- Email: [email protected]
Squamous Carcinoma and Dysplasia of the Conjunctiva and Cornea
Squamous Carcinoma and Dysplasia of the Conjunctiva and Cornea An Analysis of 101 Cases Yacoub A. Yousef, MD, Paul T. Finger, MD Objective: To evaluate clinical and histopathologic factors of squamous conjunctival neoplasia associated with recurrence. Design: Retrospective, clinical case series. Participants: One hundred one eyes of 99 patients with squamous conjunctival dysplasia, carcinoma in situ, or squamous cell carcinoma. Methods: Review of the medical records, pathology reports, and color photographs. Main Outcome Measures: Demographic information, laterality, tumor size, extension, pathologic diagnosis, seventh edition of the American Joint Committee on Cancer Classification (AJCC) staging system stage, treatment methods, recurrence, and duration of follow-up. Results: Malignant squamous conjunctival neoplasia was seen most commonly in males at a median age of 71 years. Recurrences were seen in 12.9% (n ⫽ 13/101), with 92.3% occurring 6 to 12 months after primary treatment. Recurrence was not correlated significantly to age, gender, laterality, clinical appearance or focality of the tumor at presentation. However, tumors larger than 5 mm in diameter, tumors extending more than 2 mm onto the cornea, and tumors with local invasion (corneal, scleral, intraocular or orbital invasion) were associated with a higher risk of recurrence. Increasing AJCC T-stage was correlated strongly to the incidence of recurrence (P ⫽ 0.0006). Rates were 1.7% for Tis-staged tumors, 0% for T1- and T2-staged tumors, 34.3% for T3-staged tumors, and 50% for T4-staged tumors. Histopathologic diagnosis was correlated to recurrence (P ⫽ 0.037). None of the tumors defined histologically as dysplasia showed recurrence, whereas 12.8% of carcinoma in situ tumors and 22.2% of squamous cell carcinoma tumors recurred. Although the overall recurrence rate was 12.9%, the rate for tumors treated primarily at the authors’ center was 4%, significantly less than the recurrence rate in previously operated tumors (P ⫽ 0.0003). Lymph nodes demonstrated positive results in 1%, and in no patient did distant metastasis develop. Conclusions: Advanced AJCC T-stage, locally invasive tumors, and more pathologically aggressive tumors were at higher risk for recurrence. Inadequate initial therapy also was an important risk factor for recurrence. Treatment strategies should be affected by tumor staging at presentation. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Ophthalmology 2012;119:233–240 © 2012 by the American Academy of Ophthalmology.
Squamous conjunctival carcinoma (SCC) is the most common ocular malignancy.1,2 Both sight and life threatening, SCC has the potential to invade locally the cornea, sclera, uvea, eyelids, orbit, sinuses, and brain as well as to metastasize.3–5 Commonly used local treatment options for SCC include excision, excision with cryotherapy, and adjuvant topical chemotherapy.1–12 Although there are no published guidelines, in North America, most small tumors are treated by excision or by excision with adjuvant cryotherapy. Larger, multifocal, or aggressive tumors have required combinations of resection, adjuvant cryotherapy, and topical chemotherapy (i.e., interferon, mitomycin, 5-fluoracil, or cyclosporine) and electron-based external beam radiation therapy.2,8 –16 Recently, subconjunctival ranibizumab6,13 and external beam radiation therapy10,14 using photons, electrons, or protons have been proposed as palliative treatments for nonresponsive carcinomas of the eye, lids, and © 2012 by the American Academy of Ophthalmology Published by Elsevier Inc.
orbit.14 When all else fails, orbital exenteration, typically with adjuvant external beam radiation therapy, may be used.10.14 Surgical excision of corneal and conjunctival intraepithelial dysplasia as well as squamous cell carcinoma in situ has been associated with high rates of recurrence. Reported ranges include 17% to 24% for dysplasia and 30% to 41% for squamous cell carcinoma.15,18 This is likely because of tumor edges and deep margins that often are difficult to determine.19 For example, a tumor’s advancing edges often are light gray and avascular. This can be seen easily when the edges extend onto the clear corneal epithelium. However, with a backdrop of white sclera, a light gray avascular extension is close to or entirely invisible. In clinical practice, this can lead to underestimation of tumor size. To compensate, ophthalmic oncologists increasingly have used ISSN 0161-6420/12/$–see front matter doi:10.1016/j.ophtha.2011.08.005
233
Ophthalmology Volume 119, Number 2, February 2012
Figure 1. Tumor characteristics. A, Patient with multifocal conjunctival tumor (carcinoma in situ) showing 2 tumors (nasal and temporal) separated by ⬎5 mm tumor-free tissue. B, Focal nodular appearance of conjunctival tumor (squamous dysplasia). C, Diffuse conjunctival tumor (squamous cell carcinoma). D, Pigmented conjunctival carcinoma in situ.
combination therapy (e.g., cryotherapy, topical chemotherapy, irradiation).18 This study staged tumors according to the seventh edition of the American Joint Committee on Cancer (AJCC) staging system.20 This allowed for standardized comparison of methods of both diagnosis and treatment. It also allowed for standardized collection of information re-
garding the histopathologic diagnosis, tumor size, local invasion, lymph node involvement, and distant metastasis (Table 1). This information was used to study methods of treatment, incidence of local control, and recurrence rate and their relation to tumor characteristics. Herein, the authors’ experience in the treatment of 101 consecutive cases is presented.
Figure 2. Tumor size and corneal extension. A, Small conjunctival squamous dysplasia (⬍5 mm) with no corneal extension. B, Small conjunctival carcinoma in situ (⬍3 clock hours) with corneal extension (⬍2 mm). C, Medium conjunctival squamous cell carcinoma (⬎3 clock hours and ⬍6 clock hours) with corneal extension (⬎2 mm). D, Large conjunctival squamous cell carcinoma (⬎6 clock hours) and corneal extension (⬎2 mm).
234
Yousef and Finger 䡠 Squamous Carcinoma of the Conjunctiva and Cornea Table 1. Carcinoma of the Conjunctiva: Definition of Tumor, Lymph Node, Metastasis (TNM) Staging* Definitions Apply to Both Clinical and Pathologic Staging Primary tumor (T) TX T0 Tis T1 T2 T3 T4 T4a T4b T4c T4d Regional lymph nodes (N) NX N0 N1 Distant metastasis (M) MX M0 M1
Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ Tumor ⱕ5 mm in greatest dimension Tumor ⬎5 mm in greatest dimension, without invasion of adjacent structures Tumor invades adjacent stuctures, excluding the orbit Tumor invades orbit with or without further extension Tumor invades orbital soft tissues, without bone invasion Tumor invades bone Tumor invades adjacent paranasal sinuses Tumor invades brain Regional lymph nodes cannot be assessed No regional lymph node metastasis Regional lymph node metastasis Distant metastasis cannot be assessed No distant metastasis Distant metastasis
*Reprinted with permission from The AJCC Ophthalmic Oncology Task Force. Carcinoma of the conjunctiva. In: Edge SE, Byrd DR, Carducci MA, Compton CA, eds. AJCC Cancer Staging Manual. 7th ed. New York: Springer; 2009:531–3.20
Patients and Methods This study adhered to the tenets of the Declaration of Helsinki of 1975, as revised in 2000, and the Health Insurance Portability and Accountability Act of 1996 and was approved by the institutional review board of The New York Eye Cancer Center. The study was a retrospective case review of 101 eyes of 99 consecutive patients who were diagnosed pathologically with conjunctival squamous dysplasia, conjunctival carcinoma in situ, or SCC. Selection also required access to patient medical records, pathologic reports, and color photographs. Data included patient age, gender, laterality, visual acuity, tumor size, clinical appearance, corneal extension, tumor invasiveness (corneal, scleral, uveal, orbital, lymphatic, and distant metastasis), and AJCC stage for each tumor. Data recorded included the
pathologic diagnosis, primary and secondary treatment methods, recurrence rate, as well as duration of follow-up. Recurrence was recorded only if there was no clinical evidence of tumor for at least 6 months after the end of the planned treatment.
Focused Ophthalmic Examination At presentation, all patients underwent slit-lamp examination of all conjunctival surfaces (including eversion of the upper eyelid and tarsus). Slit-lamp photography was used to record all conjunctival surfaces. Periodic gonioscopic photography and ultrasound biomicroscopy was used to monitor for and document intraocular invasion. Palpation was used to examine the preauricular and submandibular lymph nodes. Tumors found adherent to the eye wall were selected for high-frequency ultrasound Table 2. Tumor Features Not Correlated to Recurrence Feature
Figure 3. Kaplan-Meier method showing the recurrence rate of malignant conjunctival squamous neoplasia over the time (months).
