WO2000070083A1 - Assay for myelosuppression by farnesyl transferase inhibitors - Google Patents

Abstract

A bone marrow toxicity assay provides an in vitro screening tool that aids in the design of clinically useful farnesyl transferase inhibitors (FTI's) by providing a predictive measure of the potential of a FTI to induce bone marrow toxicity (myelosuppression) in man.

Description

TITLE OF THE INVENTION

ASSAY FOR MYELOSUPPRESSION BY FARNESYL TRANSFERASE

INfflB-πORS.

RELATED APPLICATION

The present patent application is a continuation-in-part application of copending provisional application Serial No. 60/134,700, filed May 12, 1999.

BACKGROUND OF THE INVENTION The Ras proteins (Ha-Ras, Ki4a-Ras, Ki4b-Ras and N-Ras) are part of a signalling pathway that links cell surface growth factor receptors to nuclear signals initiating cellular proliferation. Mutated ras genes (Ηa-ras, Ki4a-rα-s, -Mb-ra-s' and N-ras) are found in many human cancers, including colorectal carcinoma, exocrine pancreatic carcinoma, and myeloid leukemias. The protein products of these genes are defective in their GTPase activity and constitutively transmit a growth stimulatory signal.

At least 3 post-translational modifications are involved with Ras membrane localization, required for normal and oncogenic function, and all 3 modifications occur at the C-terminus of Ras. The Ras C-terminus contains a sequence motif termed a "CAAX" or "Cys-Aaa¹-Aaa²-Xaa" box, which, depending on the specific sequence, serves as a signal sequence for the enzymes farnesyl-protein transferase (also known as farnesyltransferase) or geranylgeranyl-protein transferase, which catalyze the alkylation of the cysteine residue of the CAAX motif with a C15 or C20 isoprenoid, respectively. (S. Clarke, Ann. Rev. Biochem. 67:355-386 (1992); W.R. Schafer and J. Rine, Ann. Rev. Genetics 30:209-231 (1992)). The Ras protein is one of several proteins that are known to undergo post-translational farnesylation. Other famesylated proteins include the Ras-related GTP-binding proteins such as Rho, fungal mating factors, the nuclear lamins, and the gamma subunit of transducin. James, et al., J. Biol. Chem. 269, 14182 (1994) have identified a peroxisome associated protein Pxf which is also famesylated. James, et al., have also suggested that there are famesylated proteins of unknown structure and function in addition to those listed above.

Inhibition of famesyl-protein transferase has been shown to block the growth of Ras-transformed cells in soft agar and to modify other aspects of their transformed phenotype. Recently, it has been shown that an inhibitor of famesyl-protein transferase blocks the growth of rαs-dependent tumors in nude mice (N.E. Kohl et al., Proc. Natl. Acad. Sci U.S.A., 97:9141-9145 (1994) and induces regression of mammary and salivary carcinomas in ras transgenic mice (N.E. Kohl et al. , Nature Medicine, 1:192-191 (1995). Farnesyl transferase inhibitors (FTI's) represent a new pharmacological approach to the treatment of cancer that is mechanism-based and does not rely on a cytotoxic mechanism of action. Ideally, therapeutically effective doses of FTI's will not be limited by cytotoxic side effects and these compounds will have a much larger therapeutic window than currently available an ti tumor drugs. Famesyl-protein transferase inhibitors may also be useful for inhibiting other proliferative diseases, both benign and malignant, wherein Ras proteins are aberrantly activated as a result of oncogenic mutation in other genes (i.e., the Ras gene itself is not activated by mutation to an oncogenic form) with said inhibition being accomplished by the administration of an effective amount of the instant composition to a mammal in need of such treatment. For example, a component of NF-1 is a benign proliferative disorder.

Use of famesyl-protein transferase inhibitors in the prevention of restenosis after percutaneous transluminal coronary angioplasty by inhibiting neointimal formation has recently been described (C. Indolfi et al. Nature medicine, 1:541-545(1995)). It has been disclosed that famesyl-protein transferase inhibitors may also be useful in the treatment and prevention of polycystic kidney disease (D.L. Schaffner et al. American Journal of Pathology, 142:1051-1060 (1993) and B. Cowley, Jr. et al. FASEB Journal, 2: A3160 (1988)). It has recently been reported that famesyl-protein transferase inhibitors are inhibitors of proliferation of vascular smooth muscle cells and are therefore useful in the prevention and therapy of arteriosclerosis and diabetic disturbance of blood vessels (JP H7-112930). In contrast to cytotoxic chemotherapeutic agents, a more rational approach for estimating clinically-effective doses can be used with FTI's, i.e. it may not be necessary to titrate doses in a patient until toxic effects are observed if that dose is considerably higher than needed to provide the desired effect on tumor growth. Alternatively, if dose-limiting toxicity is observed with the clinical FTI, the dose may not be sufficient to inhibit the enzyme to the extent needed to block tumor growth. Demonstration of clinical efficacy using inhibition of tumor growth or regression in tumor size as an endpoint will require considerable time (weeks to months) and, therefore, dose selection for these trials is very important. Plasma drug concentrations are often used for clinical dose selection, however this endpoint may be a poor surrogate for the drug concentration at the pharmacological target, especially when the site of action is intracellular, such as is the case with farnesyl transferase. PET (Positron Emission Tomography) radiotracers and imaging technology may provide a powerful method for clinical evaluation and dose selection of FTI's. Using a carbon- 11 or fluorine- 18 labeled radiotracer that enters cells and provides a farnesyl transferase (FPTase) enzyme-specific image in tumors and other tissues, the dose required to saturate FPTase can be determined by the blockade of the PET radiotracer image in humans. The rationale for this approach is as follows: anti-tumor efficacy of FTI's is a consequence of the extent of enzyme inhibition, which in turn is a function of the degree of drug-enzyme occupancy. Bone marrow toxicity (myelosuppression) by farnesyl protein transferase inhibitors (also known as farnesyl transferase inhibitors) (FTI's) is a newly recognized phenomenom, although the phenomenom of bone marrow toxicity has long been known for "classical" chemotherapeutic agents. Myelosuppression is the most common dose-limiting toxicity of anticancer agents in man. This toxicity can interfere with effective cancer chemotherapy as a result of delays in subsequent courses and/or reduction in the treatment dose. Severe myelosuppression can lead to infection due to prolonged inhibition of host-defense mechanisms.

There is considerable interest in determining the myelosuppressive potential of new agents that have favorable antitumor activity in experimental tumor systems. The data can be used to rank the myelosuppressive effect of these drugs and identify ones that appear to be less myelosuppressive than those currently used in man. A favorable candidate for development would be an agent with better experimental antitumor activity and less myelosuppressive action than the reference agent(s). Effective evaluation of new antitumor drugs for myelosuppressive effects requires development of appropriate test systems in experimental animals.

In vitro models for predicting myelosuppression induced by "classical" chemotherapeutics have been described. One such model involves assessing the effect of a compound on the expansion of marrow hematopoietic progenitors in culture (CFU-C assay) (J.E. Schurig et al., Experimental Hematology, 13: 101-105 (1985)). However, the CFU-C assay was not completely predictive of marrow toxicity caused by FTI's in nude mouse xenograft models, as this assay could not discriminate between compounds that display differences in the amount of marrow toxicity they induce.

It is therefore an object of the instant invention to develop an in vitro bone marrow toxicity assay that is predictive of in vivo marrow toxicity, and can correctly discriminate between -FTI's that are known to induce different levels of marrow toxicity in animal models.

SUMMARY OF THE INVENTION

The bone marrow toxicity assay of this invention provides an in vitro screening tool that will aid in the design of clinically useful FTI's by providing a predictive measure of the potential of a FTI to induce bone marrow toxicity (myelosuppression) in man. The assay measures the affinity of an FTI for farnesyltransferase (Ftase) binding sites in unpurified, isolated bone marrow cells from a mammal. The affinity of the FTI for Ftase binding sites is determined by measuring the IC50 for competition between the test compound and a radiolabeled FTI tracer for binding to Ftase binding sites in bone marrow cells. The determined IC50 is shown to correlate with the degree of marrow toxicity induced by different -FTI's in a mouse model. The utility of the screening assay of this invention relies on the assumption that, as has been observed for existing chemotherapeutic agents, myelosuppression induced by FTI's in the mouse model is predictive of its potential for inducing myelosuppression in man.

DETAILED DESCRIPTION OF THE INVENTION

The assay of this invention determines the affinity of an FTI test compound for FTase binding sites in living bone marrow cells by measuring the competition between the FTI and a radiolabeled FTI tracer for the FTase binding sites. The assay comprises the steps of: a) isolating and culturing bone marrow cells; b) exposing the bone marrow cells to growth media containing the radiolabeled farnesyl transferase inhibitor in the presence or absence of the farnesyl transferase inhibitor test compound; c) washing the cells; d) counting the radiation emitted by the cells; and e) comparing the radiation emitted by the cells exposed to both the radiolabeled farnesyl transferase inhibitor and the test compound to the radiation emitted by the cells exposed to only the radiolabeled farnesyl transferase inhibitor. The assay uses freshly isolated bone marrow cells derived from the bones of animals such as hamsters, rabbits, rats, mice or the like; preferably mice and more preferably nude mice. Any bones from these animals are useful as a source of bone marrow cells, but leg bones and particularly femurs are preferred, especially: hind-leg femurs of nude mice.

