Mitomycin Concentration in Rabbit and Human Ocular Tissues after Topical Administration

Kazuhide Kawase, MD, Hiromi Matsushita, MD, T etsuya Yamamoto, MD, Y oshiaki Kitazawa, MD

The authors measured mitomycin C (MMC) concentrations in ocular tissues with high-performance liquid chromatography. Mitomycin C concentration after a single

subconjunctival injection of the drug in rabbit eyes showed a rapid decrease with a half­life of 0.18 to 0.30 hours for the conjunctiva and 0.20 to 0.45 hours for the sclera at the injection site. Irrigating the ocular surface with 200 ml of saline after MMC application reduced the initial drug concentration to one fifth in the sclera and to one fifteenth in the conjunctiva but did not change the half-life. The MMC concentration in human tra­beculectomy specimens obtained immediately after MMC application (0.2 mg/0.5 ml) and irrigation was 5.4 to 12.0 p-g/g with a mean. of 8.4 p-g/g, a level similar to that in the rabbit sclera immediately after the irrigation after administration of the same MMC dose. These results indicate that MMC disappears rapidly from the ocular tissues and that irrigating the tissues significantly reduces the tissue concentration of MMC. Ophthalmology 1992; 99:203-207

Mitomycin C (MMC), an antibiotic-antimetabolite, is known to have a potent antiproliferative effect on cultured rabbit subconjunctival fibroblasts. 1 Chen et al2 and Palmer3 have reported remarkable results using MMC as adjunctive therapy to trabeculectomy in glaucoma pa­tients. They applied MMC in the following manner: after preparation of the scleral flap, a small sponge soaked in MMC solution (0.1 to 0.4 mg/ml) was held in contact with the exposed scleral surface and the bed of the scleral flap for 5 minutes, and the entire area was irrigated with balanced salt solution. They found overall success rates

Originally received: August 12, 1991. Revision accepted: October 22, 1991.

From the Department of Ophthalmology, Gifu University School of Medicine, Gifu, Japan.

The authors have no proprietary interest in any of the products mentioned in this article.

Reprint requests to Y oshiaki Kitazawa, MD, Department of Ophthal­mology, Gifu University School of Medicine, Tsukasa-machi 40, Gifu­shi, 500, Japan.

of 91 % and 84% after a single intraoperative administra­tion of the drug and observed very few cases with resultant corneal epithelial defects.2

,3 Our prospective, randomized study in glaucoma patients with poor surgical prognosis demonstrated that 88% of the MMC-treated eyes and 47% of the 5-fluorouracil (5-FU)-treated eyes achieved an in­traocular pressure (lOP) of 20 mmHg or less without an­tiglaucoma medication.4 Therefore, MMC might induce less corneal epithelial toxicity than 5-FU, which would make repeated administration of an antiproliferative agent to provide a successful lOP control rate unnecessary. Two recent, well-controlled studies showed the efficacy of MMC in promoting filtration in rabbits.5

,6 All of these favorable results indicated that MMC might be a better adjunct to trabeculectomy than 5-FU.

To clarify the pharmacokinetics of MMC in the eye, we studied the concentration change of the drug after a single application. We also investigated in rabbits the drug concentration effect of ocular surface irrigation after MMC application-a practice generally performed intraopera­tively on patients. For purposes of comparison with con-

203

Ophthalmology Volume 99, Number 2, February 1992

centrations in rabbit tissue, we measured the MMC con­centration in tissue specimens obtained by trabeculectomy from glaucoma patients.

