Importance of tumor affinity of nitroazoles in hypoxic radiosensitization

Inl. J. Rodiamn Oncobgy Biol. Phys., Vcl. Printed in the U.S.A. All rights reserved.

16. pp. 1033-1037

0360-3016/89 $3.00 + .JO Copyright 0 1989 Pergtmon Press plc

??Session II IMPORTANCE OF TUMOR AFFINITY OF NITROAZOLES IN HYPOXIC RADIOSENSITIZATION TSUTOMU

KAGIYA,

JUN WANG,

PH.D.,’ SEI-ICHI NISHIMOTO, PH.D.,’ YUTA

B.SC.,~ LIN ZHOU, B.Sc.,’ YAN-LING MASAJI TAKAHASHI,

‘Department of Hydrocarbon

M.D.2

SHIBAMOTO, M.D.,2

HE, B.Sc., ’ KEISUKE

AND MITSUYUKI

SASAI, M.D.,2

ABE, M.D.2

Chemistry, Faculty of Engineering, and *Department of Radiology, Faculty of Medicine, Kyoto University, Kyoto 606, Japan

In vitro and in viva sensitizing activities of a variety of nitroazole derivatives including misonidazole (MISO), SR2508, and RSU-1069 were correlated by the aid of pharmacokinetic measurements of the drug uptake in animal solid tumors. The sensitizer enhancement ratio in viva (SER& on solid tumors increased linearly with the square root of administrated dose (Ds). The specific activity (A) in viva of nitroazoles was evaluated from the square-root empirical relationship, SER,i,. = 1.00 + A DsI” . The intratumor concentration of nitroazoles at a given time t after administration was in proportion to the Ds, in which the proportional constant was defined as the tumor-affinity factor FT,,. The absolute molar activity (YMdefined by A(M/FT,$/*, where M is the molecular weight of nitroazoles, showed a linear relationship with the SER in vitro (SERvi,,) at 1 mM of sensitizers. The sensitizer dose required to achieve an SE&i,, of 1.5 (Ds& decreased and thus the overall sensitizing efficiency on animal solid tumors increased as the FT., became greater. Hypoxic cell sensdtizer, Fluorinated nitroazole, Specific activity, Tumor-affinity 1NTR:ODUCTION

and 3-nitro- 1,2,4-triazole and 4-nitro- 1,2,3-triazole derivatives (AK- and KU-2300 series) were synthesized by the Kyoto University research group. The synthesis of fluorinated nitroazoles has been described in the succeeding paper (8). Misonidazole*, SR-2508, and Ro 03-8799t were obtained from various commercial sources; SR-2508 was supplied by the NC1 (USA). RK-28 and RP- 170 were also obtained from a commercial source.+

Although

a number of drugs are now available that enhance the radiosensitivity of hypoxic cells in vitro, most of them are ineffective fior in vivo sensitization of animal solid tumors. In vivo sensitization may fail because of insufficient uptake of drugs in solid tumors. Thus, the development of radiosensitizers with higher tumor affinity and less toxicity is essential for achieving a high sensitizer enhancement ratio (SER) in clinical use. The present study ha.s focused on a pharmacokinetic determination of intratumor concentration of various nitroazoles including MISO, SR-2508 (3), RSU-1069 (l), and newly developed fluorinated derivatives. Two basic quantities, an absolute activity (a) and a tumor-affinity factor (FT), were introduced to obtain insight into the structure-activity relationship of nitroazole radiosensitizers. METHODS

factor, Absolute molar activity.

In vitro and in vivo assays Reduction potentials (Ey,$‘(S/S;)) of the nitroazole derivatives relative to the Ag/AgCl(saturated)/3.5-M KC1 electrode in N,N-dimethylformamide (DMF) solution were evaluated by the reported method (8, 9). EMT6 cells were used to evaluate the in vitro sensitizing effects of nitroazoles. The in vivo sensitizing effects on the EMT6 and SCCVII tumors (1 cm in diameter) in the thighs of 3-month-old female Balb/c mice and 8- to loweek-old female C3H/He mice, respectively, were determined by an in vivo-in vitro assay. Both the EMT6 and SCCVII cells showed similar response to radiation combined with nitroazole sensitizers. Further details of the in

AND MATERIALS

Compounds Representative compounds investigated are listed in Table 1. 2-Nitroimidazole derivatives (KU-2200 series), Reprint requests to: Sei-ichi Nishimoto, Ph.D., Department of Hydrocarbon Chemistry, Faculty of Engineering, Kyoto University, Kyoto 606, Japan. Supported in part by Grants-in-Aid from the Japanese Ministry of Education, Science, and Culture (6 1010070,620 10070).

