- Email: [email protected]
Fluorescence diagnosis of cervical squamous intraepithelial lesions: A clinical feasability study
Photodiagnosis and Photodynamic Therapy (2007) 4, 112—116
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/pdpdt
Fluorescence diagnosis of cervical squamous intraepithelial lesions: A clinical feasability study Pierre Collinet a,b, Marie Delemer a,b, Eva Jouve a,b, Claudia Regis a,b, Marie Odile Farine c, Denis Vinatier a, Michel Cosson a, Serge Mordon PhD b,∗ a
Gynaecologic Surgery Department, Jeanne de Flandre Hospital, Lille University Hospital-CHRU, 59037 Lille Cedex, France INSERM, IFR 114, Pavillon Vancostenobel, Lille University Hospital-CHRU, 59037 Lille Cedex, France c Anatomopathology Department, Lille University Hospital-CHRU, 59037 Lille Cedex, France Available online 8 December 2006 b
KEYWORDS Colposcopy; Aminolevulinic acid; Protoporphyrin IX; CIN; Fluorescence
Summary The potential of fluorescence diagnosis (FD) is still undeveloped in gynaecology. In order to diagnose and localize squamous intraepithelial lesion (SIL) of the cervix, a new method improving the low specificity of colposcopy, would be useful. Objective: The goal of this study was to assess the feasability and safety of colposcopic FD of SIL after topicaly application of methyl aminolevulinate (MAL). Materials and methods: Patients with histologic proved cervical intraepithelial neoplasia (CIN) and planned for loop electrosurgical excision procedure (LEEP) under general anesthesia, were included in a prospective study. Before general anesthesia, a thick layer of MAL (Metvix® —– 160 mg/mL cream) was applied on the cervix for 35—150 min. Fluorescent colposcopic inspection of the cervix was performed using a rigid 10-mm laparoscope inserted in the vaginal cavity and connected to D-light AF system (Karl Storz Endoskope, Tuttlingen Germany). In order to confirm neoplasic status, fluorescent foci underwent directed punch biopsy(ies). Results: Fourteen patients were included in the study. Colposcopic fluorescence imaging revealed red fluorescent foci in 71.4% of cases (10/14) (neoplasic status of fluorescent foci was confirmed in 80%). Concerning ME-ALA, the mean of application time was 73 min (35—150). Two cases presented a false-positive fluorescence and four cases of false-negative fluorescence. For all cases of false-negative fluorescence, application time of MAL was less than 60 min. We observed no systemic or local toxicity of MAL application in any of the groups. Conclusion: Using topical application of MAL to the cervix, we demonstrated that FD of SIL is feasible. This study justifies the further development of fluorescence imaging that combines classical white light colposcopy with the use of a photosensitizer. © 2006 Elsevier B.V. All rights reserved.
Introduction
∗
Corresponding author. Tel.: +33 320 446 708; fax: +33 320 446 708. E-mail address: [email protected] (S. Mordon).
Despite numerous cervical screening programmes in the world, cervical cancer still remains the second most common cause of cancer deaths in women. However, prevention is possible with early identification and treatment of infectious risk factors and of squamous intraepithelial lesion (SIL).
1572-1000/$ — see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.pdpdt.2006.10.003
Fluorescence diagnosis of cervical squamous intraepithelial lesions: A clinical feasability study Cervical cytology (Pap smear) is recommended for screening of cervical cancer. When Pap smear is abnormal, colposcopy is mandatory. This exploration allows localization of SIL and to perform biopsy(ies) under visual control. Colposcopy requires a long learning curve and specific training. Experts estimate that 3 or 4 months of training is necessary to adequately recognize SIL, 1 year to determine the optimal place for biopsy and probably many years to predict histologic result with colposcopic examination of cervix. Accuracy and diagnosis rentability are limited even in expert hands. Due to these difficulties of interpretation, variability between colposcopists is high. The junction may be poorly visible due to atrophia or infection, and colposcopy is not able to conclude in 15% of cases. A new method that can reduce the low specificity of colposcopy, would be useful and could permit a simplification and standardization of this exploration. Fluorescence diagnosis (FD) also called photodynamic diagnosis has been developed in urology for epithelial lesion of the bladder (Hexvix® ) [2]. Presently, FD is under development in gynaecology for cervical premalignancies and peritoneal micro-metastatis of ovarian cancer [3] and in gastroenterology for preinvasive colic tumors [1]. In these studies, screening is improved from 17.5 to 35%. Fluorescence can be spontaneous (autofluorescence of tissues) or enhanced by photoactive porphyrin (PAP) precursor such as 5-aminolevulinic acid (ALA). Exogenously applied ALA, results in endogenous production of the potent, fluorescent PAPs including protoporphyrin IX (PpIX) [4]. ALA by itself, is not a photosensitizing agent, but is the precursor of PAPs in the biosynthetic pathway of heme. Physiologically, the production of heme regulates the synthesis of ALA via a mechanism of negative feedback. Administration of exogenous ALA bypasses this mechanism and induces a temporary accumulation of PAPs in malignant cells of epithelial origin [5]. Because of the significant difference in the activities of key enzymes in the heme pathway between tumor and normal tissue, the PAP accumulation induced by ALA in tumor cells is higher than that in normal cells [5]. PAPs show a maximum light absorption around 410 nm and emit fluorescence with a broad peak at 635—705 nm. PAPs selectively concentrates in tumor cells and generates red fluorescence when excited by blue light revealing a distinct contrast to the surrounding normal tissues [1]. However, ALA is hydrophilic and does not easily penetrate through intact skin or cell membranes [6], thus the efficiency of PAP production is low. In order to overcome this problem, a number of ALA esters with more lipophilic property has been tested [7]. Hexaminolevulinate (HAL), gives a more rapid and higher increase of intracellular PAP and could enhance both fluorescence intensity and inter tissular contrast [4,8]. For diagnosis of superficial bladder cancer, Kriegmair et al. demonstrated that FD after intravesical instillation of HAL improved sensitivity of cystoscopy (n = 24/Se: 72% versus 97%) [9—11]. MAL is used as a topical photosensitizer in dermatology. For treatment of solar keratoses and basal cell carcinomas, clinical trials using MAL PDT demonstrated that the fluorescence of induced porphyrins is effective in detecting and delineating neoplasic skin areas [12,13]. The goal of this study was to assess the safety and efficacy of colposcopic FD of SIL after topical application of MAL.
