•

Phase 1 Study of High-Dose Metronidazole: ASpecific In Vivo and In Vitro Radiosensitizer of Hypoxic Cells 1

•

Therapeutic Radiology

Raul C. Urtasun, M.D., J. Donald Chapman, Ph.D.,2 Pierre Band, M.D., Harvey R. Rabin, M.D., Carol G. Fryer, M.Sc., and Jack Sturmwind, B.Sc. Metronidazole was administered orally to 25 patients and its maximum concentration in blood and tumor tissues, its ability to cross the blood/brain barrier and concentrate in the cerebrospinal fluid and brain tumor tissue, its immediate and long-term toxicity, and its enhancement of irradiation in normal tissue were studied. Maximum blood concentrations of 700-1,200 J,!M (120-220 J,!g/ml) were obtained at four hours with doses of 6 g/m 2 . Moderate and transient nausea and vomiting were the only immediate signs of toxicity. No long-term toxicity was found up to 18 months after administration of the drug. These data indicate that metronidazole can feasibly be administered in clinical trials of fractionated radiotherapy using dosages ranging from 9.5 to 11 9 three times a week for three to four weeks. INDEX TERMS: Hypoxia. Metronidazole • Neoplasms, therapy

•

C

Radiology 117:129-133, October 1975

HRONICALLY HYPOXIC cells are present in solid tu-

mors in man and other animals (9, 14, 15, 19). Whether Or not the radioresistance of such cells is a limiting factor in the local control of some solid tumors treated with fractioned radiotherapy is a subject of current debate; if they are, two different techniques currently under investigation could potentially lead to improved results. High-LET irradiation provides a lower oxygen enhancement ratio for cells irradiated in vitro than the low-LET radiations currently being used and consequently could prove to be more effective in the treatment of solid tumors. Alternatively, the combination of chemical radiosensitizers which are selective for hypoxic cells with low-LET irradiation might reduce the oxygen enhancement ratio of tumor cells to that observed with high-LET irradiation. The development of chemical radiosensitizers for application in radiotherapy has been reviewed (1, 8). Recently, several drugs used clinically as antibacterial and antiprotozoal agents have been found to have radiosensitizing properties against hypoxic cells in vitro (6, 7, 11, 16, 17). Of the drugs tested to date, those of the nitroimidazole class appear to hold promise for clinical application in radiotherapy because of their superior pharmacological properties in animal models (2). Metronidazole has been shown to be an active radiosensitizer of hypoxic cells in vitro (6, 11) and in vivo (4, 5, 10, 12, 16, 18) and was chosen for this study because of its apparent low toxicity at low doses when used as an antiprotozoal agent in humans. Preliminary pharmacological studies with doses of up to 2.5 g/m 2 have been reported (20). In this communication, pharmacological studies with doses of up to 6 g/m 2 are described, as weil as observations of long-term toxicity and immediate

• and short-term effects in irradiated normal tissues of patients undergoing radiation therapy. Studies of the radiosensitizing effect of metronidazole on solid human tumors are currently in progress and will be reported elsewhere when completed. MATERIALS AND METHODS

Observations on the immediate and long-term toxicity, normal-tissue radiation enhancement effect, and daily maximum blood concentrations were derived from further follow-up of a previously described group of patients who received metronidazole at a dose of 2.5 g/m 2 (20) and from ongoing prospective randomized pilot studies aimed at assessing the radiosensitizing properties of metronidazole at a dose of 6 g/m 2 in human solid tumors.

