Radiotherapy for localized sebaceous carcinoma of the eyelid: a retrospective analysis of 83 patients

Radiotherapy for localized sebaceous carcinoma of the eyelid: a retrospective analysis of 83... Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 Journal of Radiation Research, Vol. 60, No. 5, 2019, pp. 622–629 doi: 10.1093/jrr/rrz046 Advance Access Publication: 4 July 2019 Radiotherapy for localized sebaceous carcinoma of the eyelid: a retrospective analysis of 83 patients 1, 1,2 3 1 Yoshiaki Takagawa , Wakana Tamaki , Shigenobu Suzuki , Koji Inaba , 1 1 1 1 Naoya Murakami , Kana Takahashi , Hiroshi Igaki , Yuko Nakayama , 4 1 Naoyuki Shigematsu and Jun Itami Department of Radiation Oncology, National Cancer Center Hospital, Tokyo, Japan Department of Radiation Oncology, Prefectural Chubu Hospital, Okinawa, Japan Department of Ophthalmology, National Cancer Center Hospital, Tokyo, Japan Department of Radiology, Keio University School of Medicine, Tokyo, Japan *Corresponding author. Responsible for statistical analyses, Department of Radiation Oncology, National Cancer Center Hospital, Tsukiji 5-1-1, Chuo-ku, Tokyo 104-0045, Japan. Tel: +81-3-3542-2511; Fax: +81-3-3547-5291; Email: yoshiaki.takagawa@rad.med.keio.ac.jp (Received 12 April 2019; revised 22 May 2019; editorial decision 24 May 2019) ABSTRACT The current study retrospectively analyzed the results of radiotherapy for clinically localized sebaceous carcin- oma of the eyelid.We reviewed records of 83 patients with histologically confirmed sebaceous carcinoma who were treated radiotherapeutically between 1983 and 2015. Sixty-five patients (78%) were initially treated with radiotherapy of curative intent, while the remaining 18 patients underwent postoperative radiotherapy due to tumor recurrence or positive surgical margins. Thirty-seven patients belonged to T1–2, while 46 belonged to T3–4. All 83 patients were treated with radiotherapy with a median radiation dose of 60 Gy. The median follow-up period was 92.1 months (range, 2.8–310.3 months). At the time of analysis, 13 patients (15.1%) died, and 36 patients (43.3%) had local recurrence. The 7-year overall survival, freedom from neck lymph node recurrence, and local control (LC) rates for all patients were 83.5%, 75.5%, and 52.3%, respectively. Patients with a tumor size ≤10 mm had a higher 7-year LC rate than those with a tumor size >10 mm (58.8% vs 46.6%, P = 0.054). Neck lymph node recurrence was observed in 17 patients (20%) and significantly related to the tumor size. Late toxicity of an eyelid dysfunction of grade 3 was observed in 1 patient with T3 tumor. Radiotherapy for sebaceous carcinoma of the eyelid is a reasonable alternative to surgical resection for tumors <10 mm in size with few severe complications, while larger tumors should be treated with surgery if feasible. Keywords: radiotherapy; sebaceous carcinoma; eyelid; prognostic factor; toxicity INTRODUCTION invasion of adjacent organs, such as the eyeball and brain, and dis- Sebaceous carcinoma of the eyelid is an uncommon neoplasm that tant metastasis in the advanced stages, sebaceous carcinoma results accounts for <1% of all eyelid tumors and approximately 5% of all in a disease-related mortality rate of 6−30% [4, 5]. Accordingly, eyelid malignancies [1, 2]. This tumor develops most commonly early diagnosis and treatment are imperative. among Asian women of around 70 years of age. Sebaceous carcin- Radical surgical excision with a frozen section control by either a oma most commonly arises from the Meibomian glands anterior to standard method or Mohs micrographic surgery is the most com- the gray line and occasionally from the glands of Zeis or Moll and mon and effective treatment method of sebaceous carcinoma. sebaceous glands in the caruncle [3]. Typical finding of sebaceous However, due to advanced age, presence of coexisting diseases, or carcinoma of the eyelid is shown in Figure 1. All too frequently, refusal of surgery, some patients are unsuitable for surgery. sebaceous carcinoma is misdiagnosed as chalazion, blepharoconjunc- Furthermore, despite a recent progress in reconstructive surgery, tivitis, and basal or squamous cell carcinoma. Through the direct eyelid tumors may be difficult to excise completely without © The Author(s) 2019. Published by Oxford University Press on behalf of The Japan Radiation Research Society and Japanese Society for Radiation Oncology. This is an Open Access article article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. � 622 Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 Radiotherapy for sebaceous carcinoma � 623 Table 1. Patient and treatment characteristics Total patients 83 Female/male 46 (55%)/37 (45%) Median age/range 67 years/28–94 years Performance status 0–1/2–3 76/7 Indication for radiation therapy Definitive therapy as an initial treatment 65 (78%) Postoperative salvage therapy for 11 (13%) recurrent tumor Postoperative adjuvant therapy for 7 (9%) positive surgical margins Tumor location Fig. 1. Typical finding of sebaceous carcinoma of the eyelid. Upper eyelid 50 (60%) Lower eyelid 30 (36%) functional and cosmetic impairment [6, 7]. Therefore, radiotherapy is a treatment option for patients who refuse or are otherwise Upper and lower eyelid 3 (4%) unsuitable for surgery. However, limited information is currently Maximum tumor size available on the role of radiotherapy in treating these tumors. In the present study, we retrospectively analyzed the efficacy Median/range 12 mm/4–35 mm and safety of radiotherapy for the treatment of clinically localized sebaceous carcinoma of the eyelid. T stage (UICC 6th) T1 (≤5 mm) 7 (9%) MATERIALS AND METHODS T2 (5–10 mm) 30 (36%) Patient characteristics A retrospective review of medical records in our institution from T3 (>10 mm) 38 (45%) 1983 to 2015 identified 89 patients with a histologically confirmed sebaceous carcinoma of the eyelid. Four patients who had lymph T4 (invades adjacent structures) 8 (10%) node metastases and/or distant metastases at the time of diagnosis Radiation dose (EQD )(α/β = 10) 2Gy were excluded. Additionally, one patient with a diagnosis of heredi- tary retinoblastoma was excluded because of prior ocular irradiation <60 Gy 9 (11%) with a total dose of 40 Gy, and another patient who was irradiated 60 Gy 64 (77%) palliatively with only 30 Gy in 15 fractions was also excluded. The remaining 83 patients with clinically localized tumors were included in >60 Gy 10 (12%) this analysis. All patients were treated with radiotherapy with curative Radiation modality, n(%) intent. Patient and treatment characteristics are shown in Table 1. There were 37 men and 46 women. The patients’ age ranged from 28 Electron 77 (93%) to 94 years with a median of 67 years, and their Eastern Cooperative Oncology Group performance status ranged from 0 to 3 (median, 1). Photon 6 (7%) The maximum tumor size ranged from 4 to 35 mm (median, 12 Bolus 35 (42%) mm), while a tumor size <10 mm was noted in 25 patients. Epithelial invasion was identified in 34 patients (41%). Lead-based lens block 68 (82%) Sixty-five patients (78%) were initially treated with radiotherapy, EQD = equivalent dose in 2 Gy. 2Gy while the remaining 18 patients (22%) initially underwent surgical resection. The principle of treatment of the eyelid sebaceous carcin- for positive surgical margins. In the current study, we used the TNM oma was considered to be surgery for any T stage, however, if there staging system of the Union Internationale Contre le Cancer (UICC) was a serious systemic disease, large plastic surgery was required, or 6th edition [8]. In patients undergoing postoperative salvage radio- for refusal of surgery, radiotherapy was used as an initial treatment. Of the 18 patients who underwent surgical resection, 11 patients therapy, T stage was allocated to the recurrent tumor size, while patients undergoing adjuvant radiotherapy were classified by the pre- underwent salvage radiotherapy