113
Materials and methods Patients Each patient was referred to our colposcopic clinic with abnormal cytology. They all underwent a gynaecologic examination, including HPV testing, cytology, colposcopy and biopsy(ies). The cervical biopsies were assessed according to the cervical intraepithelial neoplasia (CIN) classification system as mild (CIN1), moderate (CIN2), or severe (CIN3) dysplasia. Patients with histologic proved CIN and planned for loop electrosurgical excision procedure (LEEP) under general anaesthesia, were included in a prospective study. Protocol of the study was approved by our local ethical committee. A written informed consent was obtained prior to inclusion for each patient.
Photosensitizer Metvix® cream (MAL) is a topical photosensitizer that is commercialy available [14]. The surface of the cervix was prepared by washing with saline solution. Half of the MAL tube (1 g) diluted with 2 mL of sterile saline solution was used for each patient. Using a spatula, a 1 mm thick layer of MAL cream was applied on the cervix. Application of the cream was performed before general anesthesia, FD and LEEP.
Fluorescence colposcopic examination Under general anesthesia and after cream application of 35—150 min, fluorescent colposcopic inspection of the cervix was performed. A rigid 10-mm laparoscope (Karl Storz, Tuttlingen, Germany) connected to an increased red color sensitivity camera (Karl Storz Endovision Tricam), was introduced in the vaginal cavity through the speculum trocart to visualize the uterine cervix. The D-light AF system (Karl Storz Endoskope) was used to perform both conventional and fluorescence examination. Using this system, light emission could be switched easily from conventional whitelight mode to fluorescence blue-light mode (wavelength of 380—440 nm) with a foot pedal. The color video camera was connected to a DVCAM Video Recorder (DVR 30, Sony, Tokyo, Japan) in order to analyze the video images off-line. Visual observation of fluorescence was examined on the TV screen. Fluorescence of the cervix was defined as negative or positive. A mapping of fluorescent foci was recorded for each patient. In order to confirm neoplasic status, all fluorescent foci underwent directed punch biopsy(ies). Then, LEEP was performed in order to remove transformation zone and to treat SIL. All biopsies and piece of conization were submitted for histologic examination. Histologic results were classified into the most severe category of the CIN terminology. For each patient, fluorescence profiles were compared with the histology of the punch biopsies taken from the corresponding fluorescent foci, with the histology of conization and with the initial classical colposcopic examination. Patients were monitored for systemic and local toxicity with a postoperative examination at 6—8 h and at 4—6 weeks. They were also followed up at 3 months and after, at 6
114 Table 1
P. Collinet et al. Patients’ characteristics
Patients
Cytology
Histology (colposcopy)
Time (min)
Fluorescence
Histology (PDD)
Histology (LEETZ)
1 2 3 4 5 6 7 8 9 10 11 12 13 14
HSIL HSIL LSIL ASCUS ASCUS HSIL LSIL ASCUS LSIL HSIL LSIL LSIL ASCUS HSIL
CIN3 CIN2 CIN1 CIN3 CIN2 CIN1 CIN2 CIN1 CIN2 CIN2 CIN2 CIN2 CIN2 CIN3
35 39 40 40 50 59 60 74 77 82 90 110 120 150
Negative Positive Negative Negative Negative Positive Positive Positive Positive Positive Positive Positive Positive Positive
NP CIN3 NP NP NP CIN2 CIN1 Negative CIN3 CIN2 CIN3 CIN2 Negative CIN1
CIN2 CIN3 CIN1 CIN3 CIN2 CIN2 CIN1 CIN3 CIN3 CIN2 CIN3 CIN3 Negative CIN1
NB—–CIN: cervical intraepithelial neplasia; HSIL: high grade squamous intraepithelial lesion; LSIL: low grade squamous intaepithelial lesion; ASCUS: atypical squamous cells of undeterminated significance.
months intervals with HPV testing, cytologic smear test, colposcopic assessment, and, if indicated, colposcopically directed punch biopsy of the cervix.
Results Between June 2005 and March 2006, 14 patients were included in the study. The mean age was 37.5 years (27—54). Initial cytology revealed four patients with ASCUS (28.6%), five with low grade SIL (LSIL) (35.7%) and five with high grade SIL (HSIL) (35.7%) (Table 1). During initial classical colposcopic examination, the transformation zone was not completely visible in four cases (28.6%). All colposcopical examinations were abnormal with areas of atypical transformation zone (TAG): 28.6% of TAG grade 1 (moderate colposcopic modifications of the transformation zone) and 71.4% of TAG grade 2 (severe colposcopic modifications of the transformation zone). Histology from punch biopsies preformed under classical colposcopy showed three patients with CIN1 (21.4%), eight with CIN2 (57.1%) and three with CIN3 (21.4%). Seven patients underwent high risk HPV testing preoperatively (50%) that was positive in all cases. Using the endoscope associated with D-light AF system, colposcopic fluorescence imaging proved to be applicable for real time diagnosis of SIL of the cervix. Red fluorescent foci were observed in 71.4% of cases (10/14) (Figs. 1 and 2). In these 10 cases, foci of red fluorescence of suspected SIL were clearly identified from the blue fluorescence of normal ectocervix. Neoplasic status of fluorescent foci was confirmed in 80%: two CIN1 (25%), three CIN2 (37.5%) and three CIN3 (37.5%) (Table 1). Concerning ME-ALA, the mean of application time was 73 min (35—150). With application time <45 min, foci of fluorescence were observed only in one case out of four (25%). With application time >75 min, foci of fluorescence were observed only in 100% (6/6). Two cases presented a false-positive fluorescence. One case presented a false-positive fluorescence because histology of conization revealed no SIL (probably all removed by the initial biopsy). The other case presented a fluores-
Figure 1 Example no. 1 of fluorescent foci (case no. 10); 1 mm layer of cream (500 mg of Me—–ALA); 75 min before colposcopic examination; final histology was CIN2.
cent focus with negative histology of the biopsy and positive histology of conization. We observed four cases of false-negative fluorescence: one patient with a histology of CIN1, two with a histology of CIN2 and one with a histology of CIN3 (Table 1). For one case, we experienced difficulties in applying the cream due to strong retroversion of the uterus. For another case, SIL was located in the endocervix and transformation zone was not visible under white light colposcopy. For all of cases of false-negative fluorescence, application time of MAL cream was less than 60 min. We observed no toxicity of MAL application (no local or systemic side effect).