Gender Male Female Laterality Right eye Left eye Both eyes Focality Monofocal Multifocal Morphologic features Focal nodular Diffuse Pigmented
No. of Tumors
%
No. of Recurrences
%
P Value
67 32
67.7 32.3
9 4
13.4 12.5
0.9102
49 48 2
48.5 47.5 2.0
7 6
14.3 12.5
0.8215
78 23
77.2 22.8
7 6
8.9 26.1
0.0685
43 52 6
42.6 51.5 5.9
5 8 0
11.6 15.4 0
0.6435
235
Ophthalmology Volume 119, Number 2, February 2012 Table 3. Tumor Features That Are Statistically Associated with Higher Risk of Recurrence Feature
No. of Tumors
%
No. of Recurrences
%
P Value
59 7 1 32 2
58.4 6.9 1 31.7 2
1 0 0 11 1
1.7 0 0 34.4 50
0.0006
42 37 22
41.6 36.6 21.8
1 5 7
2.4 13.5 31.8
0.0176
25 73 5 1 2 1 0
24.8 72.3 5 1 2 1 0
0 9 3 0 1 0 0
0 12.3 60 0 50 0 0
0.0423
42 31
57.5 42.5
1 12
2.4 38.7
0.0009
26 39 36
25.7 38.6 35.6
0 5 8
0 12.8 22.2
0.0373
AJCC Tis T1 T2 T3 T4 Size (mm) Small (⬍ 5) Medium (5–10) Large (⬎10 mm) Invasion Conjunctiva Cornea Sclera Uvea Orbit Lymph node Metastasis Corneal extension (mm) ⬍2 ⬎2 Pathologic features Dysplasia Carcinoma in situ SCC
margin), excision with cryotherapy, topical chemotherapy (interferon, mitomycin, and subconjunctival ranibizumab), irradiation, and exenteration. Cryotherapy was performed with Finger-tip spatulated cryotherapy applicators (Dutch Ophthalmic Research Corporation, Zuidland, The Netherlands) that are capable of producing homogeneous freezing over large surface areas, allowing for more uniform treatment of tissue (compared with standard cryotherapy applicators).7
Clinical Characteristics
AJCC ⫽ Seventh edition of the American Joint Committee staging system; SCC ⫽ squamous cell carcinoma.
biomicroscopy imaging (UBM). Adherence was evaluated by proparacaine-assisted Q-tip palpation under slit-lamp observation.
Treatment Methods Local treatments of conjunctival tumors included primary excision (excision of the entire visible tumor in addition to a 2-mm safety
Multifocal tumors were defined as having at least 2 tumors separated by 5 mm of normal-appearing tissue (Fig 1). According to the clinical appearance of the tumor, they were classified into focal nodular tumors (thickened focal lesion with welldefined margins), diffuse tumors (flat lesion with ill-defined margins), or pigmented tumors (Fig 1). Tumor sizes were divided into small, medium, and large tumors. Small tumors were defined as those less than 3 clock hours (for limbal tumors) or less than 5 mm in largest diameter. Medium tumors were defined as those 3 to 6 clock hours (for limbal tumors) or 5 to 10 mm in largest diameter, and large tumors were defined as those more than 6 clock hours (for limbal tumors) or more than 10 mm in largest diameter (Fig 2). Furthermore, tumors with corneal extension were divided into 2 groups: group 1 included tumors with 2 mm or less of corneal extension beyond the limbus and group 2 included tumors with more than 2 mm of corneal extension beyond the limbus (Fig 2). Tumor invasiveness was defined as follows: (1) corneal invasion consisted of clinically detected tumor tissue within the corneal tissue that was confirmed by histopathologic evaluation of corneal biopsy, (2) scleral invasion consisted of tumor tissue detected by UBM to be within the anatomic location of scleral tissue, (3) intraocular invasion consisted of tumor tissue detected by UBM to be within the intraocular structures, and (4) orbital invasion consisted of tumor tissue detected within the orbit posterior to the conjunctiva by orbital magnetic resonance imaging or computed tomography scan. In addition, the AJCC staging system was used to classify all cases of conjunctival squamous neoplasia in this case series (Table 1).20 According to the AJCC staging system, the same definition of tumor, node, and metastasis staging apply for both clinical and pathologic staging; therefore a pathologic classification was added Table 4. Recurrent Cases of
Patient No.
Pathologic Diagnosis
Gender
American Joint Committee on Cancer Classification Stage
Size
Extension
Lymph Node Involvement
1 2 3 4 5 6 7 8 9 10 11 12 13
SCC SCC CIS SCC SCC CIS SCC CIS CIS SCC SCC SCC CIS
M F M M M M M F M f M M F
T3c T3c Tis T3c T3 T3c T3c T3c T3C T3c T3 T3 T4
Large Medium Large Large Large Medium Small Medium Large Large Medium Medium Large
Cornea Cornea Cornea Cornea Sclera Cornea Cornea Cornea Cornea Sclera Sclera Cornea Orbit
No No No No No No No No No No No No No
CIS ⫽ carcinoma in situ; EBRT ⫽ external beam radiation therapy; Pd-103 ⫽ palladium-103; SCC ⫽ squamous cell carcinoma. *Exenteration performed because of endophthalmitis. † Exenteration performed because of fibrous histiocytoma.
236
Yousef and Finger 䡠 Squamous Carcinoma of the Conjunctiva and Cornea based on full-thickness conjunctival biopsy to include squamous dysplasia, carcinoma in situ, and squamous cell carcinoma.20
Statistical Analysis Statistical analysis of tumor clinical appearance, size, invasiveness, histopathologic results, and AJCC stage were correlated to the recurrence rate of the tumor after a 6-month period of cure after the available methods of treatment (i.e., surgical excision, cryotherapy, topical chemotherapy, subconjunctival anti–vascular endothelial growth factor injection, and radiation therapy). The P value was measured using the exact Fisher test to test the predictive power of each factor.
Results One hundred one eyes from 99 patients with an age range of 37 to 96 years (median, 71 years) were studied. Lesions were present on the right eye in 49 patients and on the left in 48 patients, and 2 patients had bilateral disease. Sixty-seven percent of patients (n ⫽ 67/99) were male. Clinical appearance evaluation showed that 42.6% (n ⫽ 43/ 101) of SCC cases were focal nodular tumors, 51.5% (n ⫽ 52/101) were diffuse tumors, and 5.9% (n ⫽ 6/101) were pigmented tumors. Multifocality was seen in 22.8% (n ⫽ 33/101). The tumor was limited to the superior tarsus in 2% (n ⫽ 2/101) and was limited to the palpebral fornix in 1% (n ⫽ 1/101). Tumor was localized to bulbar conjunctiva in 24.8% (n ⫽ 25/101). Invasion onto the cornea was present in 72.3% (n ⫽ 73/101), invasion onto the sclera was present in 4.9% (n ⫽ 5/101), invasion onto the uvea was present in 1% (n ⫽ 1/101), invasion onto the orbit was present in 2% (n ⫽ 2/101), and lymph node metastases were present in 1% (n ⫽ 1/101). In no patients did distant metastasis develop. In addition, corneal invasion was wider than 2 mm in 30.7% (n ⫽ 31/73). Histopathologic analysis revealed conjunctival squamous dysplasia (n ⫽ 26/101; 25.7%), carcinoma in situ (n ⫽ 39/101; 38.6%), and squamous cell carcinoma (n ⫽ 36/101; 35.6%). According to AJCC classification (which correlates to the pathologic diagnosis, size of the tumor, and tumor invasiveness; Table 1), in this series, 58.4% (n ⫽ 59/101) of cases were TisN0M0; 6.9% (n ⫽ 7/101) of cases were T1N0M0; 1%
(n ⫽ 1/101) of cases were T2N0M0; and 31.7% (n ⫽ 32/101) of cases were T3N0M0. Of these, 27 cases were staged T3 because of corneal invasion and 5 were staged T3 because of scleral invasion. Finally, 1% (n ⫽ 1/101) was T4aN0M0 and 1% (n ⫽ 1/101) was T4aN1M0. Of the 101 tumors in this study, there was a mean of 22 months follow-up (range, 6 –96 months). Of these, 18.8% (n ⫽ 19/101) were followed up for 6 months, 27.7% (n ⫽ 28/101) were followed up for 7 to 12 months, and 53.5% (n ⫽ 54/101) had a median follow-up of 26 months (range, 13–96 months). Primary treatment decisions were based on tumor size, location, focality, and invasiveness. Analysis of this series revealed that excision and cryotherapy was used as sole treatment in 49% (n ⫽ 50/101) of cases. In 47.5% (n ⫽ 48/101) of cases, adjuvant chemotherapy was added to excision and cryotherapy. Overall, excision and cryotherapy were used as part of therapy in 98 cases. One patient was treated by excision without cryotherapy outside the authors’ center and adjuvant topical interferon was added, another patient was treated by exenteration because of orbital invasion at presentation, and a third patient was treated with topical mitomycin 0.02% (she was receiving synchronous systemic chemotherapy for lymphoma). Recurrences were seen in 13 patients (n ⫽ 13/101; 12.9%). Subgroup analysis revealed that 12 recurrences took place during the first 12 months of follow-up and that 1 took place between and 12 and 24 months of follow-up. The probability of recurrence over 10 years (using the Kaplan-Meier method) is demonstrated in Figure 3. The recurrence rate was not correlated significantly to age, gender, laterality, or clinical appearance or focality of the tumor at presentation (Table 2). However, tumor size larger than 5 mm (P ⫽ 0.0176), larger than 2 mm extension onto the cornea (P ⫽ 0.0009), and tumor local invasiveness (P ⫽ 0.042) were associated with a statistically significant risk of tumor recurrence. This finding that larger and more invasive tumors were at greater risk for recurrence also is reflected by the probability related to AJCC stage (Table 3).
