Photodynamic Therapy Using Topically Applied Dihematoporphyrin Ether in the Treatment of Cervical Intraepithelial Neoplasia

.D.,† length delivering an energy density of 140 J/cm 2 is safe and effective in treating CIN. Phase II studies using PDT at the prescribed Objective: To perform a phase I study of topically applied di-application schedule and dose are indicated. (cid:113) Press hematoporphyrin ether (DHE) in the photodynamic treatment (PDT) of cervical intraepithelial neoplasia (CIN) using ﬁxed DHE doses and application schedules, and a variable dose of 630 nm red INTRODUCTION light delivered by an argon-pumped dye laser. Methods: Between February 1993 and April 1994, 24 nonpregnant women with a histologic diagnosis of CIN were enrolled. All patients had lesions involving at least 25% of the cervix that were colposcopically visi-ble. Using a cervical cap, 2 ml of a 1% solution of DHE (Photofrin) in a 4% Azone and isopropyl alcohol vehicle were applied diag-cervix 24 hr to PDT. argon-pumped dye light at 630 nm used perform PDT. a 400- m m silica ﬁber optic terminating in a microlens focused the laser radiation onto a circular ﬁeld of uniform light intensity perpendicular the single ﬁeld constant power density mW/cm phase I fashion 140 the ectocervix, and the possibility for of the 7 failures or recurrences occurred at energy densities of 80 For exam- J/cm 2 or less, while 8 of 11 (73%) patients were treated successfully ple, cold knife cone and LLETZ are commonly with PDT at an energy density of 100 to 140 J/cm 2 . Conclusions: with postoperative bleeding but provide tissue for histologic PDT with DHE and an argon-pumped dye laser at 630-nm wave-evaluation. We report the dorsal lithotomy position and without anesthesia. The laser system was coupled into a 400- m m silica ﬁber optic (QLT the ﬁrst phase I investigation of locally applied DHE in the PDT management of CIN. Through local application, the Phototherapeutics, Inc., Vancouver, British Columbia, Canada) terminating in a microlens which focused the laser radiation common systemic toxicity of cutaneous photosensitivity associated with intravenous DHE might be avoided without a onto a 10- to 30-mm-diameter circular ﬁeld of uniform light intensity perpendicular to the tissue. A custom-made vaginal compromise in efﬁcacy.


INTRODUCTION light delivered by an argon-pumped dye laser. Methods: Between February 1993 and April 1994, 24 nonpregnant women with a
Cancer of the cervix is the second most common cancer histologic diagnosis of CIN were enrolled. All patients had lesions affecting women worldwide and is the most common maliginvolving at least 25% of the cervix that were colposcopically visible. Using a cervical cap, 2 ml of a 1% solution of DHE (Photofrin) nancy in developing countries [1]. Although more than in a 4% Azone and isopropyl alcohol vehicle were applied to the 400,000 new cases of invasive cervical carcinoma are diagcervix 24 hr prior to PDT. An argon-pumped dye laser providing nosed each year, the predominant trend over the past three light at 630 nm was then used to perform PDT. Light was coupled decades has been a steady decline in the incidence and the into a 400-mm silica fiber optic terminating in a microlens which mortality from this disease [2]. Undoubtedly, the explanation focused the laser radiation onto a circular field of uniform light for this dramatic decrease in cervical cancer incidence and intensity perpendicular to the tissue. The entire ectocervix was mortality is a result of widespread cytologic screening foltreated in a single field including a margin of 3-5 mm of normal lowed by effective eradication of premalignant intraepithecervix. Using a constant power density (150 mW/cm 2 ) to avoid lial neolplasia [3].