Discussion The objective of the study was to assess feasibility of colposcopic FD, using different MAL application times, of SIL. In addition, we demonstrate the successful identifi-
Fluorescence diagnosis of cervical squamous intraepithelial lesions: A clinical feasability study
Figure 2 Example no. 2 of fluorescent foci (case no. 12); 1 mm layer of cream (500 mg of Me—–ALA); 120 min before colposcopic examination; final histology was CIN3.
cation of dysplastic lesions. The main parameter was the confirmation of dysplastic status of fluorescents foci by histology. We observed a higher porphyrin fluorescence intensity within the CIN lesions, as compared to normal adjacent epithelium. This result was previously described by Andikyan et al. [15]. They also observed that fluorescence selectivity was the highest in CIN3. Weingandt et al. also demonstrated that fluorescence intensity was correlated to the severity of CIN [16]. These results demonstrated that FD of SIL using topical application of MAL is feasible. The current study was designed as a dose-finding feasability study. All patients were planned for conization due to histologic proved SIL. For this reason, we observed one patient with false-positive fluorescence (no residual lesion). Because of the absence of true-negative case in this study, we were able to calculate only sensitivity and positive predictive value. In the study, colposcopic FD had a sensitivity of 69% and positive predictive value was 90%. In order to compare sensitivity and specificity of classical and FD colposcopies, next step would be to include only patients with pathologic cytology and normal classical colposcopy. The successful outcome of FD depends on parameters including dose of MAL, time interval between administration and light emission and wavelength and source of light. Concerning the photosensitizer, we selected Metvix® for this study because of its European approval in Dermatology. We observed the feasibility of using MAL cream as a drug for topical application on the cervix and for colposcopic FD. PAP precursors penetrate the normal and tumoral epithelium and induce PAP fluorescence. We did observe selective PAP accumulation in neoplasic tissue as reported by Hillemanns et al. [17]. The selectivity is partially explained by a lower ferrochelatase activity in tumor cells and a higher tumor level of porphobilinogen deaminase, which are the two key enzymes in the regulation of the heme pathway [4]. Thus, the photosensitizing effects of PAP produced by topical application of PAP precursor show a high degree of tissue specificity. ALA has been studied extensively because it is a safe and effective photosensitizing agent [18]. The results for FD and
115
PDT of SIL/CIN are variable, and to increase efficacy—–new and more potent photosensitizers should be tested. More lipophilic derivatives of ALA, such as MAL, with better tissue penetration (bioavailability suggests systemic bioavailability) should therefore show good performance. MAL and HAL were found to be more efficient than ALA itself to induce PAP in some cell lines in vitro [6] and in tumor in vivo in urology [19,20]. Recently, it has been shown that esterification of ALA into more lipophilic derivatives results in up to a 25-fold increase in PAP levels in malignant cells in culture [6] and a more homogeneous distribution across a normal pig urothelium in an in vitro model [21]. Concerning time application, the importance of this parameter was previously reported by Hillemans et al. [22]. Using topical application of 1% ALA, the optimal time interval was 60—90 min. In this study, no fluorescence could be diagnosed with intervals less than 30 min. Using ex vivo fluorescence microscopy, application of HAL 0.5% cream to the cervix induced selective fluorescence in epithelial cells of the cervix [23]. The optimal ratio with a homogeneous PAP distribution was obtained after 100 (±10) min cream application. Using MAL, we observed the same result. For the four cases of false-negative fluorescence, time intervals were 35, 40, 40 and 50 min, respectively. These three studies confirmed the need of a minimal interval between application and exploration of 60 min. In order to improve PAP distribution, Duska et al. demonstrated the feasibility of administering ALA per os (10 mg/kg) in 14 patients with abnormal Pap smears [24]. No systemic side effects were recorded. They observed an optimal fluorescence after 3 h. Since this publication in 2002, no study was reported in the literature. This result needs to be confirmed and compared to topical application. It could improve patient’s comfort compliance. Finally, several studies demonstrated potential advantage of photodynamic therapy (PDT) for the treatment of SIL with reduced risk of bleeding after conization, and more favorable long-term results avoiding cervical stenosis and preterm labor [22,25—30]. The potential of FD and PDT is promising but still experimental in gynaecology. This technique could permit to ‘‘see and treat’’ in real time without invasive procedure.
Conclusion Using topical application of MAL cream, we demonstrated that FD of SIL is feasible with promising efficacy and no local or systemic adverse events. A minimal application time of 60 min is required for optimal results. This study justifies further work to develop classical white light colposcopy including fluorescence imaging.
References [1] Gahlen J, Stern J, Laubach HH, Pietschmann M, Herfarth C. Improving diagnostic staging laparoscopy using intraperitoneal lavage of delta-aminolevulinic acid (ALA) for laparoscopic fluorescence diagnosis. Surgery 1999;126(September (3)):469—73. [2] Kriegmair M, Baumgartner R, Lumper W, Waidelich R, Hofstetter A. Early clinical experience with 5-aminolevulinic acid for the photodynamic therapy of superficial bladder cancer. Br J Urol 1996;77(May (5)):667—71.