American Joint Committee on Cancer Classification T Stage versus Recurrence Simply, the higher the AJCC stage, the more likely the tumor recurred (Table 1).20 For example, the recurrence rate was 1.7% in Tis tumors, 0% in T1 and T2 tumors, and 34.3% for T3 tumors.
Conjunctival Squamous Carcinoma Excision ⴙ Cryotherapy
Interferon
Mitomycin
Time to Recurrence (months)
Lucentis
Radiation
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes No Yes Yes Yes No Yes No Yes No
Yes No No No Yes Yes Yes Yes Yes Yes No Yes No
8 24 11 7 6 6 6 6 7 7 8 9 6
Yes No Yes No No Yes Yes No Yes No No No No
No No No No Pd-103 EBRT EBRT EBRT EBRT EBRT EBRT EBRT EBRT
Final Outcome Cure Mitomycin (cure) Cure Re-excision Cure Failed Cure Cure Exenteration* Exenteration† Failed Failed Cure
237
Ophthalmology Volume 119, Number 2, February 2012 The AJCC stage T3 was subdivided into tumors staged as T3 because of corneal invasion (29.6% recurrence rate) and tumors staged T3 because of sclera invasion (60% recurrence rate). Finally, there was a 50% risk of recurrence for T4 tumors (Table 3). Compared with Tis tumors, T1 and T2 tumors and T3 and T4 tumors were at a significantly higher risk of recurrence (P ⫽ 0.0006; Table 3).
Histopathologic Results versus Recurrence The severity of histopathologic findings were correlated positively to recurrence. For example, none of the tumors defined histologically as squamous dysplasia recurred after 6 months of tumor-free period, whereas 12.8% and 22.2% of carcinoma in situ tumors and SCC tumors demonstrated recurrence, respectively (P ⫽ 0.037).
Treatment Method versus Recurrence Of the 101 tumors in this series, 98 cases were treated by excision and cryotherapy. Subgroup analysis revealed that in 50 of 98 patients in whom primary excision and cryotherapy were used as the sole definitive treatment, recurrence developed in 2 patients (n ⫽ 2/50; 4%). In the other subgroup (n ⫽ 48/98), adjuvant chemotherapy after excision and cryotherapy was used and 22.9% (n ⫽ 11/48) of patients showed recurrence. Further analysis of the first subgroup revealed that these tumors tended to be smaller and less invasive and in earlier AJCC stages (n ⫽ 42/50; 84% classified as AJCC Tis, T1, or T2). However, tumors received adjuvant chemotherapy after excision and cryotherapy tended to be larger and more invasive and in advanced AJCC stages (n ⫽ 22/48; 45.8% classified as AJCC T3). Further analysis of the management and recurrence of the tumors in the second subgroup revealed that 31.2% (n ⫽ 15/48) were treated with adjuvant topical 0.02% mitomycin and 80% (n ⫽ 12/15) of them were cured. An additional 68.8% (n ⫽ 33/48) of that same subgroup were treated with adjuvant topical interferon. Of that interferon-treated subgroup, 60.1% (n ⫽ 20/33) were cured, whereas in 39.4% (n ⫽ 13/33), tumor reduction by excision and cryotherapy followed by planned first-line adjuvant interferon was inadequate in resolving residual disease. In these cases, additional topical mitomycin 0.02% then was applied to induce remission in all 13 cases. Unfortunately, 38.4% (n ⫽ 5/13) of these cases treated with excision, cryotherapy, interferon, and mitomycin eventually demonstrated recurrence after at least 6 months of a clinically tumor-free period. Finally, the 13 tumors that recurred after a 6-month disease-free period were treated with combination therapies (primary excision, cryotherapy, topical chemotherapy, subconjunctival ranibizumab, and radiation therapy) and were not amenable to statistical analysis. Table 4 shows the main sequence of management and final outcomes for these cases.
Adequate Primary Treatment versus Recurrence Biopsy or attempted cure before referral was analyzed as a risk factor for recurrence. Overall, 26.7% (n ⫽ 27/101) of patients sought treatment after being treated primarily by excision with or without cryotherapy at another center. Of the 13 failures of local control (12.9%) noted in this series, 10 (77%) occurred in these previously treated tumors. Conversely, only 3 of the remaining 74 cases (4%) primarily treated at the authors’ center recurred during the duration of this study. The effect of previous treatment was noted to be a significant risk factor for recurrence (P ⫽ 0.0003). Subgroup analysis of referred cases revealed that 5 cases had scleral invasion. Lymph nodes were biopsy positive results in 1 patient (who was pregnant at the time of diagnosis), and was associated with
238
both orbital and intraocular invasion. She was treated by exenteration. No patient in this series developed distant metastasis.