Patient Material Group I: Untreated patients with squamous-cell bronchogenic carcinoma limited to the lung and mediastinum were randomized to receive either irradiation alone or irradiation plus metronidazole. A total tumor dose of 2,880 rads from the 60Co source was given in six fractions (twice a week) over a period of three weeks (1,360 rets), using parallel opposed fields including the mediastinum. Group 1/: Patients with high-grade supratentorial astrocytoma were also randomized two weeks after biopsy and given irradiation alone or irradiation plus metronidazole. A total tumor dose of 3,000 rads from the 60Co source was given in nine fractions (three times a week) over a period of three weeks (1,288 rets), using

1 From the Departments of Therapeutic Radiology (R. C. U., C. F., J. S.) and Medicine (P. B., H. R.), University of Alberta and Cross Cancer Institute, Edmonton, Alberta, Canada. Accepted for publication in April 1975. Supported in part by a research grant from the Provincial Cancer Hospitals Board of Alberta. 2 Medical Biophysics Branch, Atomic Energy of Canada Ltd., Pinawa, Manitoba, Canada. sjh

129

RAUL C. URTASUN AND OTHERS

130

parallel opposed fields including 2/3 of the brain. Patients in both groups received 6 g/m 2 of metronidazole four hours before irradiation. Ten patients qualified for long-term drug toxicity evaluation over periods from 6 to 18 months after drug ingestion; these included 4 patients receiving 2.5 g/m2 daily for two to four weeks and 6 patients receiving 6 g/m 2 three times a week for three weeks. Prior to entering the study, these patients had normal base-line liver and kidney functions and normal peripheral hematological values, including blood determinations of alkaline phosphatase, serum glutamic oxaloacetic transaminase, bilirubin, blood urea nitrogen, creatinine, uric acid, total and differential white blood cell counts, platelet count, hemoglobin, and hematocrit. All of these studies were done twice a week during administration of metronidazole and at monthly intervals thereafter for up to one year. Urinalysis was done with the same frequency. Nine patients with different types of tumors were studied with regard to 22-hour blood concentration curves and tumor tissue and cerebrospinal fluid concentration of the drug.

Methods Metronidazole (Flagyl) was administered orally in the form of 250-mg tablets which were crushed and diluted in a glass of orange juice. Patients fasted for seven hours prior to drug administration. B/ood Concentration Studies: Blood was obtained hourly for 15-26 hours by means of an indwelling venous catheter in 4 patients given 2.5 g/m 2 (20) and in 2 patients given 6 g/m 2 . In patients receiving irradiation plus metronidazole, routine blood determinations were made on the treatment days at the time of irradiation (four hours after drug ingestion) and at six hours after drug ingestion. The concentration of metronidazole in the blood was determined by spectrophotometric assay as described previously (20). The solids in heparinized whole blood were precipitated by mixing it with ethanol in a ratio of 1:9. The mixture was allowed to stand for 10 minutes and then centrifuged at 2,500 g for 10 minutes. The optical density (00) of the supernatant was measured at 314 nm, and that of a control blood sample treated in an identical manner was subtracted from it. A standard calibration curve was constructed using the blood of. several patients taken prior to the drug regimen. This assay was suitable for measuring blood levels of 50-2,000 ~M with a confidence interval of ±50 ~M in the higher ranges (800-1 ,900 ~M) and ±30 in the lower ranges. Tumor Tissue and CSF Concentration Studies: Patients undergoing biopsy or excision of tumor were given a single dose of 5-6 g of metronidazole six to seven hours before surgery. Blood was obtained before drug ingestion and at the time of tumor excision. Metastatic subcutaneous nodules were removed four hours after ingestion of metronidazole and the concentrations