plasia (CIN) is not known but numerous techniques have been
Thirteen patients with CIN I, 7 patients with CIN II, and 4 patients with CIN III were treated. The maximal energy density was described which have little morbidity and reasonable efficacy. well tolerated. Toxicity was minimal with no patients experiencing Diverse methods including cold knife excision, electrocautery, local necrosis, sloughing, or scarring; however, a mild vaginal discryosurgery, laser ablation, and large loop excision of the transcharge was noted in several patients. Systemic effects were absent. formation zone (LLETZ) have all been efficacious with success After 12 months of follow-up at 3-month intervals, 22 patients are rates between 90 and 98% being common [4]. Differences evaluable of whom 15 (68%) are disease free. One patient was lost among these techniques are primarily related to cost, unique to follow-up and in another the cervical cap was dislodged. Four side effects, distortion of the ectocervix, and the possibility for of the 7 failures or recurrences occurred at energy densities of 80 histologic review of the transformation zone [4,5]. For exam-J/cm 2 or less, while 8 of 11 (73%) patients were treated successfully ple, cold knife cone and LLETZ are more commonly associated with PDT at an energy density of 100 to 140 J/cm 2 . Conclusions: with postoperative bleeding but provide tissue for histologic PDT with DHE and an argon-pumped dye laser at 630-nm waveevaluation. Cryotherapy on the other hand frequently makes subsequent colposcopy unsatisfactory and is associated with 1 Present address: Department of Obstetrics and Gynecology, Division more vaginal discharge while laser is more costly. Finally, all of Gynecologic Oncology, Texas Tech University Health Sciences Center, methods currently employed involve substantial cervical stroma and preserve normal cervical tissue and architecture, we in-nancy test was obtained on all women of reproductive age prior to enrolling in the study. Women with a known hyper-vestigated the use of photodynamic therapy (PDT) as a treatment for CIN. Photodynamic therapy refers to the light acti-sensitivity to DHE, Azone, isopropyl alcohol, or any components of the formulation were excluded. Patients with abnor-vation of a photosensitizer to generate highly reactive oxygen intermediates. These intermediates irreversibly oxidize mal renal or liver function were not eligible.
Patients enrolled in this phase I protocol were referred to essential cellular components causing local injury and necrosis [6,7]. The process typically involves intravenous admin-and treated at the Beckman Laser Institute and Medical Clinic (Irvine, CA). All participants had cystologic evidence istration of a photosensitizing drug that is retained longer in vascular and highly proliferative tissue [8]. Unfortunately, of a low-grade or high-grade squamous intraepithelial lesion on pap smear and a biopsy-proven intraepithelial lesion in-skin photosensitivity can be a significant side effect limiting the practical use of PDT. For this reason, topical application volving at least 25% of the ectocervix. The cervical biopsy was consistent with the colposcopic and cytologic descrip-of photosensitizers has been proposed as a method of minimizing this untoward effect [9]. However, it is not yet clear tion of the lesion. All lesions were completely visualized colposcopically. Women with endocervical extension as well if mechanisms of tissue destruction following systemic and topical administration of photosensitizers are identical be-as those with any cytologic, histologic, or colposcopic suspicion for invasion were excluded. Patients were not eligible cause the efficiency of topical regimes is particularly dependent upon drug distribution and retention in critical cells and for study if they had undergone treatment for CIN within the previous 3 months. subcellular organelles.
Currently, the most commonly used photosensitizers are Twenty-four hours prior to PDT, and immediately prior to use, 100 mg of lyophilized DHE (light-protected Pho-hematoporphyrins such as dihematoporphyrin ether (DHE), the active fraction of hematoporphyrin, which in clinical tofrin, QLT Phototherapeutics, Inc., Vancouver, British Columbia, Canada) was reconstituted with 10 ml of a vehicle practice is usually injected intravenously. DHE is concentrated in various tissues after injection. This predominantly containing 4% Azone (American Cyanamid Co., Lederle Laboratories, Pearl River, NY), 48% isopropyl alcohol, and occurs in the liver followed by the spleen, kidney, dysplastic and frankly invasive tumor cells, skin, muscle, brain, and 48% water to yield a 1% solution of DHE. Two milliliters of the 1% DHE/4% Azone solution was then applied to a lungs [10]. Due to the differential localization of the photosensitizing agent in neoplastic versus normal epithelial cells, circular piece of gauze placed in the appropriate size cervical cap (Prentif Cavity Rim Cervical Cap, Cervical Cap Ltd., PDT can be used to selectively ablate preinvasive and invasive neoplasms. Photodynamic therapy requires activation Los Gatos, CA).