for postoperative recurrent tumors, while 7 patients were treated with postoperative adjuvant radiotherapy operative tumor status. Seven patients had a T1 tumor, 30 patients Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 624 � Y. Takagawa et al. neck lymph node recurrence (FFNR), and local control (LC) rates were calculated according to the Kaplan−Meier method [9], start- ing from the initiation of radiotherapy. In the univariate analysis, statistical differences were estimated using the log-rank test [10]. Multivariate analyses for LC, FFNR, and OS were performed by Cox proportional hazards model using potential prognostic factors employed in the univariate analysis. A probability level <0.05 indicated statistical significance. Statistical analysis was performed using the SPSS software (version 23.0; IBM, Armonk, NY, USA). Late toxicities were graded according to the CTCAE version 4.0 [11]. This retro- spective study was approved by the institutional review board. Fig. 2. Recurrence pattern of 83 patients with sebaceous carcinoma after radiotherapy. RESULTS At the time of this analysis, 36 patients (43.3%) had disease recur- rences. Twenty-three patients had only local recurrence, 3 had only had a T2 tumor, 38 patients had a T3 tumor, and 8 patients had a neck lymph node recurrence, 13 had both local and neck lymph T4 tumor. node recurrences, 1 had neck lymph node recurrence and distant metastases, and 4 had local and neck lymph node recurrences and distant metastases (Figure 2). Therefore, local recurrence was noted Treatments in 36 patients (43.3%), and neck lymph node recurrence in 17 All 83 patients were treated with radiotherapy with curative intent. patients (20%). The treatment volume was determined based on physical examina- The 7-year LC rate of all 83 patients was 52.3% [95% confi- tions, pretreatment computed tomography (CT), or magnetic res- dence interval (CI) :0.40–0.64]. Patients with a tumor size ≤10 mm onance imaging (MRI) scans, and the planning target volume had a higher 7-year LC rate than those with a tumor size >10 mm (PTV) included at least 5-mm margins from the primary tumor. (58.8% vs 46.6%, P = 0.054) (Table 2, Fig. 3). If the analysis Radiotherapy was administered with 3–12 MeV electron appos- was confined to patients undergoing radiotherapy of ≥60 Gy in itional field in 77 patients (93%), and 4 MV X-rays were used in 6 EQD , 7-year LC rates of T1–2 (33 patients) and T3–4 (41 2Gy patients (7%). X-ray irradiation was delivered with a single anterior patients) were 60.0% and 44.2%, respectively, with an almost statis- portal in 3 patients and two anterior angled portals in the remaining tically significant difference (P = 0.053). Six patients had a late local 3 patients. Two patients undergoing X-ray irradiation had T4 recurrence more than 5 years after radiation therapy and only local tumor, one had invasion of striated muscles of the orbit, and recurrence without lymph node and distant metastases. Three of another had invasion of the nasal bone and bulbar conjunctiva. them were treated with local resection, and the other 3 patients During radiotherapy, a bolus was used in 35 patients (42.2%). were treated with exenteration of the eyelid and orbital contents, Thickness of the bolus was 5 mm in 15 patients, 10 mm in 1 with all 6 patients successfully salvaged without further recurrence. patient, and records of the bolus thickness were lacking in the We also analyzed the LC rate of patients undergoing postoperative remaining 19 patients. A lead-based lens block was inserted between salvage radiotherapy for recurrent tumor and those undergoing the eyelid and eyeball with local anesthetics and lubricant in 68 postoperative adjuvant radiotherapy for positive surgical margins. patients (81.9%). In the remaining patients, the eye was fixed by Although the 7-year LC rate in postoperative adjuvant radiotherapy gazing at an object in the contralateral side to exclude the lens from group was favorable compared with those of postoperative salvage the direct radiation beam. The dose of electron beam irradiation radiotherapy group (71.4% vs 51.9%, respectively; P = 0.736), a was prescribed to the peak dose point of the beam central axis, and statistically significant difference in LC was not shown because 1 the electron energy was selected to enclose the PTV with at least patient in the postoperative adjuvant radiotherapy group had local 80% of the peak dose. In X-ray irradiation, the dose was prescribed recurrence 9 years after radiotherapy. In the multivariate analysis, to the isocenter. Daily fractional dose of 1.8–2.0 Gy, administered 5 tumor size and radiation dose did not have a statistically significant days per week, was used in 77 of 83 patients (92.8%), while daily impact on LC (Table 2). fractional dose of 2.2–2.5 Gy was used in the remaining 6 patients. The 7-year FFNR rate of all 83 patients was 75.5% (95%CI: The total dose for the primary tumor in all patients ranged from 0.65–0.86). The relatively high rate of neck lymph node recurrence 48.0 to 70.4 Gy (median, 60.0 Gy). To compare various radiothera- (17 of 83 patients, 20%) prompted us to investigate the risk factors pies with different fractional doses, the equivalent dose in a frac- for its recurrence. Table 2 shows the FFNR rates according to vari- tional dose of 2 Gy (EQD ) was calculated according to the 2Gy ous potential risk factors. In the univariate analysis, patients with a Linear-Quadratic (LQ) model assuming a/b = 10 Gy. tumor size ≤10 mm had a significantly higher 7-year FFNR rate than those with a larger tumor size (88.4% and 63.3%, respectively; Statistical analysis P = 0.037; Table 2, Fig. 3). However, in the multivariate analysis, The median follow-up period for all 83 patients was 92.1 months tumor size did not have a statistically significance effect on FFNR (range, 2.8–310.3 months). Overall survival (OS), freedom from (Table 2). Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 Radiotherapy for sebaceous carcinoma � 625 Table 2. Univariate and multivariate analyses of various potential prognostic factors for LC, FFNR and OS in patients with sebaceous carcinoma of the eyelid treated with radiotherapy No. 7-year LC 7-year FFNR 7-year OS UVA MVA UVA MVA UVA MVA Rate (%) P value HR (95 %CI) P value Rate (%) P value HR (95 %CI) P value Rate (%) P value HR (95 %CI) P value PS 0.439 −− 0.925 −− 0.629 2.861 0.059 0–1 76 53.8 75.5 84.3 (0.959–8.537) 2–3 7 34.3 80.0 66.7 Sex 0.141 −− 0.201 −− 0.065 0.431 0.217 Female 46 57.7 81.8 89.5 (0.113–1.638) Male 37 45.2 67.5 76.5 Age 0.577 −− 0.475 −− 0.012 1.251 0.010 <70 49 56.5 77.6 90.5 (1.055–1.484) ≥70 34 45.4 73.3 71.7 T stage (UICC 6th) 0.232 0.972 0.948 0.134 1.821 0.305 0.224 2.033 0.200 T1 (≤5 mm) 7 66.7 (0.415–2.274) 100 (0.579–5.729) 100 (0.686–6.020) T2 (5–10 mm) 30 56.0 85.8 89.8 T3 (>10 mm) 38 44.1 66.5 74.2 T4 8 56.3 35.0 87.5 Tumor size 0.054 1.031 0.332 0.037 1.014 0.758 0.051 1.148 0.286 ≤10 mm 37 58.8 (0.969–1.098) 88.4 (0.928–1.108) 92.2 (0.891–1.481) >10 mm (including T4) 46 46.6 63.3 76.1 RT dose (EQD ) 0.537 0.930 0.882 0.926 1.943 0.524 0.506 3.941 0.300 2Gy ≥60 Gy 73 52.4 (0.353–2.445) 74.7 (0.252–14.976) 83.0 (0.294–52.768) <60 Gy 10 50.0 80.0 87.5 Indication for RT 0.719 1.142 0.583 0.310 1.156 0.658 0.521 0.662 0.493 Initial treatment 65 51.1 (0.712–1.831) 77.9 (0.608–2.197) 86.5 (0.204–2.151) Salvage RT 11 51.9 77.1 76.2 Adjuvant RT 7 71.4 51.4 68.6 EQD = equivalent dose in 2 Gy, FFNR = freedom from neck lymph node recurrence, HR = hazard ratio, LC = local control, MVA = multivariate analysis, OS = overall survival, PS = performance status, RT = radiation therapy, UVA = univariate analysis. 