Clinical mapping of conjunctival melanomas

Transcription

1 1 Liverpool Oular Onology Centre, St Paul s Eye Clini, Royal Liverpool University Hospital, Liverpool, UK; 2 Departments of Pathology/ Ophthalmology, University of Liverpool, Liverpool, UK Correspondene to: Professor B Damato, Oular Onology Servie, Royal Liverpool University Hospital, Presot Street Liverpool L7 8XP, UK; bertil@damato.o.uk Presented at the International Conferene on Oular Onology, Siena, Italy, June 2007 and at the European Vision and Eye Researh (EVER) Assoiation meeting, Portoroz, Slovenia, Otober Aepted 31 Otober 2007 Published Online First 31 July 2008 Clinial mapping of onjuntival melanomas B Damato, 1 S E Coupland 2 ABSTRACT Aim: To develop a system for linially mapping onjuntival melanomas, taking into aount the tumour s stage of development, loation and extent. Methods: Baseline linial features of 40 patients with onjuntival melanoma, first treated at our entre between 1993 and 2006, were studied. Maps showing setorial loation and irumferential spread in all areas of onjuntiva were digitally generated from numerial data. Results: The patients (18 female; 22 male) had a median age of 59 years. Three had multiple tumours and 20 had mixed, invasive and intraepithelial disease. The Tumour Node and Metastasis (TNM) T stage was I in 11 patients, II in 20, III in eight and IV in one. The tumour was medial in 19% of eyes with T stages I and II as ompared with 67% of eyes with stages III and IV (Mann Whitney,,0.001). Within eah T stage of the TNM system, there was onsiderable variation in tumour dimensions, irumferential spread and longitudinal extent. Conlusions: A system for linially mapping onjuntival tumour loation and extent has been developed. This revealed tumour variation within TNM T stages and may failitate better olletion of data for multientre trials. Conjuntival melanomas are rare. 1 They may be assoiated with primary aquired melanosis (PAM), whih we prefer to all onjuntival in situ melanoma. Exenteration has been superseded by loal exision of any nodules together with adjuntive therapy onsisting of ryotherapy, topial hemotherapy and radiotherapy, individually or in ombination. 2 7 Loal reurrene is ommon. 8 Metastasis an our to regional nodes and systemially. Sentinel node biopsy is ontroversial. 9 The 10- year melanoma-related mortality is 25 30% and is higher with involvement of non-bulbar onjuntiva, loal tumour reurrene, multifoal tumours, inreased tumour thikness, high mitoti rate, epithelioid ells and lymphati invasion The sixth version of the Tumour Node and Metastasis (TNM) staging system of the Amerian Joint Committee on Caner/International Union Against Caner (AJCC/UICC) ategorises onjuntival tumours only aording to their degree of invasion and radial extent but has not gained widespread aeptane. 14 The aim of this study was to develop a system for mapping onjuntival melanomas. PATIENTS AND METHODS Patients were inluded in this study if treated for onjuntival melanoma between January 1993 and Deember They were exluded if they had undergone any surgery before referral or if the harts were no longer available (one patient). Oular examination inluded visual auity and intraoular pressure measurements, inspeting the Innovations external eye, anterior segment and fundus with slit-lamp biomirosopy and indiret ophthalmosopy. Eah setor of the bulbar and inferior tarsal onjuntiva was examined by retrating the eyelid and asking the patient to look in the opposite diretion. The superior tarsal onjuntiva and fornix were examined with a magnifying lens, gently pulling the eyelid away from the eye (hene avoiding double eversion of the eyelid). If a more detailed examination of the superior tarsal onjuntiva was indiated, this was done with a slitlamp after everting the upper lid. Regional lymph nodes were palpated routinely. Patients were photographed routinely. Towards the end of the study, a formal protool was instituted, imaging eah setor of the onjuntiva in turn, with the patient looking away from that setor, retrating the eyelids to expose the fornies, and everting the eyelids to photograph the superior and inferior palpebral onjuntiva. A ruler was held against the lower lid to alibrate photographs. The patients were identified by searhing the omputerised database of the Liverpool Oular Onology Centre and arhives of the pathology department at the Royal Liverpool University Hospital. All ase notes, omputerised reords, histologial setions and linial photographs were reviewed to define tumour loation and extent. If not doumented adequately in the ase notes at the time of initial assessment, measurements suh as basal dimensions and distanes from limbus and lid margin were made from photographs. Calibration for suh measurements was based on orneal diameter, whih was assumed to be 12 mm. Tumour thikness and width were determined histologially. 