Accepted for publication 21 October 1988. * Nippon Roche, Japan. t Roche Products, UK. $ Pola, Japan. 1033

1034

1. J. Radiation Oncology 0 Biology 0 Physics

April 1989, Volume 16, Number 4

Table 1. Structure, molecular weight (M), reduction potential [ERED ,,2 (V vs. Ag/Ag+ in DMF)], specific activity [A (kg”*/mg”*)], tumor-affinity factor [FT (kg/kg-tumor)], and absolute molar activity [cQ., (kg-tumor”2/mmol”2)] of various nitroazole derivatives Structure,

Compound

substituent:

R

M

ER”D I/Z

A*

FT*

%I

MIS0 AK-21 16

I -CH2CH(OH)CH20CH3 II I

201 202

-1.04 -1.06

0.040 0.029

0.58 0.25

0.74 0.83

KU-22 11 KU-231 1

I -CH2CHFCH20CHs II I

203 204

-1.06 -1.04

0.06 1 0.056

0.59 0.48

1.13 1.15

KU-23 10 KU-22 14

II I

190 189

-1.08 -1.03

0.054 0.050

0.65 0.29

0.92 1.28

KU-22 15 KU-2315

I -CH2CF2CH20H II I

207 208

-0.97 -1.03

0.04 1 0.038

0.23 0.36

1.24 0.92

RK-28 RP- 170

I I

-CH20CH2CH=CHCH20H -CH20CH(CH20H)2

213 217

-0.97 -1.01

0.040 0.04 1

0.17 0.46

I .42 0.89

RSU- 1069

I

-CH2CH(OH)CH2N3

212

-1.03

0.104

1.15

1.41

Ro 03-8799

I

-CH2CH(OH)CH2N3

291

-1.09

0.029

0.10

1.56

SR-2508 AK-2 146

I III -CH2CONHCH2CH20H I

214 215

-1.05 -1.05

0.040 0.025

0.44 0.31

0.89 0.66

KU-2285 AK-2 123

I II

-CH2CF2CONHCH2CH20H -CH$ZONHCH2CH&XZHs

264 229

-0.96 -1.08

0.050 0.03 1

0.36 0.33

1.35 0.82

KU-2388 KU-2386

II II

-CH2CHFCONHCH2CH20CH3 -CH2CF2CONHCH2CH20CH3

261 279

-1.07 -1.01

0.018 0.036

0.39 0.31

0.47 1.08

KU-2397 KU-23 17

II II

-CH2CF2CONH2 -CH~CFZCH~NHCOCH~

221 249

-1.03 -1.05

0.027 0.024

0.04 0.36

2.10 0.63

* Evaluated

-CH2CH(OH)CH2F -CH2CHFCH20H

HCl

from in vivo data, SER,,“” or CT, obtained

i.p. administration of the nitroazoles except for KU-22 I 1 (i.p.,

40 min after

30 min), RP-170 (i.v., 20 min), and SR-2508 (i.v., 20 min).

F’-

74 NO,

vitro and in vivo experiments have been reported elsewhere (7-9). Pharmacokinetic measurements Female C3H/He mice, 8-10 weeks old, weighing 1922 g and bearing SCCVII tumors of 1 cm in diameter in the thigh, were used for the pharmacokinetic study. All sensitizers were dissolved in saline immediately before use, given to the tumor-bearing mice i.p., in a volume of 0.02 cm’/g. For comparison, saline solutions of SR-2508, RP- 170 and KU-2285 were administered i.v. The mice were killed by cervical dislocation at various times 5-90 min after the sensitizer was administered. Blood was collected from the eye of the mice into heparinized tubes. Plasma samples obtained by centrifuging the blood were mixed with methanol (I:5 v/v). The tumor was dissected and the brain was removed after killing the mice: both 3A.

/N

c

I

$ Shimadzu

NO,

N4

N \

1 II

r(I \ N N ‘N’ k III

were subjected to immediate HPLC analysis. The tissue was weighed, homogenized in distilled water (10% w/v), and vertically mixed with methanol (I:2 v/v). The sample mixture was centrifuged and the supematant was analyzed by HPLC on a liquid chromatograph§ equipped with a reversed phase column (4.6 mm X 15 cm) containing octadecyl chemically bonded silicagel (ODS(t&J). The 20% methanol/O. 1-M ammonium phosphate at pH 4.5 was delivered as the mobile phase at a flow rate of 0.6 ml/min. The column eluents were monitored by the UV absorbance at 254 nm for 3-nitro- 1,2,4-triazole derivatives and 320 nm for 2-nitroimidazole derivatives. RESULTS

AND DISCUSSION

The sensitizing activity of various types of nitroazole derivatives (1 .O mM) on EMT6 single cells, as measured

Tumor affinity of nitroazoles 0 T.