In an embodiment of the instant assay each compound is assayed in a nine- point, half-log dilution series (plus a vehicle control), with each titration point done in triplicate. Each replicate requires at least about 5 x 105 bone marrow cells, for a total of about 1.5 x 107 cells for each compound assayed. Approximately 5 x 106 cells can be isolated from each of the preferred nude mouse hind-leg femurs, so that a typical experiment involving the assay of 4 compounds requires the use of about 6 animals. Prior to isolation of the cells, the radiotracer is diluted in cell culture medium to a concentration of ~lnM (Xμ Ci/ml), mixed with the serially diluted FTI, and then transferred into 24-well tissue culture plates. The marrow cells are then isolated, pooled and added to the tracer/FTI mixtures in the 24-well plates. After a four-hour incubation, the cells are recovered on glass-fiber filters and washed four times with PBS using a vacuum manifold. The cell-associated radioactivity on the filters is quantified by gamma counting. Dose-inhibition curves and IC5θ's are derived using a 4-parameter curve-fitting equation. Two reference FTI's are run as standards in the assay.

The radiotracer useful in this assay may be a radiolabeled analog of any FTI. In particular, a radiolabeled analog of the compounds described in the following patents, patent applications and publications may be employed. WO 95/32987 published on 7 December 1995; U.S. Pat. No. 5,420,245; U.S. Pat. No. 5,523,430; U.S. Pat. No. 5,532,359; U.S. Pat. No. 5,510,510; U.S. Pat. No. 5,589,485; U.S. Pat. No. 5,602,098; European Pat. Publ. 0 618 221; European Pat. Publ. 0675 112

European Pat. Publ. 0 604 181

European Pat. Publ. 0 696 593

WO 94/19357; WO 95/08542;

WO 95/11917;

WO 95/12612;

WO 95/12572;

WO 95/10514 and U.S. Pat. No. 5,661,152; WO 95/10515;

WO 95/10516;

WO 95/24612;

WO 95/34535;

WO 95/25086; WO 96/05529;

WO 96/06138;

WO 96/06193;

WO 96/16443;

WO 96/21701; WO 96/21456;

WO 96/22278;

WO 96/24611;

WO 96/24612;

WO 96/05168; WO 96/05169;

WO 96/00736 and U.S. Pat. No. 5,571,792 granted on November 5, 1996;

WO 96/17861;

WO 96/33159

WO 96/34850 WO 96/30018 WO 96/30362 5 WO 96/30363 WO 96/31111 WO 96/31477 WO 96/31478 WO 96/31501

10 WO 97/00252 WO 97/03047 WO 97/03050 WO 97/04785 WO 97/02920

15 WO 97/17070 WO 97/23478 WO 97/26246 WO 97/30053 WO 97/44350

20 WO 97/43437 WO 97/49700 WO 98/00409 WO 98/00411 WO 98/02436

25 WO 98/04545 WO 98/09641 WO 98/07692 WO 98/11091 WO 98/11092 WO 98/11097 WO 98/11098 WO 98/11099 WO 98/11100 WO 98/11106 WO 98/15556 WO 98/17629 WO 98/20001 WO 98/27109 WO 98/29390 WO 98/30558 WO 98/32741 WO 98/34921 WO 98/38162 GB 2323841 GB 2323842 GB 2323783 WO 98/40383 WO 98/42676 WO 98/46625 WO 98/49157 WO 98/50029 WO 98/50030 WO 98/50031 EP 810223; KR 97/006208; and U.S. Pat. No. 5,532,359 granted on July 2, 1996 .

Radiolabeled analogs of the following compounds, which are inhibitors of famesyl-protein transferase, are particularly useful as radiotracers in the assay of this invention:

(+)-6-[amino(4-chlorophenyl)(l-methyl-lH-imidazol-5-yl)methyl]-4-(3- chlorophenyl)-l-methyl-2(lH)-quinolinone (Compound J)

(-)-6-[amino(4-chlorophenyl)(l-methyl-lH-imidazol-5-yl)methyl]-4-(3-chlorophenyl)- l-methyl-2(lH)-quinolinone (Compound J-A; designated "comp. 74" in WO 97/21701)

(+)-6-[amino(4-chlorophenyl)(l-methyl-lH-imidazol-5-yl)methyl]-4-(3- chlorophenyl)-l-methyl-2(lH)-quinolinone (Compound J-B; designated "comp. 75" in WO 97/21701)

or a pharmaceutically acceptable salt thereof. The syntheses of these compounds are specifically described in PCT Publication WO 97/21701, in particular on pages 19-28. The following compound which is an inhibitor of famesyl-protein transferase is also useful when radiolabeled in the assay described herein:

or a pharmaceutically acceptable salt thereof. The synthesis of this compound is specifically described in PCT Publication WO 97/23478, in particular on pages 18-56. In WO 97/23478, the above compound is designated compound "39.0" and is specifically described in Example 10.

Compounds which are inhibitors of famesyl-protein transferase and whose radiolabeled analogs are therefore useful in the present invention, and methods of synthesis thereof, can be found in the following patents, pending applications and publications, which are incorporated herein by reference:

U.S. Pat. No. 5,238,922 granted on August 24, 1993;

U.S. Pat. No. 5,340,828 granted on August 23, 1994;

U.S. Pat. No. 5,480,893 granted on January 2, 1996;

U.S. Pat. No. 5,352,705 granted on October 4, 1994;

U.S. Pat. No. 5,504,115 granted on April 2, 1996;

U.S. Pat. No. 5,536,750 granted on July 16, 1996; U.S. Pat. No. 5,504,212 granted on April 2, 1996;

U.S. Pat. No. 5,439,918 granted on August 8, 1995;

U.S. Pat. No. 5,686,472 granted on November 11, 1997;

U.S. Pat. No. 5,736,539 granted on April 4, 1998;

U.S. Pat. No. 5,576,293 granted on November 19, 1996;

U.S. Pat. No. 5,468,733 granted on November 21, 1995;

WO 96/06609 (March 3, 1996) and USSN 08/298,478 filed on August 24, 1994;

U.S. Pat. No. 5,585,359 granted on December 17, 1996;

U.S. Pat. No. 5,523,456 granted on June 4, 1996;

U.S. Pat. No. 5,661,161 granted on August 26, 1997;

U.S. Pat. No. 5,571,835 granted on November 5, 1996;

U.S. Pat. No. 5,491,164 granted on February 13, 1996;

U.S. Pat. No. 5,652,257 granted on July 29, 1997;

U.S. Pat. No. 5,631,280 granted on May 20, 1997; U.S. Pat. No. 5,578,629 granted on November 26, 1996;

U.S. Pat. No. 5,627,202 granted on May 6, 1997;

U.S. Pat. No. 5,856,326 granted on January 5, 1999; WO 96/30343 (October 3, 1996); USSN 08/412,829 filed on March 29, 1995; and USSN 08/470,690 filed on June 6, 1995; and USSN 08/600,728 filed on February 28, 1996;

U.S. Pat. No. 5,624,936 granted on April 29, 1997;

U.S. Pat. No. 5,534,537 granted on July 9, 1996;

U.S. Pat. No. 5.710.171 granted on April 29. 1997;

WO 96/39137 (December 12, 1996); USSN 08/468,160 filed on June 6, 1995; USSN 08/652,055 filed on May 23, 1996; USSN 08/960,248 filed October 29, 1997;

U.S. Pat. No. 5,703,241 granted on December 30. 1997;

WO 97/18813; USSN 08/749,254 filed on November 15, 1996;

WO 97/27854 (August 7, 1997); USSN 60/010,799 filed on January 30, 1996; USSN 08/786,520 filed on January 21, 1997; USSN 09/015,823 filed on January 29, 1998;

WO 97/27752 (August 7, 1997): USSN 60/010,860 filed on January 30, 1996; USSN 08/784,556 filed on January 21, 1997; USSN 09/030,223 filed on February 25, 1998;

WO 97/27853 (August 7. 1997); USSN 60/011,081 filed on January 30, 1996; USSN 08/786,519 filed on January 21, 1997; WO 97/27852 (August 7. 1997); USSN 60/010,798 filed on January 30, 1996; USSN 08/786,516 filed on January 21, 1997;

WO 97/36888 (October 9, 1997); USSN 60/014,587 filed on April 3, 1996; USSN 08/823,919 filed on March 25, 1997;

WO 97/36889 (October 9. 1997); USSN 60/014,589 filed on April 3, 1996; USSN 08/823,923 filed on March 25, 1997;

WO 97/36876 (October 9, 1997); USSN 60/014,592 filed on April 3, 1996; USSN 08/834,671 filed on April 1, 1997;

WO 97/36593 (October 9. 1997); USSN 60/014,593 filed on April 3, 1996; USSN 08/827,485, filed on March 27, 1997;

WO 97/36879 (October 9. 1997); USSN 60/014,594 filed on April 3, 1996; USSN 08/823,920 filed on March 25, 1997;

WO 97/36583 (October 9. 1997); USSN 60/014,668 filed on April 3, 1996; USSN 08/824,588 filed on March 26, 1997;

WO 97/36592 (October 9. 1997); USSN 60/014,775 filed on April 3, 1996; USSN 08/826,292 filed on March 27, 1997;

WO 97/36584 (October 9. 1997); USSN 60/014,776 filed on April 3, 1996; USSN 08/824,427 filed on March 26, 1997;

USSN 60/014,777 filed on April 3, 1996; USSN 08/826,317 filed on March 27, 1997; O 97/38665 (October 23, 1997); USSN 60/014,791 filed on April 3, 1996; USSN /831,308 filed on April 1, 1997;

O 97/36591 (October 9, 1997); USSN 60/014,792 filed on April 3, 1996; USSN /827,482, filed on March 27, 1997;

O 97/36605 (October 9. 1997); USSN 60/014,793 filed on April 3, 1996; USSN /823,934 filed on March 25, 1997;