15 Setions were stained using H&E and immunohistohemially. Clinial and histologial data were summarised on proformas and omputerised in the oular onology database, whih was ustomised for onjuntival tumours. The data were then exported to SPSS (Chiago) for analysis. Tumour loation and extent were mapped using an Exel spreadsheet (Mirosoft Windows, Seattle, WA). This was done by generating a table in SPSS, with a sore of 1 or 0 alloated to eah lok hour of every region of onjuntiva, aording to whether or not this was affeted by the tumour in question. This was done for up to three tumours per patient. The tables were transferred from SPSS to Exel so that bubble plots ould be generated, with invasive melanoma depited in grey and pigmented intraepithelial disease with a striped texture. Cumulative bubble plots with data from groups of patients (eg, with or without metastati death) were generated, with the bubble sizes representing either the number or the proportion of patients in the group having invasive tumour in the relevant onjuntival area. Br J Ophthalmol: first published as /bjo on 31 July Downloaded from on 4 July 2018 by guest. Proteted by opyright. Br J Ophthalmol 2008;92: doi: /bjo

2 The entre of the onjuntiva was defined as the entral ornea, with limbus, bulbar onjuntiva, fornix, tarsal onjuntiva, eyelid margin and skin onsidered as being progressively more peripheral. The plia was regarded as being an extension of fornieal onjuntiva, and the arunle was onsidered to be at the same eentriity as tarsal onjuntiva. Cirumferential spread along limbus, bulbar onjuntiva, fornix, tarsal onjuntiva, eyelid margin and adjaent skin was measured in lok minutes (whih are different from minutes of ar). Longitudinal and transverse basal tumour dimensions were doumented in millimetres. For tumours involving bulbar onjuntiva, distane between entral tumour margin and limbus was measured in millimetres, with negative values if the tumour extended into ornea. For tarsal tumours, distane between peripheral tumour margin and inner eyelid margin was measured, with negative values if this margin was loated within onjuntiva and positive values if it was in skin. Our proforma was designed to store details on up to five tumours per patient. The largest tumour was regarded as the primary lesion. Invasive melanoma and intraepithelial disease in the same eye were onsidered separately, even if onfluent. Clinially, only pigmented intraepithelial melanoyti proliferation ould be mapped, with suh primary aquired melanosis (PAM) subsequently ategorised histologially as (1) PAM without atypia, (2) PAM with atypia resulting in invasive melanoma or (3) seondary, diffuse, superfiial spread from a nodular melanoma. We now refer to PAM without atypia and PAM with atypia as onjuntival melanoyti hyperplasia and onjuntival in situ melanoma respetively. Invasive tumours were defined as being pigmented or amelanoti. Seondary effets inluded features suh as feeder vessels and inflammation. The Tenets of the Helsinki Delaration were followed. Patient onsent was routinely obtained for the use of data, images and tissues for researh, audit and teahing. Ethial Committee approval was not required. RESULTS Baseline data The ohort omprised 40 patients (18 female; 22 male) with a median age of 59 years (range 24 to 88 years). The tumour involved the right eye in 18 patients and the left eye in 22. Three patients had multiple tumours, and 20 had mixed tumours (ie, invasive melanoma and intraepithelial disease) (fig 1). The entre of the largest tumour was loated in the following setors: superior (3); supero-temporal (1); temporal (18); inferotemporal (4); inferior (2); infero-nasal (4); nasal (4); and superonasal (4). The