Discussion Significant predictors for tumor recurrence were the size of the tumor (⬎5 mm), extension of the corneal involvement (⬎2 mm), local invasiveness, pathologic characteristics, higher AJCC T stage of the disease at presentation, and inadequate primary surgical treatment before referral. Recurrence was not correlated significantly to age, gender, laterality, clinical appearance, or focality of the tumor at presentation. In line with previous reports, a slight male predominance (67.7% were males) was noted, likely related to increased occupational exposure to sunlight.21–23 Deep corneal stromal invasion and involvement of the anterior chamber structures suggest intraocular invasion.3,4,5,24 Intraocular invasion has been reported to be between 2% and 8% of cases.5,24,25 The present case series showed 1% (n ⫽ 1/101) had intraocular extension. In the authors’ center, high-frequency anterior segment imaging, also called UBM, was performed for all patients with conjunctival tumor adherent to the underlying sclera for evidence of intraocular extension (angle blunting, iris disinsertion, or uveal thickening).5 Local invasion has been reported to be the most prevalent mechanism of tumor spread.24 For example, previous studies reported orbital invasion in 12% to 16% of invasive cases.23,24 A lower incidence was noted our series, in that orbital involvement was seen in 2% (n ⫽ 2/101) of cases. Metastases are rare. Preauricular and submandibular regional lymph nodes typically are affected first, but additional metastatic sites include the parotid gland, submandibular and submaxillary glands, lungs, and bone.24,26,27 In this series, only 1 patient (1%) showed lymph node involvement, and none demonstrated distant metastasis. Pizzarello and Jakobiec17 found that the more severe grades of conjunctival squamous carcinoma seem to recur at higher rates: for example those termed conjunctival intraepithelial neoplasia (CIN) had a recurrence rate of 17% to 24%, whereas squamous cell carcinoma recurs in 30% to 41% of cases after surgical excision. In the present series, the overall recurrence rate was 12.9%, 12.8% for carcinoma in situ tumors, 22.2% for squamous cell carcinoma, and 0% for cases with mild or moderate dysplasia. These recurrence rates are lower than those of most of the previous reports.18,23,24,26 Pizzarello and Jakobiec17 also found that if dysplastic tissues were left at the surgical margin, there was a 69% recurrence rate. When adjunctive cryotherapy is used with surgical excision, recurrence rates were reduced, ranging between 7% and 22%.7,26 –30 In general, it is difficult to define scleral involvement accurately (which is a risk factor for both tumor recurrence and intraocular invasion), so in the authors’ center, the scleral base and conjunctival margins routinely are treated with double freeze cryotherapy using Finger-tip cryotherapy probes.7 In that most tumors are perilimbal and epibulbar, the risk of scleral cryotherapy using this technique has been minimal.7
Yousef and Finger 䡠 Squamous Carcinoma of the Conjunctiva and Cornea In this series, all the cases with scleral invasion that demonstrated recurrence were initially treated outside the authors’ center (by inadequate excision with or without cryotherapy). In contrast, the recurrence rate for the patients who were initially treated by excision and cryotherapy after presentation to the authors’ center showed recurrence in only 4% (n ⫽ 3/74). This can be further contrasted to the overall study recurrence rate (n ⫽ 13/101) or 12.9%. This suggests that those patients primarily treated by a general ophthalmologist may fare worse than those treated by an ophthalmic oncologist. This may be because of inadequate primary treatment, inadvertent invagination of tumor tissue, difficulty related to the specialist selecting the best secondary treatment without an initial assessment of the tumor, a selection bias toward referring advanced tumors, or a combination of the above circumstances. Therefore, future care could be aided by initial slit lamp photography of the tumor and all conjunctival surfaces prior to initial biopsy or resection. According to the seventh edition of the AJCC staging system, the present case series showed that the recurrence rate was higher in advanced AJCC T-stage tumors. This suggests the need for more aggressive treatment for tumors in advanced stages. However, this study suggests that tumors with corneal extension should not be given the same stage as sclera-invading tumors, because the latter demonstrated a higher rate of recurrence. Clearly, AJCC staging was helpful in this analysis and will allow others to compare their scientific results with those presented in this study.31 In this series, the recurrence rate in patients who required topical chemotherapy in addition to excision and cryotherapy was 23%, higher than the 4% recurrence rate for those cured by excision and cryotherapy alone. This finding is likely related to a selection bias in which tumors with more advanced stages were more likely to receive topical chemotherapy as part of the treatment plan. In these cases, it is likely that the higher recurrence rate related to the advanced tumor stage rather than to the selected treatment. Furthermore, this study shows that follow-up examinations are necessary for patients diagnosed and treated for squamous conjunctival and corneal malignant neoplasia. Using the authors’ methods of excision and Finger-tip cryotherapy, smaller, less invasive tumors require at least 1 year of follow-up. Larger tumors (⬎5 mm), those with corneal extension (⬎2 mm), and local (e.g., scleral) invasion may require more long-term surveillance. This study examined predictive factors for recurrence of squamous cell neoplasia. Although this is unique work, it is retrospective and of limited size. Therefore, a larger and more comprehensive multicenter study should be performed to analyze better the efficacy of the various treatment methods and treatment-specific recurrence rates.
References 1. Zaki AA, Farid SF. Management of intraepithelial and invasive neoplasia of the cornea and conjunctiva: a long-term follow up. Cornea 2009;28:986 – 8.
2. Shields CL, Shields JA. Tumors of the conjunctiva and cornea. Surv Ophthalmol 2004;49:3–24. 3. Walsh-Conway N, Conway RM. Plaque brachytherapy for the management of ocular surface malignancies with corneoscleral invasion. Clin Experiment Ophthalmol 2009;37:577– 83. 4. Berenbom A, Milman T, Finger PT. FIT biopsy for conjunctival squamous carcinoma with extensive intraocular invasion. Graefes Arch Clin Exp Ophthalmol 2008; 246:467–9. 5. Finger PT, Tran HV, Turbin RE, et al. High-frequency ultrasonographic evaluation of conjunctival intraepithelial neoplasia and squamous cell carcinoma. Arch Ophthalmol 2003;121: 168 –72. 6. Teng CC, Chin KJ, Finger PT. Subconjunctival ranibizumab for squamous cell carcinoma for the conjunctiva with corneal extension. Br J Ophthalmol 2009;93:837– 8. 7. Finger PT. “Finger-tip” cryotherapy probes: treatment of squamous and melanocytic conjunctival neoplasia. Br J Ophthalmol 2005;89:942–5. 8. Midena E, Angeli CD, Valenti M, et al. Treatment of conjunctival squamous cell carcinoma with topical 5-fluorouracil. Br J Ophthalmol 2000;84:268 –72. 9. Shields CL, Demirci H, Marr BP, et al. Chemoreduction with topical mitomycin C prior to resection of extensive squamous cell carcinoma of the conjunctiva. Arch Ophthalmol 2005; 123:109 –13. 10. Graue GF, Tena LB, Finger PT. Electron beam radiation therapy for conjunctival squamous carcinoma. Ophthal Plast Reconstr Surg. 2011;27:277– 81. 11. Frucht-Pery J, Rozenman Y, Pe’er J. Topical mitomycin-C for partially excised conjunctival squamous cell carcinoma. Ophthalmology 2002;109:548 –52. 12. Akpek EK, Ertoy D, Kalayci D, et al. Postoperative topical mitomycin C in conjunctival squamous cell neoplasia. Cornea 1999;18:59 – 62. 13. Finger PT. The 7th edition AJCC staging system for eye cancer: an international language for ophthalmic oncology. Arch Pathol Lab Med 2009;133:1197– 8. 14. Finger PT. Radiation therapy for orbital tumors: concepts, current use and ophthalmic radiation side effects. Surv Ophthalmol 2009;54:545– 68. 15. Karp CL, Galor A, Chhabra S, et al. Subconjunctival/ perilesional recombinant interferon ␣2b for ocular surface squamous neoplasia: a 10-year review. Ophthalmology 2010; 117:2241– 6. 16. Newton R, Ferlay J, Reeves G, et al. Effect of ambient solar ultraviolet radiation on incidence of squamous-cell carcinoma of the eye. Lancet 1996;347:1450 –1. 17. Pizzarello LD, Jakobiec FA. Bowen’s disease of the conjunctiva: a misnomer. In: Jakobiec FA, ed. Ocular and Adnexal Tumors. Birmingham, AL: Aesculapius; 1978:553–71. 18. Yang J, Foster CS. Squamous cell carcinoma of the conjunctiva. Int Ophthalmol Clin 1997;37:73– 85. 19. Semenova EA, Milman T, Finger PT, et al. The diagnostic value of exfoliative cytology vs histopathology for ocular surface squamous neoplasia. Am J Ophthalmol 2009;148: 772– 8. 20. The AJCC Ophthalmic Oncology Task Force. Carcinoma of the conjunctiva. In: Edge SE, Byrd DR, Carducci MA, Compton CA, eds. AJCC Cancer Staging Manual. 7th ed. New York: Springer; 2009:531–3. 21. Ni C, Searl SS, Kriegstein HJ, et al. Epibulbar carcinoma. Int Ophthalmol Clin 1982;22:1–33. 22. Ng J, Coroneo MT, Wakefield D, Di Girolamo N. Ultraviolet radiation and the role of matrix metalloproteinases in the
239
Ophthalmology Volume 119, Number 2, February 2012
23. 24. 25. 26. 27.
pathogenesis of ocular surface squamous neoplasia. Invest Ophthalmol Vis Sci 2008;49:5295–306. Erie JC, Campbell RJ, Liesegang TJ. Conjunctival and corneal intraepithelial neoplasia. Ophthalmology 1986;93:176 – 83. Illiff WJ, Marback R, Green R. Invasive squamous cell carcinoma of the conjunctiva. Arch Ophthalmol 1975;93:119 –20. Buuns DR, Tse DT, Folberg R. Microscopically controlled excision of conjunctival squamous cell carcinoma. Am J Ophthalmol 1994;117:97–102. Peksayar G, Soyturk MK, Demiryont M. Long-term results of cryotherapy on malignant epithelial tumors of the conjunctiva. Am J Ophthalmol 1989;107:337– 40. Bhattacharyya N, Wenokur RK, Rubin PA. Metastasis of squamous cell carcinoma of the conjunctiva: case report
28. 29. 30. 31.
and review of the literature. Am J Otolaryngol 1997;18:217–9. Char DH, Crawford JB, Howes EL, et al. Resection of intraocular squamous cell carcinoma. Br J Ophthalmol 1992;76: 123–5. Divine RD, Anderson RL. Nitrous oxide cryotherapy for intraepithelial epithelioma of the conjunctiva. Arch Ophthalmol 1983;101:782– 6. Dutton JJ, Anderson RL, Tse DT. Combined surgery and cryotherapy for scleral invasion of epithelial malignancies. Ophthalmic Surg 1984;15:289 –94. Finger PT. The 7th edition AJCC staging system for eye cancer: an international language for ophthalmic oncology. Arch Pathol Lab Med 2009;133:1197– 8.