O 97/37877 (October 9, 1997); USSN 60/014,794 filed on April 3, 1996; USSN /834,675 filed on April 1, 1997;

O 97/37900 (October 9, 1997); USSN 60/014,798 filed on April 3, 1996; USSN /823,929 filed on March 25, 1997;

O 97/36891 (October 9, 1997); USSN 60/014,774 filed on April 3, 1996; USSN /826,291 filed on March 27, 1997;

O 97/36886 (October 9, 1997); USSN 60/022,332 filed on July 24, 1996; USSN /823,919, filed on March 27, 1997;

O 97/36881 (October 9, 1997); USSN 60/022,340 filed on July 24, 1996; USSN /827,486, filed on March 27, 1997;

O 97/36585 (October 9. 1997); USSN 60/022,341 filed on July 24, 1996; USSN /826,251 filed on March 27, 1997; O 97/36898 (October 9, 1997); USSN 60/022,342 filed on July 24, 1996; USSN /825,293 filed on March 27, 1997;

O 97/36897 (October 9, 1997); USSN 60/022,558 filed on July 24, 1996; USSN /827,476, filed on March 27, 1997;

O 97/36874 (October 9, 1997);

O 97/36585 (October 9, 1997); USSN 60/022,586 filed on July 24, 1996; USSN /827,484, filed on March 27, 1997;

O 97/36890 (October 9, 1997); USSN 60/022,587 filed on July 24, 1996; USSN /831,105 filed on April 1, 1997;

O 97/36901 (October 9, 1997); USSN 60/022,647 filed on July 24, 1996; USSN /827,483, filed on March 27, 1997;

O 99/27929 (June 10, 1999); USSN 60/032,126 filed on December 5, 1996; USSN /985,732, filed on December 4, 1997;

O 99/27933 (June 10, 1999); USSN 60/032,428 filed on December 5, 1996; USSN /985,124, filed on December 4, 1997;

O 99/28314 (June 10, 1999); USSN 60/032,578 filed on December 5, 1996; USSN /985,337, filed on December 4, 1997;

O 99/28313 (June 10. 1999); USSN 60/032,579 filed on December 5, 1996; USSN /985,320, filed on December 5, 1997; WO 98/29119 (July 9, 1998); USSN 60/033,990, filed on December 30, 1996; USSN 08/995,744, filed on December 22, 1997;

WO 98/29980 (July 9, 1998); USSN 60/033,991, filed on December 30, 1996; USSN 08/985,124, filed on December 5, 1997;

WO 99/10329 (March 3. 1999); USSN 60/057,097, filed on August 27, 1997; USSN 09/140,919, filed on August 26, 1998;

WO 99/09985 (March 3, 1999): USSN 60/057,080, filed on August 27, 1997; USSN 09/140,584, filed on August 26, 1998;

WO 99/17777 (April 15, 1999); USSN 60/062,660, filed on October 8, 1997; USSN 09/167,180, filed on October 6, 1998;

WO 99/20609 (April 29, 1999); USSN 60/064,342, filed on October 17, 1997; USSN 09/170,951, filed on October 13, 1998;

WO 00/01702 (January 13, 2000); USSN 09/342,701, filed on June 29, 1999; USSN 60/091,629, filed on July 2, 1998;

WO 00/01701 (January 13. 2000); USSN 09/347,673, filed on June 29, 1999; USSN 60/091,596, filed on July 2, 1998;

WO 00/01382 (January 13, 2000); USSN 09/344,577, filed on June 29, 1999; USSN 60/091,513, filed on July 2, 1998;

USSN 09/516,756, filed on March 1,2000; USSN 60/122,968, filed on March 3, 1999, 60/127,132, filed March 31, 1999. USSN 09/516,945, filed on March 1,2000; USSN 60/122,970, filed on March 3, 1999, 60/127,259, filed March 31, 1999;

USSN 09/516,750, filed on March 1,2000; USSN 60/122,768, filed on March 3, 1999, 60/127,253, filed March 31, 1999;

USSN 09/516,757, filed on March 1,2000; USSN 60/122,771, filed on March 3, 1999, 60/127,257, filed March 31, 1999; and

USSN 09/516,944, filed on March 1,2000; USSN 60/123,620, filed on March 3, 1999, 60/127,252, filed March 31, 1999.

Suitable radionuclides that may be incorporated in the compound useful in the instant method of treatment include ^H (also written as T), ^C, ^F, 125j 82jj_r) 123 131 124_I; 75_Br, 1 o, 13JN, ^{21 1}At or ⁷⁷Br. The radionuclide that is incorporated in the instant radiolabeled compounds will depend on the specific analytical or pharmaceutical application of that radiolabeled compound. Thus, for in vitro FPTase labeling and competition assays, inhibitor compounds that incorporate -1H, 125τ _or 82β_{r w}ju generally be most useful.

The labeled famesyl-protein transferase inhibitor should bind with a high affinity to FPTase. Examples of such compounds are described in PCT Publ. No. WO

99/00654 (January 7, 1999). Preferably, the labeled inhibitor has an IC50 ≤ lOnM, and most preferably the labeled inhibitor has an IC50 ≤ 5nM. Preferably, the radiolabeled farnesyl protein transferase inhibiting compound is selected from a class of piperazinones of structural formula:

wherein:

R² and R^ are independently selected from: H; unsubstituted or substituted Ci -8 alkyl, unsubstituted or substituted C2-8 alkenyl, unsubstituted or substituted C2-8 alkynyl, unsubstituted or substituted aryl, unsubstituted or substituted heterocycle,

wherein the substituted group is substituted with one or more of:

1) aryl or heterocycle, unsubstituted or substituted with: a) Ci -4 alkyl,

d) halogen, e) a radionuclide,

2) C3_6 cycloalkyl,

3) OR⁶, 4) SR⁶, S(O)R⁶, SO2R⁶, 5) — NR⁶R⁷

10) \ .NR⁶R⁷ 0

11) — S0₂-NR⁶R⁷ R⁶

1

12) — N-S0₂— R⁷

15) a radionuclide; or R4 is selected from H and CH3; and any two of R², R^ and R^ are optionally attached to the same carbon atom;

R⁶, R7 and R^7a are independently selected from: H; C1 -4 alkyl, llC-methyl, C3-6 cycloalkyl, heterocycle, aryl, aroyl, heteroaroyl, arylsulfonyl, heteroarylsulfonyl, unsubstituted or substituted with: a) C1 -4 alkoxy, b) aryl or heterocycle, c) halogen, ) HO,

f) — S0₂R¹¹

g) N(R¹⁰)2, h) HC-methyl, or i) a radionuclide; or

R⁶ and R⁷ may be joined in a ring;

R7 and R7^a may be joined in a ring;

R8 is independently selected from: a) hydrogen, b) aryl, heterocycle, C3-C10 cycloalkyl, C2-C6 alkenyl, C2-C6 alkynyl, perfluoroalkyl, F, Cl, Br, R¹⁰O-, Rl lS(O)_m-, R¹⁰C(O)NR¹⁰-, CN, NO2, R¹⁰2N-C(NR¹⁰)-, R¹⁰C(O)-, R¹⁰OC(O)-, N3, -N(R¹⁰)2, or

RHOC^NR¹⁰-, c) C1-C6 alkyl unsubstituted or substituted by aryl, heterocycle, C3-C10 cycloalkyl, C2-C6 alkenyl, C2-C6 alkynyl, perfluoroalkyl, F, Cl, Br, R¹⁰O-, RHS(O)_m-, R¹⁰C(O)NH-, CN, H2N-C(NH)-, R¹⁰C(O)-,

R!0OC(O)-, N3, -N(R!0)2, or R¹⁰OC(O)NH-, and d) a radionuclide;

RIO is independently selected from hydrogen, C1 -C6 alkyl, ^H-methyl, HC-methyl, benzyl and aryl;

RU is independently selected from C1-C6 alkyl and aryl;

Z is phenyl, unsubstituted or substituted with one or more of:

1) Cl-4 ^; alkyl, unsubstituted or substituted with: a) Ci-4 alkoxy, b) NR⁶R⁷, c) C3-6 cycloalkyl, d) aryl or heterocycle, e) HO, ) -S(O)_mR⁶,

g) -C(O)NR⁶R⁷, or h) a radionuclide;

2) aryl or heterocycle,

3) halog< -n,

4) OR⁶ _'

5) NR⁶R7, 6) CN,

7) NO ,

8) CF₃;

9) -S(O)_mR⁶, 10) -C(O)NR⁶R⁷,

11) C3-C6 cycloalkyl, or

12) a radionuclide;

m is 0, 1 or 2; and r is 0 to 5; and

wherein at least one radionuclide or ^H-methyl is present in the molecule;

or a pharmaceutically acceptable salt thereof.

A sub-class of the piperazinones includes the following specific examples of radiolabeled famesyl-protein transferase inhibitors:

or a pharmaceutically acceptable salt thereof.

Preferably the radiolabeled famesyl-protein transferase inhibitor is the compound of the formula:

EXAMPLES Examples provided are intended to assist in a further understanding of the invention. Particular materials employed, species and conditions are intended to be further illustrative of the invention and not limitative of the reasonable scope thereof.