After 24 hr of continuously exposing the ectocervix to the of the photosensitizing agent by a monochromatic light, usually a red light at a wavelength of 630 nm delivered by DHE/Azone mixture, the cervical cap was removed and PDT was performed with an argon-pumped dye laser (Coherent, dye an argon-pumped dye laser. Because of its selectivity and tumoricidal effect, we expanded on the traditional methods Model 920, Palo Alto, CA) providing light at 630 nm (maximum output 0.8 W). This was performed with the patient in of systemic DHE administration and PDT by developing a technique to apply DHE directly to the cervix. We report the dorsal lithotomy position and without anesthesia. The laser system was coupled into a 400-mm silica fiber optic (QLT the first phase I investigation of locally applied DHE in the PDT management of CIN. Through local application, the Phototherapeutics, Inc., Vancouver, British Columbia, Canada) terminating in a microlens which focused the laser radiation common systemic toxicity of cutaneous photosensitivity associated with intravenous DHE might be avoided without a onto a 10-to 30-mm-diameter circular field of uniform light intensity perpendicular to the tissue. A custom-made vaginal compromise in efficacy. speculum was manufactured (U.S. Patent No. 5,458,595) which allowed stabilization of the optical fiber and focusing of the MATERIALS AND METHODS light spot size by changing the distance of the fiber from the treatment area (Fig. 1). The entire transformation zone was Approval was obtained from the Human Subjects Review Committee (Institutional Review Board) of the University treated in a single field including a margin of 3-5 mm of normal ectocervix. The density of the PDT energy was in-of California, Irvine and Investigational New Drug (IND) status from the Food and Drug Administration (No. 40,296) creased every 4 patients in a phase I fashion (40, 60, 80, 100, 120, and 140 J/cm 2 ) by using a constant power density (150 was granted. Twenty-four consenting nonpregnant women over the age of 18 with histologically documented CIN were mW/cm 2 ) in order to avoid thermal injury. Incident energy densities were escalated by increasing the exposure time at a recruited from community sources for a phase I study of locally applied DHE. Patients who were less than 1 year given power output (Table 1).
Patients were monitored for systemic as well as local tox-postmenopausal were required to use an acceptable method of birth control during the entire study period. No intravagi-icity 1 and 4 weeks after therapy and the degree of cervical damage and local injury was evaluated grossly as well as nal contraceptives were permitted. A negative serum preg-biopsy proven CIN II or III was detected or if progression (new lesions) was documented.

RESULTS
Between February 1993 and April 1994, 13 patients with CIN I, 7 patients with CIN II, and 4 patients with CIN III were treated with locally applied DHE in a 4% Azone and isopropyl alcohol vehicle and PDT as previously described. As expected, injury to the underlying normal cervical stroma was minimal and none of the patients experienced local necrosis, sloughing, scarring, or a significant change in the location of the squamocolumnar junction. However, a mild vaginal discharge was noted for the first 2-4 days following PDT. Thus, an energy density of 140 J/cm 2 was well tolerated. Finally, systemic effects were absent at all energy densities.
At the completion of the 12-month follow-up period, 22 patients ware evaluable of whom 15 (68%) are disease free (Table 2). One patient (No. 11 in Table 2) was lost to followup after 6 months and in another (No. 24 in Table 2) the cervical cap was dislodged and presumably there was not or less, while 8 of 11 (73%) patients were treated successfully with PDT at an energy density of 100 to 140 J/cm 2 . Four of the failures/recurrences (patients 1, 7, 15, and 21 in colposcopically. Local toxicity was graded according to the size, depth, and time to healing of the area of destruction. Table 2) were CIN II or III, while 3 were CIN I (patients 6, 10, and 23 in Table 2). One patient (No. 8 in Table 2) The energy density was then increased only if local necrosis, pain, or nonhealing ulcers were absent in three of the four developed CIN I 12 months after PDT. When she returned for treatment, no lesion could be identified colposcopically patients treated at the previous energy density. Response was assessed colposcopically and cytologically 3, 6, 9, and and her cytologic smear was normal. Four other patients (patients 9, 14, 17, and 19 in Table 2) had equivocal colpos-12 months after PDT. Colposcopically directed biopsies of all abnormal areas were performed. Patients were removed copic findings associated with normal exfoliated cytology and yet had biopsy-proven CIN I and koilocytotic atypia at from study and treated by other means (laser or LLETZ) if  Note. Failures are highlighted in boldface. Normal is defined as the absence of biopsy-proven dysplasia. NA means that the results are not available (i.e., patient missed appointment). LTFU means that the patient was lost to follow-up.