Materials and Methods

Determination of Mitomycin C Concentration

We determined the MMC concentration by high-perfor­mance liquid chromatography according to the following procedures. For the conjunctiva and sclera, we added 5 ml of cold acetonitrile and 100 mg of silica gel to the 200-mg specimen and homogenized it. For the aqueous hu­mor, we added 3.2 ml of ethyl acetate to 0.2 ml of the fluid and stirred it. After centrifugation at 3000 rpm for 10 minutes, the organic phase was evaporated completely at 400 C. After dissolving the residue in 1 ml of a 98:2 mixture of chloroform and methanol, sonication was per­formed. We injected 500 J.LI of this solution onto the silica gel column (Nucleosill00-S, 4.5 X 250 mm) and analyzed it using the mobile phase consisting of chloroform-meth­anol-water (90: 10:0.15) at a flow rate of 1 ml/minute. The MMC concentration was measured by ultraviolet ab­sorption at 360 nm. With this high performance liquid chromatography method, the minimum detectable MMC concentration was 5 X 10-3 J.Lg/g for conjunctiva and sclera, and 5 X 10-3 J.Lg/ml for aqueous humor. When the tissue weight was less than 200 mg, the minimum de­tectable concentration was estimated as l/w J.Lg/g, where w was the tissue weight (mg).

Subconjunctival Injection Study

To determine MMC concentration change over time after a single application in ocular tissues, we used 54 New Zealand albino rabbits of both sexes, weighing between 2.0 and 3.0 kg each. All of the animal experiments were conducted in accordance with the ARVO Resolution on the Use of Animals in Research. In both eyes of the rabbits, we subconjunctivally injected 0.5 ml of sterile distilled water containing 0.2, 0.02, or 0.002 mg ofMMC (Kyowa Hakko Kogyo Co, Tokyo, Japan) after administering top­ical anesthesia with 0.4% oxybuprocaine hydrochloride. We killed the animals with intravenous injections of 10 ml of 5% pentobarbital sodium immediately after the subconjunctival administration or 0.5, 1, 2, 3, 6, 24, 72, or 168 hours thereafter. Then we drew 100 to 200 J.LI of aqueous humor via a 26-gauge needle, and excised a sec­tion of conjunctiva measuring 15 X 15 mm and a section of sclera measuring 15 X 7 mm; each section weighed 150 to 250 mg. Four eyes of two rabbits were used at each designated time and each drug concentration. We im­mediately froze the specimens at -800 C and determined the MMC concentration using the previously described high-performance liquid chromatography technique. When three of the four specimens showed the MMC con­centration to be less than the minimum detectable level at a designated time, the MMC concentration was rated

204

as trace; when all of the specimens showed less than that level, the concentration was estimated to be nil.

Irrigation Study

To determine if irrigating the ocular surface after MMC application had an effect on drug concentration in the ocular tissues, we used 1 eye from each of 41 albino rab­bits. After administering intravenous anesthesia with 1 ml of 5% pentobarbital sodium, we exposed the sclera in the superior temporal quadrant by incising the conjunc­tiva 5 mm posterior to the limbus and applied cut sponges (Spongel, Yamanouchi Pharmaceutical Co, Tokyo, Japan) soaked in 0.5 ml MMC solution containing 0.2 mg of the drug to the exposed sclera. Five minutes later, the exposed tissues were irrigated with 200 ml of saline. We designed this experiment to simulate the procedures done during trabeculectoml where MMC was used as an adjunct. We replaced the conjunctiva in its original position and left the wound unsutured. The methods of obtaining the ocu­lar tissues and determining the MMC concentration and the timing of the measurements were the same as in the subconjunctival injection study. Four or five eyes were used each time. When three or more, but not all, speci­mens showed the MMC concentration to be less than the minimum detectable level for a designated time, the MMC concentration was rated as trace; when all showed less than that level, the concentration was estimated to be nil.