2Gy Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 626 � Y. Takagawa et al. Of 83 patients, 13 (15.1%) died during the follow-up period. Of and 44.1%, respectively; Table 2), for which we could not deter- these 13 patients, 4 died of sebaceous carcinoma, and the remaining mine the reason. However, a favorable LC rate of T4 tumors might 9 died without clinical recurrence: lung cancer (n = 3) and pancre- be related to the small tumor size <20 mm in 5 of 7 patients with atic cancer (n = 1) were the causes of death, and five patients died T4 tumors. of unknown cause (n = 5). The 7-year OS rate for all patients was The optimal dose for radiotherapy with curative intent for seba- 83.5% (95%CI: 0.74–0.93), and age had a significant impact on OS ceous carcinoma of the eyelid remains unknown. Nunery et al. in the univariate and multivariate analyses (Table 2). found that all 6 patients with sebaceous carcinoma of the eyelid Table 3 outlines the late toxicities in all patients. Grade 3 or experienced local recurrence with radical radiotherapy at total doses more late toxicities were observed in only 1 patient (grade 3 eyelid of 33–54 Gy (median, 50 Gy) in 1.8–3.1 Gy fraction size [19]. dysfunction). This patient had T3 tumor that was treated with Hendley et al. treated 3 patients with sebaceous carcinoma of the radiotherapy with a total dose of 60 Gy. Grade 2 cataract developed eyelid with radiotherapy at 48 Gy and observed local recurrence in in 1 patient. Twenty-five patients had cataract already before radi- a single patient [20]. In contrast, Hata et al. conducted definitive ation therapy due to old age. We could not find any relationship radiotherapy with total doses of 60–66 Gy in 5 patients with seba- between lead-based lens block and late toxicities. Moreover, ceous carcinoma of the eyelid with fraction sizes of 1.8–2Gyin4 approximately 10–30% of patients had grade 1–2 late toxicities patients and 3 Gy in 1 patient. The authors found no recurrences including watery eye, dry eye, eyelid dysfunction, and keratitis. within the radiation field in any of the treated patients [13]. In the current study, a 7-year LC rate of 60.0% was obtained in patients with T1-2 tumors treated with ≥60 Gy. The results suggest that a dose of ≥60 Gy in conventional fractionation may be insufficient to DISCUSSION This is the largest study on radiotherapy of sebaceous carcinoma of eradicate even T1-2 tumors. The 7-year LC rate in a postoperative adjuvant radiotherapy group was favorable compared with those of the eyelid, which includes 83 patients. Surgical resection has been considered as the standard treatment of sebaceous carcinoma of the a postoperative salvage radiotherapy group and initial definitive radiotherapy group (71.4% vs 51.9%, 51.1%). In the postoperative eyelid. However, the current study revealed that radiotherapy yields comparable LC in sebaceous carcinoma with a tumor size ≤10 mm adjuvant radiotherapy group, 3 of 7 patients had local recurrence. One patient treated with 50 Gy recurred in 108.9 months after (58.8% at 7 years). Although the proportions of patients with early- stage tumors were not documented, several studies have reported radiotherapy. The other 2 patients treated with 60 Gy recurred in 6.3 months and 14.3 months after radiotherapy, respectively. While the efficacy of radiotherapy in LC of sebaceous carcinoma [12, 13]. Hata et al. treated 5 patients with radical radiotherapy for gross the remaining 4 patients treated with >60 Gy in EQD (1 patient 2Gy was 59.4 Gy in 27 fractions, 2 patients were 60 Gy in 30 fractions tumors with a median maximum diameter of 12 mm and found that all 5 patients achieved LC [13]. Pardo et al. analyzed 4 patients trea- and 1 patient was 66 Gy in 33 fractions) achieved LC. Therefore, we consider that a dose of ≤60 Gy was insufficient to eradicate ted with radical radiotherapy (total doses of 45–63 Gy) and found that no patients had recurrences at the primary tumor sites during even in a postoperative adjuvant radiotherapy setting. The bolus was used only in 35 of 77 patients who underwent electron beam the follow-up period of 60–17 months [12]. After wide local exci- sion or Mohs micrographic surgery, sebaceous carcinoma of the eye- therapy. Without bolus, the surface dose is lower, especially in 3-MeV electron irradiation. Additionally, PTV was enclosed by at lid can be controlled locally in 60–90% [14–17]. Erovic et al. analyzed 33 patients with periorbital sebaceous carcinomas (76% least 80% of the prescription dose in electron irradiation. Therefore, were classified as T1 or T2) treated with primary surgery. The the dose to PTV appear be lower than the prescribed dose in some 5-year LC rate was 63%, and the 5-year regional control rate was patients. Further optimization of electron beam irradiation might improve the LC rate of sebaceous carcinoma. 58% [17]. The radiotherapeutic results of sebaceous carcinoma with a tumor size ≤10 mm in the current study seemed to be compar- The optimal radiation field for sebaceous carcinoma of the eyelid is also unknown. When performing surgery, several authors recom- able to the surgical series. However, LC rate in the current study was lower than that in the recent series on Mohs micrographic sur- mended that surgical margins of 5–6 mm are required to completely gery. Previous reports on Mohs micrographic surgery had a small excise tumors at a microscopic level [6, 21]. For example, Dogru sample size and shorter follow-up period than the current study et al. reported that recurrence occurred in the primary lesion in 36% of patients who underwent surgical resection with 1–3-mm margins [15]. Sebaceous carcinoma of the eyelid can recur locally very late even after 5 years as stated. Therefore, long-term follow-up is also from the edge of gross tumors but that no local recurrence was observed in patients with at least 5-mm margins [21]. Hata et al. needed in the surgery reports. Moreover, the current study indicated that radiotherapy achieved treated 5 patients with gross tumors by radiotherapy with 10-mm a 7-year LC rate of 46.6% even for sebaceous carcinoma of the eye- margins and found that no patients had local recurrences [13]. In lid with a larger tumor size (>10 mm, including T4). Several the current study, we could not obtain adequate information regard- ing the patterns of local failure, such as in-field, marginal, or out- authors emphasized that a tumor size >10 mm and extensive inva- sion (represented as T3-4 tumors) are closely related to the risk of field recurrence. However, considering the previous reports, the gross tumor with at least 5-mm margins appears to be preferable for recurrence and metastases [18]. This study revealed that radiother- apy can be a viable alternative to surgery even in the patients with clinical target volume when conducting radiotherapy. Sebaceous carcinoma of the eyelid has often been reported to T3-4 tumors. In our study, patients with T4 tumor had a higher 7-year LC rate than those with T2 and T3 tumors (56.3%, 56.0%, lead to neck lymph node recurrence, particularly in the preauricular, Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 Radiotherapy for sebaceous carcinoma � 627 Fig. 3. Outcomes according to tumor size and T stage in patients with sebaceous carcinoma of the eyelid. (A) Local control. (B) Freedom from neck lymph node recurrence (FFNR). (C) Overall survival. parotid, or submandibular lymph nodes in 8−32% of patients Table 3. Late toxicities (n = 83) [4, 5]. In agreement with the previous series, 20% of patients trea- ted with radical radiotherapy in the current study developed neck No. of patients Grade lymph node recurrence. It remains unknown whether prophylactic 12 3 irradiation to the neck lymph nodes is necessary for patients with clinically N0 tumors. Our results indicated that tumor size was a sig- Cataract 1 (1.2%) 0 1 0 nificant risk factor for FFNR in the univariate analysis. Esmaeli et al. Dry eye 8 (9.6%) 6 2 0 reported that T stage was significantly associated with lymph node metastasis in surgery cases [22]. To our knowledge, the current study Watery eye 11 (13.2%) 3 8 0 is the first report to indicate the risk factors for neck lymph node Keratitis 15 (18.1%) 11 4 0 recurrence in patients treated with radiotherapy. The 7-year FFNR rate was 88.4% in patients with a tumor size ≤10 mm but only 63.3% Eyelid dysfunction 29 (34.9%) 14 14 1 in patients with a larger tumor size. Additionally, of 17 patients with neck lymph node recurrences, 13 patients had both local and neck No grade 4 or 5 toxicities were observed. Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 628 � Y. Takagawa et al. lymph node recurrences. Ten of 13 patients had local recurrence ACKNOWLEDGEMENTS before or concurrent with neck lymph node recurrence (Figure 2). We would like to thank Editage (www.editage.jp) for English lan- This suggests that improvement in LC could favorably affect FFNR. guage editing. However, prophylactic neck irradiation appears to be warranted in patients with a tumor size >10 mm, including T4 tumors. A total dose of 45–50 Gy appears to be appropriate for prophylactic neck REFERENCES irradiation in patients with N0 diseases. Indeed, Hata et al. performed 1. Shields JA, Shields CL. Sebaceous adenocarcinoma of the eyelid. prophylactic neck irradiation with doses of 44–50.5 Gy and reported Int Ophthalmol Clin 2009;49:45–61. no neck lymph node recurrences within the radiation field [13]. 2. Doxanas MT, Green WR. Sebaceous gland carcinoma. Review Recently, systemic chemotherapy with various agents, such as of 40 cases. Arch Ophthalmol 1984;102:245–9. doxorubicin, cisplatin, and 5-fluorouracil, is reported to have 3. Boniuk M, Zimmerman LE. Sebaceous carcinoma of the eyelid, achieved a significant reduction in the volume of bulky eyelid eyebrow, caruncle, and orbit. Trans Am Acad Ophthalmol tumors and contributed to prolonged survival [23]. Therefore, Otolaryngol 1968;72:619–42. multimodal treatment, including surgery and radiotherapy with or 4. Song A, Carter KD, Syed NA et al. Sebaceous cell carcinoma of without chemotherapy, may be appropriate for T3-4 tumors. the ocular adnexa: clinical presentations, histopathology, and In the current study, late morbidity of grade 3 or more was outcomes. Ophthal Plast Reconstr Surg 2008;24:194–200. observed in only 1 patient, suggesting that radiotherapy in the treat- 5. Shields JA, Demirci H, Marr BP et al. Sebaceous carcinoma of ment of these tumors is quite safe. Only 1 patient developed the ocular region: a review. Surv Ophthalmol 2005;50:103–22. radiation-induced cataract. Several authors have also indicated that 6. Paridaens D, van den Bosch WA. Orbicularis muscle advance- radical radiotherapy is associated with few severe late toxicities [1]. ment flap combined with free posterior and anterior lamellar In the current study, approximately 10–30% of patients experienced grafts: a 1-stage sandwich technique for eyelid reconstruction. grade 1–2 late toxicities, including watery eye, dry eye, eyelid dys- Ophthalmology 2008;115:189–94. function, and keratitis with a median total dose of 60 Gy. These 7. Ohshiro T, Tsuchida Y, Harashina T et al. Reconstruction of results indicate that radiotherapy is associated with few grade 3 or the levator function and conjunctiva using an inferiorly based higher late toxicities, while some patients may have grade 1–2 late orbital septal flap for subtotal full-thickness defects of the upper toxicities. However, the true incidence of radiation-induced cataract eyelid: two cases of ocular sebaceous carcinoma. Ann Plast Surg might be underestimated because cataract had already been present 2006;56:336–9. before radiation therapy in 25 patients due to old age. Lead shields 8. Sobin LH, Wittekind C. TNM classification of malignant tumors. may be effective in reducing late toxicities, but we could not find 6th ed. New York: John Wiley & Sons, 2002. the correlation between lead shields and late toxicities. Because of 9. Kaplan EL, Meier P. Nonparametric estimation from incom- the retrospective nature of this study and the study period extending plete observations. J Am Stat Assoc 1958;53:457–81. to more than 30 years, some patients were lost to follow-up, and 10. Mantel N. Evaluation of survival data and two new rank order the incidence of late toxicities may be underestimated. statistics arising in its consideration. Cancer Chemother Rep 1966;50:163–70. 11. National Cancer Institute, Common Terminology Criteria for CONCLUSION Adverse Events (CTCAE) version 4.0. http://ctep.cancer.gov/ Radiotherapy for sebaceous carcinoma of the eyelid is a reasonable protocolDevelopment/electronic_applications/ctc.htm#ctc_40 alternative to surgical resection for tumors <10 mm in size with few (11 May 2019, date last accessed). severe complications, while larger tumors should be treated with 12. Pardo FS, Wang CC, Albert D et al. Sebaceous carcinoma of surgery with or without radiotherapy, if feasible. Radiotherapy alone the ocular adnexa: radiotherapeutic management. Int J Radiat can be a reasonable alternative even for patients with T3–4 tumors Oncol Biol Phys 1989;17:643–7. (>10 mm) if surgical removal is not possible. Prophylactic neck 13. Hata M, Koike I, Omura M et al. Noninvasive and curative radi- irradiation appears to be required for patients with T3-4 tumors. ation therapy for sebaceous carcinoma of the eyelid. Int J Radiat Further investigations are required to determine the optimal radi- Oncol Biol Phys 2012;82:605–11. ation dose and field for radiotherapy of sebaceous carcinoma. 14. Callahan EF, Appert DL, Roenigk RK et al. Sebaceous carcin- oma of the eyelid: A review of 14 cases. Dermatol Surg 2004;30: CONFLICT OF INTEREST 1164–8. Jun Itami reports personal fees from HekaBio, ItoChu, Tsumura, 15. Kyllo RL, Brady KL, Hurst EA. Sebaceous carcinoma: Review Taiho and M3, non-financial support from Keys-Japan outside the of the literature. Dermatol Surg 2015;41:1–15. submitted work. Yuko Nakayama reports personal fees from Astra 16. Spencer JM, Nossa R, Tse DT et al. Sebaceous carcinoma of Zenaka, Chugai, Taiho, Dai-ichi Sankyo, Ono, Rino Medical, Elekta, the eyelid treated with Mohs micrographic surgery. J Am Acad Toshiba, Kirin, Johnson & Johnson, NHK educational, Eli Lilly, Dermatol 2001;44:1004–9. Bristol-Myers Squibb, Bayer Yakuhin and Novartis outside the sub- 17. Erovic BM, Goldstein DP, Kim D et al. Sebaceous gland carcin- mitted work. Hiroshi Igaki reports personal fees from Itochu, oma of the head and neck: The Princess Margaret Hospital Secom Sonpo outside the submitted work. experience. Head Neck 2013;35:316–20. Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 Radiotherapy for sebaceous carcinoma � 629 18. Zürcher M, Hintschich CR, Garner A et al. Sebaceous carcin- 21. Dogru M, Matsuo H, Inoue M et al. Management of eyelid oma of the eyelid: a clinicopathological study. Br J Ophthalmol sebaceous carcinomas. Ophthalmologica 1997;211:40–3. 1998;82:1049–55. 22. Esmaeli B, Nasser QJ, Cruz H et al. American Joint Committee on 19. Nunery WR, Welsh MG, McCord CD Jr. Recurrence of seba- Cancer T category for eyelid sebaceous carcinoma correlates with nodal metastasis and survival. Ophthalmology 2012;119:1078–82. ceous carcinoma of the eyelid after radiation therapy. Am J Ophthalmol 1983;96:10–5. 23. Murthy R, Honavar SG, Burman S et al. Neoadjuvant chemo- 20. Hendley RL, Rieser JC, Cavanagh HD et al. Primary radiation therapy in the management of sebaceous gland carcinoma of therapy for meibomian gland carcinoma. Am J Ophthalmol the eyelid with regional lymph node metastasis. Ophthal Plast 1979;87:206–9. Reconstr Surg 2005;21:307–9. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Radiation Research Oxford University Press

Radiotherapy for localized sebaceous carcinoma of the eyelid: a retrospective analysis of 83 patients

Loading next page...
 
/lp/oxford-university-press/radiotherapy-for-localized-sebaceous-carcinoma-of-the-eyelid-a-a5wxMxJXR0
Publisher
Oxford University Press
Copyright
© The Author(s) 2019. Published by Oxford University Press on behalf of The Japan Radiation Research Society and Japanese Society for Radiation Oncology.