116 [3] Sabban F, Collinet P, Cosson M, Mordon S. Fluorescence imaging technique: diagnostic and therapeutic interest in gynecology. J Gynecol Obstet Biol Reprod (Paris) 2004;33(December (8)):734—8. [4] Kennedy JC, Pottier RH. Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 1992;14(July (4)):275—92. [5] Peng Q, Warloe T, Berg K, et al. 5-Aminolevulinic acid-based photodynamic therapy. Clinical research and future challenges. Cancer 1997;79(June (12)):2282—308. [6] Gaullier JM, Berg K, Peng Q, et al. Use of 5-aminolevulinic acid esters to improve photodynamic therapy on cells in culture. Cancer Res 1997;57(April (8)):1481—6. [7] Brunner H, Hausmann F, Knuechel R. New 5-aminolevulinic acid esters—–efficient protoporphyrin precursors for photodetection and photodynamic therapy. Photochem Photobiol 2003;78(November (5)):481—6. [8] Kloek J, Akkermans W, Beijersbergen van Henegouwen GM. Derivatives of 5-aminolevulinic acid for photodynamic therapy: enzymatic conversion into protoporphyrin. Photochem Photobiol 1998;67(January (1)):150—4. [9] Kriegmair M, Baumgartner R, Knuechel R, et al. Fluorescence photodetection of neoplasic urothelial lesions following intravesical instillation of 5-aminolevulinic acid. Urology 1994;44(December (6)):836—41. [10] Kriegmair M, Stepp H, Steinbach P, et al. Fluorescence cystoscopy following intravesical instillation of 5-aminolevulinic acid: a new procedure with high sensitivity for detection of hardly visible urothelial neoplasias. Urol Int 1995;55(4):190—6. [11] Kriegmair M, Zaak D, Knuechel R, Baumgartner R, Hofstetter A. Photodynamic cystoscopy for detection of bladder tumors. Semin Laparosc Surg 1999;6(June (2)):100—3. [12] Foley P. Clinical efficacy of methyl aminolevulinate (Metvix) photodynamic therapy. J Dermatol Treat 2003;14(Suppl. 3):15—22. [13] Freeman M, Vinciullo C, Francis D, et al. A comparison of photodynamic therapy using topical methyl aminolevulinate (Metvix) with single cycle cryotherapy in patients with actinic keratosis: a prospective, randomized study. J Dermatol Treat 2003;14(June (2)):99—106. [14] Gardlo K, Ruzicka T. Metvix (PhotoCure). Curr Opin Investig Drugs 2002;3(November (11)):1672—8. [15] Andikyan V, Kronschnabl M, Hillemanns M, Wang X, Stepp H, Hillemanns P. Fluorescence diagnosis with 5-ALA thermogel of cervical intraepithelial neoplasia. Gynakol Geburtshilfliche Rundsch 2004;44(January (1)):31—7. [16] Weingandt H, Stepp H, Baumgartner R, Diebold J, Xiang W, Hillemanns P. Autofluorescence spectroscopy for the diagnosis of cervical intraepithelial neoplasia. Bjog 2002;109(August (8)):947—51. [17] Hillemanns P, Weingandt H, Baumgartner R, Diebold J, Xiang W, Stepp H. Photodetection of cervical intraepithelial neoplasia using 5-aminolevulinic acid-induced porphyrin fluorescence. Cancer 2000;88(May (10)):2275—82.
P. Collinet et al. [18] Kennedy JC, Pottier RH, Pross DC. Photodynamic therapy with endogenous protoporphyrin IX: basic principles and present clinical experience. J Photochem Photobiol B 1990;6(June (1—2)):143—8. [19] Lange N, Jichlinski P, Zellweger M, et al. Photodetection of early human bladder cancer based on the fluorescence of 5aminolaevulinic acid hexylester-induced protoporphyrin IX: a pilot study. Br J Cancer 1999;80(April (1—2)):185—93. [20] Marti A, Jichlinski P, Lange N, et al. Comparison of aminolevulinic acid and hexylester aminolevulinate induced protoporphyrin IX distribution in human bladder cancer. J Urol 2003;170(August (2 Pt 1)):428—32. [21] Marti A, Lange N, van den Bergh H, Sedmera D, Jichlinski P, Kucera P. Optimisation of the formation and distribution of protoporphyrin IX in the urothelium: an in vitro approach. J Urol 1999;162(August (2)):546—52. [22] Hillemanns P, Korell M, Schmitt-Sody M, et al. Photodynamic therapy in women with cervical intraepithelial neoplasia using topically applied 5-aminolevulinic acid. Int J Cancer 1999;81(March (1)):34—8. [23] Andrejevic-Blant S, Major A, Ludicke F, et al. Time-dependent hexaminolaevulinate induced protoporphyrin IX distribution after topical application in patients with cervical intraepithelial neoplasia: a fluorescence microscopy study. Lasers Surg Med 2004;35(4):276—83. [24] Duska LR, Wimberly J, Deutsch TF, et al. Detection of female lower genital tract dysplasia using orally administered 5aminolevulinic acid induced protoporphyrin IX: a preliminary study. Gynecol Oncol 2002;85(April (1)):125—8. [25] Ichimura H, Yamaguchi S, Kojima A, et al. Eradication and reinfection of human papillomavirus after photodynamic therapy for cervical intraepithelial neoplasia. Int J Clin Oncol 2003;8(October (5)):322—5. [26] Bodner K, Bodner-Adler B, Wierrani F, et al. Coldknife conization versus photodynamic therapy with topical 5-aminolevulinic acid (5-ALA) in cervical intraepithelial neoplasia (CIN) II with associated human papillomavirus infection: a comparison of preliminary results. Anticancer Res 2003;23(March—April (2C)):1785—8. [27] Barnett AA, Haller JC, Cairnduff F, Lane G, Brown SB, Roberts DJ. A randomised, double-blind, placebo-controlled trial of photodynamic therapy using 5-aminolaevulinic acid for the treatment of cervical intraepithelial neoplasia. Int J Cancer 2003;103(Match (6)):829—32. [28] Keefe KA, Tadir Y, Tromberg B, et al. Photodynamic therapy of high-grade cervical intraepithelial neoplasia with 5-aminolevulinic acid. Lasers Surg Med 2002;31(4):289—93. [29] Wierrani F, Kubin A, Jindra R, et al. 5-aminolevulinic acidmediated photodynamic therapy of intraepithelial neoplasia and human papillomavirus of the uterine cervix—–a new experimental approach. Cancer Detect Prev 1999;23(4):351—5. [30] Muroya T, Suehiro Y, Umayahara K, et al. Photodynamic therapy (PDT) for early cervical cancer. Gan To Kagaku Ryoho 1996;23(January (1)):47—56.