EXAMPLE 1

Preparation of (S)-l-(3-iodo-5-fluoromethylphenyl)-4-[l-(4-cyanobenzyl)-5- imidazolylmethyl]-5-r2-(methanesulfonyl)ethyl1-2-piperazinone dihydrochloride

Step A: 3-Amino-5-nitrobenzyl alcohol

A three neck flask fitted with a reflux condenser and an addition funnel was charged with 3,5-dinitrobenzyl alcohol (10 g, 50 mmol), 10% Pd/C (630 mg) and triethylamine (35 mL, 251mmol). This was placed in a 70°C oil bath and through the addition funnel was added 90% formic acid (10 mL, 234 mmol) dropwise. The reaction was heated for 30 minutes and cooled to room temperature. The clear yellow supernatant was filtered through celite and the lower layer containing the catalyst was treated with acetone, filtered through celite and rinsed with acetone. The filtered organic layers were combined and concentrated to give an orange oil. The oil was diluted in ethyl acetate which was washed with H2O, dried (MgSO4), filtered and concentrated to give crude 3-amino-5-nitrobenzyl alcohol as an orange solid.

Step B: 3-Iodo-5-nitrobenzyl alcohol A mixture of crude 3-amino-5-nitrobenzyl alcohol (8.8 g, 52.3 mmol) described in Step A, in H2O (140 mL) was cooled to 0°C and treated with H2SO4 (31.2 mL) followed by an aqueous solution of NaNO2 (3.9 g, 56.5 mmol in 55 mL H2O) added over approximately 10 minutes. After stirring this mixture for 35 minutes at 0°C, an aqueous solution of KI (7.6 g, 65.5 mmol in 55 mL of H2O) was added via pipette over approximately 10 minutes. The resulting mixture was placed in a 60°C oil bath and heated for 1.5 hours. HPLC analysis [C18 μBondapak, 3.9 x 300 mm, 50:50 AcCN:H2θ(0.1% TFA) at 1 mlJmin, 254 nm] showed the desired product at 6.5 minutes. The reaction was cooled to room temperature, placed in a separatory funnel and extracted with ethyl acetate. The organic layer was dried (MgSO4), filtered and concentrated to give a purple oil. By TLC, the desired product has Rf=0.45 (3:1 hexane:ethyl acetate). Purification by flash chromatography (hexane, 10:1 hexane:ethyl acetate, 5:1 hexane:ethyl acetate) gave the 3-iodo-5- nitrobenzyl alcohol as a yellow solid: mp 85.5-86.5°C.

Step C: 3-Iodo-5-nitrobenzyl fluoride

A solution of 3-iodo-5-nitrobenzyl alcohol (7.2 g, 25.8 mmol) described in Step B, in CH2CI2 (325 mL) was cooled to 0°C and treated with iPr2NEt (6.57 mL, 37.7 mmol) and methane sulfonyl chloride (2.19 mL, 28.2 mmol). After stirring for 1.5 hours at 0°C, the reaction was poured into a separatory funnel containing aqueous 5% citric acid. The layers were separated and the aqueous layer was extracted with CH2CI2. The organic layers were combined and dried (MgSO4), filtered and concentrated to give an orange oil. This material was dissolved in acetonitrile (325 mL), cooled to 0°C and treated with tetrabutyl ammonium fluoride (47 mL, 1 M in

THF, 47 mmol) via an addition funnel over about 10 minutes giving a dark blue-green mixture. This was stirred overnight as the reaction warmed to room temperature. TLC analysis (1:1 hexane:ethyl acetate) showed the desired product (Rf=0.56). The reaction was diluted with H2O, concentrated and then extracted with ethyl acetate. The ethyl acetate was dried (MgSO4), filtered and concentrated to give a dark brown oil. This material was dissolved in methanol, treated with 7 g of silica gel, concentrated to a free flowing powder and purified by flash chromatography (hexane then 10:1 hexane:ethyl acetate) to give 3-iodo-5-nitrobenzyl fluoride as a tan solid: mp 50.5-51.5°C.

Step D: 3-Amino-5-iodobenzyl fluoride A solution of 3-iodo-5-nitrobenzyl fluoride (9.2 g, 32.7 mmol) described in

Step C, in methanol (311 mL) was treated with ΗCI3 (311 mL, >10 wt. % in 20-30 wt % HCl) and stirred at room temperature for 20 minutes. The reaction was poured into a separatory funnel containing 95:5 CH2Cl2:MeOH and the aqueous layer was made basic with 15% aq. NaOH. The layers were separated and the aqueous layer was extracted with 95:5 CH2Cl2:MeOH. The organic layers were combined, dried (MgSO4), filtered and concentrated to give an orange semi-solid. Purification by flash chromatography (3:1 to 1:1 hexane: methylene chloride) provided 3-amino-5- iodobenzyl fluoride as a tan solid: mp 48.5-50°C.

Step E: (S)-N-(tert-butoxycarbonyl)homoserine lactone

To a solution of (S)-homoserine lactone hydrochloride (11.0 g, 79.9 mmol) and di-tert-butylpyrocarbonate (19.2 g, 88.0 mmol) in 160 mL of dichloromethane at 0°C was added diisopropyl-ethylamine (13.9 mL, 79.9 mmol) over 3 min. The solution was allowed to warm to room temperature. After 3 hours, another portion of di-tert-butylpyrocarbonate (1.75 g, 8.0 mmol) and diisopropylethylamine (0.70 mL, 4.0 mmol) were added, and the mixture was stirred for an additional 2.5 hours. The solution was washed with 10% citric acid, sat. ΝaHCθ3, and brine, dried (Na2SO4), filtered, and concentrated in vacuo. The resulting material was purified by silica gel chromatography (50% EtOAc/hexane) to provide the title compound.

Step F: (S)-N-(tert-butoxycarbonyl)homoserine lactol

To a solution of the lactone from Step E (7.0 g, 35 mmol) in 175 mL of THF at -7°C was added diisobutylaluminum hydride (72.0 mL, 1M in THF, 72 mmol) dropwise, while maintaining the reaction temperature below -7°C. After 3 hours, another portion of diisobutylaluminum hydride (10.0 mL, 10 mmol) was added, followed by another after 1 hour (20.0 mL, 20 mmol). After an additional hour, the reaction was quenched with EtOAc at -7°C, followed by sat. Na-K-tartrate soln., then warmed to room temperature. The solution was poured into EtOAc, washed with brine, dried (Na2SO4), filtered, and concentrated in vacuo. The resulting material was purified by silica gel chromatography (50% EtOAc/hexane) to give the titled lactol.

Step G: (S)-l-N-(3-Fluoromethyl-5-iodophenyl)-2-N-t-butoxy-carbonyl-l,2- di amino-4-butanol

A solution of 3-amino-5-iodobenzyl fluoride from Step D (997 mg, 3.97 mmol) and aminolactol described in Step F (882 mg, 4.34 mmol) in 1,2- dichloroethane (10.4 mL) at room temperature was treated with acetic acid (0.23 mL), stirred for 10 minutes, and then was treated with NaBH(OAc)3 (1.15 g, 5.43 mmol). After stirring at room temperature for 2.75 hours, the reaction was poured into a separatory funnel containing CH2CI2, washed with aq. Sat'd NaHCO3- The organic layer was dried (MgSO4), filtered and concentrated to give 1.71 g of a yellow foam. TLC analysis (3:1 hexane:ethyl acetate) showed unreacted benzyl fluoride (Rf=0.27) along with an unknown spot (Rf=0.17) and the desired product (Rf=0.05). Purification by radial chromatography (10:1 to 1:2 hexane:ethyl acetate) gave the title benzyl di amine.

Step H: (S)-l-(3-Fluoromethyl-5-iodophenyl)-4-(t-butoxycarbonyl)-5- hydroxyethyl-2-piperazinone A mixture of the diamine described in Step G (630 mg, 1.44 mmol) in ethyl acetate (9.6 mL) and aq. Sat'd NaHCO3 (9.6 mL) was cooled to 0°C and treated with chloroacetyl chloride (0.126 mL, 1.58 mmol). After stirring for 2.25 hours at 0°C, the reaction was diluted with ethyl acetate/H2θ. The layers were separated and the aqueous layer was extracted with ethyl acetate. The organic layers were combined, dried (MgSO4), filtered and concentrated to give a foam. This material was dissolved in DMF (13 mL), cooled to 0°C and treated with CS2CO3 (3.1 g, 9.51 mmol). After stirring for 2.25 hours at 0°C, the reaction was diluted with ethyl acetate/brine/aq. Sat'd NH4CI. The layers were separated and extracted with ethyl acetate. The organic layers were combined, washed with brine, dried (MgSO4), filtered and concentrated to give a yellow oil.

TLC analysis (1:1 hexane:ethyl acetate) showed the main spot with Rf=0.12. Purification by radial chromatography (1:1 to 1:2 hexane:ethyl acetate) gave the title hydroxyethyl piperazinone as a yellow oil.