Human Trabeculectomy Specimen Study

Human trabeculectomy specimens were obtained from 13 eyes of 11 glaucoma patients at the time of surgery. Informed consent was obtained from each patient and the study protocol was approved by the Investigational Review Committee of Gifu University School of Medi­cine. The types of glaucoma were as follows: primary open-angle glaucoma (4 cases, 4 eyes), primary angle-clo­sure glaucoma (4 cases, 5 eyes), glaucoma secondary to uveitis (1 case, 1 eye), developmental glaucoma (2 cases, 3 eyes). The patients' ages ranged from 17 to 85 years. Five of the 13 eyes had already had 1 or more filtering surgeries performed before our study. The technique for trabeculectomy was essentially the same as Cairns' tech­nique. After preparing a half-layer scleral flap, we applied the sponges containing 0.2 mg of MMC dissolved in 0.5 ml of sterile distilled water to the exposed subconjunctival tissue and sclera, including the tissue under the scleral flap, for 5 minutes. Then we irrigated the tissues with 250 ml of balanced salt solution.4 Immediately after the irri­gation, a limbal tissue block containing trabecula was ex­cised, as in a routine trabeculectomy procedure. The ex­cised block was approximately 0.5 X 0.5 X 3.0 mm in size. The MMC concentration of the block was deter­mined using the high-performance liquid chromatography technique. Because the weight of each excised tissue was too small to allow a reliable measurement, we combined three or four tissue blocks to constitute one specimen. As

Kawase et al . Mitomycin Concentration in Ocular Tissues

a result, the original 13 tissue blocks yielded 4 specimens. The wet weight of the specimens was 0.60 to 1.78 mg.

Results

Subconjunctival Injection Study

The results of the subconjunctival injection study are summarized in Table 1 and Figure 1. The intraindividual differences of the measured MMC concentration were al­most identical to the interindividual differences. The MMC concentration showed a rapid decrease within 24 hours after the drug application in all tissues and aqueous humor examined. It fell below the minimum detectable level within 72 hours in the rabbits injected with 0.2 mg, and within 24 and 3 hours in those injected with 0.02 and 0.002 mg, respectively. The initial concentration detected was proportional to the applied dosage. The concentration over time was approximated to the single exponential curve of C. = Co EXP (-at) (Equation 1), where C. was the MMC concentration (p.gjg) at time t, Co was the MMC concentration at time 0, a was the time constant ofMMC disappearance (h- 1

), and t was the time after MMC ap­plication (h). Calculation using the data obtained within 2 hours after the drug application yielded the time con­stants of 2.28 to 3.92 h- 1 with the correlation coefficient (r) of -0.96 to -0.90 for the conjunctiva and 1.54 to 3.41 h- 1 with r of -0.96 to -0.81 for the sclera. The half-life was 0.18 to 0.30 hours for the conjunctiva and 0.20 to 0.45 hours for the sclera (Table 2).

The aqueous humor concentration peaked at 0.5 hours after the injection. It also decreased rapidly and fell below the detectable level within 3 hours. Since the aqueous humor concentrations in log unit showed linearity be­tween 0.5 and 2 hours after the application in the rabbits with 0.2 mg injection, a linear regression line was calcu­lated to determine the decay of MMC from the anterior chamber. This yielded a disappearance time constant of 1.70 h- 1 and a half-life of 0.41 hours (r = -0.85). In the rabbits with 0.02 and 0.002 mg injection, the concentra­tions were too low to allow a reliable calculation.

Irrigation Study

The results of this study are shown in Table 1 and Figure 2. The MMC concentration decreased rapidly within 24 hours after drug application. However, the mean initial concentrations in the conjunctiva and the sclera were one fifteenth and one fifth, respectively, of those of the rabbits in which the same dose was injected subconjunctivally. Calculation using Equation 1 yielded a disappearance time constant of 2.33 h- 1 for the conjunctiva (r = -0.96) and 2.19 h- 1 for the sclera (r = -0.87). The half-life was cal­culated to be 0.30 hours for the conjunctiva and 0.32 hours for the sclera (Table 2).

Human Trabeculectomy Specimen Study

All of the MMC concentrations in the human trabecu­lectomy specimens were above the minimum detectable level. They averaged 8.4 ± 2.8 p.g/g (mean ± standard deviation) and ranged from 5.4 to 12.0 p.g/g (n = 4).