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/pdpdt
Fluorescence diagnosis of cervical squamous intraepithelial lesions: A clinical feasability study Pierre Collinet a,b, Marie Delemer a,b, Eva Jouve a,b, Claudia Regis a,b, Marie Odile Farine c, Denis Vinatier a, Michel Cosson a, Serge Mordon PhD b,∗ a
Gynaecologic Surgery Department, Jeanne de Flandre Hospital, Lille University Hospital-CHRU, 59037 Lille Cedex, France INSERM, IFR 114, Pavillon Vancostenobel, Lille University Hospital-CHRU, 59037 Lille Cedex, France c Anatomopathology Department, Lille University Hospital-CHRU, 59037 Lille Cedex, France Available online 8 December 2006 b
KEYWORDS Colposcopy; Aminolevulinic acid; Protoporphyrin IX; CIN; Fluorescence
Summary The potential of fluorescence diagnosis (FD) is still undeveloped in gynaecology. In order to diagnose and localize squamous intraepithelial lesion (SIL) of the cervix, a new method improving the low specificity of colposcopy, would be useful. Objective: The goal of this study was to assess the feasability and safety of colposcopic FD of SIL after topicaly application of methyl aminolevulinate (MAL). Materials and methods: Patients with histologic proved cervical intraepithelial neoplasia (CIN) and planned for loop electrosurgical excision procedure (LEEP) under general anesthesia, were included in a prospective study. Before general anesthesia, a thick layer of MAL (Metvix® —– 160 mg/mL cream) was applied on the cervix for 35—150 min. Fluorescent colposcopic inspection of the cervix was performed using a rigid 10-mm laparoscope inserted in the vaginal cavity and connected to D-light AF system (Karl Storz Endoskope, Tuttlingen Germany). In order to confirm neoplasic status, fluorescent foci underwent directed punch biopsy(ies). Results: Fourteen patients were included in the study. Colposcopic fluorescence imaging revealed red fluorescent foci in 71.4% of cases (10/14) (neoplasic status of fluorescent foci was confirmed in 80%). Concerning ME-ALA, the mean of application time was 73 min (35—150). Two cases presented a false-positive fluorescence and four cases of false-negative fluorescence. For all cases of false-negative fluorescence, application time of MAL was less than 60 min. We observed no systemic or local toxicity of MAL application in any of the groups. Conclusion: Using topical application of MAL to the cervix, we demonstrated that FD of SIL is feasible. This study justifies the further development of fluorescence imaging that combines classical white light colposcopy with the use of a photosensitizer. © 2006 Elsevier B.V. All rights reserved.
Introduction
∗
Corresponding author. Tel.: +33 320 446 708; fax: +33 320 446 708. E-mail address: [email protected] (S. Mordon).
Despite numerous cervical screening programmes in the world, cervical cancer still remains the second most common cause of cancer deaths in women. However, prevention is possible with early identification and treatment of infectious risk factors and of squamous intraepithelial lesion (SIL).
1572-1000/$ — see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.pdpdt.2006.10.003
Fluorescence diagnosis of cervical squamous intraepithelial lesions: A clinical feasability study Cervical cytology (Pap smear) is recommended for screening of cervical cancer. When Pap smear is abnormal, colposcopy is mandatory. This exploration allows localization of SIL and to perform biopsy(ies) under visual control. Colposcopy requires a long learning curve and specific training. Experts estimate that 3 or 4 months of training is necessary to adequately recognize SIL, 1 year to determine the optimal place for biopsy and probably many years to predict histologic result with colposcopic examination of cervix. Accuracy and diagnosis rentability are limited even in expert hands. Due to these difficulties of interpretation, variability between colposcopists is high. The junction may be poorly visible due to atrophia or infection, and colposcopy is not able to conclude in 15% of cases. A new method that can reduce the low specificity of colposcopy, would be useful and could permit a simplification and standardization of this exploration. Fluorescence diagnosis (FD) also called photodynamic diagnosis has been developed in urology for epithelial lesion of the bladder (Hexvix® ) [2]. Presently, FD is under development in gynaecology for cervical premalignancies and peritoneal micro-metastatis of ovarian cancer [3] and in gastroenterology for preinvasive colic tumors [1]. In these studies, screening is improved from 17.5 to 35%. Fluorescence can be spontaneous (autofluorescence of tissues) or enhanced by photoactive porphyrin (PAP) precursor such as 5-aminolevulinic acid (ALA). Exogenously applied ALA, results in endogenous production of the potent, fluorescent PAPs including protoporphyrin IX (PpIX) [4]. ALA by itself, is not a photosensitizing agent, but is the precursor of PAPs in the biosynthetic pathway of heme. Physiologically, the production of heme regulates the synthesis of ALA via a mechanism of negative feedback. Administration of exogenous ALA bypasses this mechanism and induces a temporary accumulation of PAPs in malignant cells of epithelial origin [5]. Because of the significant difference in the activities of key enzymes in the heme pathway between tumor and normal tissue, the PAP accumulation induced by ALA in tumor cells is higher than that in normal cells [5]. PAPs show a maximum light absorption around 410 nm and emit fluorescence with a broad peak at 635—705 nm. PAPs selectively concentrates in tumor cells and generates red fluorescence when excited by blue light revealing a distinct contrast to the surrounding normal tissues [1]. However, ALA is hydrophilic and does not easily penetrate through intact skin or cell membranes [6], thus the efficiency of PAP production is low. In order to overcome this problem, a number of ALA esters with more lipophilic property has been tested [7]. Hexaminolevulinate (HAL), gives a more rapid and higher increase of intracellular PAP and could enhance both fluorescence intensity and inter tissular contrast [4,8]. For diagnosis of superficial bladder cancer, Kriegmair et al. demonstrated that FD after intravesical instillation of HAL improved sensitivity of cystoscopy (n = 24/Se: 72% versus 97%) [9—11]. MAL is used as a topical photosensitizer in dermatology. For treatment of solar keratoses and basal cell carcinomas, clinical trials using MAL PDT demonstrated that the fluorescence of induced porphyrins is effective in detecting and delineating neoplasic skin areas [12,13]. The goal of this study was to assess the safety and efficacy of colposcopic FD of SIL after topical application of MAL.