Step I: (S)-l-(3-Fluoromethyl-5-iodophenyl)-4-(t-butoxycarbonyl)-5- r(methanesulfonyl)ethyll-2-piperazinone

A solution of hydroxyethyl piperazinone described in Step H (52 mg, 0.11 mmol) in CH2CI2 (1 mL) was cooled to 0°C and treated with N,N- diisopropylethylamine (0.038 mL, 0.22 mmol) and methanesulfonyl chloride (0.012 mL, 0.15 mmol). After being stirred for 1.25 hours at 0°C, the reaction was poured into a separatory funnel containing 5% citric acid and the layers were separated. The aqueous layer was extracted with CH2CI2, dried (MgSO4), filtered and concentrated to give 66.4 mg of a white foam. The H NMR (CDCI3) showed the mesylate singlet at 63.01. The mesylate was dissolved in DMF (1.5 mL), cooled to 0°C and treated with sodium thiomethoxide (15.4 mg, 0.22 mmol) to give a clear magenta solution. The reaction was stirred for 30 minutes at 0°C, stored overnight in the freezer and stirred an additional hour at 0°C. The reaction was poured into a separatory funnel containing ethyl acetate and washed with aq. Sat'd NaHCO3 and brine. The organic layer was dried (MgSO4), filtered and concentrated to give 59.5 mg of an orange oil. The 1H NMR (CDCI3) showed the methyl singlet at 52.12. The methyl sulfide was dissolved in methanol (1 mL) at room temperature and treated with monoperoxyphthalic acid, magnesium salt (163 mg, 0.33 mmol) in methanol (1.5 mL) and stirred at room temperature for 1.75 hours. The reaction was quenched with 2N Na2S2θ3, poured into a separatory funnel containing ethyl acetate/aq. Sat'd NaHCO3 and the layers were separated. The organic layer was washed with brine, dried (MgSO4), filtered and concentrated to give the crude product. TLC analysis (2:1 ethyl acetate:hexane) showed one main spot at Rf=0.25. Purification by radial chromatography (1:1 hexane:ethyl acetate) gave the title methyl sulfone as an oil.

Step J: Preparation of l-triphenylmethyl-4-(hvdroxymethyl)-imidazole

To a solution of 4-(hydroxymethyl)imidazole hydrochloride (35.0 g, 260 mmol) in 250 mL of dry DMF at room temperature was added triethylamine (90.6 mL, 650 mmol). A white solid precipitated from the solution.

Chlorotriphenylmethane (76.1 g, 273 mmol) in 500 mL of DMF was added dropwise. The reaction mixture was stirred for 20 hours, poured over ice, filtered, and washed with ice water. The resulting product was slurried with cold dioxane, filtered, and dried in vacuo to provide the titled product as a white solid which was sufficiently pure for use in the next step.

Step K: Preparation of l-triphenylmethyl-4-(acetoxymethyl)-imidazole

Alcohol described in Step J (260 mmol, prepared above) was suspended in 500 mL of pyridine. Acetic anhydride (74 mL, 780 mmol) was added dropwise, and the reaction was stirred for 48 hours during which it became homogeneous. The solution was poured into 2 L of EtOAc, washed with water (3 x 1 L), 5% aq. HCl soln. (2 x 1 L), sat. aq. NaHCO3, and brine, then dried (Na2SO4), filtered, and concentrated in vacuo to provide the crude product. The acetate was isolated as a white powder which was sufficiently pure for use in the next reaction. Step L: Preparation of l-(4-cyanobenzyl)-5-(acetoxymethyl)-imidazole hydrobromide

A solution of the product described in Step K (85.8 g, 225 mmol) and - bromo-p-tolunitrile (50.1 g, 232 mmol) in 500 mL of EtOAc was stirred at 6°C for 20 hours, during which a pale yellow precipitate formed. The reaction was cooled to room temperature and filtered to provide the solid imidazolium bromide salt. The filtrate was concentrated in vacuo to a volume 200 mL, reheated at 60°C for two hours, cooled to room temperature, and filtered again. The filtrate was concentrated in vacuo to a volume 100 mL, reheated at 60°C for another two hours, cooled to room temperature, and concentrated in vacuo to provide a pale yellow solid. All of the solid material was combined, dissolved in 500 mL of methanol, and warmed to 60°C. After two hours, the solution was reconcentrated in vacuo to provide a white solid which was triturated with hexane to remove soluble materials. Removal of residual solvents in vacuo provided the titled product hydrobromide as a white solid which was used in the next step without further purification.

Step M: Preparation of l-(4-cyanobenzyl)-5-(hydroxymethyl)-imidazole

To a solution of the acetate described in Step L (50.4 g, 150 mmol) in 1.5 L of

3:1 THF/water at 0°C was added lithium hydroxide monohydrate (18.9 g, 450 mmol). After one hour, the reaction was concentrated in vacuo, diluted with EtOAc (3 L), and washed with water, sat. aq. NaHCO3 and brine. The solution was then dried (Na2SO4), filtered, and concentrated in vacuo to provide the crude product as a pale yellow fluffy solid which was sufficiently pure for use in the next step without further purification.

Step N: Preparation of l-(4-cyanobenzyl)-5-imidazolecarboxaldehyde

To a solution of the alcohol described in Step M (21.5 g, 101 mmol) in 500 mL of DMSO at room temperature was added triethylamine (56 mL, 402 mmol), then Sθ3-pyridine complex (40.5 g, 254 mmol). After 45 minutes, the reaction was poured into 2.5 L of EtOAc, washed with water (4 x 1 L) and brine, dried (Na2SO4), filtered, and concentrated in vacuo to provide the aldehyde as a white powder which was sufficiently pure for use in a next step without further purification.

Step O: (S)-l-(3-Fluoromethyl-5-iodophenyl)-4-[l-(4-cyanobenzyl)-5- imidazoylmethvH-5-r(methanesulfonyl)-ethyl.-2-piperazinone A solution of methyl sulfone described in Step I (48 mg, 0.094 mmol) in CH2CI2 (1.2 mL) at room temperature was treated with TFA (0.33 mL, 4.28 mmol) and stirred for 30 minutes at room temperature. The reaction was concentrated, treated with 1,2-dichloroethane (1 mL), triethylamine (10-15 drops until pH>7.5), aldehyde describe in Step N (40 mg, 0.19 mmol), sodium triacetoxyborohydride (46 mg, 0.22 mmol) and molecular sieves. The reaction was stirred at room temperature and followed by HPLC [C18 μBondapak, 3.9 x 300 mm, 10% AcCN:H2θ (0.1% TFA) to 90% AcCN over 30 minutes with linear gradient, 1 m-IJmin, 220 and 254 nm] until product was maximized (retention time = 21 minutes). Two additional portions of sodium triacetoxyborohydride had to be added. The reaction was stirred a total of 4 days. The reaction was diluted with CH2CI2, washed with aq. Sat;d NaHCO3 (emulsion forms) and dried (MgSO4), filtered and concentrated to give an oil. Purification by radial chromatography (2% MeOH:CHCl3 until the aldehyde eluted and then 5% MeOH) gave the title compound (Compound 1).

EXAMPLE 2

Preparation of (S)-l-(3-iodophenyl)-4-[l-(4-cyanobenzyl)-5-imidazolylmethyl]-5- r2-(methanesulfonyl)ethyll-2-piperazinone dihydrochloride (Compound 2)

The title compound was prepared using the procedure described in Example 1, Steps G-O, but replacing 3-amino-5-iodobenzyl fluoride with 3- iodoaniline in Step G EXAMPLE 3

Preparation of [123j](S)_ι _(3_i_{0( 0}.5.fi_uoromemyip]_ιenyi)_4.[i.(4._Cy_anobenzyl)-5- imidazolylmethyll-5-r2-(methanesulfonyl)ethvll-2-piperazinone dihydrochloride

(^,r¹²³Il-Compound 1)

Step A: (S)-l-(3-Fluoromethyl-5-trimethylstannylphenyl)-4-[l-(4- cyanobenzyl)-5-imidazoylmethyl]-5-[(methanesulfonyl)-ethyl]-2- piperazinone

A solution of Compound 1 (12.9 mg, 0.02 mmol, described in Example 1) in dioxane (0.5 mL) at room temperature was treated with hexamethylditin (0.015 mL, 0.03 mmol), a catalytic amount of catalyst and placed in a 100°C oil bath for 2 hours giving a black opaque mixture. The reaction was filtered through celite, rinsed with ethyl acetate and concentrated to give an oil. TLC analysis (5% MeOH:CH3Cl) showed the desired product (Rf=0.06). Purification by radial chromatography (5% to 10% MeOH:CH3θ) gave the title compound as a clear colorless oil.

Step B: [123τ](S)-l-(3-Fluoromethyl-5-iodophenyl)-4-[l-(4-cyanobenzyl)-5- imidazoylmethvn-5-r(methanesulfonyl)-ethyl1-2-piperazinone A shipping vial of Nal23i (20 mCi, dry, Nordion) containing a stir bar was treated with an iodobead, MeOH (0.05 mL) and Nal27τ solution (0.013 mg/mL H2O) and stirred for 5 minutes at room temperature. A solution of trimethylstannane described in Step A (0.5 mg) in MeOH (0.05 mL) was treated with trifluoroacetic acid (0.02 mL) and immediately added to the Nal23τ/j₀dobead vial. After stirring for 5 minutes at room temperature, the reaction was quenched with concentrated NH4OH (0.02 mL) and aqueous Na2S2θ5 (0.01 mL of a 10 mg/mL solution). The reaction mixture was drawn into an HPLC syringe and injected for purification [Nydac C18 Protein and Peptide column, 3.9 x 250 mm, 1 m-lJmin, 30:70 AcCΝ:H2θ (0.1% TFA), retention time = 8 minutes]. A fraction collector was used to collect 0.5 mL fractions. The fractions containing the labeled product were pooled and partially concentrated to give a solution [123i] -Compound 1.

[123i](S)-i-^.(3-iodo-phenyl)-4-[l-(4-cyanobenzyl)-5-imidazolylmethyl]-5-[2-

(methanesulfonyl)ethyl]-2-piperazinone dihydrochloride ([1 3i]_Compound 2) was also prepared by this method but starting with Compound 2 (prepared as described in Example 2) instead of Compound 1.