a Subsequent evaluation was normal, therefore this patient was considered free of disease. b Cervical cap dislodged.
the location of there previous lesion. These patients ulti-Because DHE is selectively sequestered in neoplastic tissues, PDT has been used extensively in the treatment of mately became completely free of disease without further therapy and thus were considered cures. Finally, 3 patients bladder, skin, head and neck, brain, and esophageal tumors with minimal surrounding normal tissue toxicity [12][13][14][15]. had steady regression of their disease occurring slowly dur-Although less well studied, a few gynecologic malignancies ing the study period (patients 3, 4, and 20).
hampered by limited optical penetration with effective cytotoxicity of only 4-10 mm in depth [11]. For this reason, DISCUSSION clinicians have suggested that PDT is ideally suited for the Photodynamic therapy is a technique whereby light is used treatment of intraepithelial lesions such as CIN [16], Barto achieve a toxic effect in photosensitized cells. When excited rett's esophagus [24], and carcinoma in situ of the bladder by red light in the presence of oxygen, photosensitizers such [14]. In addition, the development of PDT has been limited as the hematoporphyrin, DHE, undergo a series of reactions by the systemic toxicity of the photosensitizing agent when producing a selective cytotoxic effect in areas where the photoadministered intravenously, predominantly cutaneous photochemical reaction occurs. Energy from the excited triplet state sensitivity [11, 16,[12][13][14][15][16][17][18][19][20][21][22][23]. In order to overcome these side of the photosensitizer, induced by the absorption of light of effects, investigators have recently began to study topically the appropriate energy and wavelength, is directly transferred applied photosensitizers [25][26][27]. Indeed, animal model to molecular oxygen producing singlet oxygen. Although it is studies indicate that there is selective retention of hematovery unstable and short-lived, singlet oxygen is extremely reacporphyrins in areas of chemically induced squamous dysplative and capable of oxidizing biologic molecules causing irresia after topical application of hematoporphyrin derivatives versible damage to subcellular organelles such as the cell mem-(HPD) [16].
On the basis of these animal model studies demonstrating brane, mitochondria, and lysosomes [11]. effective topical delivery and selective concentration in pre-may instead represent recurrences and thus not accurately reflected the true effectiveness of PDT in the treatment of malignant epithelial lesions, we initiated a pilot study of PDT in human intraepithelial neoplasia of the female genital CIN. Therefore, phase II studies of high-grade lesions and/ or phase III studies comparing PDT to existing established tract [16]. We demonstrated that PDT was more effective in treating nonkeritinizing epithelium such as the vagina and modalities will determine the role of PDT in the current management of intraepithelial neoplasia of the cervix. In cervix, while it was relatively less effective on the vulva. In addition, further study revealed that Azone with isopropyl addition, second generation photosensitizers such as 5aminolevulinic acid, and Benzo-Prophyrin derivative, may alcohol was a more effective photosensitizer vehicle for PDT than Eucerin cream (Biersdorf Inc., South Norwalk, CT) [25, offer improved penetration following topical application and should also be tested for this indication. 28]. Therefore, we began to study the treatment of CIN and VAIN, since they were not keritinized, with PDT and locally applied photosensitizers in the Azone vehicle. In unpub-REFERENCES lished trials, we successfully treated a patient with VAIN less, numerous dramatic responses to PDT were documented in the current study. Moreover, many of the PDT failures 15. Hayata, Y., Kato, H., Okitsu, H., Kawaguchi, M., and Konaka, C.