by the SER in vitro (SER,,,,), increased as the one-electron RED~:S/S;)(V vs. Ag/Ag+ in DMF) reduction potential EIjZ of sensitizers became more positive and the ability of oneelectron oxidation increased, in accord with the previous results (2). Particularly, rapid increase in the SER,i,,, was observed in the EP,5D(S/S;) range of greater than -0.90 V. The sensitizing effect on the radiolytic hydroxylation of thymine to thymine glycol in deoxygenated aqueous solution was also correlated to the reduction potential (6). Comparing with N,- and N2-substituted 3-nitro- 1,2,4triazole derivatives bearing the same side-chain structures, the E?E’(S/S;) of N,-sub’stituted isomers (approximately -0.80 V) was more positive than that of N,-substituted isomers (-1.10 to -0.95 V). 5-Bromo-substituted 3nitro- 1,2,4-triazole derivatives also indicated higher EkED(S/S;) values (approximately -0.90 V). The EFED(S/S;) of fluorinated nitroazoles was the same level as, or slightly higher than that of non-fluorinated analogues. Typically, KU-22 11 bearing a side-chain group of -CH2CHFCH20CH3 at the Ni-position possessed an Ey,5D(S/S;) value of - 1.06 V, similar to MIS0 ( -CH2CH(OH)CH20CH3 group; - 1.04 V). On the other hand, KU-2285 bearing ;a -CH2CF2CONHCH2CH20H group indicated an E:zD(S/S;) value of -0.96 V, which is more positive than that of SR-2508 ( -CH2CONHCH2CH2C)H group; - 1.05 V). As a common behavior, the SER in vivo (SER,,,,) increased not linearly but with an upward convex curve upon increasing the administration dose (Ds) of various nitroazole derivatives. This observation could be represented by a linear relationship between the SER,iVo and the square root of Ds, as in equation (1); SER,i,, := 1.00 + A D&‘2

KAGIYA

1035

et al.

In

.._

0

IO

5

15

20

25

Fig. 1. Linear relationship between the SER in vivo (SER,i,,) and the square root of administered dose (Ds) of several nitroazoles: onen and closed svmbols reoresent fluorinated and nonfluorinated nitroazoles, respectively.

3oo_.

(1)

where A is defined as a specific activity in vivo (Fig. 1). Among the nitroazole derivatives shown in Figure 1, the specific activity A that is the overall activity increased in the order of AK-2123 < KU-2386 < MIS0 < KU-2285 < KU-2310 < RSU-1069 (see also Table 1). Figure 2 shows the re:sult of pharmacokinetic studies indicating that the sensitizer concentration in solid tumor (C,,) at a given time t after i.p. administration was proportional to Ds. The slope of the straight line in Figure 2 is defined as the tumor-affinity factor (FT,l); CT,~=

FTJ Ds

(2)

A trend is evident that the uptake of alcohol-type derivatives (e.g., MISO) and their fluorine modifications (e.g., KU-231 1) in solid tumors occurs more readily than amide-type derivatives such as KU-2285, AK-2123, and KU-2386 (see Fig. 2 and Table 1). When measured 20 min after iv. administration, however, KU-2285 resulted in high FT,t value of 0.64 which is about 1.5-times greater than that (0.44) of SR-2508. The highest FT.t value of 1.15 can be obtained with R:SU-1069, as reported previously

0

200

400

600

D, /w/ks~

Fig. 2. Linear dependence ofthe intratumor concentration (CT,,) of several nitroazoles, observed at a given time t (40 min in these cases) after i.p. administration, on the administered dose (Ds): open and closed symbols represent fluorinated and non-fluorinated nitroazoles, respectively.

1036

1. J. Radiation

ov* 0

n

’

’

SERvit,,

Oncology

’

’

0.5 (1.0

mhl)

0 Biology 0 Physics

’

’

’

n

1.0

1

- 1.00

Fig. 3. Relationship between the absolute molar activity ((YN) and the SER in vitro (SER,i,,,) of various nitroazoles: open and closed symbols represent fluorinated and non-fluorinated nitroazoles, respectively.