EXAMPLE 4

Preparation of [1 5i](S)_ι .(3.j₀₍j₀_5.fi_uorometj₁ylph_enyl)-.4-.[i -.(4--_Cy_anol-_)enZyl).5_ imidazolvlmethvH-5-r2-(methanesulfonvl)ethyl^~l-2-piperazinone dihydrochloride

Step A: [125i](S) -(3-Fluoromethyl-5-iodophenyl)-4-[l-(4-cyanobenzyl)-5- imidazoylmethyll-5-r(methanesulfonyl)-ethyn-2-piperazinone A shipping vial of Nal25τ (i o _mCi, 100 mCi/mL 10^"5M NaOH, pH 8-11, NEN) containing a stir bar was treated with an iodobead and stirred for 5 minutes at room temperature. A solution of the trimethylstannane described in Example 3, Step A (0.5 mg) in MeOH (0.05 mL) was treated with trifluoroacetic acid (0.02 mL) and immediately added to the Nal25τ/i₀₍jobead vial. After being stirred for 5 minutes at room temperature, the reaction was quenched with concentrated NH4OH (0.02 mL) and aqueous Na2S2θ5 (0.02 mL of a 10 mg/mL solution). The reaction mixture was drawn into an HPLC syringe and injected for purification [Vydac C18 Protein and Peptide column, 3.9 x 250 mm, 1 mlJmin, 30:70 AcCN:H2θ (0.1% TFA), retention time = 8 minutes]. A fraction collector was used to collect 0.2 mL fractions. The fractions containing the labeled product were pooled, concentrated almost to dryness and reconstituted in ethanol (1.0 mL) to give a solution of [125i]-derivative of the title compound.

EXAMPLE 5

Preparation of [1 51] (S)-l-(3-iodophenyl)-4-[l-(4-cyanobenzyl)-5- imidazolylmethyl1-5-r2-(methanesulfonvl)ethyl]-2-piperazinone dihydrochloride

(Compound I) Employing the procedure substantially as described in Example 4 but substituting for the starting material used therein the product of Example 2, there was produced the title compound.

BIOLOGICAL ASSAYS. The biological activities of the above compounds and the use of the compounds in the instant assay can be demonstrated using the following assays.

In vitro inhibition of ras farnesyl transferase

Transferase Assays. Isoprenyl-protein transferase activity assays are carried out at 30 °C unless noted otherwise. A typical reaction contains (in a final volume of 50 μL): [³H]farnesyl diphosphate, Ras protein , 50 mM HEPES, pH 7.5, 5 mM MgCl2, 5 mM dithiothreitol, 10 μM ZnCl2, 0.1% polyethyleneglycol (PEG)

(15,000-20,000 mw) and isoprenyl-protein transferase. The FPTase employed in the assay is prepared by recombinant expression as described in Omer, C.A., Krai, A.M., Diehl, R.E., Prendergast, G.C., Powers, S., Allen, CM., Gibbs, J.B. and Kohl, N.E. (1993) Biochemistry 32:5167-5176. After thermally pre-equilibrating the assay mixture in the absence of enzyme, reactions are initiated by the addition of isoprenyl- protein transferase and stopped at timed intervals (typically 15 min) by the addition of 1 M HCl in ethanol (1 mL). The quenched reactions are allowed to stand for 15 m (to complete the precipitation process). After adding 2 mL of 100% ethanol, the reactions are vacuum-filtered through Whatman GF/C filters. Filters are washed four times with 2 mL aliquots of 100% ethanol, mixed with scintillation fluid (10 mL) and then counted in a Beckman LS3801 scintillation counter.

For inhibition studies, assays are ran as described above, except inhibitors are prepared as concentrated solutions in 100% dimethyl sulfoxide and then diluted 20-fold into the enzyme assay mixture. Substrate concentrations for inhibitor IC50 determinations are as follows: FTase, 650 nM Ras-CVLS (SEQ.ID.NO.: 1), 100 nM farnesyl diphosphate.

The radiolabeled compounds useful in the instant invention described in the above Examples are tested for inhibitory activity against human farnesyl transferase by the assay described above.

In vivo ras prenylation assay

The cell lines used in this assay consist of either Ratl or NTH3T3 cells transformed by either viral Ha-ras; an N-r -y chimeric gene in which the C-terminal hypervariable region of v-Ha-rαs was substituted with the corresponding region from the N-ras gene; or ras-CVLL (SEQ.ID.NO.: 2), a v-Ha-ras mutant in which the C- terminal exon encodes leucine instead of serine, making the encoded protein a substrate for geranylgeranylation by GGPTase I. The assay can also be performed using cell lines transformed with human Ha-ras, N-ras or Ki4B-ra-s. The assay is performed essentially as described in DeClue, J.E. et al., Cancer Research 51:712- 717, (1991). Cells in 10 cm dishes at 50-75% confluency are treated with the test compound(s) (final concentration of solvent, methanol or dimethyl sulfoxide, is 0.1%). After 4 hours at 37°C, the cells are labelled in 3 ml methionine-free DMEM supplemented with 10% regular DMEM, 2% fetal bovine serum, 400 μCi[35s]methionine (1000 Ci/mmol) and test compound(s). Cells treated with lovastatin, a compound that blocks Ras processing in cells by inhibiting the rate- limiting step in the isoprenoid biosynthetic pathway (Hancock, J.F. et al. Cell, 57:1167 (1989); DeClue, J.E. et al. Cancer Res., 51:712 (1991); Sinensky, M. et al. J. Biol. Chem., 265:19937 (1990)), serve as a positive control in this assay. After an additional 20 hours, the cells are lysed in 1 ml lysis buffer (1% NP40/20 mM HEPES, pH 7.5/5 mM MgCl2/lmM DTT/10 mg/ml aprotinen/2 mg/ml leupeptin/2 mg/ml antipain 0.5 mM PMSF) and the lysates cleared by centrifugation at 100,000 x g for 45 min. Alternatively, four hours after the additon of the labeling media, the media is removed, the cells washed, and 3 ml of media containing the same or a different test compound added. Following an additional 16 hour incubation, the lysis is carried out as above. Aliquots of lysates containing equal numbers of acid-precipitable counts are bought to 1 ml with IP buffer (lysis buffer lacking DTT) and immunoprecipitated with the ras-specific monoclonal antibody Y13-259 (Furth, M.E. et al., J. Virol. 43:294-304, (1982)). Following a 2 hour antibody incubation at 4°C, 200 μl of a 25% suspension of protein A-Sepharose coated with rabbit anti rat IgG is added for 45 min. The immunoprecipitates are washed four times with IP buffer (20 nM HEPES, pH 7.5/1 mM EDTA/1% Triton X- 100.0.5% deoxycholate/0.1%/SDS/0.1 M NaCl) boiled in SDS-PAGE sample buffer and loaded on 13% acrylamide gels. When the dye front reached the bottom, the gel is fixed, soaked in Enlightening, dried and auto- radiographed. The intensities of the bands corresponding to prenylated and nonprenylated Ras proteins are compared to determine the percent inhibition of prenyl transfer to protein.

Partition Coefficient Protocol Prior to assay, equal volumes of 1-octanol and pH 7.4 buffer are mutually saturated, allowed to settle, and separated into individual containers until use.

Standard solutions are prepared by dissolving an accurately weighed amount (1-2 mg) of the compound under study in a suitable solvent. Sample is dissolved in 10.00 ml spectral grade methanol although ethanol, acetonitrile or water may also be used provided the compound is completely soluble in that solvent. The solutions or appropriate dilutions are scanned by UV (HP 8452A Diode Array Spectrophotometer) to determine the wavelength of maximum absorbance (λmax). The scan can also be obtained from the HPLC chromatogram of the methanol standard. Partition sample solutions are prepared by placing an accurately weighed 1-2 mg of sample into a 20 ml scintillation vial and adding 10.00 ml of pH 7.4 buffer and 10.00 ml of 1-octanol which have been mutually saturated with each other. Vials are placed in the ultrasonic bath for 5 min. and then onto a flatbed shaker for at least 2 to 4 hrs. After agitation, vials are centrifuged at 1500 rpm for 10 min. and approximately 1 ml of each layer is removed to separate HPLC vials for analysis.

Measurement of concentration of the standard, octanol, and buffer solutions can be conducted either by absorbance in ultraviolet spectroscopy (UV) or by peak areas in high performance liquid chromatography (HPLC) with the detector set at the compound's λmax. Currently, HPLC measurement is the method of choice.

Appropriate dilutions, if necessary, should be made into the appropriate solvent.

Standard conditions of analysis are as follows:

Instrument—HP 1090 HPLC using either the diode array (DAD) or the variable wavelength detectors (VWD) set at the compound's λmax. Column- Vydac Protein/Peptide C-18

Solvent system— Gradient from 5-95% Acetonitrile/H3PO4 Buffer over 10 min.; flow rate-3.0 ml/min.

Sample size— 5 μl injection; larger injections (50, 100, 200 μl) can be used subsequently should no detectable peak result from the smaller injection. Temperature— Ambient conditions.

Partition coefficients (lipophiliciies) are calculated by the following equation:

(a) Partition Coefficient (PC) = (Octanol HPLC Area)(Octanol Dilution)

(Buffer HPLC Area)(Buffer Dilution)

In vivo Enzyme Binding Studies

To determine the biodistribution of a radiotracer compound, 3-5 μCi of radiotracer is injected (i.v. in 30% PEG/10% EtOH/H2θ) in male Sprague-Dawley rats (200-250 g). The animals are euthanized at 30 min., 120 min. and 360 min. after radiotracer injection. To determine the extent of specific binding to FPTase, an unlabeled FPTase inhibitor, such as the title compound of Example 1, (10 mpk or 5 mpk) is injected 30 min. prior to injection of the radiotracer, then the animals are euthanized at the times provided above. After euthanasia by cervical dislocation under light anesthesia, the thoracic aorta are cut and 1 ml of blood collected in heparinized syringes. Centrifugation for 2.5 minutes at 5000 rpm provides the plasma samples. The left ventricular muscle, left upper lobe of the lung, liver, kidney (cortex), both adrenals, spleen, pancreas, descending colon and prostate are removed, placed on ice and two 50 mg samples biopsied. For radiotracers with 125τ labels, samples are counted, without further treatment, in an autogamma counter for 2 min. each. For tritium-containing radiotracers, tissues are dissolved by shaking in Biosolv (New England Nuclear) overnight. After neutralization of base with 0.5 N HCl, scintillation cocktail is added (in subdued light), samples dark adapted for several hours, then counted in a scintillation counter through several cycles to insure chemiluminescene is not significant. Data are expressed as %-injected dose/gm wet weight tissue.

