or as solutions appropriate for parenteral administration, for instance by intramuscular, subcutaneous or intravenous routes. Additionally, the compounds are well suited to formulation as sustained release dosage forms arid the like.
The formulations can be so constituted that they release the active ingredient only or preferably in a particular part of the intestinal tract, possibly over a period of time. The coatings, envelopes, and protective matrices may be made, for example, from polymeric substances or waxes.
The particular dosage of a compound of-formula I
required-toinhibit endometrial cancer according to this invention will depend upon the-severity of the condition, the route of administration, and related factors that will -be decided by the attending physician. Generally, accepted and effective daily doses will be from about 0.1 to about 100D-mg/day, and more typically..from about 50 to about 200 mg/day. Such dosages will be administered to a subject in need thereof from once to about three times each day, or ' more often as needed, and for a-time to effectively treat or prevent endometrial cancer.

F n _g _ 2t is usually preferred to administer a compound of formula 1 in the form of an acid addition salt, as is customary in the administration of pharmaceuticals bearing a basic group, such as the piperidino ring. For such purposes the following oral-dosage forins are available:
Formulations In the formulations which foI-low, "Active ingredient" means-a compound of formula-I.
Formulation 1: Gelatin Capsules Hard gelatin capsules are prepared using the following:
Inslredient Quantity (mc/capsule) Active ingredient - ~ 0.1 -~10D0~~
Starch, NF D - 650 Starch flowable powder 0 - 650 Silicone fluid 350 centistokes 0 - 15 The ingredients are blended,-passed through a No. 45 mesh U.S. sieve, and filled into hard gelatin capsules.
_Examples of specific capsule formulations of raloxifene that have been made include those shown below:
Formulation 2: Raloxifene capsule Ingredient ~- Quantity-(mc7Tcapsule) Raloxifene 1 Starch, NF 112 Starch flowable powder - 225_3 ..
Silicone fluid 35D centistokes 1.7 w 219$119 j,,;;~:~ PCf/US95I10651 Formulation 3: Raloxifene capsule Ingredient -Quantity (m /capsule) Raloxifene Starch, NF - 108 Starch flowable powder 225.3 Silicone fluid 350 centistokes 1.7 Formulation 4: Raloxifene capsule Ingredient - -uantitv(ma/capsule) Raloxifene 10 -.
Starch, NF - - 103 Starch flowable powder 225.3 Silicone fluid 350 centistokes 1 7 Formulation 5: Raloxifene capsule -Ingredient - - uantity (mo/capsule) -Raloxifene Sp-Starch, NF 150 -Starch flowable powder 397 Silicone fluid 350 centistokes 3.0 The specific formulations above may be changed in compliance with the reasonable variations provided.
A tablet formulation is prepared using--the ingredients below:
w WO96105833 '~. PCT/US9511065I

Formulation 6: Tablets Ingredient - Quantity (m /tablet) Active ingredient 0.1 - 1000 Cellulose, microcrystalline 0 - 650 Silicon dioxide;- fumed 0 - 650 Stearate acid 0 - 15 The components are blended and compressed to form tablets.
Alternatively, tablets each containing 0.1 -1000 mg of Active ingredient are made up as follows:
Formulation 7: Tablets Ingredient - - Quantity-(mc/tablet) Active ingredient 0.1 - 1000 Starch &5 Cellulose, miorocrystalline - 35~

Polyvinylpyrrolidone (as 10$ solution in water) Sodium carbaxymethpl cellulose~ 4_5 Magnesium stearate 0.5 Talc 1 The Active ingredient, starch, and cellulose are passed through a No. 45 mesh U.S. sieve and mixed thoroughly. The solution of polyvinylpyrrolidone is mixed with the resultant powders which are then passed through a No. 14 mesh U.S. sieve. The granules so produced are dried at 50°-60° C and passed through a No. 18 mesh U.S. sieve.
The sodium carboxymethyl starch, magnesium stearate, and talc, previously passed through a No. 60 U.S. sieve, are then added to the granules which, after-mixing, are compressed on a Lablet machine to yield tablets.