ISSN
0449-3060
eISSN
1349-9157
DOI
10.1093/jrr/rrz046
Publisher site
See Article on Publisher Site

Abstract

Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 Journal of Radiation Research, Vol. 60, No. 5, 2019, pp. 622–629 doi: 10.1093/jrr/rrz046 Advance Access Publication: 4 July 2019 Radiotherapy for localized sebaceous carcinoma of the eyelid: a retrospective analysis of 83 patients 1, 1,2 3 1 Yoshiaki Takagawa , Wakana Tamaki , Shigenobu Suzuki , Koji Inaba , 1 1 1 1 Naoya Murakami , Kana Takahashi , Hiroshi Igaki , Yuko Nakayama , 4 1 Naoyuki Shigematsu and Jun Itami Department of Radiation Oncology, National Cancer Center Hospital, Tokyo, Japan Department of Radiation Oncology, Prefectural Chubu Hospital, Okinawa, Japan Department of Ophthalmology, National Cancer Center Hospital, Tokyo, Japan Department of Radiology, Keio University School of Medicine, Tokyo, Japan *Corresponding author. Responsible for statistical analyses, Department of Radiation Oncology, National Cancer Center Hospital, Tsukiji 5-1-1, Chuo-ku, Tokyo 104-0045, Japan. Tel: +81-3-3542-2511; Fax: +81-3-3547-5291; Email: yoshiaki.takagawa@rad.med.keio.ac.jp (Received 12 April 2019; revised 22 May 2019; editorial decision 24 May 2019) ABSTRACT The current study retrospectively analyzed the results of radiotherapy for clinically localized sebaceous carcin- oma of the eyelid.We reviewed records of 83 patients with histologically confirmed sebaceous carcinoma who were treated radiotherapeutically between 1983 and 2015. Sixty-five patients (78%) were initially treated with radiotherapy of curative intent, while the remaining 18 patients underwent postoperative radiotherapy due to tumor recurrence or positive surgical margins. Thirty-seven patients belonged to T1–2, while 46 belonged to T3–4. All 83 patients were treated with radiotherapy with a median radiation dose of 60 Gy. The median follow-up period was 92.1 months (range, 2.8–310.3 months). At the time of analysis, 13 patients (15.1%) died, and 36 patients (43.3%) had local recurrence. The 7-year overall survival, freedom from neck lymph node recurrence, and local control (LC) rates for all patients were 83.5%, 75.5%, and 52.3%, respectively. Patients with a tumor size ≤10 mm had a higher 7-year LC rate than those with a tumor size >10 mm (58.8% vs 46.6%, P = 0.054). Neck lymph node recurrence was observed in 17 patients (20%) and significantly related to the tumor size. Late toxicity of an eyelid dysfunction of grade 3 was observed in 1 patient with T3 tumor. Radiotherapy for sebaceous carcinoma of the eyelid is a reasonable alternative to surgical resection for tumors <10 mm in size with few severe complications, while larger tumors should be treated with surgery if feasible. Keywords: radiotherapy; sebaceous carcinoma; eyelid; prognostic factor; toxicity INTRODUCTION invasion of adjacent organs, such as the eyeball and brain, and dis- Sebaceous carcinoma of the eyelid is an uncommon neoplasm that tant metastasis in the advanced stages, sebaceous carcinoma results accounts for <1% of all eyelid tumors and approximately 5% of all in a disease-related mortality rate of 6−30% [4, 5]. Accordingly, eyelid malignancies [1, 2]. This tumor develops most commonly early diagnosis and treatment are imperative. among Asian women of around 70 years of age. Sebaceous carcin- Radical surgical excision with a frozen section control by either a oma most commonly arises from the Meibomian glands anterior to standard method or Mohs micrographic surgery is the most com- the gray line and occasionally from the glands of Zeis or Moll and mon and effective treatment method of sebaceous carcinoma. sebaceous glands in the caruncle [3]. Typical finding of sebaceous However, due to advanced age, presence of coexisting diseases, or carcinoma of the eyelid is shown in Figure 1. All too frequently, refusal of surgery, some patients are unsuitable for surgery. sebaceous carcinoma is misdiagnosed as chalazion, blepharoconjunc- Furthermore, despite a recent progress in reconstructive surgery, tivitis, and basal or squamous cell carcinoma. Through the direct eyelid tumors may be difficult to excise completely without © The Author(s) 2019. Published by Oxford University Press on behalf of The Japan Radiation Research Society and Japanese Society for Radiation Oncology. This is an Open Access article article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. � 622 Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 Radiotherapy for sebaceous carcinoma � 623 Table 1. Patient and treatment characteristics Total patients 83 Female/male 46 (55%)/37 (45%) Median age/range 67 years/28–94 years Performance status 0–1/2–3 76/7 Indication for radiation therapy Definitive therapy as an initial treatment 65 (78%) Postoperative salvage therapy for 11 (13%) recurrent tumor Postoperative adjuvant therapy for 7 (9%) positive surgical margins Tumor location Fig. 1. Typical finding of sebaceous carcinoma of the eyelid. Upper eyelid 50 (60%) Lower eyelid 30 (36%) functional and cosmetic impairment [6, 7]. Therefore, radiotherapy is a treatment option for patients who refuse or are otherwise Upper and lower eyelid 3 (4%) unsuitable for surgery. However, limited information is currently Maximum tumor size available on the role of radiotherapy in treating these tumors. In the present study, we retrospectively analyzed the efficacy Median/range 12 mm/4–35 mm and safety of radiotherapy for the treatment of clinically localized sebaceous carcinoma of the eyelid. T stage (UICC 6th) T1 (≤5 mm) 7 (9%) MATERIALS AND METHODS T2 (5–10 mm) 30 (36%) Patient characteristics A retrospective review of medical records in our institution from T3 (>10 mm) 38 (45%) 1983 to 2015 identified 89 patients with a histologically confirmed sebaceous carcinoma of the eyelid. Four patients who had lymph T4 (invades adjacent structures) 8 (10%) node metastases and/or distant metastases at the time of diagnosis Radiation dose (EQD )(α/β = 10) 2Gy were excluded. Additionally, one patient with a diagnosis of heredi- tary retinoblastoma was excluded because of prior ocular irradiation <60 Gy 9 (11%) with a total dose of 40 Gy, and another patient who was irradiated 60 Gy 64 (77%) palliatively with only 30 Gy in 15 fractions was also excluded. The remaining 83 patients with clinically localized tumors were included in >60 Gy 10 (12%) this analysis. All patients were treated with radiotherapy with curative Radiation modality, n(%) intent. Patient and treatment characteristics are shown in Table 1. There were 37 men and 46 women. The patients’ age ranged from 28 Electron 77 (93%) to 94 years with a median of 67 years, and their Eastern Cooperative Oncology Group performance status ranged from 0 to 3 (median, 1). Photon 6 (7%) The maximum tumor size ranged from 4 to 35 mm (median, 12 Bolus 35 (42%) mm), while a tumor size <10 mm was noted in 25 patients. Epithelial invasion was identified in 34 patients (41%). Lead-based lens block 68 (82%) Sixty-five patients (78%) were initially treated with radiotherapy, EQD = equivalent dose in 2 Gy. 2Gy while the remaining 18 patients (22%) initially underwent surgical resection. The principle of treatment of the eyelid sebaceous carcin- for positive surgical margins. In the current study, we used the TNM oma was considered to be surgery for any T stage, however, if there staging system of the Union Internationale Contre le Cancer (UICC) was a serious systemic disease, large plastic surgery was required, or 6th edition [8]. In patients undergoing postoperative salvage radio- for refusal of surgery, radiotherapy was used as an initial treatment. Of the 18 patients who underwent surgical resection, 11 patients therapy, T stage was allocated to the recurrent tumor size, while patients undergoing adjuvant radiotherapy were classified by the pre- underwent salvage radiotherapy for postoperative