Footnotes and Financial Disclosures Originally received: May 13, 2011. Final revision: July 30, 2011. Accepted: August 2, 2011. Available online: December 22, 2011.
Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Manuscript no. 2011-722.
The New York Eye Cancer Center, The New York Eye and Ear Infirmary and the New York University School of Medicine, New York, New York. Dr. Yousef received fellowships from The Eye Cancer Foundation, Inc, New York, New York (http://eyecancerfoundation.net), and the International Council of Ophthalmology Switzerland and San Francisco, California.
240
Correspondence: Paul T. Finger, MD, The New York Eye Cancer Center, 115 East 61st Street, NewYork, NY 10065. E-mail: [email protected].
Squamous conjunctival carcinoma (SCC) is the most common ocular malignancy.1,2 Both sight and life threatening, SCC has the potential to invade locally the cornea, sclera, uvea, eyelids, orbit, sinuses, and brain as well as to metastasize.3–5 Commonly used local treatment options for SCC include excision, excision with cryotherapy, and adjuvant topical chemotherapy.1–12 Although there are no published guidelines, in North America, most small tumors are treated by excision or by excision with adjuvant cryotherapy. Larger, multifocal, or aggressive tumors have required combinations of resection, adjuvant cryotherapy, and topical chemotherapy (i.e., interferon, mitomycin, 5-fluoracil, or cyclosporine) and electron-based external beam radiation therapy.2,8 –16 Recently, subconjunctival ranibizumab6,13 and external beam radiation therapy10,14 using photons, electrons, or protons have been proposed as palliative treatments for nonresponsive carcinomas of the eye, lids, and © 2012 by the American Academy of Ophthalmology Published by Elsevier Inc.
orbit.14 When all else fails, orbital exenteration, typically with adjuvant external beam radiation therapy, may be used.10.14 Surgical excision of corneal and conjunctival intraepithelial dysplasia as well as squamous cell carcinoma in situ has been associated with high rates of recurrence. Reported ranges include 17% to 24% for dysplasia and 30% to 41% for squamous cell carcinoma.15,18 This is likely because of tumor edges and deep margins that often are difficult to determine.19 For example, a tumor’s advancing edges often are light gray and avascular. This can be seen easily when the edges extend onto the clear corneal epithelium. However, with a backdrop of white sclera, a light gray avascular extension is close to or entirely invisible. In clinical practice, this can lead to underestimation of tumor size. To compensate, ophthalmic oncologists increasingly have used ISSN 0161-6420/12/$–see front matter doi:10.1016/j.ophtha.2011.08.005
233
Ophthalmology Volume 119, Number 2, February 2012
Figure 1. Tumor characteristics. A, Patient with multifocal conjunctival tumor (carcinoma in situ) showing 2 tumors (nasal and temporal) separated by ⬎5 mm tumor-free tissue. B, Focal nodular appearance of conjunctival tumor (squamous dysplasia). C, Diffuse conjunctival tumor (squamous cell carcinoma). D, Pigmented conjunctival carcinoma in situ.
combination therapy (e.g., cryotherapy, topical chemotherapy, irradiation).18 This study staged tumors according to the seventh edition of the American Joint Committee on Cancer (AJCC) staging system.20 This allowed for standardized comparison of methods of both diagnosis and treatment. It also allowed for standardized collection of information re-
garding the histopathologic diagnosis, tumor size, local invasion, lymph node involvement, and distant metastasis (Table 1). This information was used to study methods of treatment, incidence of local control, and recurrence rate and their relation to tumor characteristics. Herein, the authors’ experience in the treatment of 101 consecutive cases is presented.
Figure 2. Tumor size and corneal extension. A, Small conjunctival squamous dysplasia (⬍5 mm) with no corneal extension. B, Small conjunctival carcinoma in situ (⬍3 clock hours) with corneal extension (⬍2 mm). C, Medium conjunctival squamous cell carcinoma (⬎3 clock hours and ⬍6 clock hours) with corneal extension (⬎2 mm). D, Large conjunctival squamous cell carcinoma (⬎6 clock hours) and corneal extension (⬎2 mm).
234
Yousef and Finger 䡠 Squamous Carcinoma of the Conjunctiva and Cornea Table 1. Carcinoma of the Conjunctiva: Definition of Tumor, Lymph Node, Metastasis (TNM) Staging* Definitions Apply to Both Clinical and Pathologic Staging Primary tumor (T) TX T0 Tis T1 T2 T3 T4 T4a T4b T4c T4d Regional lymph nodes (N) NX N0 N1 Distant metastasis (M) MX M0 M1
Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ Tumor ⱕ5 mm in greatest dimension Tumor ⬎5 mm in greatest dimension, without invasion of adjacent structures Tumor invades adjacent stuctures, excluding the orbit Tumor invades orbit with or without further extension Tumor invades orbital soft tissues, without bone invasion Tumor invades bone Tumor invades adjacent paranasal sinuses Tumor invades brain Regional lymph nodes cannot be assessed No regional lymph node metastasis Regional lymph node metastasis Distant metastasis cannot be assessed No distant metastasis Distant metastasis
*Reprinted with permission from The AJCC Ophthalmic Oncology Task Force. Carcinoma of the conjunctiva. In: Edge SE, Byrd DR, Carducci MA, Compton CA, eds. AJCC Cancer Staging Manual. 7th ed. New York: Springer; 2009:531–3.20
Patients and Methods This study adhered to the tenets of the Declaration of Helsinki of 1975, as revised in 2000, and the Health Insurance Portability and Accountability Act of 1996 and was approved by the institutional review board of The New York Eye Cancer Center. The study was a retrospective case review of 101 eyes of 99 consecutive patients who were diagnosed pathologically with conjunctival squamous dysplasia, conjunctival carcinoma in situ, or SCC. Selection also required access to patient medical records, pathologic reports, and color photographs. Data included patient age, gender, laterality, visual acuity, tumor size, clinical appearance, corneal extension, tumor invasiveness (corneal, scleral, uveal, orbital, lymphatic, and distant metastasis), and AJCC stage for each tumor. Data recorded included the
pathologic diagnosis, primary and secondary treatment methods, recurrence rate, as well as duration of follow-up. Recurrence was recorded only if there was no clinical evidence of tumor for at least 6 months after the end of the planned treatment.
Focused Ophthalmic Examination At presentation, all patients underwent slit-lamp examination of all conjunctival surfaces (including eversion of the upper eyelid and tarsus). Slit-lamp photography was used to record all conjunctival surfaces. Periodic gonioscopic photography and ultrasound biomicroscopy was used to monitor for and document intraocular invasion. Palpation was used to examine the preauricular and submandibular lymph nodes. Tumors found adherent to the eye wall were selected for high-frequency ultrasound Table 2. Tumor Features Not Correlated to Recurrence Feature
Figure 3. Kaplan-Meier method showing the recurrence rate of malignant conjunctival squamous neoplasia over the time (months).