R'O 96105833 ~ ~ ~~ , , PCTIUS95110651 Suspensions each containing 0.1 - 1000-mg of Active ingredient.per 5 mL dose are--made as follows:
Formulation 8: Suspensions Ingredient uantitv (ma/5 ml) Active ingredient 0.1 - 1000 mg -Sodium carboxymethyl cellulose 50 mg Syrup 1.25 mg Benaoic acid solution - 0.10 mL
Flavor -q.v.
Color q.v.
Purified water to The Activs ingredient is passed-through a No. 45 mesh U:S.
sieve and mixed with the sodium-c3rboxymethyl cellulose and syrup to form a smooth paste. The benzoic acid solution, flavor, and color are diluted with some of the water and added, with stirring. Sufficient water is then added to produce the required volume.
Continuous cultures of transformed endometrial cells reflective o~ en.dometrial.carcinoma are maintained in culture. Estrogen receptor containing cells (such as the Ishikawa line) are assessed for proliferation in response to estrogen and compounds of Formula I. Responsivenessis assessed by monitoring transcription of known estrogen/anti-estrogen-regulated genes such as progesterone receptor, PS2 and others.
Induced Animal Models - Endometrial carcinoma'is induced by injections of estrogenic substances such as 17-(3-estradiol or DES during neonatal development. The WO96l05833 - ~ j~~i~''Jt ~ p~/pS95110651 presence of carcinoma in the adult animal is confirmed by biopsy of affected animals and/or sacrifice and biopsy of syngeneic animals identically treated. After evaluation of the lesions, affected animals are treated with estrogen, compounds of formula I, or progestins for 1-8 weeks. After treatment, lesions of surviving animals are examined for progression, regression or stasis.
Five to fifty women are selected for the clinical study. The women suffer from endometrial cancer.
The study has a placebo control group, i.e., the women are divided into two groups, one of which receives a compound of formula 1 as the active agent and the other receives a -standard treatmeht ~or endoinetrial cancer_ Women in the test group receive-between 50-200 mg of the drug per day by the oral route. They continue-this therapy for 3-12 months. Accurate records are kept as to the symptoms and status of the cancer in both groups and at the end of the study these results are--compared. The results are compared both between members of each group and also the results for each patient are compared to the status reported for ea-ch patient before the-study began.. _.

A total of 251 healthy, postmenopausal women are recruited. Each subject has had her last menstrual period more than 6 months but less than 6 years prior to beginning the treatment phase of the study. Postmenopausal status of each subject is -confirmed before beginning treatment by - serum estradiol <120 pmol/L and by F5H >30 IU/L. Subjects will not have been-treated with-estrogen-over at least the ° last 3 months before the study and have never been treated with fluoride, calcitonin, or bisphosphonate. Subjects are in good health and range in age from 46.-to 60 years.

R'O 96105833 ~ . PCT/US95I10651 The-study is a multi-center, randomized, controlled, double-blind study. Qualified subjects who consent are randomized to one c~f four treatment-groups:-- , placebo, a compound of formula I 200 mg once daily, a '-compound offormula I 600-mg once daily, or estrogen-0.625 , mg once daily. All subjects also receive daily oral calcium carbonate supplements 0520 mglday elemental calcium). All medications and supplements are taken daily in the morning during the 8-week treatment period_ Once treatment is completed (Visit 5), each suject receives Provera~ 5 mg/day for 12 days.
Using a Pipelle-catheter, a uterine biopsy is performed at baseline and after.8 weeks of treatment. The -biopsies are performed in a routine manner and the tissue specimens are placed in 10~ buffered formalin. Specimens are retrieved by pouring them into tissue paper ~iltersand then are grossly examined and classified as to appearance (color, texture, and consistency) and volume. Standard histologic processing into paraffin blocks is used and the tissues are serially sectioned onto a minimum of- two slides which results in serial strips of 6 to 20 cross sections.
Since subjects with clinically significant endometrial -abnormalities are to be excluded from the study or are discontinued from the study if they develop abnormalities, the biopsies are evaluated immediately for a descriptive diagnosis, This is performed by one of two pathologists and immediately reported tothe-clinical physicians. The primary purpose ofthe biopsies is-to deteziuine the degree of morphologic estrogenic effect of study treatment. Two .
patholor~ists are trained to-read the biopsies by reviewing a series of-Pipelle biopsies obtained outside-the study ' that represent the full spectrum of endometrial morphology.
Using standard morphologic criteria associated with -- ' estrogen-induced proliferation, a scoring system is devised to quantitate this estrogenic effect and-include-the more subtle changes that may be encountered. Ten of these WO96105833 2198 i 19 PCT/US95/10651 ~t y outside cases are then scorEd with this system by each pathologist, and, the cases are reviewed together to assure uniform understanding and use of the criteria After the .