Table 1. Mitomycin C Concentration Changes (Mean ± Standard Deviation)

MMC Hours after Administration Dosage (mg) O· 0.5 2 3 6 24 72 168

Conjunctiva 0.2 183.9 ± 38.6 40.7 ± 8.1 2.1 ± 1.5 0.1 ± 0.2 0.04 ± 0 .05 0.1 ± 0 .2 0.01 ± 0.02 Trace 0 (ltg/g) 0.02 12.5 ± 2.5 1.2 ± 0 .4 0.3 ± 0.1 0.1 ± 0.1 0 0.01 ± 0 .02 0 0 0

0.002 1.4 ± 0.8 0.4 ± 0 .3 0.1 ±0.04 0.01 ± 0.004 0 0 0 0 0 0.2 and

irrigated 12.3 ± 6.4 5.7 ± 2 .8 1.4 ±0.6 0.1 ± 0 .1 0 .1 ± 0 .1 0.03 ± 0 .04 0.02 ± 0 .02 Trace Trace

Sclera 0.2 21.8 ± 9.1 7.4 ± 2.4 0.4 ± 0.3 0.02 ±0.03 Trace 0.02 ± 0 .02 0 0 Trace (ltg/g) 0.02 1.4 ± 0.8 0.2 ± 0 .1 0.03 ±0.02 0.02 ±0.03 0 0 0 0 0

0.002 0.09 ± 0.03 0.12 ± 0 .04 0.1 ±0.2 0.002 ± 0.003 0 Trace 0 0 0 0.2.and

irrigated 4.3 ± 2.4 0.9 ± 0.4 0.14 ± 0.1 0.05 ±0.03 0.03 ± 0 .03 Trace Trace Trace 0

Aqueous 0.2 0 0.2 ± 0 .1 0.06 ±0.04 0.01 ±0.02 0 Trace Trace 0 0 humor 0.02 0 om ± 0.01 0.004 ± 0.005 Trace 0 0 0 0 0 (ltg/ml) 0.002 0 0 Trace 0 0 0 0 0 0

0.2 and irrigated -t 0.09 ± 0 .02 0.11 ± 0 .09 0.03 ±0.04 0.02 ± 0 .02 0 0 0 0

• Immediately after administration.

t MMC concentration was not measured at 0 hrs.

205

Ophthalmology Volume 99, Number 2, February 1992

3 10

10 \ \

\ 10"- 1\ \

i~

10-3 \"

- Conjunctiva

.. ----... Sclera

~ Aqueous humor

10.

2 J \ /' , "

trace.J ' ... ''-/-----\\_.>.-_----orO

rl ""--,--r-----,--\~\~'-----.'----" o 0.5 1 24 72 168

A Time after MMC Administration (h)

B

::1 ~ '" :1-

C5 -g: .a. c 0

~ "i: .. u c: 0 U

U ::;; ::;;

3 10

102

10

10"

10.2

10-3

trace 5 orO

c

I I I

00.51

-- Conjunctiva

......... Sclera

~ Aqueous humor

.--.-..... ------\~-~-----. \\--.,-------,r----,

24 72 168

Time after MMC Administration (h)

-- Conjunctiva ......... Sclera

---. Aqueous humor

_ ...... -..... _ .. _._.\\ .. _ .. -.---_ .. -.---1 1 I 1 II I I I

o 0.5 1 3 6 24 72 168

Time after MMC Administration (h)

Figure 1. Mitomycin C concentration after a single subconjunctival in­jection. The unit of MMC concentration is Jlg/g for conjunctiva and sclera and Jlg/ml for aqueous humor. Dosage of MMC injected is: A, 0.2 mg; B, 0.02 mg; C, 0.002 mg.