October1975

compared with those in blood drawn from the same patient at the time of the biopsy. Tissue was obtained from the center of tumors measuring more than 3 cm, and contaminating blood was removed from smaller samples as much as possible before assaying. Tumor tissue samples were minced with scissors and diluted in ethanol (1 g/9 ml). Samples were allowed to stand for 24 hours before centrifugation at 2,500 g for 10 minutes. The 00 of the supernatant was measured at 314 nm and a control value established from surgical specimens was subtracted from it. The calibration curve for blood was then used to determine the concentration. For cerebrospinal fluid drug concentration studies, patients received metronidazole four to six hours before lumbar puncture. Blood was obtained before drug ingestion and at the time of the spinal tap. The technique for measuring drug concentration in the cerebrospinal fluid was identical to that used for blood concentrations except that fluid from normal individuals who had not taken the drug was used as a control. Assessment of Norma/-Tissue Radiation Enhancement Effect: The groups of patients receiving irradiation to either the brain or lung and mediastinum were randomized as per the above protocol-irradiation alone or irradiation plus the drug-with each group receiving the same amount of irradiation to a field of the same size over the same period of time. In those whose brain was irradiated, the degree and timing of allopecia were noted during the course of irradiation and the time of hair regrowth was estimated in the course of monthly follow-up examinations. In those who received irradiation to the lung, anteroposterior and lateral chest tomograms were obtained at monthly intervals for up to one year after irradiation and every three to six months thereafter. The timing and degree of radiation lung fibrosis were noted. Radiation dermatitis and subcutaneous reactions were assessed during treatment and at monthly intervals thereafter for up to one year. Neurological examinations were carried out before treatment and at monthly intervals thereafter for up to one year in 10 patients in order to assess evidence of transient radiation myelitis. These patients will continue to be assessed for late radiation changes every three to six months after the first year for as long as survival permits. RESULTS

Metronidazo/e Concentration in the B/ood With single doses of 6 g/m 2 given after an overnight fast, blood concentrations of approximately 800-1,000 ~M were obtained four to six hours after drug ingestion in 2 patients (Fig. 1). With dosages of 6 g/m 2 administered after an overnight fast two to three times a week, a mean blood concentration of 980 ~M (6-9 determinations per patient) was obtained at four hours in 5 patients (range, 642-1 ,290 ~M). At 48 hours after the ingestion of metronidazole, the blood concentration was

PHASE 1 STUDY OF HIGH-DOSE METRONIDAZOLE

Vol. 117

Table I:

Time After Drug Ingestion (hr.)

Surface Area (rn")

Dose (g)

Source

Y.M.

1.5

6

F. B.

2.0

5

P. K.

1.6

5

D. M.

1.2

4

Blood Subcutaneous tumor (center of tumor) Blood Blood CSF Blood Brain tumor tissue Blood Subcutaneous tumor (center of tumor)

~

I&l

Metronidazole Concentration (MM)

4 4

695 640

6 5 5 6 6 4.5 4.5

610 400 305 465 430 905 840

2:

.!-

~900

• ••

e 700

z

o

•

~ 500 ~

LLI :::l ...J-

21000 800

g

•• •

~a::

....o 300

600

I-

~ 400

~ 200

5 0::

&I.

t-

ffi

Therapeutic Radiology

Metronidazole Concentrations in Tumor Tissue, Cerebrospinal Fluid, and Blood

Patient

5

131

100+-...,...-,---'T--,:--r----r-...,...-,--.,.--.-,.-..,.........

U

~

o

2 4 10gm.'

6

8

10 12 14 16 18 20 22 24 26

HOURS AFTER DRUG ADM.

o 0: 200 IZ LLI

o Z

o

!

o

~900

j700

~500

.,

~

"0 30 0

.

•

••

• •

•

•

~200

E .,c

1001-J---r-----.,.---,--0 I 2 3 HOURS AFTER DRUG ADM.

Fig. 2. Blood concentrations of metronidazole in the same patient at different times during the first three hours following drug administration repeated on at least three different days. Note that the variability of concentration values is less pronounced close to three hours after drug ingestion.

g 100...................................---..--.-r--......-........-.--.........-..........,;

B

A 2 4 6 8 ~ ~ ~ ~ ~ OOU M 10gm. HOURS AFTER DRUG ADM.

