Doxil-Induced Chemomyectomy: Effectiveness for Permanent Removal of Orbicularis Oculi Muscle in Monkey Eyelid

Abstract

purpose. To test the safety and effectiveness of Doxil chemomyectomy in monkey eyelids using treatment schedules and doses similar to those proposed for the human blepharospasm patients in Phase I and II trials.

methods. Bupivacaine/hyaluronidase and Doxil were injected sequentially into the eyelids of five Cynomolgus monkeys. Eyelids received 1, 2, or 3 sets of injections. The monkeys were euthanatized at a minimum of 2 months and a maximum of 12 months after the final treatment. The eyelids were prepared for histologic examination, and muscle fiber loss was quantified.

results. All Doxil injections resulted in a significant loss of myofibers. No bruising, ulceration, or other skin injuries occurred, even after a third injection regimen within a single treated eyelid. Two-day preinjury with a bupivacaine/hyaluronidase mixture had a significant adjuvant effect.

conclusions. Doxil chemomyectomy is an effective protocol to permanently remove muscle from injected eyelids in nonhuman primates. Serial injections over the course of several months using the preinjury protocol combined with Doxil treatment significantly increased Doxil’s myotoxic effects. Additionally, the injection of the liposome-encapsulated form of doxorubicin did not result in skin injury or ulceration. Species differences demonstrated the importance of testing these drugs in nonhuman primates. Thus, repeated doses of Doxil may prove to be as clinically effective as free doxorubicin injections in reducing muscle spasms in blepharospasm patients but with increased safety to the skin and tissue around the injection site.

Currently doxorubicin chemomyectomy is the only nonsurgical, permanent treatment option for blepharospasm and hemifacial spasm patients. ¹ Repeated injections of 1 to 1.5 mg free doxorubicin have proven to be permanent and extremely effective for reducing both muscle volume and muscle spasms in patients. ² However, the eyelids were sore to the touch after the doxorubicin injections. In eyelids that received two or three injections, the chance of developing skin ulceration and scabbing over the injection site increased. Although these side effects slowly resolved in both the laboratory studies in rabbits and in the human patients, they were a deterrent to patient and physician acceptance of the procedure. 

In an effort to develop a protocol that would reduce or eliminate the risk of skin injury and hyperalgesia over the injection site, we examined the efficacy of using doxorubicin in its liposome-encapsulated form in rabbit eyelid. ³ Injection of liposome-encapsulated doxorubicin resulted in significant muscle loss after treatment. No skin injury was seen in any of the treated rabbits. This represented a significant improvement over the use of free doxorubicin. Previous studies also demonstrated that preinjury of the orbicularis oculi muscle with bupivacaine resulted in an amplification of the myotoxic effect of subsequent doxorubicin or Doxil injections. ^{3 4 5} Thus, these studies showed that pretreatment with local anesthetic would be desirable in Doxil injection protocols to ensure the maximum chemomyectomy effect of each injection series. 

In human patients, the single-dose limit for free doxorubicin treatment was lower than was used in the rabbit dose-ranging experiments. ¹ Effective treatment would necessitate multiple injections over time. These observations were incorporated into the protocol for a Phase I dose-ranging trial for the use of liposome-encapsulated doxorubicin in blepharospasm and hemifacial spasm patients. In an effort to determine potential efficacy and possible negative sequellae of this multiple-injection protocol before treatment of patients, the present study in nonhuman primates was initiated. Five monkeys received injections of Doxil, a commercially available liposome-encapsulated form of doxorubicin, following the same treatment schedule and doses proposed for the human blepharospasm patients. After each set of eyelid injections, the monkeys were examined daily for up to 1 month for inflammatory changes or changes in the skin over the injection site. Previous studies demonstrated that the chemomyectomy effect of both doxorubicin and Doxil on skeletal muscle is both rapid ⁶ and permanent. ² At a minimum of 2 months and a maximum of 12 months after the termination of the full sequence of treatments, the orbicularis oculi muscles of these monkeys were examined for permanent muscle loss using histologic and morphometric procedures. 