(4). Thus, the higher sensitizing effect of RSU-1069 on solid tumors is attributable not only to the absolute activity but also to the tumor-affinity factor. Equations (1) and (2) indicate that the value Of(SER,i,, - 1.OO)is proportional to the effective sensitizer concentration CT., in solid tumor: the SER,i,, was evaluated at a given time t after administration of sensitizers. We defined the corresponding proportional constant as an ab-

April 1989, Volume

16, Number

4

solute activity ((Y)of sensitizers, thus obtaining the form of LYE= A/F?: in kg-tumor”*/mg”* unit or ffM = A(M/ FT.,) 'O = aWM”* (M is the molecular weight of sensitizers) in kg-tumor”*/mmol”* unit. The numerical values of absolute molar activity (YMare summarized in Table 1. Figure 3 shows a linear relationship between the (YMand the SER in vitro (SER,i,,,) at 1 mM of sensitizers. Figure 3 also demonstrates that fluorinated nitroazole derivatives possess in general higher absolute activities than non-fluorinated derivatives. This result is accounted for partly by the increased electron affinity, as in the case of KU2285 relative to SR-2508. It is also plausible that the fluorine modification of the side chain may cause some biological action, e.g., a block effect on the metabolic process (5). The sensitizer doses required to achieve an SER of 1.5 in vivo (Ds,,.J were evaluated for a series of nitroazole derivatives by interpolation of the square-root plot as in Figure 1. Figure 4 shows the relationship between the Ds., .5 value thus evaluated and the tumor-affinity factor FT.t of sensitizers. The solid and broken curves in Figure 4 rep resent averaged isoactivity curves for fluorinated and nonfluorinated nitroazole derivatives, respectively. Evidently, the Ds., .5 of each series of sensitizers, fluorinated or nonfluorinated nitroazoles, decreases and thus the overall sensitizing activity in vivo increases as the Fr.1 becomes greater. This result demonstrates that the uptake of nitroazoles in solid tumors plays a key-function in the radiosensitization. It is also noteworthy that the fluorination of both 2nitroimidazole and 3-nitro- 1,2,4-triazole sensitizers can modify their absolute activity, as is seen from the difference in isoactivity curve. Application of the presented empirical method to the development of highly active and less toxic nitroazole sensitizers is in progress.

400-

300 2 \ i! \ ;

200-

0”

loo-

01 0

1.0

0.5

FT.t Fig. 4. Dependence of the sensitizer dose (Ds.,.s) leading to an SER,im value of 1.5 on the tumor-affinity (FT.,): the sensitizers were administered i.p. except for SR-2508 and RF 170 (i.v.).

factor

Tumor affinity of nitroazoles 0 T.

KAGIYA

et al.

1037

REFERENCES 1. Adams, G. E.; Ahmed, I.; Sheldon, P. W.; Stratford, I. J. Radiation sensitization a.nd chemopotentiation: RSU- 1069, a compound more efficient than misonidazole in vitro and in vivo. Br. J. Cancer 49571-577; 1984. 2. Adams, G. E.; Clarke, E:. D.; Flockert, I. R.; Jacobs, R. S.; Shemi, D. S.; Stratford, I. J.; Wardman, P.; Watts, M. E.; Panick, J.; Wallance, R. G.; and Smithen, C. E. Structureactivity relationships in the development of hypoxic cell radiosensitizers I. Sensitization efficiency. Int. J. Radiat. Biol. 35:133-150; 1979. 3. Brown, J. M.; Yu, N. Y,; Brown, D. M.; Lee, W. W. SR2508: A 2-nitroimidazole amide which should be superior to misonidazole as a radiosensitizer for clinical use. Int. J. Radiat. Oncol. Biol. Phys. 7:695-703; 198 1. 4. Deacon, J. M.; Holliday, S. B.; Ahmed, I.; Jenking, T. C. Experimental phannacokinetics of RSU-1069 and its analogues: high tumor/plasma ratios. Int. J. Radiat. Oncol. Biol. Phys. 12:1087-1090; 1986.

5. Filler R. Ed. Biochemistry Involving Carbon-Fluorine Bonds. Washington, D.C.: Am. Chem. Sot.; 1976. 6. Nishimoto, S.; Ide, H.; Wada, T.; Kagiya, T. Radiationinduced hydroxylation of thymine promoted by electronaffinic compounds. Int. J. Radiat. Biol. 44:585-600; 1983. 7. Shibamoto, Y.; Sakano, K.; Kimura, R.; Nishidai, T.; Nishimoto, S.; Ono, K.; Kagiya, T.; Abe, M. Radiosensitization in vitro and in vivo by 3-nitrotriazoles. Int. J. Radiat. Oncol. Biol. Phys. 12:1063-1066; 1986. 8. Shibamoto, Y.; Nishimoto, S.; Shimokawa, K.; Hisanaga, Y.; Zhou, L.; Wang, J.; Sasai, K.; Takahashi, M.; Abe, M.; Kagiya, T. Characteristics of fluorinated nitroazoles as hypoxic cell radiosensitizers. Int. J. Radiat. Oncol. Biol. Phys. (In press). 9. Shibamoto, Y.; Nishimoto, S.; Mi, F.; Sasai, K.; Kagiya, T.; Abe, M. Evaluation of various types of new hypoxic cell sensitizers using the EMT6 single cell-spheroid-solid tumor system. Int. J. Radiat. Biol. 52:347-357; 1987.