113
Materials and methods Patients Each patient was referred to our colposcopic clinic with abnormal cytology. They all underwent a gynaecologic examination, including HPV testing, cytology, colposcopy and biopsy(ies). The cervical biopsies were assessed according to the cervical intraepithelial neoplasia (CIN) classification system as mild (CIN1), moderate (CIN2), or severe (CIN3) dysplasia. Patients with histologic proved CIN and planned for loop electrosurgical excision procedure (LEEP) under general anaesthesia, were included in a prospective study. Protocol of the study was approved by our local ethical committee. A written informed consent was obtained prior to inclusion for each patient.
Photosensitizer Metvix® cream (MAL) is a topical photosensitizer that is commercialy available [14]. The surface of the cervix was prepared by washing with saline solution. Half of the MAL tube (1 g) diluted with 2 mL of sterile saline solution was used for each patient. Using a spatula, a 1 mm thick layer of MAL cream was applied on the cervix. Application of the cream was performed before general anesthesia, FD and LEEP.
Fluorescence colposcopic examination Under general anesthesia and after cream application of 35—150 min, fluorescent colposcopic inspection of the cervix was performed. A rigid 10-mm laparoscope (Karl Storz, Tuttlingen, Germany) connected to an increased red color sensitivity camera (Karl Storz Endovision Tricam), was introduced in the vaginal cavity through the speculum trocart to visualize the uterine cervix. The D-light AF system (Karl Storz Endoskope) was used to perform both conventional and fluorescence examination. Using this system, light emission could be switched easily from conventional whitelight mode to fluorescence blue-light mode (wavelength of 380—440 nm) with a foot pedal. The color video camera was connected to a DVCAM Video Recorder (DVR 30, Sony, Tokyo, Japan) in order to analyze the video images off-line. Visual observation of fluorescence was examined on the TV screen. Fluorescence of the cervix was defined as negative or positive. A mapping of fluorescent foci was recorded for each patient. In order to confirm neoplasic status, all fluorescent foci underwent directed punch biopsy(ies). Then, LEEP was performed in order to remove transformation zone and to treat SIL. All biopsies and piece of conization were submitted for histologic examination. Histologic results were classified into the most severe category of the CIN terminology. For each patient, fluorescence profiles were compared with the histology of the punch biopsies taken from the corresponding fluorescent foci, with the histology of conization and with the initial classical colposcopic examination. Patients were monitored for systemic and local toxicity with a postoperative examination at 6—8 h and at 4—6 weeks. They were also followed up at 3 months and after, at 6
114 Table 1
P. Collinet et al. Patients’ characteristics
Patients
Cytology
Histology (colposcopy)
Time (min)
Fluorescence
Histology (PDD)
Histology (LEETZ)
1 2 3 4 5 6 7 8 9 10 11 12 13 14
HSIL HSIL LSIL ASCUS ASCUS HSIL LSIL ASCUS LSIL HSIL LSIL LSIL ASCUS HSIL
CIN3 CIN2 CIN1 CIN3 CIN2 CIN1 CIN2 CIN1 CIN2 CIN2 CIN2 CIN2 CIN2 CIN3
35 39 40 40 50 59 60 74 77 82 90 110 120 150
Negative Positive Negative Negative Negative Positive Positive Positive Positive Positive Positive Positive Positive Positive
NP CIN3 NP NP NP CIN2 CIN1 Negative CIN3 CIN2 CIN3 CIN2 Negative CIN1
CIN2 CIN3 CIN1 CIN3 CIN2 CIN2 CIN1 CIN3 CIN3 CIN2 CIN3 CIN3 Negative CIN1
NB—–CIN: cervical intraepithelial neplasia; HSIL: high grade squamous intraepithelial lesion; LSIL: low grade squamous intaepithelial lesion; ASCUS: atypical squamous cells of undeterminated significance.
months intervals with HPV testing, cytologic smear test, colposcopic assessment, and, if indicated, colposcopically directed punch biopsy of the cervix.
Results Between June 2005 and March 2006, 14 patients were included in the study. The mean age was 37.5 years (27—54). Initial cytology revealed four patients with ASCUS (28.6%), five with low grade SIL (LSIL) (35.7%) and five with high grade SIL (HSIL) (35.7%) (Table 1). During initial classical colposcopic examination, the transformation zone was not completely visible in four cases (28.6%). All colposcopical examinations were abnormal with areas of atypical transformation zone (TAG): 28.6% of TAG grade 1 (moderate colposcopic modifications of the transformation zone) and 71.4% of TAG grade 2 (severe colposcopic modifications of the transformation zone). Histology from punch biopsies preformed under classical colposcopy showed three patients with CIN1 (21.4%), eight with CIN2 (57.1%) and three with CIN3 (21.4%). Seven patients underwent high risk HPV testing preoperatively (50%) that was positive in all cases. Using the endoscope associated with D-light AF system, colposcopic fluorescence imaging proved to be applicable for real time diagnosis of SIL of the cervix. Red fluorescent foci were observed in 71.4% of cases (10/14) (Figs. 1 and 2). In these 10 cases, foci of red fluorescence of suspected SIL were clearly identified from the blue fluorescence of normal ectocervix. Neoplasic status of fluorescent foci was confirmed in 80%: two CIN1 (25%), three CIN2 (37.5%) and three CIN3 (37.5%) (Table 1). Concerning ME-ALA, the mean of application time was 73 min (35—150). With application time <45 min, foci of fluorescence were observed only in one case out of four (25%). With application time >75 min, foci of fluorescence were observed only in 100% (6/6). Two cases presented a false-positive fluorescence. One case presented a false-positive fluorescence because histology of conization revealed no SIL (probably all removed by the initial biopsy). The other case presented a fluores-
Figure 1 Example no. 1 of fluorescent foci (case no. 10); 1 mm layer of cream (500 mg of Me—–ALA); 75 min before colposcopic examination; final histology was CIN2.
cent focus with negative histology of the biopsy and positive histology of conization. We observed four cases of false-negative fluorescence: one patient with a histology of CIN1, two with a histology of CIN2 and one with a histology of CIN3 (Table 1). For one case, we experienced difficulties in applying the cream due to strong retroversion of the uterus. For another case, SIL was located in the endocervix and transformation zone was not visible under white light colposcopy. For all of cases of false-negative fluorescence, application time of MAL cream was less than 60 min. We observed no toxicity of MAL application (no local or systemic side effect).