Materials and Methods

All animal studies were approved by the University of Minnesota Institutional Animal Care and Use Committee. Animal studies were performed in accordance to the published guidelines of the NIH on the use of animals in research and with adherence to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Five Cynomolgus monkeys were treated with Doxil (Sequus Pharmaceuticals, Menlo Park, CA) using the following set of injection parameters (Table 1) . Doxil was injected at a dose of 0.5, 1, or 2 mg in a volume of 1 ml saline into the eyelid region. As in our other studies, ⁷ care was taken to ensure that the injection covered the length of the eyelid from medial to lateral canthus. The injection was placed within the palpebral region of the muscle. Eight lids were subjected to one course of treatment, and 12 eyelids were subjected to three courses of treatment. Upper and lower eyelids were never injected simultaneously in an individual monkey. Rather, upper and lower eyelid injections were alternated at a minimum interval of 2 months between injection sessions. Two distinct preinjury treatment paradigms were tested to maximize the chemomyectomy effect of the liposome-encapsulated doxorubicin. ^{4 5} Approximately 50% of the monkey eyelids received an injection of a mixture of 0.75% bupivacaine (Sensorcaine; Astra Pharmaceuticals, Westborough, MA) containing 1:200,000 epinephrine and hyaluronidase (150 units, Wydase; Wyeth Laboratories, Philadelphia, PA) in 1 ml 20 minutes before administration of Doxil into that eyelid. The second half of the eyelids was also treated with bupivacaine with epinephrine and hyaluronidase 2 days before injection of Doxil into that eyelid. All eyelids were monitored daily for the first month after injection. In all reinjection series, a minimum of 2 months was allowed between subsequent injections of Doxil. Thus, the postinjection survival times were a minimum of 2 months for a subset of the treated lids, a minimum of 4 months for the other set of lids on the same monkey, and in several specimens, the monkeys were euthanatized 6, 8, or 10 months after the last injection. In a final monkey, ⁷ two eyelids received no injections, and two eyelids received saline only to serve as a control. A statistical analysis was done using the Least Significance Difference for All Pairwise Comparisons. For each dose of Doxil (0.5, 1, or 2 mg), no significant differences were seen in the myofiber counts when muscle loss for the different postinjury survival times was compared statistically. 

At the end of the sequence of injections for each monkey, the animals were euthanatized with an overdose of barbiturate anesthesia. Eyelid samples were removed and immediately frozen in methylbutane chilled to a slurry over liquid nitrogen. Sections were prepared at 12 μm on a cryostat, and the tissue was stained using the histochemical myosin ATPase procedure to visualize the myofibers. Myofiber number was determined using a computer-aided morphometry program (Bioquant; R & M Biometrics, Nashville, TN). Statistical significance was determined using a Student’s t-test, with assistance from the Prism statistical software (Graphpad, San Diego, CA). 

Results

The eyelids were monitored daily after the Doxil injections, and all eyelids appeared to be completely normal. No skin injury or overt swelling of the lids was seen in any of the days after the initial injection at any of the Doxil doses administered. Some redness often was present within the treated eyelid, but usually dissipated within 2 to 3 days. The Doxil itself is red, and thus this redness was thought to be Doxil remaining within the injected eyelids. However, we did not observe or recognize the erythematosis and erythrodysesthesia that can occur as a side effect of Doxil extravasation in humans. ^{8 9} Some of the redness we observed could be due to the development of a short-lived erythema in the treated eyelids. In any event, this reaction was in dramatic contrast to the skin redness, inflammation, and injury we observed after injection of free doxorubicin. ⁷ Thus, injection of liposome-encapsulated doxorubicin, or Doxil, completely eliminated these undesirable side effects. No observable changes were seen in blink, nor were any irregularities seen in the cornea. The same was true for the many studies using either free doxorubicin ⁵ or Doxil ³ in rabbits and in the human blepharospasm and hemifacial spasm patients treated with either free doxorubicin ¹ or Doxil. 

All the Doxil-treated eyelids were analyzed morphometrically for muscle loss. Muscle loss was significant for all doses of Doxil tested. As in previous studies, ⁹ muscle loss was greatest in the preseptal regions of the orbicularis oculi, although muscle loss was seen in the pretarsal region as well (Fig. 1) . There was a slight dose-related effect on muscle loss after single injections of Doxil at doses of 0.5, 1, and 2 mg averaging 63%, 70%, and 79%, respectively (Fig. 2A ). Muscle loss was examined in the upper eyelids and compared with that seen for a comparable dose and treatment protocol in the lower eyelids. No significant difference was seen in the myotoxic effect of Doxil injections, based on the eyelid injected (Fig. 2B) . 

The range of muscle loss in the palpebral portion of the orbicularis oculi resulting from Doxil injections was from 64% to 94% (Fig. 3) . Although all muscle was significantly decreased compared with controls, only when a protocol was followed that included a 2-day preinjury protocol with multiple injections of Doxil was muscle loss significantly increased compared with other Doxil treatments (Fig. 2) . Thus, the maximal muscle loss was seen when the orbicularis oculi was preinjured with the local anesthetic bupivacaine followed 2 days later by injection of the Doxil and repeated two or three times in the same lid (Fig. 2) . It should be noted that in no instance was skin injury or signs of a localized inflammatory reaction present, even after a third exposure to the Doxil. 