recurrent tumors, while 7 patients were treated with postoperative adjuvant radiotherapy operative tumor status. Seven patients had a T1 tumor, 30 patients Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 624 � Y. Takagawa et al. neck lymph node recurrence (FFNR), and local control (LC) rates were calculated according to the Kaplan−Meier method [9], start- ing from the initiation of radiotherapy. In the univariate analysis, statistical differences were estimated using the log-rank test [10]. Multivariate analyses for LC, FFNR, and OS were performed by Cox proportional hazards model using potential prognostic factors employed in the univariate analysis. A probability level <0.05 indicated statistical significance. Statistical analysis was performed using the SPSS software (version 23.0; IBM, Armonk, NY, USA). Late toxicities were graded according to the CTCAE version 4.0 [11]. This retro- spective study was approved by the institutional review board. Fig. 2. Recurrence pattern of 83 patients with sebaceous carcinoma after radiotherapy. RESULTS At the time of this analysis, 36 patients (43.3%) had disease recur- rences. Twenty-three patients had only local recurrence, 3 had only had a T2 tumor, 38 patients had a T3 tumor, and 8 patients had a neck lymph node recurrence, 13 had both local and neck lymph T4 tumor. node recurrences, 1 had neck lymph node recurrence and distant metastases, and 4 had local and neck lymph node recurrences and distant metastases (Figure 2). Therefore, local recurrence was noted Treatments in 36 patients (43.3%), and neck lymph node recurrence in 17 All 83 patients were treated with radiotherapy with curative intent. patients (20%). The treatment volume was determined based on physical examina- The 7-year LC rate of all 83 patients was 52.3% [95% confi- tions, pretreatment computed tomography (CT), or magnetic res- dence interval (CI) :0.40–0.64]. Patients with a tumor size ≤10 mm onance imaging (MRI) scans, and the planning target volume had a higher 7-year LC rate than those with a tumor size >10 mm (PTV) included at least 5-mm margins from the primary tumor. (58.8% vs 46.6%, P = 0.054) (Table 2, Fig. 3). If the analysis Radiotherapy was administered with 3–12 MeV electron appos- was confined to patients undergoing radiotherapy of ≥60 Gy in itional field in 77 patients (93%), and 4 MV X-rays were used in 6 EQD , 7-year LC rates of T1–2 (33 patients) and T3–4 (41 2Gy patients (7%). X-ray irradiation was delivered with a single anterior patients) were 60.0% and 44.2%, respectively, with an almost statis- portal in 3 patients and two anterior angled portals in the remaining tically significant difference (P = 0.053). Six patients had a late local 3 patients. Two patients undergoing X-ray irradiation had T4 recurrence more than 5 years after radiation therapy and only local tumor, one had invasion of striated muscles of the orbit, and recurrence without lymph node and distant metastases. Three of another had invasion of the nasal bone and bulbar conjunctiva. them were treated with local resection, and the other 3 patients During radiotherapy, a bolus was used in 35 patients (42.2%). were treated with exenteration of the eyelid and orbital contents, Thickness of the bolus was 5 mm in 15 patients, 10 mm in 1 with all 6 patients successfully salvaged without further recurrence. patient, and records of the bolus thickness were lacking in the We also analyzed the LC rate of patients undergoing postoperative remaining 19 patients. A lead-based lens block was inserted between salvage radiotherapy for recurrent tumor and those undergoing the eyelid and eyeball with local anesthetics and lubricant in 68 postoperative adjuvant radiotherapy for positive surgical margins. patients (81.9%). In the remaining patients, the eye was fixed by Although the 7-year LC rate in postoperative adjuvant radiotherapy gazing at an object in the contralateral side to exclude the lens from group was favorable compared with those of postoperative salvage the direct radiation beam. The dose of electron beam irradiation radiotherapy group (71.4% vs 51.9%, respectively; P = 0.736), a was prescribed to the peak dose point of the beam central axis, and statistically significant difference in LC was not shown because 1 the electron energy was selected to enclose the PTV with at least patient in the postoperative adjuvant radiotherapy group had local 80% of the peak dose. In X-ray irradiation, the dose was prescribed recurrence 9 years after radiotherapy. In the multivariate analysis, to the isocenter. Daily fractional dose of 1.8–2.0 Gy, administered 5 tumor size and radiation dose did not have a statistically significant days per week, was used in 77 of 83 patients (92.8%), while daily impact on LC (Table 2). fractional dose of 2.2–2.5 Gy was used in the remaining 6 patients. The 7-year FFNR rate of all 83 patients was 75.5% (95%CI: The total dose for the primary tumor in all patients ranged from 0.65–0.86). The relatively high rate of neck lymph node recurrence 48.0 to 70.4 Gy (median, 60.0 Gy). To compare various radiothera- (17 of 83 patients, 20%) prompted us to investigate the risk factors pies with different fractional doses, the equivalent dose in a frac- for its recurrence. Table 2 shows the FFNR rates according to vari- tional dose of 2 Gy (EQD ) was calculated according to the 2Gy ous potential risk factors. In the univariate analysis, patients with a Linear-Quadratic (LQ) model assuming a/b = 10 Gy. tumor size ≤10 mm had a significantly higher 7-year FFNR rate than those with a larger tumor size (88.4% and 63.3%, respectively; Statistical analysis P = 0.037; Table 2, Fig. 3). However, in the multivariate analysis, The median follow-up period for all 83 patients was 92.1 months tumor size did not have a statistically significance effect on FFNR (range, 2.8–310.3 months). Overall survival (OS), freedom from (Table 2). Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 Radiotherapy for sebaceous carcinoma � 625 Table 2. Univariate and multivariate analyses of various potential prognostic factors for LC, FFNR and OS in patients with sebaceous carcinoma of the eyelid treated with radiotherapy No. 7-year LC 7-year FFNR 7-year OS UVA MVA UVA MVA UVA MVA Rate (%) P value HR (95 %CI) P value Rate (%) P value HR (95 %CI) P value Rate (%) P value HR (95 %CI) P value PS 0.439 −− 0.925 −− 0.629 2.861 0.059 0–1 76 53.8 75.5 84.3 (0.959–8.537) 2–3 7 34.3 80.0 66.7 Sex 0.141 −− 0.201 −− 0.065 0.431 0.217 Female 46 57.7 81.8 89.5 (0.113–1.638) Male 37 45.2 67.5 76.5 Age 0.577 −− 0.475 −− 0.012 1.251 0.010 <70 49 56.5 77.6 90.5 (1.055–1.484) ≥70 34 45.4 73.3 71.7 T stage (UICC 6th) 0.232 0.972 0.948 0.134 1.821 0.305 0.224 2.033 0.200 T1 (≤5 mm) 7 66.7 (0.415–2.274) 100 (0.579–5.729) 100 (0.686–6.020) T2 (5–10 mm) 30 56.0 85.8 89.8 T3 (>10 mm) 38 44.1 66.5 74.2 T4 8 56.3 35.0 87.5 Tumor size 0.054 1.031 0.332 0.037 1.014 0.758 0.051 1.148 0.286 ≤10 mm 37 58.8 (0.969–1.098) 88.4 (0.928–1.108) 92.2 (0.891–1.481) >10 mm (including T4) 46 46.6 63.3 76.1 RT dose (EQD ) 0.537 0.930 0.882 0.926 1.943 0.524 0.506 3.941 0.300 2Gy ≥60 Gy 73 52.4 (0.353–2.445) 74.7 (0.252–14.976) 83.0 (0.294–52.768) <60 Gy 10 50.0 80.0 87.5 Indication for RT 0.719 1.142 0.583 0.310 1.156 0.658 0.521 0.662 0.493 Initial treatment 65 51.1 (0.712–1.831) 77.9 (0.608–2.197) 86.5 (0.204–2.151) Salvage RT 11 51.9 77.1 76.2 Adjuvant RT 7 71.4 51.4 68.6 EQD = equivalent dose in 2 Gy, FFNR = freedom from neck lymph node recurrence, HR = hazard ratio, LC = local control, MVA = multivariate analysis, OS = overall survival, PS = performance status, RT = radiation therapy, UVA = univariate analysis. 