Discussion The objective of the study was to assess feasibility of colposcopic FD, using different MAL application times, of SIL. In addition, we demonstrate the successful identifi-
Fluorescence diagnosis of cervical squamous intraepithelial lesions: A clinical feasability study
Figure 2 Example no. 2 of fluorescent foci (case no. 12); 1 mm layer of cream (500 mg of Me—–ALA); 120 min before colposcopic examination; final histology was CIN3.
cation of dysplastic lesions. The main parameter was the confirmation of dysplastic status of fluorescents foci by histology. We observed a higher porphyrin fluorescence intensity within the CIN lesions, as compared to normal adjacent epithelium. This result was previously described by Andikyan et al. [15]. They also observed that fluorescence selectivity was the highest in CIN3. Weingandt et al. also demonstrated that fluorescence intensity was correlated to the severity of CIN [16]. These results demonstrated that FD of SIL using topical application of MAL is feasible. The current study was designed as a dose-finding feasability study. All patients were planned for conization due to histologic proved SIL. For this reason, we observed one patient with false-positive fluorescence (no residual lesion). Because of the absence of true-negative case in this study, we were able to calculate only sensitivity and positive predictive value. In the study, colposcopic FD had a sensitivity of 69% and positive predictive value was 90%. In order to compare sensitivity and specificity of classical and FD colposcopies, next step would be to include only patients with pathologic cytology and normal classical colposcopy. The successful outcome of FD depends on parameters including dose of MAL, time interval between administration and light emission and wavelength and source of light. Concerning the photosensitizer, we selected Metvix® for this study because of its European approval in Dermatology. We observed the feasibility of using MAL cream as a drug for topical application on the cervix and for colposcopic FD. PAP precursors penetrate the normal and tumoral epithelium and induce PAP fluorescence. We did observe selective PAP accumulation in neoplasic tissue as reported by Hillemanns et al. [17]. The selectivity is partially explained by a lower ferrochelatase activity in tumor cells and a higher tumor level of porphobilinogen deaminase, which are the two key enzymes in the regulation of the heme pathway [4]. Thus, the photosensitizing effects of PAP produced by topical application of PAP precursor show a high degree of tissue specificity. ALA has been studied extensively because it is a safe and effective photosensitizing agent [18]. The results for FD and
115
PDT of SIL/CIN are variable, and to increase efficacy—–new and more potent photosensitizers should be tested. More lipophilic derivatives of ALA, such as MAL, with better tissue penetration (bioavailability suggests systemic bioavailability) should therefore show good performance. MAL and HAL were found to be more efficient than ALA itself to induce PAP in some cell lines in vitro [6] and in tumor in vivo in urology [19,20]. Recently, it has been shown that esterification of ALA into more lipophilic derivatives results in up to a 25-fold increase in PAP levels in malignant cells in culture [6] and a more homogeneous distribution across a normal pig urothelium in an in vitro model [21]. Concerning time application, the importance of this parameter was previously reported by Hillemans et al. [22]. Using topical application of 1% ALA, the optimal time interval was 60—90 min. In this study, no fluorescence could be diagnosed with intervals less than 30 min. Using ex vivo fluorescence microscopy, application of HAL 0.5% cream to the cervix induced selective fluorescence in epithelial cells of the cervix [23]. The optimal ratio with a homogeneous PAP distribution was obtained after 100 (±10) min cream application. Using MAL, we observed the same result. For the four cases of false-negative fluorescence, time intervals were 35, 40, 40 and 50 min, respectively. These three studies confirmed the need of a minimal interval between application and exploration of 60 min. In order to improve PAP distribution, Duska et al. demonstrated the feasibility of administering ALA per os (10 mg/kg) in 14 patients with abnormal Pap smears [24]. No systemic side effects were recorded. They observed an optimal fluorescence after 3 h. Since this publication in 2002, no study was reported in the literature. This result needs to be confirmed and compared to topical application. It could improve patient’s comfort compliance. Finally, several studies demonstrated potential advantage of photodynamic therapy (PDT) for the treatment of SIL with reduced risk of bleeding after conization, and more favorable long-term results avoiding cervical stenosis and preterm labor [22,25—30]. The potential of FD and PDT is promising but still experimental in gynaecology. This technique could permit to ‘‘see and treat’’ in real time without invasive procedure.
Conclusion Using topical application of MAL cream, we demonstrated that FD of SIL is feasible with promising efficacy and no local or systemic adverse events. A minimal application time of 60 min is required for optimal results. This study justifies further work to develop classical white light colposcopy including fluorescence imaging.
References [1] Gahlen J, Stern J, Laubach HH, Pietschmann M, Herfarth C. Improving diagnostic staging laparoscopy using intraperitoneal lavage of delta-aminolevulinic acid (ALA) for laparoscopic fluorescence diagnosis. Surgery 1999;126(September (3)):469—73. [2] Kriegmair M, Baumgartner R, Lumper W, Waidelich R, Hofstetter A. Early clinical experience with 5-aminolevulinic acid for the photodynamic therapy of superficial bladder cancer. Br J Urol 1996;77(May (5)):667—71.