Discussion

These results demonstrate that Doxil injections into monkey eyelids are a safe and effective strategy to permanently remove muscle from within the eyelid. Previous studies have shown that even 2 years after an injection sequence of free doxorubicin, the muscle that had been destroyed did not regenerate. ² The additional benefit of using Doxil compared with free doxorubicin is that the skin and eyelid tissue are not injured beyond the appearance of mild redness. This represents a major improvement to the doxorubicin chemomyectomy protocol. 

There was a species difference between the relative effectiveness of chemomyectomy as a result of the injection of free doxorubicin compared with liposome-encapsulated doxorubicin. In rabbit studies, a single injection of Doxil produced less of a chemomyectomy effect than a single injection of free doxorubicin. ³ A single dose of 2 mg Doxil resulted in a 60% loss of myofibers compared with a loss of 87% of the myofibers after a single injection of 2 mg doxorubicin. ³ In the monkey the average muscle loss in the palpebral portion (i.e., the pretarsal and preseptal portions combined) of the orbicularis oculi muscle from a single 1- or 2-mg injection of Doxil was 74.5%. This is compared with an average myofiber loss of 70% after a single eyelid injection of 1 or 2 mg free doxorubicin. ⁷ Thus, in monkey, the use of doxorubicin in its liposome-encapsulated form was equally as effective as the use of free doxorubicin. 

In the rabbit we have found that preinjury of the orbicularis oculi with bupivacaine 2 days before Doxil treatment was useful in increasing the chemomyectomy effect. These studies confirmed in nonhuman primates that preinjury of the orbicularis oculi with bupivacaine 2 days before Doxil treatment also significantly increased its chemomyectomy effect. We have previously hypothesized that this phenomenon can be correlated with the time when muscle satellite cells are maximally activated by the local anesthetic injury ^{4 5} and that the doxorubicin treatment arrives at a time of increased susceptibility to injury. 

Within the palpebral portion of the eyelid, the pretarsal portion of the orbicularis oculi muscle has proven to be very recalcitrant to injury both in monkeys and in rabbits. ⁷ This study demonstrates that it is indeed possible for Doxil to be effective in the pretarsal region, but it can take a minimum of three injections to do so. It may be related to the ability of the drug to spread within the injected eyelids, because rabbit eyelids are noticeably thicker, with denser connective tissue, than either nonhuman primate or human eyelids. This finding underscores the importance of testing treatments destined for human patients in nonhuman primates before proceeding with a human trial. 

All studies thus far have shown that direct, local injection of doxorubicin, either free in solution or in liposome-encapsulated form, results in permanent muscle loss. ^{2 7} This study confirms the apparent permanence of this effect for liposome-encapsulated doxorubicin in nonhuman primate lids, with the maximum postinjection interval being 12 months. We have previously demonstrated that muscle loss after doxorubicin administration is fairly rapid and relatively complete within 7 days. ⁶ Histologic examination of biopsy material removed from an eyelid 2 years after a final doxorubicin treatment of a blepharospasm patient demonstrated that muscle killed by doxorubicin did not regenerate; the treated eyelids remained essentially muscle free. ² Long-term follow up of blepharospasm patients treated with free doxorubicin also confirm that the muscle loss is permanent. ¹ Blepharospasm patients treated with free doxorubicin have now gone more than 9 years without requiring further botulinum toxin treatment. ¹  

The most significant finding in this study is the complete absence of skin ulceration and injury after any of the Doxil injections, even after three separate injection regimens in the same lid. This is in marked contrast to the skin reaction after injection of free doxorubicin, where each additional exposure increased the risk of skin ulceration. ¹ Although the side-effects to the skin over the injection site caused by free doxorubicin slowly resolved in both the laboratory studies and in the human patients, they were a deterrent to patient and physician acceptance of the procedure. The absence of significant injury to the overlying skin after local Doxil injections correlates with the distinctly different classification of these two drugs pharmacologically. Free doxorubicin is considered a vesicant, and this drug has a long history of causing skin ulcers with sufficient exposure to the skin after an extravasation from IV drug therapy for cancer treatment. ¹⁰ Doxil, however, is classified as an irritant. The only skin-related side effect from its use in high systemic doses in cancer patients has been the development of palmar-plantar erythematosis or erythrodysesthesia. ^{8 9} Although some redness over the injection site was seen in the treated monkeys, it may have been due to the Doxil itself, which is red. This redness always resolved within 2 to 3 days in the injected monkey eyelids. 