Fig. 1. Blood concentrations of metronidazole following a single oral dose of 5.5-6 g/m 2 in 2 patients (* F. B., t:.. W. C.).

only slightly above control levels (approximately 30 J.LM). A slight trend toward an increase in peak concentration was observed during each of the three weeks making up the course of treatment (approximately 50- 100 J.LM). In one patient given two consecutive daily doses of 6 g/m 2 , the peak blood concentration rose from 1,100 MM in the first day to 1,700 J.LM in the second day. Blood concentrations of the drug determined one, two, and three hours after drug ingestion in the same patient on several days showed some variation, which was least pronounced at three hours (Fig. 2).

Concentrations of Metronidazole in Solid Tumors and Cerebrospinal Fluid Drug concentration within subcutaneous metastases from bronchogenic and esophageal squamous-cell carcinoma and measuring more than 3 em was assessed in 2 patients. On removing the lesions in toto, tissue

was dissected from the center of the tumors and assayed. No blood was seen, and we assume that no gross blood contamination occurred. The drug concentration in the center of the tumor was nearly equal to the concentrations of blood samples obtained when the tumor was removed (TABLE I). Drug concentrations in the cerebrospinal fluid and brain tumor tissue were assessed in 2 patients and found to be close to that in the blood five and six hours after drug ingestion (TABLE I). There was no apparent CSF contamination of the brain tissue specimen. Immediate Toxicity: The only immediate signs of toxicity were transient moderate nausea and vomiting. Only one patient exhibited vomiting so severe that it was not possible to continue the study. All patients received anti-emetic medication (Stemetil, 10 mg intramuscularly) half an hour before ingesting the metronidazole. Long-term toxicity: Ten patients were observed for periods ranging from six to 18 months after drug administration. Six received 6 g/m 2 of metronidazole daily for two to four weeks. No cerebellar or cerebral dysfunction was found, and there was no evidence of peripheral neuropathy. Normal- Tissue Radiation Enhancement Effect:

RAUL C. URTASUN AND OTHERS

132

900

...

Mr. J.R.

700

-

~500

900 700

Mr. F.W.

October 1975

,.

500

=L

.g°300

300

-6200 c:

200

g

LLJ ...J 100

2

~

0 4 8 12 16 20 24 A3.5gm. HOURS AFTER DRUG ADM. ~.Ogm.

100+--'T"'"""'-..----.---..-......,..---..0 4 8 12 16 20 24 A4.0 gm. HOURS AFTER DRUG ADM. 4 4 .0 gm.

Z900

~700 t-

Mrs. E. J.

Blood Concentration of METRONIDAZOLE Following on Oral Dose of 2.5gm./m 2 in 3 Patients.

~500

u,

o

•••

Z300

~

•••

~200 tZ LLJ

~ 100+--..,r----r---..---r--~_..,...

o

U

0

4

4 8 ~ ~ 20 HOURS AFTER DRUG ADM. 4 .0 gm.

M

Fig. 3. Blood concentrations of metronidazole following an oral dose of 2.5 g/m 2 in 3 different patients. [Reprinted from Urtasun et al. (20) by permission of the publisher]

Seven patients with carcinoma of the lung were observed from six months to one year; 5 received irradiation alone and 2 were also given 6 g/m 2 of metronidazole with each radiation fraction. The degree of late radiation lung fibrosis in the treated areas was the same in both groups. Although the spinal cord was included in the radiation field to a length of 14-18 em, there was no evidence of transient radiation myelitis in either group during the period of observation (up to one year). No overt skin or subcutaneous radiation reaction was observed in the drug-treated group. The patients who received a low dose (2.5 gl m2 ) were followed up for a period of up to 18 months; these included 4 patients given irradiation plus 2.5 g/m 2 of metronidazole and 3 patients given irradiation alone using the same dose fractionation schedule, volume, and overall time. Both groups showed the same degree of lung fibrosis or subcutaneous reaction, and no radiation myelitis was seen. Fourteen patients were irradiated to the head; these included 6 patients given irradiation alone and 8 patients given 6 g/m 2 of metronidazole as well. No difference in the degree and timing of allopecia was found, and the time of hair regrowth was the same in both groups. DISCUSSION