median largest basal tumour diameter was 6.8 mm (range 3 to 25), and the median tumour thikness was 2.2 mm (range 0.1 to 15.5). The entral tumour margin was loated in ornea (20 patients), limbus (6), bulbar onjuntiva (12), tarsal onjuntiva (1) and arunle (1). The peripheral tumour margin was loated in bulbar onjuntiva (31), plia (1), arunle (5), tarsal onjuntiva (1), eyelid margin (1) and skin (1). Six tumours involved the arunle, with five of these also involving the plia. Non-statistially, bubble plots suggested no great interoular or gender differenes but indiated that medial onjuntival involvement orrelated with loal tumour reurrene and metastati death (fig 2). The TNM stage was: I (11 patients); II (20 patients); III (eight patients); and IV (one patient). Figure 3 shows the basal tumour dimensions, tumour height, irumferential tumour extent and, for bulbar tumours, the tumour-to-limbus distane, aording to TNM stage. In stages T1 and T2, the tumour loation was medial in six (19%), A linial mapping system for onjuntival melanoma Displays the entire onjuntiva as a flat surfae, with the entral point loated at the entre of the ornea and onentri regions suh as limbus, bulbar onjuntiva, fornix, palpebral onjuntiva and eyelid onsidered progressively more peripheral. Plots whether or not eah lok hour of every region is affeted by invasive melanoma (solid fill) or intraepithelial disease (hathed drawing). Maps are drawn digitally from numerial data, allowing umulative plots omparing groups of patients. Clinial mapping improves doumentation of tumour size, loation and extent, and should therefore failitate data olletion for multientre trials. midline in two (7%) and temporal in 23 (74%) as ompared with medial in six (67%) and midline in three (33%) in stages T3 and T4 (Mann Whitney U = 39; z = 3.68; p,0.001). DISCUSSION We have developed a system for mapping onjuntival melanomas linially (table 1). Tumours were drawn from numerial data, whih enabled umulative plots to be generated, thereby allowing different subgroups of patients to be ompared visually. Staging of aner progression at the time of initial treatment is fundamental to prognostiation, treatment seletion and linial trials. The sixth version of the TNM staging system lassifies onjuntival melanomas as: TI for tumours involving only the bulbar onjuntiva; TII for tumours with orneal involvement; TIII for tumours involving fornix, arunle or tarsal onjuntiva; and TIV for tumours invading the eye, orbit, entral nervous system or skin. Our study revealed shortomings of the TNM staging system. First, our umulative plots (fig 2) showed how loal tumour reurrene and metastasis orrelated with involvement of medial onjuntiva. Seond, as shown in fig 3, our mapping demonstrated that within eah T stage, there was marked variation in tumour dimensions, irumferential spread and longitudinal extent, whih are known to have prognosti signifiane. We have other onerns regarding the TNM staging system. First of all, we are not aware of any evidene indiating that tumours involving ornea have a worse prognosis than those ourring only in bulbar onjuntiva; if anything, the prognosis should be better, beause of the absene of lymphatis. 18 Seond, we intuitively suspet that tumours invading the entral nervous system have a worse prognosis from those involving only the eyelid skin. For all these reasons, we suspet that important prognosti information is not fully utilised by the TNM system. The TNM system has reently been revised, and the next edition 19 will take aount of our findings. Large, multientre studies are needed to determine how irumferential tumour loation and spread orrelate with survival and how ontinuous variables should be ategorised. Suh studies will require aurate doumentation of tumour loation and extent, whih we hope will be failitated by our proposed mapping system. Several authors have shown a orrelation between nonepibulbar tumour loation and metastati disease, but without measuring irumferential spread. In their evaluation of risk fators for metastati death and other outomes, Shields et Br J Ophthalmol: first published as /bjo on 31 July Downloaded from on 4 July 2018 by guest. Proteted by opyright Br J Ophthalmol 2008;92: doi: /bjo