To determine in vivo dissociation rates of radiotracer compound, 3-5 μCi of radiotracer is injected (i.v. in 30% PEG/10% EtOH/H2θ) in male Sprague-Dawley rats (200-250 g). One hour after radiotracer injection, an unlabeled FPTase is injected i.v., and the animals are euthanized 60 min. after this "chase." For comparison, uninhiibited binding is also determined at 60 and 120 min. post-injection of radiotracer (corresponds to 0 and 60 min. post-chase). Total enzyme signal is examined by preinjection of unlabeled FPTase inhibitor (5 mpk) 30 min. prior to radiotracer injection. After euthanasia, tissues samples are obtained and processed as described above.

Gamma Camera Imaging

Two rats are anesthetized (ketamine/ace-promazine), positioned on the camera head, and their tail veins canulated for ease of injection. One rat is preinjected with an unlabeled FPTase inhibitor (10% EtOH/27% PEG/63% H2O) 30 min. prior to injection of radiotracer to demonstrate non-specific binding. 150 mCi/rat of an 123j labeled FPTase inhibitor is injected via its tail vein, and the catheters flushed with several mis of normal saline. Acquisition of images is started as the radiotracer was injected. Sixty, one minute images were acquired and the rats are subsequently euthanized with sodium pentobarbital. The images acquired during the first minute are dominated by blood flow, and as a result, provide good depiction of the heart, liver and kidneys. Regions of interest (ROIs) are drawn on the first image which includes a region defined as soft tissue (upper left chest), then used to analyze the count rates in subsequent images. The ROIs do not include the entire liver since radioactivity in adjacent tissues partially obscures these structures. Therefore, ROIs are defined to remain fairly clear during the course of the study, and are assumed to be representative of the entire organ. Radioactivity in the bladder was obtained from the final image. Count-rates are converted to %-dose/ROI by dividing the count-rate in the ROI by that of the whole rat, which is then multiplied by 100.

In vivo Occupancy Studies

Kinetics of enzyme occupancy by an unlabeled FPTase inhibitor (test compound) is determined by this assay. Thus, FNB mice are injected the test compound at either 40 mpk or 10 mpk subcutaneously (0.1 mis). Thirty minutes prior to euthanasia, 3 nCi of tritiated farnesyl protein transferase inhibitor (radiotracer) was injected i.p. (0.2 mis 10%EtOH in 0.9% saline). Uninhibited binding of the radiotracer is determined by injection of the radiotracer in mice which have not received injection of the test compound. The mice are euthanized either 2 hours of 14 hours after injection of the test compound and lung, spleen, pancreas and blood are removed and processed as described above. Data are plotted as %-dose/g of tissue (wet weight). PET Imaging in Dogs

Female beagle dogs weighing 7.7 -14.6 kg (11.0 ± 2.3 kg) are fasted for at least 12 hours allowing water intake ad libitum, and are premedicated with 0.3 - 0.4 mL Acepromazine injected i.m. on the day of the experiment. A 20 G two inch venous catheter is placed into the right front leg ulnar vein through which anesthesia is introduced by sodium pentobarbital 25-30 mg/kg in 3-4 ml and maintained with additional pentobarbital at an average dose of 3 mg/kg/hr. Another catheter is inserted into the contralateral ulnar vein for radiotracer administration.

Oxygen saturation of circulating blood is measured with a pulse oximeter (Nellcor Inc., Hayward, CA) placed on the tongue of the animal. Circulatory volume is maintained by intravenous infusion of isotonic saline. A 22 G cannula is inserted into the anterior tibial or distal femoral artery for continuous pressure monitoring (Spacelabs, model 90603 A). EKG, heart rate, and core temperature are monitored continuously. In particular, EKG is observed for ST segment changes and arrhythmias.

The animal is positioned in the PET camera and a tracer dose of [15θ]H2θ administered via i.v. catheter. The image thus obtained is used to insure that the dog is positioned correctly to include liver, kidneys and pancreas. Subsequently [1 lC]- Compound 1 (<20 mCi) is administered via i.v. catheter. Following the acquisition of the total radiotracer image, an infusion is begun of the unlabeled FTI (test compound) at one of three dose rates (0.1, 1 or 10 mpk/day). After infusion for 2.5 hrs, [1 lC]- Compound 1 is again injected via the catheter. Images are again acquired for up to 90 min. Within ten minutes of the injection of radiotracer and at the end of the imaging session, 1 ml blood samples are obtained for determining the plasma concentration of test compound. At the conclusion of the study, animals are recovered and returned to animal housing.

For uninhibited distribution of radiotracer, regions of interest (ROIs) are drawn on the reconstructed image includes the kidney cortex and a region of liver which is removed from the gallbladder images. These regions are used to generate time activity curves obtained in the absence of test compound or in the presence of test compound at the various infusion doses examined. Inhibition curves are generated from the data obtained in a region of interest obtained starting at 70 min. post-injection of radiotracer. At this time, clearance of non-specific binding has reached steady state. The ID50 values were obtained by curve fitting the dose- rate/inhibition curve, hereinabove.

Cell Radiotracer Assay of Farnesyl Transferase Inhibitors (CRAFTI )

This assay measures the competition between a farnesyl transferase inhibitor (FTI) and a radiolabeled FTI for binding to high affinity sites (presumably farnesyl transferase) in living cells.

Fresh radiotracer ([123i]_c_omp_0uncl 2, described in Example 3) is synthesized monthly, with a specific activity of -350-1400 Ci/mmole. CRAFTI is run routinely using a Ratl fibroblast line transformed by v-Ha-ras (Hras/ratl). CRAFTI is performed by growing cells under anchorage-dependent conditions in 24 -well tissue culture plates overnight, to achieve near confluent monolayers of cells. Radiotracer is diluted in cell growth media to a concentration of ~1 nM (~1 mCi/ml), and vehicle or test FTI (in log dilutions, 6 point titration) is added to the diluted tracer. The growth media is removed from the cell monolayers, and 0.65 ml of the diluted radiotracer/test FTI mixture is applied. After 4 hr incubation, the tracer is removed by aspiration, the monolayers are rinsed quickly with 2 ml PBS, and the cells are trypsinized and transferred to tubes for gamma counting. Dose-inhibition curves and ICso's are determined by curve-fitting the equation: Radiotracer Bound = A₀ - A₀*[I_o]/([I₀) + IC50) + NS, where A₀ is the count-rate of radiotracer in the absence of inhibitor, IQ is the concentration of added FTI, IC₅₀ is the concentration of FTI that inhibits 50% of radiotracer binding and NS is the extent of non-specific binding. Cell Radiotracer Assay of Farnesyl Transferase Inhibitors Using Mouse Bone

Marrow Cells

Materials

Nude Mice (Charles River Labs, females, 3-6 months old)

Dissection tools (scissors, forceps)

Plastic chamber for CO2 asphyxiation of mice

Dry ice

5, 10, 20 ml disposable syringes (Becton Dickinson)

21G, 25G needles (Becton Dickinson) • Sterile 50 ml conical tubes

24-well tissue culture plates (Costar #3524) sterile reagent reservoir (Costar #4870) cell culture medium

• 90% (v/v) DMEM (Dulbeccoϊs Modified Eagle Medium) (Cellgro #10- 013-LM)

• 10% Fetal Bovine Serum (heat inactivated 30' @ 56°C)

• Penicillin/Streptomycin

• hemacytometer

• trypan blue • Phosphate Buffered Saline (PBS) (RCM #563)

• DMSO

• Wescodyne® germicidal detergent

• Multichannel Pipette (from Matrix Technologies Corporation, Lowell, MA (800- 345-0206)) • Electrapette® 1-25 μl, and 1250 μl multichannel pipette

• Electrapette-EXP® 1250 μl pipette, converts between 6-well and 24-well plates

• 96-well round bottom seroclusters (Costar #3794) • Uniblock 96- 2ml wells (Matrix Technologies Corporation, #8200-96)

• Cell culture incubator, 5% CO2

• Vacuum manifold (for 2.5 cm filters; Millipore Corporation Bedford, MA; cat.#09-804-24C) • G4 Glass fiber filter circles (2.4 cm); Fisher Cat.#09-804-24C

• Dispenser Bottle (Brinkmann Dispensette)

• Disposable transfer pipettes (Samco® Scientific CO., catalog #202-20S)

• gamma counter (for example, Cobra π from Packard Instrument Company)

• test tubes (polypropylene, 5 ml) and caps (#8562-1 and #5232 from Packard) Procedure

Each test compound is tested in a 9-point titration, in half-log dilutions, plus a vehicle control. 500X stock solutions of half-log dilutions of each compound is prepared in DMSO in a 96-well round bottom plate, (a robotics apparatus can be used to do this). Radiotracer I is diluted in cell culture media to 2 μCi/ml (2 nM, assuming a specific activity of 1000 Ci/mmol). About 16 mis of diluted radiotracer is required for each compound to be assayed.