206

Discussion

Our single subconjunctival injection study clearly showed rapid disappearance of MMC from the injected and ad­jacent tissues and also from the anterior chamber, with its half-life in the conjunctiva, the sclera, and the aqueous humor ranging from 0.18 to 0.45 hours. Although we used both eyes of each rabbit subconjunctivally in the injection study, it seems unlikely that the study design has affected the results, because the intraindividual dif­ferences of the MMC concentration were almost identical to the interindividual differences. To compare the half­life of MMC with that of 5-FU, we used Kondo and Araie's7 rabbit data derived within 5 hours after a single subconjunctival injection: the half-life of 5-FU was esti­mated to be 0.38 hours in the conjunctiva at the injected site and 0.43 hours at the corresponding site in the sclera. Thus, the half-life of 5-FU in ocular tissues is quite similar to that ofMMC noted in our study. The half-life ofMMC in ocular tissues is similar to the serum half-life after in­travenous administration in humans and animals, cal­culated on the basis of a two-compartment open model (0.55 to 0.84 hours8,9 and 0.16 to 0.59 hours,IO-12 respec­tively). The results suggest that there is little possibility that the concentration change of MMC occurs more slowly than does that of 5-FU in the ocular tissues and that the pharmacokinetics of MMC in the eye is similar to that in the serum. As a result of this rapid disappear­ance, the MMC concentration in ocular tissues was found to be less than 10-2 f,lg/g within 24 hours after a single 0.2-mg subconjunctival injection of the drug. At 72 hours, it was below the minimum detectable level of 5 X 10-3

f,lg/g, which was demonstrated by Yamamoto et all to induce 75% inhibition of the growth of cultured rabbit subconjunctival fibroblasts. The same authors also re­ported 50% inhibition at an MMC concentration of 2 X 10-3 f,lg/ml. Although their 1Dso value is lower than the minimum detectable concentration in our in vivo study and the tissue affinity ofMMC is probably different among the various ocular tissues, the disappearance constant we observed seems to strongly indicate that the MMC con­centrations in the conjunctiva and the sclera, the tissues involved with the filtering bleb formation, are not likely to be maintained above the IDso value longer than a few days.

Chen et al2 and Palmer3 irrigated the MMC-applied ocular tissues intraoperatively. Our irrigation study sim­ulated their procedure to determine the effect of irrigation on tissue concentrations of the drug. We found that the irrigation 5 minutes after MMC application reduced the initial tissue concentration to one fifth to one fifteenth, but did not affect its half-life. These results, however, do not necessarily mean that the use ofMMC in the reduced dosage can obviate the need to irrigate the ocular surface to avoid possible corneal complications, as practiced in clinical settings.2-4 Corneal epithelial defects are the most common complications of topical 5-FU administration,13 but little is known about the relationship between the dose of MMC and this adverse corneal effect.

Kawase et al . Mitomycin Concentration in Ocular Tissues

Table 2. Initial Concentration, Disappearance Time Constant, and Half-life*

Dose of Conjunctiva Sclera

MMC Initial Time Initial Time Applied Concentration t Constant Half-life Concentration t Constant Half-life

Study (mg) (p.g/g) (h-l) (h)

Subconjunctival 0.2 158.0 3.92 injection 0.02 5.76 2.33

0.002 1.05 2.28

Irrigation 0.2 12.6 2.33 MMC = mitomycin C; r = correlation coefficient .

• Calculated on the basis of Equation 1.

t MMC concentration at 0 hrs.

Our results also demonstrated that in human trabec­ulectomy cases, after irrigating the ocular surface, the MMC concentration in the limbal block at the adminis­tered site was identical to that of the rabbit sclera in the irrigation study. Although we cannot extrapolate our an­imal data to human trabeculectomy cases, it seems prob­able that concentrations of intraoperatively administered MMC decrease rapidly, even in the tissues where MMC has been applied, and that they fall below the minimum effective antiproliferative concentration within a few days.JO-121t appears, therefore, from the reports ofMMC's favorable antiproliferative effect as an adjunct to glaucoma filtering surgery,2-4 that this effect can last much longer in clinical situations than tissue concentrations would suggest. Moreover, tissue levels do not fully support the well-accepted concept that the first 14 days after surgery form a critical period for fibroblast proliferation and that continuous use of an antiproliferative agent such as 5-FU during those days is desirable. 13.14 Further investigations are needed to determine the optimum dosing schedule of antiproliferative agents to obtain functioning filtering blebs.