first twenty cases from the baseline biopsies in the study are blindly scored by each-pathologist, the scores are reviewed to verify proper use of the system. The .

pathologists evaluate the biopsy samples, for the following components: 1) specimen adequacy, 2) glandular morphology, 3) stromal morphology, and 9) other changes. Additional findings are entered as textual comments. Point scores are generated from the glandular and stroznal morphologic features and are totaled and graded on a 4-point estrogenicity scale where a grade of--0- indicates typical postmenopausal endometrium and a grade of 2 indicates a marked estrogenic effect. Total scores for both pathologists are averaged and then assigned a final grade of 0 to 3_ Scoring occurs well after the initial immediate diagnosis and usually 10 to 20 cases are scored sequentially.

It is expected that the rate of scant tissue is relatively high on the initial biopsy since the typical postmenopausal endometrium is inactive and consists of a very shallow (5 mm or less) tissue lining and the Pipelle biopsy is a limited, blind biopsy method. Because . endometrial glands are required to score features of glandular and stromal morphology, the final biopsy must have-contained glands before any conclusions can be drawn in individual subjects. Specimen adequacy is defined as follows:

If no tissue or debatable tissue of endometrial ' origin is present, the specimen is-deemed inadequate and not included in the evaluation.

R'O 96!05833 21981 ~ 9 If multiple fragmenLS of endnmetrial Surface epithelium are obtained, the -specimen-can not be scored= However the biopsy is deemed adequate and is assigned a grade of D nn the 4-point scale indicating no estrogen effect. _ If disrupted endometrium with glands are obtained, the biopsy is adequate and is scored for the glandular and Stromal='features.-If intact endometrial tissue is obtained, the-biopsy is adequate and is scored for the glandular and stromal changes.- In addition, the volume of the tissues is taken into account as an indication of estrbgen effect.
Glandular morphology is the primary scoring factor for adequate biopsy specimens. Stromal morphola~y is the secondary scoring factor-for adequate biopsy 2D specimens. Tables l and 2 display the features to be used to score=each specimen that_have glands aridlor-stroma -present. Four features are used to classify the glands:
shape, -cellular nuclear to cytoplasmic cross sectional areas, nuclear pseudostrati~ication, and mitotic activity.

WO 96105833 ~ ~ ~ ~ r~ ~ ~ PCTlUS95/10651 Table 1. Glandular Features: Scoring of Estrogenicity Estrogenic Effect/POint Value No Limited High Glandular Estrogenicity Estrogeneicity Estrogenicity Feature (0 Points) (1 Point) (2 Points) Shape Small, Open, straight Open, cystic, tubular tortuous straight Nucleus-to- Very High Moderate (75$ Low (<50~) cytoplasm ,(> 75~) to 50$) ratio Nuclear None Limited Diffuse pseudostrati -fication Mitoses None Bare Scattered to many Note: At least 20 gland profiles are used to grade for mitotic activity (four serial sections of scant specimens).
In more scanty specimens a minimum of 20 gland profiles in-serial sections are viewed before concluding no mitoses are evident. In Table 2 the stromal and "other"
features are listed. Four features are also used to classify the stroma: density, mitoses, metaplastic changes in epithelia, and tissue volume.
C

W096105833 ~~ . , PCfIUS95110651 Table-2. Stromal Features: Scoring of Estrogenicity Estrogenic Effect/POint Value . t No Limited High Stromal Estrogenicity Estrogeneicity Estrogenicity Feature (0 Points) (1 Point) (2 Points) Compact, Loosely Edematous Density fibrous cellular Mitoses None - Rare- Few/Many Metaplasiag None Rare Scattered, diffuse-Tissue Volumeb Disrupted or Moderate, much Abundant, few intact being intact intact._ aMetaplasia includes tubular, eo-sinophilic, and squamous type.
bLJSed only if glands show some estrogenic effect (1 or 2 points).
Morphologic features that indicate a lack of estrogenicity generate a score of 0 points and features indicating a limited or significant estrogenic effect generate a score of 1 to 2 points, respectively. Using this approach, a biopsy can receive between 0 to 16 points.
In-addition to glandular and stromal morphology, and other-important morphologic features including progestational effect, inflammatory processes-, breakdown bleeding, polypoid-growth, or other pathologic findings are described but are not included in the scoring of proliferative effects since the other changes are,primarily nonproliferative.
The sum of the scores-obtained from grading the ' glandular and stromal morphology features resultin a 4-point estrogenicity grading scale which is assigned to each sample as follows:

W096105833 ~ 1 ~ PCT/US95110651 En v' Grade 0 = 0 to 3 points Typical postmenopausal endometrium with little or-no estrogenic effect Grade 1 = 4 to 6 points Definite but limited estrogenic effect _._.. . _.
Grade 2 = 7 to 10 points Moderte estrogenic ef f ect Grade 3 _ >10 points Marked estrogenic -effect As noted earlier, if biopsy specimens contain multiple fragments of endometrial surface epithelium, those specimens are assigned a grade of 0.

For each biopsy, there are eight scores: four assessments of the glandular morphology, two assessments of the stromal morphology (density and mitoses scores are combined and metaplasia and tissue volume are combined for statistical analyses), the sum of these six scores and the grade as defined above.

Intraclass correlation coefficients are calculated to assess agreement between the two readers on the sum of the scores obtained at baseline and at 8 weeks (Fleiss, JL (1981) ,S a ~ s a~ M hod or R- c -nd P1'ODOr ion . New York: John Wiley and Sons, p. 218.]

The baseline, week 8 and change-from-baseline to week 8 scores for each of-the eight scores are analyzed for treatment differences using Cochran-Mantel-Haenazel statistical techniques adjusting for investigator [Landis, RJ Heyman, ER and Rock, GG (1978). "Average Partial ' Association in Three-Way Contingency Tables: A Review and Discussion of Alternative Tests". International Statistical Review 46:237-254.].

The occurrence-of endometrial glands in the biopsy tissue is evaluated at baseline and 8 weeks for treatment differences using the chi-square test.

R'0, 96105833 ~ PCTIUS95I10651 Pairwise treatment comparisons between each active treatment and placebo are performed if. the overall treatmentdifference is statistically significant. , Statistical significance is judged at a two-sided 0.05 -level of significance. A1 statistical analyses use the SAS , system [SAS Institute Inc. (1989),-SASISTAT Users Guide, Version 6, Fourth Edition, Volumes 1 and 2, Gary, NC: SAS
Institute Inc_7 A positive-result in-this assay is the reduction of the score for glandular mitoses indicating a decrease in cell replication-relative to placebo:- _ Table 3--illustrates important results of the_ study.
Table 3. Mean (t SEM) Scores for Glandular Fatures At Endpoint Placebo Raloxifene Raloxifene Estrogen (n 53) 200 mg 600 0.625 = mg mg Variable (n 54) (n 54) (n 47) = = =

Shape 0.44t 0.08 0.58t 0.07 0.51t 0_06.1_37t 0_06*

Pseudostrat-0.64t 0.10 0.57t 0.06 0.56Ø06 1_680.07*
- -ification Mitoses 0.19t 0_05 0.05t 0.02* 0.07t0_03f 0.98t D_OS*

Nucleus: 0.48t 0.09 0.58t0_D-fi 0.5810.05 1.56t D.OS*
-Cytoplasm * Statistically significantly different-from placebo, two-tailed test (p<.050) f Marginally significantly different from placebo, two-tailed test (p = .053) VVO 96105833 ~~ ~ PC1'IUS95110651 , , C- .

Utility of the compounds of formula I is illustrated by ~he positive impactthey have in at least one of the assays described above.

Claims (4)

I claim:

1. A method of inhibiting endometrial cancer comprising administering to a human in need thereof an effective amount of a compound having the formula wherein R1 and R3 are independently hydrogen, or or , wherein Ar is optionally substituted phenyl;
R2 is selected from the group consisting of pyrrolidine, hexamethyleneimino, and piperidino; or a pharmaceutically acceptable salt of solvate thereof.

2. The method of Claim 1 wherein said compound is the hydrochloride salt thereof.

3. The method of Claim 1 wherein said administration is prophylactic.

4. The method of Claim 1 wherein said compound is or its hydrochloride salt.