We have not studied functional changes in blink frequency with any of the various protocols in either the animals or the patients. It is possible to measure blink squeezing strength semiquantitatively in humans, which is reduced at the end of the free doxorubicin treatment cycle. However, we have not seen excessive weakness of the eyelid, paralytic ectropion, corneal dryness, or impaired lacrimal drainage, which occasionally occur in patients treated with botulinum injections. ¹¹  

As a result of this study, the human Doxil Blepharospasm Treatment Trial was initiated (www.med.umn.edu/ophthalmology/dbnews.html). Two blepharospasm patients have been treated thus far in a Phase I clinical trial (Wirtschafter JD, McLoon LK, unpublished results, 2000). The Doxil injections have proved to be very effective in providing spasm relief in these patients. Both patients have developed much of the desired weakness in their orbicularis oculi muscles. Both patients developed marked erythematosis and mild swelling of the eyelids, primarily after the first and second injections. After subsequent injections the skin was much less red and swollen. There was only mild discomfort when the eyelids were swollen. An unanticipated positive effect was a mild peeling of the skin that restored the eyelids to a“ younger” appearance that was pleasing to both patients. It is apparent, however, that species differences between young adult monkeys and older adult humans indicate that human clinical trials remain a necessary measure of safety and efficacy. 

Direct injections of Doxil for the treatment of blepharospasm and hemifacial spasm represent a major improvement over the injection of free doxorubicin. Even multiple exposures to the Doxil in the monkey eyelids did not result in any skin ulceration or injury. By following the protocol proposed in the human blepharospasm Doxil trial, these studies in nonhuman primates support its use as safe and effective for patients. 

 Supported by the Benign Essential Blepharospasm Research Foundation, National Eye Institute Grants EY07935 and EY11375, the Minnesota Lions and Lionesses, the Frank E. Burch chair (JDW), and an unrestricted grant to the Department of Ophthalmology from Research to Prevent Blindness, Inc.

 Submitted for publication October 16, 2000; revised January 16, 2001; accepted February 7, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Doxil Injection Protocols

Dose of Doxil	Single Injection			Multiple Injections
		BUP/HYAL 20 min before Doxil	BUP/HYAL 2 days before Doxil	BUP/HYAL 20 min before Doxil	BUP/HYAL 2 days before Doxil
0.5 mg	1, 1, 6	1, 1, 6	1, 3, 10	2, 3, 4
			2, 3, 8
1 mg	1, 1, 2	1, 1, 4	2, 3, 2	2, 3, 2
	1, 1, 6	1, 1, 12	1, 3, 10
2 mg	1, 1, 12	1, 1, 2	1, 3, 2	1, 3, 2

 Values are number of lids, number of repeat injection cycles, and observation time which is the number of months from last injection to time animals were euthanatized.

Figure 1.

 

(A) The pretarsal region of a saline-injected control eyelid. (B) The pretarsal region of an eyelid 2 months after the last of a series of three injections, spaced over the course of 10 months, of 2 mg Doxil 2 days after preinjury with bupivacaine and hyaluronidase. (C) The preseptal region of a saline-injected control eyelid. (D) The preseptal region of an eyelid 2 months after a one time injection of 2 mg Doxil 2 days after preinjury with bupivacaine and hyaluronidase. Arrows indicate individual myofibers. In all photomicrographs the skin side is at the top. Scale bar, 100 μm.

Figure 2.

 

(A) All doses of Doxil injected in the eyelids of monkeys significantly reduced the number of muscle fibers in the treated eyelids compared with controls. A small dose-related effect was seen when Doxil was combined with a preinjury protocol using bupivacaine treatment 20 minutes before injection of Doxil, with 2 mg being the most effective. There was a significant difference only between 0.5- and 2-mg doses. (B) After either 0.5 or 1 mg Doxil injections, there was no significant difference in the amount of muscle loss in the upper lid compared with the lower lid.

Figure 3.

 

All doses of Doxil injected in the eyelids of monkeys significantly reduced the number of muscle fibers in the treated eyelids compared with controls. Only a treatment protocol that involved preinjury (preinj) of the orbicularis oculi 2 days before Doxil exposure, repeated at least two times per lid, significantly increased the muscle loss over a single injection of Doxil at the same dose. *Significantly different from control; **significantly different from the same dose given in a single injection protocol or given after a 20-minute preinjury protocol.