2Gy Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 626 � Y. Takagawa et al. Of 83 patients, 13 (15.1%) died during the follow-up period. Of and 44.1%, respectively; Table 2), for which we could not deter- these 13 patients, 4 died of sebaceous carcinoma, and the remaining mine the reason. However, a favorable LC rate of T4 tumors might 9 died without clinical recurrence: lung cancer (n = 3) and pancre- be related to the small tumor size <20 mm in 5 of 7 patients with atic cancer (n = 1) were the causes of death, and five patients died T4 tumors. of unknown cause (n = 5). The 7-year OS rate for all patients was The optimal dose for radiotherapy with curative intent for seba- 83.5% (95%CI: 0.74–0.93), and age had a significant impact on OS ceous carcinoma of the eyelid remains unknown. Nunery et al. in the univariate and multivariate analyses (Table 2). found that all 6 patients with sebaceous carcinoma of the eyelid Table 3 outlines the late toxicities in all patients. Grade 3 or experienced local recurrence with radical radiotherapy at total doses more late toxicities were observed in only 1 patient (grade 3 eyelid of 33–54 Gy (median, 50 Gy) in 1.8–3.1 Gy fraction size [19]. dysfunction). This patient had T3 tumor that was treated with Hendley et al. treated 3 patients with sebaceous carcinoma of the radiotherapy with a total dose of 60 Gy. Grade 2 cataract developed eyelid with radiotherapy at 48 Gy and observed local recurrence in in 1 patient. Twenty-five patients had cataract already before radi- a single patient [20]. In contrast, Hata et al. conducted definitive ation therapy due to old age. We could not find any relationship radiotherapy with total doses of 60–66 Gy in 5 patients with seba- between lead-based lens block and late toxicities. Moreover, ceous carcinoma of the eyelid with fraction sizes of 1.8–2Gyin4 approximately 10–30% of patients had grade 1–2 late toxicities patients and 3 Gy in 1 patient. The authors found no recurrences including watery eye, dry eye, eyelid dysfunction, and keratitis. within the radiation field in any of the treated patients [13]. In the current study, a 7-year LC rate of 60.0% was obtained in patients with T1-2 tumors treated with ≥60 Gy. The results suggest that a dose of ≥60 Gy in conventional fractionation may be insufficient to DISCUSSION This is the largest study on radiotherapy of sebaceous carcinoma of eradicate even T1-2 tumors. The 7-year LC rate in a postoperative adjuvant radiotherapy group was favorable compared with those of the eyelid, which includes 83 patients. Surgical resection has been considered as the standard treatment of sebaceous carcinoma of the a postoperative salvage radiotherapy group and initial definitive radiotherapy group (71.4% vs 51.9%, 51.1%). In the postoperative eyelid. However, the current study revealed that radiotherapy yields comparable LC in sebaceous carcinoma with a tumor size ≤10 mm adjuvant radiotherapy group, 3 of 7 patients had local recurrence. One patient treated with 50 Gy recurred in 108.9 months after (58.8% at 7 years). Although the proportions of patients with early- stage tumors were not documented, several studies have reported radiotherapy. The other 2 patients treated with 60 Gy recurred in 6.3 months and 14.3 months after radiotherapy, respectively. While the efficacy of radiotherapy in LC of sebaceous carcinoma [12, 13]. Hata et al. treated 5 patients with radical radiotherapy for gross the remaining 4 patients treated with >60 Gy in EQD (1 patient 2Gy was 59.4 Gy in 27 fractions, 2 patients were 60 Gy in 30 fractions tumors with a median maximum diameter of 12 mm and found that all 5 patients achieved LC [13]. Pardo et al. analyzed 4 patients trea- and 1 patient was 66 Gy in 33 fractions) achieved LC. Therefore, we consider that a dose of ≤60 Gy was insufficient to eradicate ted with radical radiotherapy (total doses of 45–63 Gy) and found that no patients had recurrences at the primary tumor sites during even in a postoperative adjuvant radiotherapy setting. The bolus was used only in 35 of 77 patients who underwent electron beam the follow-up period of 60–17 months [12]. After wide local exci- sion or Mohs micrographic surgery, sebaceous carcinoma of the eye- therapy. Without bolus, the surface dose is lower, especially in 3-MeV electron irradiation. Additionally, PTV was enclosed by at lid can be controlled locally in 60–90% [14–17]. Erovic et al. analyzed 33 patients with periorbital sebaceous carcinomas (76% least 80% of the prescription dose in electron irradiation. Therefore, were classified as T1 or T2) treated with primary surgery. The the dose to PTV appear be lower than the prescribed dose in some 5-year LC rate was 63%, and the 5-year regional control rate was patients. Further optimization of electron beam irradiation might improve the LC rate of sebaceous carcinoma. 58% [17]. The radiotherapeutic results of sebaceous carcinoma with a tumor size ≤10 mm in the current study seemed to be compar- The optimal radiation field for sebaceous carcinoma of the eyelid is also unknown. When performing surgery, several authors recom- able to the surgical series. However, LC rate in the current study was lower than that in the recent series on Mohs micrographic sur- mended that surgical margins of 5–6 mm are required to completely gery. Previous reports on Mohs micrographic surgery had a small excise tumors at a microscopic level [6, 21]. For example, Dogru sample size and shorter follow-up period than the current study et al. reported that recurrence occurred in the primary lesion in 36% of patients who underwent surgical resection with 1–3-mm margins [15]. Sebaceous carcinoma of the eyelid can recur locally very late even after 5 years as stated. Therefore, long-term follow-up is also from the edge of gross tumors but that no local recurrence was observed in patients with at least 5-mm margins [21]. Hata et al. needed in the surgery reports. Moreover, the current study indicated that radiotherapy achieved treated 5 patients with gross tumors by radiotherapy with 10-mm a 7-year LC rate of 46.6% even for sebaceous carcinoma of the eye- margins and found that no patients had local recurrences [13]. In lid with a larger tumor size (>10 mm, including T4). Several the current study, we could not obtain adequate information regard- ing the patterns of local failure, such as in-field, marginal, or out- authors emphasized that a tumor size >10 mm and extensive inva- sion (represented as T3-4 tumors) are closely related to the risk of field recurrence. However, considering the previous reports, the gross tumor with at least 5-mm margins appears to be preferable for recurrence and metastases [18]. This study revealed that radiother- apy can be a viable alternative to surgery even in the patients with clinical target volume when conducting radiotherapy. Sebaceous carcinoma of the eyelid has often been reported to T3-4 tumors. In our study, patients with T4 tumor had a higher 7-year LC rate than those with T2 and T3 tumors (56.3%, 56.0%, lead to neck lymph node recurrence, particularly in the preauricular, Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 Radiotherapy for sebaceous carcinoma � 627 Fig. 3. Outcomes according to tumor size and T stage in patients with sebaceous carcinoma of the eyelid. (A) Local control. (B) Freedom from neck lymph node recurrence (FFNR). (C) Overall survival. parotid, or submandibular lymph nodes in 8−32% of patients Table 3. Late toxicities (n = 83) [4, 5]. In agreement with the previous series, 20% of patients trea- ted with radical radiotherapy in the current study developed neck No. of patients Grade lymph node recurrence. It remains unknown whether prophylactic 12 3 irradiation to the neck lymph nodes is necessary for patients with clinically N0 tumors. Our results indicated that tumor size was a sig- Cataract 1 (1.2%) 0 1 0 nificant risk factor for FFNR in the univariate analysis. Esmaeli et al. Dry eye 8 (9.6%) 6 2 0 reported that T stage was significantly associated with lymph node metastasis in surgery cases [22]. To our knowledge, the current study Watery eye 11 (13.2%) 3 8 0 is the first report to indicate the risk factors for neck lymph node Keratitis 15 (18.1%) 11 4 0 recurrence in patients treated with radiotherapy. The 7-year FFNR rate was 88.4% in patients with a tumor size ≤10 mm but only 63.3% Eyelid dysfunction 29 (34.9%) 14 14 1 in patients with a larger tumor size. Additionally, of 17 patients with neck lymph node recurrences, 13 patients had both local and neck No grade 4 or 5 toxicities were observed. Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 628 � Y. Takagawa et al. lymph node recurrences. Ten of 13 patients had local recurrence ACKNOWLEDGEMENTS before or concurrent with neck lymph node recurrence (Figure 2). We would like to thank Editage (www.editage.jp) for English lan- This suggests that improvement in LC could favorably affect FFNR. guage editing. However, prophylactic neck irradiation appears to be warranted in patients with a tumor size >10 mm, including T4 tumors. A total dose of 45–50 Gy appears to be appropriate for prophylactic neck REFERENCES irradiation in patients with N0 diseases. Indeed, Hata et al. performed 1. Shields JA, Shields CL. Sebaceous adenocarcinoma of the eyelid. prophylactic neck irradiation with doses of 44–50.5 Gy and reported Int Ophthalmol Clin 2009;49:45–61. no neck lymph node recurrences within the radiation field [13]. 2. Doxanas MT, Green WR. Sebaceous gland carcinoma. Review Recently, systemic chemotherapy with various agents, such as of 40 cases. Arch Ophthalmol 1984;102:245–9. doxorubicin, cisplatin, and 5-fluorouracil, is reported to have 3. Boniuk M, Zimmerman LE. Sebaceous carcinoma of the eyelid, achieved a significant reduction in the volume of bulky eyelid eyebrow, caruncle, and orbit. Trans Am Acad Ophthalmol tumors and contributed to prolonged survival [23]. Therefore, Otolaryngol 1968;72:619–42. multimodal treatment, including surgery and radiotherapy with or 4. Song A, Carter KD, Syed NA et al. Sebaceous cell carcinoma of without chemotherapy, may be appropriate for T3-4 tumors. the ocular adnexa: clinical presentations, histopathology, and In the current study, late morbidity of grade 3 or more was outcomes. Ophthal Plast Reconstr Surg 2008;24:194–200. observed in only 1 patient, suggesting that radiotherapy in the treat- 5. Shields JA, Demirci H, Marr BP et al. Sebaceous carcinoma of ment of these tumors is quite safe. Only 1 patient developed the ocular region: a review. Surv Ophthalmol 2005;50:103–22. radiation-induced cataract. Several authors have also indicated that 6. Paridaens D, van den Bosch WA. Orbicularis muscle advance- radical radiotherapy is associated with few severe late toxicities [1]. ment flap combined with free posterior and anterior lamellar In the current study, approximately 10–30% of patients experienced grafts: a 1-stage sandwich technique for eyelid reconstruction. grade 1–2 late toxicities, including watery eye, dry eye, eyelid dys- Ophthalmology 2008;115:189–94. function, and keratitis with a median total dose of 60 Gy. These 7. Ohshiro T, Tsuchida Y, Harashina T et al. Reconstruction of results indicate that radiotherapy is associated with few grade 3 or the levator function and conjunctiva using an inferiorly based higher late toxicities, while some patients may have grade 1–2 late orbital septal flap for subtotal full-thickness defects of the upper toxicities. However, the true incidence of radiation-induced cataract eyelid: two cases of ocular sebaceous carcinoma. Ann Plast Surg might be underestimated because cataract had already been present 2006;56:336–9. before radiation therapy in 25 patients due to old age. Lead shields 8. Sobin LH, Wittekind C. TNM classification of malignant tumors. may be effective in reducing late toxicities, but we could not find 6th ed. New York: John Wiley & Sons, 2002. the correlation between lead shields and late toxicities. Because of 9. Kaplan EL, Meier P. Nonparametric estimation from incom- the retrospective nature of this study and the study period extending plete observations. J Am Stat Assoc 1958;53:457–81. to more than 30 years, some patients were lost to follow-up, and 10. Mantel N. Evaluation of survival data and two new rank order the incidence of late toxicities may be underestimated. statistics arising in its consideration. Cancer Chemother Rep 1966;50:163–70. 11. National Cancer Institute, Common Terminology Criteria for CONCLUSION Adverse Events (CTCAE) version 4.0. http://ctep.cancer.gov/ Radiotherapy for sebaceous carcinoma of the eyelid is a reasonable protocolDevelopment/electronic_applications/ctc.htm#ctc_40 alternative to surgical resection for tumors <10 mm in size with few (11 May 2019, date last accessed). severe complications, while larger tumors should be treated with 12. Pardo FS, Wang CC, Albert D et al. Sebaceous carcinoma of surgery with or without radiotherapy, if feasible. Radiotherapy alone the ocular adnexa: radiotherapeutic management. Int J Radiat can be a reasonable alternative even for patients with T3–4 tumors Oncol Biol Phys 1989;17:643–7. (>10 mm) if surgical removal is not possible. Prophylactic neck 13. Hata M, Koike I, Omura M et al. Noninvasive and curative radi- irradiation appears to be required for patients with T3-4 tumors. ation therapy for sebaceous carcinoma of the eyelid. Int J Radiat Further investigations are required to determine the optimal radi- Oncol Biol Phys 2012;82:605–11. ation dose and field for radiotherapy of sebaceous carcinoma. 14. Callahan EF, Appert DL, Roenigk RK et al. Sebaceous carcin- oma of the eyelid: A review of 14 cases. Dermatol Surg 2004;30: CONFLICT OF INTEREST 1164–8. Jun Itami reports personal fees from HekaBio, ItoChu, Tsumura, 15. Kyllo RL, Brady KL, Hurst EA. Sebaceous carcinoma: Review Taiho and M3, non-financial support from Keys-Japan outside the of the literature. Dermatol Surg 2015;41:1–15. submitted work. Yuko Nakayama reports personal fees from Astra 16. Spencer JM, Nossa R, Tse DT et al. Sebaceous carcinoma of Zenaka, Chugai, Taiho, Dai-ichi Sankyo, Ono, Rino Medical, Elekta, the eyelid treated with Mohs micrographic surgery. J Am Acad Toshiba, Kirin, Johnson & Johnson, NHK educational, Eli Lilly, Dermatol 2001;44:1004–9. Bristol-Myers Squibb, Bayer Yakuhin and Novartis outside the sub- 17. Erovic BM, Goldstein DP, Kim D et al. Sebaceous gland carcin- mitted work. Hiroshi Igaki reports personal fees from Itochu, oma of the head and neck: The Princess Margaret Hospital Secom Sonpo outside the submitted work. experience. Head Neck 2013;35:316–20. Downloaded from https://academic.oup.com/jrr/article-abstract/60/5/622/5528219 by DeepDyve user on 06 December 2019 Radiotherapy for sebaceous carcinoma � 629 18. Zürcher M, Hintschich CR, Garner A et al. Sebaceous carcin- 21. Dogru M, Matsuo H, Inoue M et al. Management of eyelid oma of the eyelid: a clinicopathological study. Br J Ophthalmol sebaceous carcinomas. Ophthalmologica 1997;211:40–3. 1998;82:1049–55. 22. Esmaeli B, Nasser QJ, Cruz H et al. American Joint Committee on 19. Nunery WR, Welsh MG, McCord CD Jr. Recurrence of seba- Cancer T category for eyelid sebaceous carcinoma correlates with nodal metastasis and survival. Ophthalmology 2012;119:1078–82. ceous carcinoma of the eyelid after radiation therapy. Am J Ophthalmol 1983;96:10–5. 23. Murthy R, Honavar SG, Burman S et al. Neoadjuvant chemo- 20. Hendley RL, Rieser JC, Cavanagh HD et al. Primary radiation therapy in the management of sebaceous gland carcinoma of therapy for meibomian gland carcinoma. Am J Ophthalmol the eyelid with regional lymph node metastasis. Ophthal Plast 1979;87:206–9. Reconstr Surg 2005;21:307–9.

Journal

Journal of Radiation ResearchOxford University Press

Published: Oct 23, 2019

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create folders to
organize your research

Export folders, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off