Gender Male Female Laterality Right eye Left eye Both eyes Focality Monofocal Multifocal Morphologic features Focal nodular Diffuse Pigmented
No. of Tumors
%
No. of Recurrences
%
P Value
67 32
67.7 32.3
9 4
13.4 12.5
0.9102
49 48 2
48.5 47.5 2.0
7 6
14.3 12.5
0.8215
78 23
77.2 22.8
7 6
8.9 26.1
0.0685
43 52 6
42.6 51.5 5.9
5 8 0
11.6 15.4 0
0.6435
235
Ophthalmology Volume 119, Number 2, February 2012 Table 3. Tumor Features That Are Statistically Associated with Higher Risk of Recurrence Feature
No. of Tumors
%
No. of Recurrences
%
P Value
59 7 1 32 2
58.4 6.9 1 31.7 2
1 0 0 11 1
1.7 0 0 34.4 50
0.0006
42 37 22
41.6 36.6 21.8
1 5 7
2.4 13.5 31.8
0.0176
25 73 5 1 2 1 0
24.8 72.3 5 1 2 1 0
0 9 3 0 1 0 0
0 12.3 60 0 50 0 0
0.0423
42 31
57.5 42.5
1 12
2.4 38.7
0.0009
26 39 36
25.7 38.6 35.6
0 5 8
0 12.8 22.2
0.0373
AJCC Tis T1 T2 T3 T4 Size (mm) Small (⬍ 5) Medium (5–10) Large (⬎10 mm) Invasion Conjunctiva Cornea Sclera Uvea Orbit Lymph node Metastasis Corneal extension (mm) ⬍2 ⬎2 Pathologic features Dysplasia Carcinoma in situ SCC
margin), excision with cryotherapy, topical chemotherapy (interferon, mitomycin, and subconjunctival ranibizumab), irradiation, and exenteration. Cryotherapy was performed with Finger-tip spatulated cryotherapy applicators (Dutch Ophthalmic Research Corporation, Zuidland, The Netherlands) that are capable of producing homogeneous freezing over large surface areas, allowing for more uniform treatment of tissue (compared with standard cryotherapy applicators).7
Clinical Characteristics
AJCC ⫽ Seventh edition of the American Joint Committee staging system; SCC ⫽ squamous cell carcinoma.
biomicroscopy imaging (UBM). Adherence was evaluated by proparacaine-assisted Q-tip palpation under slit-lamp observation.
Treatment Methods Local treatments of conjunctival tumors included primary excision (excision of the entire visible tumor in addition to a 2-mm safety
Multifocal tumors were defined as having at least 2 tumors separated by 5 mm of normal-appearing tissue (Fig 1). According to the clinical appearance of the tumor, they were classified into focal nodular tumors (thickened focal lesion with welldefined margins), diffuse tumors (flat lesion with ill-defined margins), or pigmented tumors (Fig 1). Tumor sizes were divided into small, medium, and large tumors. Small tumors were defined as those less than 3 clock hours (for limbal tumors) or less than 5 mm in largest diameter. Medium tumors were defined as those 3 to 6 clock hours (for limbal tumors) or 5 to 10 mm in largest diameter, and large tumors were defined as those more than 6 clock hours (for limbal tumors) or more than 10 mm in largest diameter (Fig 2). Furthermore, tumors with corneal extension were divided into 2 groups: group 1 included tumors with 2 mm or less of corneal extension beyond the limbus and group 2 included tumors with more than 2 mm of corneal extension beyond the limbus (Fig 2). Tumor invasiveness was defined as follows: (1) corneal invasion consisted of clinically detected tumor tissue within the corneal tissue that was confirmed by histopathologic evaluation of corneal biopsy, (2) scleral invasion consisted of tumor tissue detected by UBM to be within the anatomic location of scleral tissue, (3) intraocular invasion consisted of tumor tissue detected by UBM to be within the intraocular structures, and (4) orbital invasion consisted of tumor tissue detected within the orbit posterior to the conjunctiva by orbital magnetic resonance imaging or computed tomography scan. In addition, the AJCC staging system was used to classify all cases of conjunctival squamous neoplasia in this case series (Table 1).20 According to the AJCC staging system, the same definition of tumor, node, and metastasis staging apply for both clinical and pathologic staging; therefore a pathologic classification was added Table 4. Recurrent Cases of
Patient No.
Pathologic Diagnosis
Gender
American Joint Committee on Cancer Classification Stage
Size
Extension
Lymph Node Involvement
1 2 3 4 5 6 7 8 9 10 11 12 13
SCC SCC CIS SCC SCC CIS SCC CIS CIS SCC SCC SCC CIS
M F M M M M M F M f M M F
T3c T3c Tis T3c T3 T3c T3c T3c T3C T3c T3 T3 T4
Large Medium Large Large Large Medium Small Medium Large Large Medium Medium Large
Cornea Cornea Cornea Cornea Sclera Cornea Cornea Cornea Cornea Sclera Sclera Cornea Orbit
No No No No No No No No No No No No No
CIS ⫽ carcinoma in situ; EBRT ⫽ external beam radiation therapy; Pd-103 ⫽ palladium-103; SCC ⫽ squamous cell carcinoma. *Exenteration performed because of endophthalmitis. † Exenteration performed because of fibrous histiocytoma.
236
Yousef and Finger 䡠 Squamous Carcinoma of the Conjunctiva and Cornea based on full-thickness conjunctival biopsy to include squamous dysplasia, carcinoma in situ, and squamous cell carcinoma.20
Statistical Analysis Statistical analysis of tumor clinical appearance, size, invasiveness, histopathologic results, and AJCC stage were correlated to the recurrence rate of the tumor after a 6-month period of cure after the available methods of treatment (i.e., surgical excision, cryotherapy, topical chemotherapy, subconjunctival anti–vascular endothelial growth factor injection, and radiation therapy). The P value was measured using the exact Fisher test to test the predictive power of each factor.
Results One hundred one eyes from 99 patients with an age range of 37 to 96 years (median, 71 years) were studied. Lesions were present on the right eye in 49 patients and on the left in 48 patients, and 2 patients had bilateral disease. Sixty-seven percent of patients (n ⫽ 67/99) were male. Clinical appearance evaluation showed that 42.6% (n ⫽ 43/ 101) of SCC cases were focal nodular tumors, 51.5% (n ⫽ 52/101) were diffuse tumors, and 5.9% (n ⫽ 6/101) were pigmented tumors. Multifocality was seen in 22.8% (n ⫽ 33/101). The tumor was limited to the superior tarsus in 2% (n ⫽ 2/101) and was limited to the palpebral fornix in 1% (n ⫽ 1/101). Tumor was localized to bulbar conjunctiva in 24.8% (n ⫽ 25/101). Invasion onto the cornea was present in 72.3% (n ⫽ 73/101), invasion onto the sclera was present in 4.9% (n ⫽ 5/101), invasion onto the uvea was present in 1% (n ⫽ 1/101), invasion onto the orbit was present in 2% (n ⫽ 2/101), and lymph node metastases were present in 1% (n ⫽ 1/101). In no patients did distant metastasis develop. In addition, corneal invasion was wider than 2 mm in 30.7% (n ⫽ 31/73). Histopathologic analysis revealed conjunctival squamous dysplasia (n ⫽ 26/101; 25.7%), carcinoma in situ (n ⫽ 39/101; 38.6%), and squamous cell carcinoma (n ⫽ 36/101; 35.6%). According to AJCC classification (which correlates to the pathologic diagnosis, size of the tumor, and tumor invasiveness; Table 1), in this series, 58.4% (n ⫽ 59/101) of cases were TisN0M0; 6.9% (n ⫽ 7/101) of cases were T1N0M0; 1%
(n ⫽ 1/101) of cases were T2N0M0; and 31.7% (n ⫽ 32/101) of cases were T3N0M0. Of these, 27 cases were staged T3 because of corneal invasion and 5 were staged T3 because of scleral invasion. Finally, 1% (n ⫽ 1/101) was T4aN0M0 and 1% (n ⫽ 1/101) was T4aN1M0. Of the 101 tumors in this study, there was a mean of 22 months follow-up (range, 6 –96 months). Of these, 18.8% (n ⫽ 19/101) were followed up for 6 months, 27.7% (n ⫽ 28/101) were followed up for 7 to 12 months, and 53.5% (n ⫽ 54/101) had a median follow-up of 26 months (range, 13–96 months). Primary treatment decisions were based on tumor size, location, focality, and invasiveness. Analysis of this series revealed that excision and cryotherapy was used as sole treatment in 49% (n ⫽ 50/101) of cases. In 47.5% (n ⫽ 48/101) of cases, adjuvant chemotherapy was added to excision and cryotherapy. Overall, excision and cryotherapy were used as part of therapy in 98 cases. One patient was treated by excision without cryotherapy outside the authors’ center and adjuvant topical interferon was added, another patient was treated by exenteration because of orbital invasion at presentation, and a third patient was treated with topical mitomycin 0.02% (she was receiving synchronous systemic chemotherapy for lymphoma). Recurrences were seen in 13 patients (n ⫽ 13/101; 12.9%). Subgroup analysis revealed that 12 recurrences took place during the first 12 months of follow-up and that 1 took place between and 12 and 24 months of follow-up. The probability of recurrence over 10 years (using the Kaplan-Meier method) is demonstrated in Figure 3. The recurrence rate was not correlated significantly to age, gender, laterality, or clinical appearance or focality of the tumor at presentation (Table 2). However, tumor size larger than 5 mm (P ⫽ 0.0176), larger than 2 mm extension onto the cornea (P ⫽ 0.0009), and tumor local invasiveness (P ⫽ 0.042) were associated with a statistically significant risk of tumor recurrence. This finding that larger and more invasive tumors were at greater risk for recurrence also is reflected by the probability related to AJCC stage (Table 3).