::J ~ 10· ::I.

o 10

~ 3 c o ! 'E GO ... C o u u :! :!

trace J or 0

- Conjunctiva ......... Sclera

~ Aqueous humor

rl ""--.--,--------,-\\~-,----,_--_, 00.51 24 72 168

Time after MMC Administration (h)

Figure 2. Mitomycin C concentration after irrigation of ocular surface 5 minutes after the application of 0.2 mg of MMC.

0.18 0.30 0.31

0.30

(p.g/g) (h-l) (h) r

-0.96 20.0 3.41 0.20 -0.92 -0.90 0.62 1.82 0.38 -0.81 -0.96 0.12 1.54 0.45 -0.89

-0.96 3.16 2.19 0.32 -0.87

References

1. Yamamoto T, Varani J, Soong HK, Lichter PRo Effects of 5-fluorouracil and mitomycin C on cultured rabbit subcon­junctival fibroblasts. Ophthalmology 1990; 97: 1204-10.

2. Chen C-W, Huang H-T, Shen M-M. Enhancement of lOP control effect of trabeculectomy by local application of an­ticancer drug. In: ACTA XXV Concilium Ophthalmolo­gicum (Rome), 1986. Vol. 2; 1487-91.

3. Palmer SS. Mitomycin as an adjunct chemotherapy with trabeculectomy. Ophthalmology 1991; 98:317-21.

4. Kitazawa Y, Kawase K, Matsushita H, Minobe M. Trabec­ulectomy with mitomycin C: a comparative study with 5-fluorouracil. Arch Ophthalmol [in press].

5. Wilson MR, Lee DA, Baker RS, et al. The effects of topical mitomycin on glaucoma filtration surgery in rabbits. J Ocul Pharmacol1991; 7:1-8.

6. Charles J-B, Ganthier R Jr, Wilson MR, et al. Use of bio­erodible polymers impregnated with mitomycin in glaucoma filtration surgery in rabbits. Ophthalmology 1991; 98: 503-8.

7. Kondo M, Araie M. Concentration change of fluorouracil in the external segment of the eye after subconjunctival in­jection. Arch Ophthalmol 1988; 106: 1718-21.

8. Fujita H. Pharmacokinetics of mitomycin C and its deriv­ative. (KW-2083). Gan To Kagaku Ryoho 1982; 9:1362-73.

9. Reich SD. Clinical pharmacology of mitomycin C. In: Carter SK, Crooke ST, eds. Mitomycin C: Current Status and New Developments. New York: Academic Press, 1979; 243-50.

10. Okumura S, Deguchi T, Nakamizo N. Studies on the phys­iological deposition and pharmacokinetics of 7-N-(p-hy­droxyphenyl)-mitomycin C. Jpn J Antibiot 1982; 35:1967-76.

11. Imai R, Morimoto M, Marumo H. Distribution and excre­tion of 7-N-(p-hydroxyphenyl)-mitomycin C in normal mice. Gann 1982; 73:675-80.

12. Engelmann U, Burger R, Jacobi GH. Experimental inves­tigations on the absorption of irttravesically instilled mito­mycin C in the urinary bladder of the rat. Eur Urol 1982; 8: 176-81.

13. The Fluorouracil Filtering Surgery Study Group. Fluoro­uracil Filtering Surgery Study one-year follow-up. Am J Ophthalmol 1989; 108:625-35.

14. Desjardins DC, Parrish RK II, Folberg R, et al. Wound healing after filtering surgery in owl monkeys. Arch Ophthalmol1986; 104:1835-9.

207