Previous pharmacological studies with metronidazole have involved dosages of 2.5 g/m 2 (20). Since no obvious immediate toxicity was found and higher blood concentrations of the drug in blood are desirable to potentiate its radiosensitizing effect, further pharmacological studies at higher dosages were felt to be necessary. When dosages of approximately 6 g/m 2 are given, the highest blood concentrations appear to occur between four and six hours after administration, indicating a shift

in the maximum concentration compared to the lower dose used previously (Figs. 1 and 3). Our observations also show more variable drug concentration values during the first two hours after drug ingestion. Both of these findings are probably caused by the large bulk of material ingested (36-40 250-mg tablets) producing a variation in the rates of dissolution and absorption. This effect was not as pronounced when a lower dose was administered (2.5 g/m 2 , or approximately 15-20 tablets). The close time relationship between the drug concentration in three solid tumors and that in the blood suggests a short lag-time between maximum blood and tumor tissue concentrations. The broad peaks of the 24-hour blood concentration curves (Fig. 1) indicate that the concentration of the drug is relatively constant from four to six hours after drug ingestion. Assuming that there is no significant lag-time from blood to tumor tissue, this period would provide a convenient interval during which irradiation could be administered. In addition, the observed concentrations of the drug in the cerebrospinal fluid and brain tumor tissue suggest that there is an effective passage of the drug across the bloodlbrain barrier. When administering the drug two to three times per week (an interval of at least 48 hours between each drug administration), there was no significant drug accumulation in the blood. Using this schedule, a high drug concentration was achieved at the time of irradiation (four to six hours after drug ingestion) and the drug was almost entirely eliminated from the blood by 48 hours. The blood concentrations in the patient who received two consecutive daily doses suggest that when the drug is administered on consecutive days there is a significant drug accumulation from day to day. Although moderate and transient nausea and vomit-

Vol. 117

PHASE

1 STUDY OF HIGH-DOSE METRONIDAZOLE

ing continue to be the immediate manifestations of drug toxicity, these were readily controlled in most of our patients by routinely administering an anti-emetic medication half an hour before giving the drug. No long-term drug toxicity was observed. None of the patients treated with metronidazole plus irradiation demonstrated any obvious immediate or short-term increase in undesirable radiation effects on the skin, subcutaneous tissue, spinal cord, or lung, though we recognize that 18 months is a relatively short period of observation: surviving patients will be observed regularly for a period of up to five years. Thus we conclude that blood concentrations of approximately 1 mM of metronidazole did not enhance the effect of irradiation on normal tissue under euoxic conditions over periods of up to 18 months. It remains to be determined whether the dose of metronidazole used in this investigation is high enough to produce radiosensitization of hypoxic cells in human solid tumors. Stone and Withers reported that the radiation response was enhanced by a factor of 1.2 in hypoxic tumor cells in murine mammary carcinoma using 100 mglkg of metronidazole (18). Dosages of 6 g/m 2 in humans (10-11 g in an average man) correspond to approximately 150 mglkg, which compares favorably with these authors' work. Furthermore, if extrapolation purelyon a weight ratio is attempted in order to apply the mouse data to man, a conversion factor based on the difference in surface area from mouse to man should probably be used, as reported elsewhere (13). Thus it appears that metronidazole can feasibly be given orally at dosages ranging from 9.5 to 11 g every 48 to 72 hours, producing peak blood concentrations of 1.0-1.2 mM (180-200 ,ug/ml) in clinical trials of fractionated radiotherapy over a period of three to four weeks, in order to assess its radiosensitizing properties in solid tumors. It is hoped that the information gained from this study will also prove to be useful in optimizing radiation regimens to be used with other active hypoxic. cell radiosensitizers of the nitroimidazole group (3). ACKNOWLEDGMENT: We wish to thank Poulenc, Inc., Montreal for supplying the metronidazole used in this study. Department of Radiology Division of Therapeutic Radiology University of Alberta 11560 University Ave. Edmonton. Alberta, Canada T6G 1Z2