American Joint Committee on Cancer Classification T Stage versus Recurrence Simply, the higher the AJCC stage, the more likely the tumor recurred (Table 1).20 For example, the recurrence rate was 1.7% in Tis tumors, 0% in T1 and T2 tumors, and 34.3% for T3 tumors.
Conjunctival Squamous Carcinoma Excision ⴙ Cryotherapy
Interferon
Mitomycin
Time to Recurrence (months)
Lucentis
Radiation
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes No Yes Yes Yes No Yes No Yes No
Yes No No No Yes Yes Yes Yes Yes Yes No Yes No
8 24 11 7 6 6 6 6 7 7 8 9 6
Yes No Yes No No Yes Yes No Yes No No No No
No No No No Pd-103 EBRT EBRT EBRT EBRT EBRT EBRT EBRT EBRT
Final Outcome Cure Mitomycin (cure) Cure Re-excision Cure Failed Cure Cure Exenteration* Exenteration† Failed Failed Cure
237
Ophthalmology Volume 119, Number 2, February 2012 The AJCC stage T3 was subdivided into tumors staged as T3 because of corneal invasion (29.6% recurrence rate) and tumors staged T3 because of sclera invasion (60% recurrence rate). Finally, there was a 50% risk of recurrence for T4 tumors (Table 3). Compared with Tis tumors, T1 and T2 tumors and T3 and T4 tumors were at a significantly higher risk of recurrence (P ⫽ 0.0006; Table 3).
Histopathologic Results versus Recurrence The severity of histopathologic findings were correlated positively to recurrence. For example, none of the tumors defined histologically as squamous dysplasia recurred after 6 months of tumor-free period, whereas 12.8% and 22.2% of carcinoma in situ tumors and SCC tumors demonstrated recurrence, respectively (P ⫽ 0.037).
Treatment Method versus Recurrence Of the 101 tumors in this series, 98 cases were treated by excision and cryotherapy. Subgroup analysis revealed that in 50 of 98 patients in whom primary excision and cryotherapy were used as the sole definitive treatment, recurrence developed in 2 patients (n ⫽ 2/50; 4%). In the other subgroup (n ⫽ 48/98), adjuvant chemotherapy after excision and cryotherapy was used and 22.9% (n ⫽ 11/48) of patients showed recurrence. Further analysis of the first subgroup revealed that these tumors tended to be smaller and less invasive and in earlier AJCC stages (n ⫽ 42/50; 84% classified as AJCC Tis, T1, or T2). However, tumors received adjuvant chemotherapy after excision and cryotherapy tended to be larger and more invasive and in advanced AJCC stages (n ⫽ 22/48; 45.8% classified as AJCC T3). Further analysis of the management and recurrence of the tumors in the second subgroup revealed that 31.2% (n ⫽ 15/48) were treated with adjuvant topical 0.02% mitomycin and 80% (n ⫽ 12/15) of them were cured. An additional 68.8% (n ⫽ 33/48) of that same subgroup were treated with adjuvant topical interferon. Of that interferon-treated subgroup, 60.1% (n ⫽ 20/33) were cured, whereas in 39.4% (n ⫽ 13/33), tumor reduction by excision and cryotherapy followed by planned first-line adjuvant interferon was inadequate in resolving residual disease. In these cases, additional topical mitomycin 0.02% then was applied to induce remission in all 13 cases. Unfortunately, 38.4% (n ⫽ 5/13) of these cases treated with excision, cryotherapy, interferon, and mitomycin eventually demonstrated recurrence after at least 6 months of a clinically tumor-free period. Finally, the 13 tumors that recurred after a 6-month disease-free period were treated with combination therapies (primary excision, cryotherapy, topical chemotherapy, subconjunctival ranibizumab, and radiation therapy) and were not amenable to statistical analysis. Table 4 shows the main sequence of management and final outcomes for these cases.
Adequate Primary Treatment versus Recurrence Biopsy or attempted cure before referral was analyzed as a risk factor for recurrence. Overall, 26.7% (n ⫽ 27/101) of patients sought treatment after being treated primarily by excision with or without cryotherapy at another center. Of the 13 failures of local control (12.9%) noted in this series, 10 (77%) occurred in these previously treated tumors. Conversely, only 3 of the remaining 74 cases (4%) primarily treated at the authors’ center recurred during the duration of this study. The effect of previous treatment was noted to be a significant risk factor for recurrence (P ⫽ 0.0003). Subgroup analysis of referred cases revealed that 5 cases had scleral invasion. Lymph nodes were biopsy positive results in 1 patient (who was pregnant at the time of diagnosis), and was associated with
238
both orbital and intraocular invasion. She was treated by exenteration. No patient in this series developed distant metastasis.
Discussion Significant predictors for tumor recurrence were the size of the tumor (⬎5 mm), extension of the corneal involvement (⬎2 mm), local invasiveness, pathologic characteristics, higher AJCC T stage of the disease at presentation, and inadequate primary surgical treatment before referral. Recurrence was not correlated significantly to age, gender, laterality, clinical appearance, or focality of the tumor at presentation. In line with previous reports, a slight male predominance (67.7% were males) was noted, likely related to increased occupational exposure to sunlight.21–23 Deep corneal stromal invasion and involvement of the anterior chamber structures suggest intraocular invasion.3,4,5,24 Intraocular invasion has been reported to be between 2% and 8% of cases.5,24,25 The present case series showed 1% (n ⫽ 1/101) had intraocular extension. In the authors’ center, high-frequency anterior segment imaging, also called UBM, was performed for all patients with conjunctival tumor adherent to the underlying sclera for evidence of intraocular extension (angle blunting, iris disinsertion, or uveal thickening).5 Local invasion has been reported to be the most prevalent mechanism of tumor spread.24 For example, previous studies reported orbital invasion in 12% to 16% of invasive cases.23,24 A lower incidence was noted our series, in that orbital involvement was seen in 2% (n ⫽ 2/101) of cases. Metastases are rare. Preauricular and submandibular regional lymph nodes typically are affected first, but additional metastatic sites include the parotid gland, submandibular and submaxillary glands, lungs, and bone.24,26,27 In this series, only 1 patient (1%) showed lymph node involvement, and none demonstrated distant metastasis. Pizzarello and Jakobiec17 found that the more severe grades of conjunctival squamous carcinoma seem to recur at higher rates: for example those termed conjunctival intraepithelial neoplasia (CIN) had a recurrence rate of 17% to 24%, whereas squamous cell carcinoma recurs in 30% to 41% of cases after surgical excision. In the present series, the overall recurrence rate was 12.9%, 12.8% for carcinoma in situ tumors, 22.2% for squamous cell carcinoma, and 0% for cases with mild or moderate dysplasia. These recurrence rates are lower than those of most of the previous reports.18,23,24,26 Pizzarello and Jakobiec17 also found that if dysplastic tissues were left at the surgical margin, there was a 69% recurrence rate. When adjunctive cryotherapy is used with surgical excision, recurrence rates were reduced, ranging between 7% and 22%.7,26 –30 In general, it is difficult to define scleral involvement accurately (which is a risk factor for both tumor recurrence and intraocular invasion), so in the authors’ center, the scleral base and conjunctival margins routinely are treated with double freeze cryotherapy using Finger-tip cryotherapy probes.7 In that most tumors are perilimbal and epibulbar, the risk of scleral cryotherapy using this technique has been minimal.7
Yousef and Finger 䡠 Squamous Carcinoma of the Conjunctiva and Cornea In this series, all the cases with scleral invasion that demonstrated recurrence were initially treated outside the authors’ center (by inadequate excision with or without cryotherapy). In contrast, the recurrence rate for the patients who were initially treated by excision and cryotherapy after presentation to the authors’ center showed recurrence in only 4% (n ⫽ 3/74). This can be further contrasted to the overall study recurrence rate (n ⫽ 13/101) or 12.9%. This suggests that those patients primarily treated by a general ophthalmologist may fare worse than those treated by an ophthalmic oncologist. This may be because of inadequate primary treatment, inadvertent invagination of tumor tissue, difficulty related to the specialist selecting the best secondary treatment without an initial assessment of the tumor, a selection bias toward referring advanced tumors, or a combination of the above circumstances. Therefore, future care could be aided by initial slit lamp photography of the tumor and all conjunctival surfaces prior to initial biopsy or resection. According to the seventh edition of the AJCC staging system, the present case series showed that the recurrence rate was higher in advanced AJCC T-stage tumors. This suggests the need for more aggressive treatment for tumors in advanced stages. However, this study suggests that tumors with corneal extension should not be given the same stage as sclera-invading tumors, because the latter demonstrated a higher rate of recurrence. Clearly, AJCC staging was helpful in this analysis and will allow others to compare their scientific results with those presented in this study.31 In this series, the recurrence rate in patients who required topical chemotherapy in addition to excision and cryotherapy was 23%, higher than the 4% recurrence rate for those cured by excision and cryotherapy alone. This finding is likely related to a selection bias in which tumors with more advanced stages were more likely to receive topical chemotherapy as part of the treatment plan. In these cases, it is likely that the higher recurrence rate related to the advanced tumor stage rather than to the selected treatment. Furthermore, this study shows that follow-up examinations are necessary for patients diagnosed and treated for squamous conjunctival and corneal malignant neoplasia. Using the authors’ methods of excision and Finger-tip cryotherapy, smaller, less invasive tumors require at least 1 year of follow-up. Larger tumors (⬎5 mm), those with corneal extension (⬎2 mm), and local (e.g., scleral) invasion may require more long-term surveillance. This study examined predictive factors for recurrence of squamous cell neoplasia. Although this is unique work, it is retrospective and of limited size. Therefore, a larger and more comprehensive multicenter study should be performed to analyze better the efficacy of the various treatment methods and treatment-specific recurrence rates.