3 Figure 1 Maps of 40 onjuntival melanomas, showing omputer-generated plots of tumour loation and extent with invasive melanoma in dark grey and intraepithelial disease in light grey, and reurrent disease (manually drawn) in blak. Eah spot represents involvement of a setor of onjuntiva and not a separate tumour. al look at irumferential spread, but only in bulbar onjuntiva and not in fornix and tarsal onjuntiva, whih seem more relevant to survival probability. 13 We measured irumferential spread in all onjuntival regions: limbus, bulbar onjuntiva, palpebral onjuntiva, eyelid margin and skin. Suh data are likely to be useful not only in prediting survival but also in evaluating oular results following treatment of the onjuntival tumour. We preferred lok minutes to lok hours, beause lok minutes may be more preise, also avoiding the need for frations or deimal points. The number of patients in our study is small ompared with other investigations. This is beause, unlike other studies, we exluded all patients who had undergone any onjuntival tumour surgery elsewhere before referral to our entre. There are several reasons for this. First, in many suh patients, we were unable to define the presenting tumour, beause it had already been exised. Seond, patients referred with reurrent tumours may have had more aggressive disease so that they may have introdued seletion bias into our results. Third, outomes may have been worsened by suboptimal surgery elsewhere, as previously suggested. 20 A shortoming of our study is that the tumour features at initial presentation were defined retrospetively, and this was beause we oneived our mapping system only reently. Nevertheless, with the exeption of the tumour-to-lid-margin, we felt we were able to desribe most tumours adequately from the patient reords and photographs. Another limitation is that the patients referred to us with untreated onjuntival melanoma may not have been representative of all individuals developing this disease. Prospetive, multientre studies are needed, with patients being examined and photographed aording to a standardised protool. Suh a protool would also need to speify how primary aquired melanosis should be defined and graded, and how histologial data should be added to any linial maps. The omputer-generated maps shown in fig 1 showed only whether a lok hour of a partiular onjuntival region was involved by tumour. Suh a low resolution did not allow a full desription of tumour shape and did not define the preise loation of the entral, peripheral and lateral tumour margins. Furthermore, it did not identify two distint tumours if these were loated lose together. More detailed mapping using our Br J Ophthalmol: first published as /bjo on 31 July Downloaded from on 4 July 2018 by guest. Proteted by opyright. Br J Ophthalmol 2008;92: doi: /bjo

4 Figure 2 Tumour loation and extent aording to (A) laterality; (B) sex; (C) loal tumour reurrene; and (D) metastati death. In (B), (C) and (D), tumours in the left eye are transposed to the right eye. In (A) and (B), the bubble size represents the number of ases, and in (C) and (D), the bubble size represents the proportion of ases. urrent methods would be feasible but may prove exessively laborious. We are hoping to develop a more user-friendly program for mapping onjuntival tumours more preisely. This will hopefully take into aount not only the urvature of the eye but also the maximal radial onjuntival dimensions that our with eentri gaze. We measured tumour thikness from histologial setions, whih is not ideal beause of a slight redution in tissue volume after fixation and beause of the risk of tangential uts. However, in the vast majority of ases, orthograde setions were ahieved, providing aurate measurements. In the ase of onjuntval melanomas assoiated with surrounding inflammation, immunohistohemial staining Br J Ophthalmol: first published as /bjo on 31 July Downloaded from on 4 July 2018 by guest. Proteted by opyright. Figure 3 Basal tumour diameter and tumour height, tumour-to-limbus distane, and irumferential tumour extent aording to the Tumour Node and Metastasis T stage. Only one patient (ase 17) had a TIV tumour (see fig 2). Tumours in the left eye were transposed to the right eye Br J Ophthalmol 2008;92: doi: /bjo