133

Therapeutic Radiology

REFERENCES 1. Adams GE: Chemical radiosensitization of hypoxic cells. Br Med Bull 29:48-53, Jan 1973 2. Asquith JC. Foster JL. Willson RL, et al: Metronidazole ("Flagyl"). A radiosensitizer of hypoxic cells. Br J Radiol 47:474481, Aug 1974 3. Asquith JC. Watts ME. Patel K, et al: Electron affinic sensitization. V. Radiosensitization of hypoxic bacteria and mammalian cells in vitro by some nitroimidazoles and nitropyrazoles. Radiat Res 60:108-118, Oct 1974 4. Begg AC: Demonstration of hypoxic cell sensitization in solid tumors using 1251UdR (abst). Radiat Res 59:144, Jul1974 5. Begg AC, Sheldon PW, Foster JL: Demonstration of radiosensitization of hypoxic cells in solid tumours by metronidazole. Br J RadioI47:399-404. Jul1974 6. Chapman JD, Reuvers AP. Borsa J: Effectiveness of nitrofuran derivatives in sensitizing hypoxic mammalian cells to x rays. Br J Radiol 46:623-630, Aug 1973 7. Chapman JD, Reuvers AP. Borsa J, et al: Nitrofurans as radiosensitizers of hypoxic mammalian cells. Cancer Res 32:26162624. Dec 1972 8. Chapman JD, Reuvers AP, Borsa J. et al: Nitroheterocyclic drugs as selective radiosensitizers of hypoxic mammalian cells. Cancer Chemother Rep 58:559-570, Jul-Aug 1974 9. Churchill-Davidson I: The oxygen effect in radiotherapy. [In] Raven RW. ed: Cancer Progress Volume. London. Butterworths, 1960, pp 164-179 10. Denekamp J 1 Michael BD. Harris SR: Hypoxic cell radiosensitizers: comparative test of some electron affinic compounds using epidermal cell survival in vivo. Radiat Res 60: 119-132. Oct 1974 11. Foster JL, Willson RL: Radiosensitization of anoxic cells by metronidazole. Br J RadioI46:234-235. Mar 1973 12. Fowler JF, Sheldon PW. Foster LJ: Radiosensitisation of hypoxic cells in solid tumours in air-breathing C3H mice (abst). Radiat Res 59:141-142, Jul1974 13. Freireich EJ, Gahan EA, Rail DP, et al: Quantitative comparison of toxicity of anticancer agents in mouse. rat. hamster. dog, monkey. and man. Cancer Chemother Rep 50:219-244. May 1966 14. Hill RP. Bush RS. Yeung P: The effect of anaemia on the fraction of hypoxic cells in an experimental tumour. Br J Radiol 44: 299-304, Apr 1971 15. Powers WE, Tolmach LJ: A multicomponent x-ray survival curve for mouse lymphosarcoma cells irradiated in vivo. Nature 197:710-711, 16 Feb 1963 16. Rauth AM: In vivo testing of hypoxic cell radiosensitizers (abst). Radiat Res 59: 165, Jul 1974 17. Reuvers AP, Chapman JD. Borsa J: Potential use of nitrofurans in radiotherapy. Nature 237:402-403, 16 Jun 1972 18. Stone HB, Withers HR: Tumor and normal tissue response to metronidazole and irradiation in mice. Radiology 113:441-444, Nov 1974 19. Thomlinson RH, Gray LH: The histological structure of some human lung cancers and the possible implications for radiotherapy. Br J Cancer 9:539-549, Dec 1955 20. Urtasun RC. Sturmwind J, Rabin H. et al: "High-Dose" metronidazole: a preliminary pharmacological study prior to its investigational use in clinical radiotherapy trials. Letter to the editor. Br J RadioI47:297-299, May 1974