Using the 1.25 mL multichannel pipette, a 0.55 mL aliquot of diluted tracer is added to each well of the 96-well Uniblock. Using the 1.25 mL multichannel pipette, 2.2 μl of the 500X stock compound is added to the 0.55 ml diluted tracer. The solution is well mixed by pipetting up and down 6X using the mixing function on the 1.25 ml Electrapette® multichannel pipette. Tracer/FTI mixture (0.5 mL) of the tracer/FTI is removed from each well of the 96-well dilution block and transferred to the 24-well plate. (The six position multichannel pipette that converts between 96- and 24-well plates (Electrapette-EXP® from Matrix Technologies) is most convenient for this purpose.) The 24-well plates are then held at 37°C in a cell culture incubator until the bone marrow cells are ready.

The mice are sacrificed by CO2 asphyxiation and transferred to a tissue culture hood and the hind leg femurs are dissected out with their ends sealed by cutting the bone above the hip and below the knee. With sterile dissecting tools, the bones are cut above the knees and below the hip joints to "open" the femurs. Using a 5 ml syringe and a 21G needle, the marrow is flushed from each femur into a 50 ml conical tube on ice with media using ~ 4 mL per femur by flushing in one direction with ~2 ml, then flushing from the other end of bone with 2 ml. The procedure is repeated for the remaining femurs, and the marrow from all of the femurs is pooled.

Using a 20 ml syringe, the suspension of cells is passed slowly IX through a 25G needle into a separate 50 ml conical tube to produce a single cell suspension. A small aliquot of the cells is diluted 1:2 with trypan blue, and the larger viable cells are counted using a hemacytometer. The cells are diluted to a final concentration of 0.5-1 x 10⁶ per mL.

The cells are poured into a sterile reagent reservoir, and 0.5 mL of cells is transferred into each well of the 24-well plate containing 0.5 mL of the Tracer/FTI mix using the Electrapette-EXP®. As the 24-well plates are seeded, the cells are mixed in the reagent reservoir using a gentle rocking motion to ensure an even distribution of cells per well. Once the cells have been added to the 24-well plates, the plates are gently tapped to mix the cells and Tracer/FTI. The 24-well plates are incubated for 4 hr at 37°C in incubator with 5% CO2-

The GFC filters are placed into the vacuum manifold and prewet by adding 1 ml PBS to each filter with the dispenser bottle and then applying a vacuum to remove the PBS. The cell/Tracer/FTI mixtures are transferred to the vacuum manifold using the disposable transfer pipettes, and the filter by applying a gentle vacuum. The vacuum is turned off and 1 mL of PBS is added to each well of the 24-well plate. Each well is washed and the wash transferred to the appropriate filter in the vacuum manifold. More PBS (5 mL) is added to each filter and drawn through the filter by re- applying the vacuum. The filters are washed 3 more times with 5 ml PBS. (The wash steps should be done in as little time as possible (i.e. less than 5 min) since the radiotracer will begin to dissociate from the high affinity sites and wash out of the cell once the wash procedure has been initiated.) The filters are placed in the bottom of the 5 mL test tubes, caps applied and each vial is counted for 2 minutes on a gamma counter (without scintillation cocktail).

IC5θ's are determined by curve fitting a plot of the CPM bound as a function of dose using a 4-parameter curve-fitting equation: Radiotracer Bound = A₀ - A₀*[Io]/([Io) + IC50) + NS, where A₀ is the count-rate of radiotracer in the absence of inhibitor, IQ is the concentration of added FTI, IC₅₀ is the concentration of FTI that inhibits 50% of radiotracer binding and NS is the extent of non-specific binding. Non-specific binding in a titration series is defined as the CPM bound at the highest concentration of compound tested

Claims

WHAT IS CLAIMED IS:

1. An assay for determining the affinity of a famesyl transferase inhibitor test compound for famesyl transferase binding sites in living bone marrow cells which comprises measuring the competition between the test compound and a radiolabeled famesyl transferase inhibitor for the famesyl transferase binding sites.

2. The assay of Claim 1 which comprises the steps of: a) isolating and culturing bone marrow cells; b) exposing the bone marrow cells to growth media containing the radiolabeled famesyl transferase inhibitor in the presence or absence of the famesyl transferase inhibitor test compound; c) washing the cells; d) counting the radiation emitted by the cells; and e) comparing the radiation emitted by the cells exposed to both the radiolabeled famesyl transferase inhibitor and the test compound to the radiation emitted by the cells exposed to only the radiolabeled famesyl transferase inhibitor

3. The assay of Claim 2 wherein the bone marrow cells are isolated from hamsters, rabbits, rats or mice.

4. The assay of Claim 3 wherein the bone marrow cells are isolated from mice.

5. The assay of Claim 4 wherein the bone marrow cells are isolated from nude mice.

6. The assay of Claim 5 wherein the bone marrow cells are isolated from the hind-leg femurs of nude mice.

7. The assay of Claim 2 wherein the radiolabeled FTI is a radiolabeled analog of a compound of structural formula:

wherein:

R2 and R are independently selected from: H; unsubstituted or substituted Ci-8 alkyl, unsubstituted or substituted C2-8 alkenyl, unsubstituted or substituted C2-8 alkynyl, unsubstituted or substituted aryl, unsubstituted or substituted heterocycle,

wherein the substituted group is substituted with one or more of:

1) aryl or heterocycle, unsubstituted or substituted with: a) Ci-4 alkyl, b) (CH₂)_pOR⁶, c) (CH2)_pNR⁶R⁷, d) halogen, e) a radionuclide,

2) C3-6 cycloalkyl,

3) OR⁶,

4) SR⁶, S(O)R⁶, SO2R⁶, 5) — NR⁶R⁷

10) \ .NR⁶R⁷ 0

11) — S0₂-NR⁶R⁷

R⁶

12) -N-S0₂-R⁷

¹⁴⁾ ^_|-^0R6 ' °^r

O

15) a radionuclide; or R4 is selected from H and CH3; and any two of R2, R3 and R^ are optionally attached to the same carbon atom;

R⁶, R7 and R7 are independently selected from: H; C1.4 alkyl, 1 ^C-methyl, C3.6 cycloalkyl, heterocycle, aryl, aroyl, heteroaroyl, arylsulfonyl, heteroarylsulfonyl, unsubstituted or substituted with: a) Ci-4 alkoxy, b) aryl or heterocycle, c) halogen, d) HO,

f) -S0₂R¹¹ g) N(RlO)₂, h) HC-methyl, or i) a radionuclide; or

R⁶ and R7 may be joined in a ring;

R7 and R7a may be joined in a ring;

R8 is independently selected from: a) hydrogen, b) aryl, heterocycle, C3-C10 cycloalkyl, C2-C6 alkenyl, C2-C6 alkynyl, perfluoroalkyl, F, Cl, Br, R!0Q-, RUS(O)_m-, R¹⁰C(O)NR¹⁰-, CN, NO2, R¹⁰2N-C(NR¹⁰)-, R¹⁰C(O)-, R¹⁰OC(O)-, N3, -N(R¹⁰)2, or

RϋOC OjNR¹⁰-, c) C1-C6 alkyl unsubstituted or substituted by aryl, heterocycle, C3-C10 cycloalkyl, C2-C6 alkenyl, C2-C6 alkynyl, perfluoroalkyl, F, Cl, Br, R¹⁰O-, R¹ iSCOJm-, R¹⁰C(O)NH-, CN, H2N-C(NH)-, R¹⁰C(O)-,

R¹⁰OC(O)-, N3, -N(R¹⁰)2, or R¹⁰OC(O)NH-, and d) a radionuclide;

RIO is independently selected from hydrogen, C1-C6 alkyl, 3H-methyl, HC-methyl, benzyl and aryl;

RU is independently selected from C1-C6 alkyl and aryl;

is phenyl, unsubstituted or substituted with one or more of:

1) Cl-4 ^; alkyl, unsubstituted or substituted with: a) Cl-4 alkoxy, b) NR⁶R⁷, c) C3-6 cycloalkyl, ) aryl or heterocycle, e) HO, ) -S(O)_mR⁶,

g) -C(O)NR⁶R⁷, or h) a radionuclide;

2) aryl or heterocycle,

3) halogi -in,

4) OR⁶ _'

5) NR⁶R7, 6) CN,

7) NO₂,

8) CF₃;

9) -S(O)_mR⁶, 10) -C(O)NR⁶R⁷,

11) C3-C6 cycloalkyl, or

12) a radionuclide;

m is 0, 1 or 2; and r is 0 to 5; and

wherein at least one radionuclide selected from 3H ( T), ^C, l^F, 125 82B_Γ; 123j₃ ¹³ ll, 75_Br? 15₀, 13N, 211 At, 77_Br; 3_Hj 124j _or 82β_r is present in the molecule;

or a pharmaceutically acceptable salt thereof.

8. The assay of Claim 7 wherein the radiolabeled FTI is selected from:

or a pharmaceutically acceptable salt thereof.

9. The assay of Claim 8 wherein the radiolabeled FTI is the compound of the formula:

or a pharmaceutically acceptable salt thereof.

10. A method of predicting the relative bone marrow toxicity of an inhibitor of famesyl transferase in a patient by assessing the affinity of the inhibitor for famesyl transferase binding sites in living bone marrow.

11. The method according to Claim 10 wherein the affinity of the inhibitor for famesyl transferase binding sites is determined by utilizing an assay which comprises measuring the competition between the test compound and a radiolabeled famesyl transferase inhibitor for the famesyl transferase binding sites.