References 1. Zaki AA, Farid SF. Management of intraepithelial and invasive neoplasia of the cornea and conjunctiva: a long-term follow up. Cornea 2009;28:986 – 8.
2. Shields CL, Shields JA. Tumors of the conjunctiva and cornea. Surv Ophthalmol 2004;49:3–24. 3. Walsh-Conway N, Conway RM. Plaque brachytherapy for the management of ocular surface malignancies with corneoscleral invasion. Clin Experiment Ophthalmol 2009;37:577– 83. 4. Berenbom A, Milman T, Finger PT. FIT biopsy for conjunctival squamous carcinoma with extensive intraocular invasion. Graefes Arch Clin Exp Ophthalmol 2008; 246:467–9. 5. Finger PT, Tran HV, Turbin RE, et al. High-frequency ultrasonographic evaluation of conjunctival intraepithelial neoplasia and squamous cell carcinoma. Arch Ophthalmol 2003;121: 168 –72. 6. Teng CC, Chin KJ, Finger PT. Subconjunctival ranibizumab for squamous cell carcinoma for the conjunctiva with corneal extension. Br J Ophthalmol 2009;93:837– 8. 7. Finger PT. “Finger-tip” cryotherapy probes: treatment of squamous and melanocytic conjunctival neoplasia. Br J Ophthalmol 2005;89:942–5. 8. Midena E, Angeli CD, Valenti M, et al. Treatment of conjunctival squamous cell carcinoma with topical 5-fluorouracil. Br J Ophthalmol 2000;84:268 –72. 9. Shields CL, Demirci H, Marr BP, et al. Chemoreduction with topical mitomycin C prior to resection of extensive squamous cell carcinoma of the conjunctiva. Arch Ophthalmol 2005; 123:109 –13. 10. Graue GF, Tena LB, Finger PT. Electron beam radiation therapy for conjunctival squamous carcinoma. Ophthal Plast Reconstr Surg. 2011;27:277– 81. 11. Frucht-Pery J, Rozenman Y, Pe’er J. Topical mitomycin-C for partially excised conjunctival squamous cell carcinoma. Ophthalmology 2002;109:548 –52. 12. Akpek EK, Ertoy D, Kalayci D, et al. Postoperative topical mitomycin C in conjunctival squamous cell neoplasia. Cornea 1999;18:59 – 62. 13. Finger PT. The 7th edition AJCC staging system for eye cancer: an international language for ophthalmic oncology. Arch Pathol Lab Med 2009;133:1197– 8. 14. Finger PT. Radiation therapy for orbital tumors: concepts, current use and ophthalmic radiation side effects. Surv Ophthalmol 2009;54:545– 68. 15. Karp CL, Galor A, Chhabra S, et al. Subconjunctival/ perilesional recombinant interferon ␣2b for ocular surface squamous neoplasia: a 10-year review. Ophthalmology 2010; 117:2241– 6. 16. Newton R, Ferlay J, Reeves G, et al. Effect of ambient solar ultraviolet radiation on incidence of squamous-cell carcinoma of the eye. Lancet 1996;347:1450 –1. 17. Pizzarello LD, Jakobiec FA. Bowen’s disease of the conjunctiva: a misnomer. In: Jakobiec FA, ed. Ocular and Adnexal Tumors. Birmingham, AL: Aesculapius; 1978:553–71. 18. Yang J, Foster CS. Squamous cell carcinoma of the conjunctiva. Int Ophthalmol Clin 1997;37:73– 85. 19. Semenova EA, Milman T, Finger PT, et al. The diagnostic value of exfoliative cytology vs histopathology for ocular surface squamous neoplasia. Am J Ophthalmol 2009;148: 772– 8. 20. The AJCC Ophthalmic Oncology Task Force. Carcinoma of the conjunctiva. In: Edge SE, Byrd DR, Carducci MA, Compton CA, eds. AJCC Cancer Staging Manual. 7th ed. New York: Springer; 2009:531–3. 21. Ni C, Searl SS, Kriegstein HJ, et al. Epibulbar carcinoma. Int Ophthalmol Clin 1982;22:1–33. 22. Ng J, Coroneo MT, Wakefield D, Di Girolamo N. Ultraviolet radiation and the role of matrix metalloproteinases in the
239
Ophthalmology Volume 119, Number 2, February 2012
23. 24. 25. 26. 27.
pathogenesis of ocular surface squamous neoplasia. Invest Ophthalmol Vis Sci 2008;49:5295–306. Erie JC, Campbell RJ, Liesegang TJ. Conjunctival and corneal intraepithelial neoplasia. Ophthalmology 1986;93:176 – 83. Illiff WJ, Marback R, Green R. Invasive squamous cell carcinoma of the conjunctiva. Arch Ophthalmol 1975;93:119 –20. Buuns DR, Tse DT, Folberg R. Microscopically controlled excision of conjunctival squamous cell carcinoma. Am J Ophthalmol 1994;117:97–102. Peksayar G, Soyturk MK, Demiryont M. Long-term results of cryotherapy on malignant epithelial tumors of the conjunctiva. Am J Ophthalmol 1989;107:337– 40. Bhattacharyya N, Wenokur RK, Rubin PA. Metastasis of squamous cell carcinoma of the conjunctiva: case report
28. 29. 30. 31.
and review of the literature. Am J Otolaryngol 1997;18:217–9. Char DH, Crawford JB, Howes EL, et al. Resection of intraocular squamous cell carcinoma. Br J Ophthalmol 1992;76: 123–5. Divine RD, Anderson RL. Nitrous oxide cryotherapy for intraepithelial epithelioma of the conjunctiva. Arch Ophthalmol 1983;101:782– 6. Dutton JJ, Anderson RL, Tse DT. Combined surgery and cryotherapy for scleral invasion of epithelial malignancies. Ophthalmic Surg 1984;15:289 –94. Finger PT. The 7th edition AJCC staging system for eye cancer: an international language for ophthalmic oncology. Arch Pathol Lab Med 2009;133:1197– 8.
Footnotes and Financial Disclosures Originally received: May 13, 2011. Final revision: July 30, 2011. Accepted: August 2, 2011. Available online: December 22, 2011.
Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Manuscript no. 2011-722.
The New York Eye Cancer Center, The New York Eye and Ear Infirmary and the New York University School of Medicine, New York, New York. Dr. Yousef received fellowships from The Eye Cancer Foundation, Inc, New York, New York (http://eyecancerfoundation.net), and the International Council of Ophthalmology Switzerland and San Francisco, California.
240
Correspondence: Paul T. Finger, MD, The New York Eye Cancer Center, 115 East 61st Street, NewYork, NY 10065. E-mail: [email protected].