5 highlighted the tumour ell population, enabling better identifiation of the tumour dimensions. A method using highfrequeny ultrasonography has been desribed. 21 We have reently used this tehnique to measure bulbar tumours, but further evaluation is needed, espeially with non-bulbar tumours. We have not reported outomes, whih were not relevant to the desription of our mapping system. These are published elsewhere. 22 The umulative plots in fig 2 were shown only to demonstrate the potential of our mapping system. Our treatment outomes, inluding loal tumour reurrene and metastati death, have been reported elsewhere. In onlusion, mapping may failitate data olletion for multientre trials. There is sope for evaluating and refining this system, not only with melanomas but also with other onjuntival aners, ideally in several different entres. Funding: The Oular Onology Servie is funded by the National Commissioning Group of the National Health Servie, whih had no role in the design or ondut of the researh. Competing interests: None. Patient onsent: Obtained. REFERENCES 1. Seregard S. Conjuntival melanoma. Surv Ophthalmol 1998;42: Shields JA, Shields CL, De Potter P. Surgial management of onjuntival tumors. The 1994 Lynn B. MMahan Leture. Arh Ophthalmol 1997;115: Stannard CE, Sealy GR, Hering ER, et al. Malignant melanoma of the eyelid and palpebral onjuntiva treated with iodine-125 brahytherapy. Ophthalmology 2000;107: Demiri H, MCormik SA, Finger PT. Topial mitomyin hemotherapy for onjuntival malignant melanoma and primary aquired melanosis with atypia: linial experiene with histopathologi observations. Arh Ophthalmol 2000;118: Finger PT. Finger-tip ryotherapy probes: treatment of squamous and melanoyti onjuntival neoplasia. Br J Ophthalmol 2005;89: Wuestemeyer H, Sauerwein W, Meller D, et al. Proton radiotherapy as an alternative to exenteration in the management of extended onjuntival melanoma. Graefes Arh Clin Exp Ophthalmol 2006;244: Fruht-Pery J, Pe er J. Use of Mitomyin C in the treatment of onjuntival primary aquired melanosis with atypia. Arh Ophthalmol 1996;114: Missotten GS, Keijser S, de Keizer RJ, et al. Conjuntival melanoma in the Netherlands: a nationwide study. Invest Ophthalmol Vis Si 2005;46: Amato M, Esmaeli B, Ahmadi MA, et al. Feasibility of preoperative lymphosintigraphy for identifiation of sentinel lymph nodes in patients with onjuntival and perioular skin malignanies. Ophthal Plast Reonstr Surg 2003;19: Desjardins L, Ponet P, Levy C, et al. Prognosti fators in malignant melanoma of the onjuntiva. An anatomo-linial study of 56 patients. J Fr Ophtalmol 1999;22: Anastassiou G, Heiligenhaus A, Behrakis N, et al. Prognosti value of linial and histopathologial parameters in onjuntival melanomas: a retrospetive study. Br J Ophthalmol 2002;86: Wershnik C, Lommatzsh PK. Long-term follow-up of patients with onjuntival melanoma. Am J Clin Onol 2002;25: Shields CL, Shields JA, Gunduz K, et al. Conjuntival melanoma: risk fators for reurrene, exenteration, metastasis, and death in 150 onseutive patients. Arh Ophthalmol 2000;118: Sobin L, Wittekind C. TNM lassifiation of malignant tumours. 6th edn. New York: Wiley-Liss, Tuomaala S, Eskelin S, Tarkkanen A, et al. Population-based assessment of linial harateristis prediting outome of onjuntival melanoma in whites. Invest Ophthalmol Vis Si 2002;43: Keijser S, Missotten GS, Bonfrer JM, et al. Immunophenotypi markers to differentiate between benign and malignant melanoyti lesions. Br J Ophthalmol 2006;90: Glasgow BJ, MCall LC, Foos RY. HMB-45 antibody reativity in pigmented lesions of the onjuntiva. Am J Ophthalmol 1990;109: Tuomaala S, Aine E, Saari KM, et al. Corneally displaed malignant onjuntival melanomas. Ophthalmology 2002;109: Seregard S, MGovern H, Coupland SE, et al. TNM staging of onjuntival melanoma. In: Sobin LH, Wittekind Ch, eds. TNM lassifiation of malignant tumours. 7th edn. New York: Wiley Liss. In Press. 20. Shields CL. Conjuntival melanoma. Br J Ophthalmol 2002;86: Ho VH, Prager TC, Diwan H, et al. Ultrasound biomirosopy for estimation of tumor thikness for onjuntival melanoma. J Clin Ultrasound 2007;35: Damato B, Coupland SE. An audit of onjuntival melanoma treatment in Liverpool. Eye. Published Online First: 6 June doi: /eye Br J Ophthalmol: first published as /bjo on 31 July Downloaded from on 4 July 2018 by guest. Proteted by opyright. Br J Ophthalmol 2008;92: doi: /bjo