116 [3] Sabban F, Collinet P, Cosson M, Mordon S. Fluorescence imaging technique: diagnostic and therapeutic interest in gynecology. J Gynecol Obstet Biol Reprod (Paris) 2004;33(December (8)):734—8. [4] Kennedy JC, Pottier RH. Endogenous protoporphyrin IX, a clinically useful photosensitizer for photodynamic therapy. J Photochem Photobiol B 1992;14(July (4)):275—92. [5] Peng Q, Warloe T, Berg K, et al. 5-Aminolevulinic acid-based photodynamic therapy. Clinical research and future challenges. Cancer 1997;79(June (12)):2282—308. [6] Gaullier JM, Berg K, Peng Q, et al. Use of 5-aminolevulinic acid esters to improve photodynamic therapy on cells in culture. Cancer Res 1997;57(April (8)):1481—6. [7] Brunner H, Hausmann F, Knuechel R. New 5-aminolevulinic acid esters—–efficient protoporphyrin precursors for photodetection and photodynamic therapy. Photochem Photobiol 2003;78(November (5)):481—6. [8] Kloek J, Akkermans W, Beijersbergen van Henegouwen GM. Derivatives of 5-aminolevulinic acid for photodynamic therapy: enzymatic conversion into protoporphyrin. Photochem Photobiol 1998;67(January (1)):150—4. [9] Kriegmair M, Baumgartner R, Knuechel R, et al. Fluorescence photodetection of neoplasic urothelial lesions following intravesical instillation of 5-aminolevulinic acid. Urology 1994;44(December (6)):836—41. [10] Kriegmair M, Stepp H, Steinbach P, et al. Fluorescence cystoscopy following intravesical instillation of 5-aminolevulinic acid: a new procedure with high sensitivity for detection of hardly visible urothelial neoplasias. Urol Int 1995;55(4):190—6. [11] Kriegmair M, Zaak D, Knuechel R, Baumgartner R, Hofstetter A. Photodynamic cystoscopy for detection of bladder tumors. Semin Laparosc Surg 1999;6(June (2)):100—3. [12] Foley P. Clinical efficacy of methyl aminolevulinate (Metvix) photodynamic therapy. J Dermatol Treat 2003;14(Suppl. 3):15—22. [13] Freeman M, Vinciullo C, Francis D, et al. A comparison of photodynamic therapy using topical methyl aminolevulinate (Metvix) with single cycle cryotherapy in patients with actinic keratosis: a prospective, randomized study. J Dermatol Treat 2003;14(June (2)):99—106. [14] Gardlo K, Ruzicka T. Metvix (PhotoCure). Curr Opin Investig Drugs 2002;3(November (11)):1672—8. [15] Andikyan V, Kronschnabl M, Hillemanns M, Wang X, Stepp H, Hillemanns P. Fluorescence diagnosis with 5-ALA thermogel of cervical intraepithelial neoplasia. Gynakol Geburtshilfliche Rundsch 2004;44(January (1)):31—7. [16] Weingandt H, Stepp H, Baumgartner R, Diebold J, Xiang W, Hillemanns P. Autofluorescence spectroscopy for the diagnosis of cervical intraepithelial neoplasia. Bjog 2002;109(August (8)):947—51. [17] Hillemanns P, Weingandt H, Baumgartner R, Diebold J, Xiang W, Stepp H. Photodetection of cervical intraepithelial neoplasia using 5-aminolevulinic acid-induced porphyrin fluorescence. Cancer 2000;88(May (10)):2275—82.
P. Collinet et al. [18] Kennedy JC, Pottier RH, Pross DC. Photodynamic therapy with endogenous protoporphyrin IX: basic principles and present clinical experience. J Photochem Photobiol B 1990;6(June (1—2)):143—8. [19] Lange N, Jichlinski P, Zellweger M, et al. Photodetection of early human bladder cancer based on the fluorescence of 5aminolaevulinic acid hexylester-induced protoporphyrin IX: a pilot study. Br J Cancer 1999;80(April (1—2)):185—93. [20] Marti A, Jichlinski P, Lange N, et al. Comparison of aminolevulinic acid and hexylester aminolevulinate induced protoporphyrin IX distribution in human bladder cancer. J Urol 2003;170(August (2 Pt 1)):428—32. [21] Marti A, Lange N, van den Bergh H, Sedmera D, Jichlinski P, Kucera P. Optimisation of the formation and distribution of protoporphyrin IX in the urothelium: an in vitro approach. J Urol 1999;162(August (2)):546—52. [22] Hillemanns P, Korell M, Schmitt-Sody M, et al. Photodynamic therapy in women with cervical intraepithelial neoplasia using topically applied 5-aminolevulinic acid. Int J Cancer 1999;81(March (1)):34—8. [23] Andrejevic-Blant S, Major A, Ludicke F, et al. Time-dependent hexaminolaevulinate induced protoporphyrin IX distribution after topical application in patients with cervical intraepithelial neoplasia: a fluorescence microscopy study. Lasers Surg Med 2004;35(4):276—83. [24] Duska LR, Wimberly J, Deutsch TF, et al. Detection of female lower genital tract dysplasia using orally administered 5aminolevulinic acid induced protoporphyrin IX: a preliminary study. Gynecol Oncol 2002;85(April (1)):125—8. [25] Ichimura H, Yamaguchi S, Kojima A, et al. Eradication and reinfection of human papillomavirus after photodynamic therapy for cervical intraepithelial neoplasia. Int J Clin Oncol 2003;8(October (5)):322—5. [26] Bodner K, Bodner-Adler B, Wierrani F, et al. Coldknife conization versus photodynamic therapy with topical 5-aminolevulinic acid (5-ALA) in cervical intraepithelial neoplasia (CIN) II with associated human papillomavirus infection: a comparison of preliminary results. Anticancer Res 2003;23(March—April (2C)):1785—8. [27] Barnett AA, Haller JC, Cairnduff F, Lane G, Brown SB, Roberts DJ. A randomised, double-blind, placebo-controlled trial of photodynamic therapy using 5-aminolaevulinic acid for the treatment of cervical intraepithelial neoplasia. Int J Cancer 2003;103(Match (6)):829—32. [28] Keefe KA, Tadir Y, Tromberg B, et al. Photodynamic therapy of high-grade cervical intraepithelial neoplasia with 5-aminolevulinic acid. Lasers Surg Med 2002;31(4):289—93. [29] Wierrani F, Kubin A, Jindra R, et al. 5-aminolevulinic acidmediated photodynamic therapy of intraepithelial neoplasia and human papillomavirus of the uterine cervix—–a new experimental approach. Cancer Detect Prev 1999;23(4):351—5. [30] Muroya T, Suehiro Y, Umayahara K, et al. Photodynamic therapy (PDT) for early cervical cancer. Gan To Kagaku Ryoho 1996;23(January (1)):47—56.