Lificiguat

Malignant Eyelid Tumors in India: A Study of 536 Asian Indian Patients

Abstract
Purpose: To study the proportion of eyelid malignant tu- mors in an Asian Indian population and to review their clini- cal features and outcomes. Methods: This is a retrospective study of 536 patients. Results: The mean age at presentation with eyelid malignancy was 58 years. Histopathology-prov- en diagnoses of these patients included sebaceous gland carcinoma (SGC) (n = 285, 53%), basal cell carcinoma (BCC) (n = 128, 24%), squamous cell carcinoma (SCC) (n = 99, 18%), and miscellaneous tumors (n = 24, 4%). The statistically sig- nificant differences between eyelid malignant tumors in- cluded age at presentation, tumor location, and tumor ex- tent. The clinicopathological correlation of SGC, BCC, SCC, and miscellaneous tumors was 91, 86, 46, and 38% (p = 0.001), respectively. Comparing SGC with BCC, SCC, and mis- cellaneous tumors, SGC was more commonly associated with tumor recurrence (21 vs. 3, 8, and 13%; p = 0.001), systemic metastasis (13 vs. 0, 4, and 13%; p = 0.001), and death (9 vs. 0, 4, and 0%; p = 0.004). Compared to SGC, BCC, and SCC, locoregional lymph node metastasis was more com- mon with miscellaneous tumors (26 vs. 16, <1, and 8%; p = 0.001) over a mean follow-up period of 19 months. Conclu- sion: In Asian Indians, SGC is twice as common as BCC and 3 times more common than SCC. SGC is associated with poor- er prognosis compared to other eyelid malignant tumors. Introduction In America and Europe, the most common malignant tumor of the eyelid is basal cell carcinoma (BCC), ac- counting for 80–95% of all eyelid malignancies [1–5]. Squamous cell carcinoma (SCC) (<5%), sebaceous gland carcinoma (SGC) (1–3%), malignant melanoma (1%), and miscellaneous tumors (<1%) constitute the rest of eyelid malignancies [1]. In a large series of 5,504 cases with eyelid tumors in a European population including 894 malignant eyelid tumors, BCC was seen in 86%, SCC in 7%, SGC in 3%, and miscellaneous tumors in 4% ofcases [2]. Similarly, in studies from the USA, BCC was seen in 82–91%, SCC in 2–9%, SGC in 0–6%, and miscel- laneous tumors in <1–9% of cases [3, 5].The reports from Asia and our personal experience are contrary to this observation. BCC is less common (11– 65%) while SCC (5–48%) and SGC (7–56%) occur more frequently than in the West [6–21]. In a large study of 1,086 malignant eyelid tumors from China, BCC ac- counted for 38%, SCC was seen in 19%, SGC in 32%, and miscellaneous tumors in 7% of cases [6]. However, there are no large case series on malignant eyelid tumors from an Asian Indian population. In this study, we will discuss the proportion of all histopathology-proven eyelid malig- nancies in an Asian Indian population and compare the clinical presentation, treatment, and outcomes of com- mon eyelid malignancies.This is a retrospective study conducted at the Operation Eyesight Universal Institute for Eye Cancer, LV Prasad Eye Institute, Hyder- abad, India. Institutional review board approval was obtained. A search was conducted in our medical records and ophthalmic pa- thology database for the diagnosis of eyelid malignancies. All caseswith histopathology-proven eyelid malignancy during the period from January 1995 to December 2016 were included in this analysis. Those patients with a single visit to our institute with a clinical di- agnosis of an eyelid malignancy but without a histopathology con- firmation of the diagnosis were excluded from the study.The following data was extracted from the medical records: age (years), gender, tumor location, tumor extent, tumor growth pat- tern, tumor size, tumor classification based on the 8th edition of the American Joint Committee Classification [22], clinical diagno- sis, treatment details, and final histopathology diagnosis. Clinical photographs were reviewed in all cases to confirm the tumor de-tails. The following data related to outcome were recorded: tumor recurrence, globe salvage, locoregional lymph node metastasis, systemic metastasis, and death. The follow-up duration of each patient was also recorded.The data were analyzed based on age at presentation and gender. They were further grouped into 4 categories and analyzed based on the diagnosis of SGC, BCC, SCC, and miscellaneous tumors.For the literature review, a search was conducted with the key- words “malignant,” “tumor,” “eyelid,” and “cancer” in PubMed and Google Scholar, and all related articles were reviewed. Publica- tions from Asian countries were analyzed further.Statistical analysis was performed using the software Origin v7.0 (OriginLab Corporation, Northampton, MA, USA). Continu- ous data were analyzed for normality by the Shapiro-Wilk test and described in terms of mean, median, and range. Among the 4 groups, the data were compared using the Kruskal-Wallis test and post hoc analysis was performed using the Mann-Whitney test (multiple comparisons). Categorical data were described in terms of proportions; among the 4 groups, comparisons were made using the χ2 test and post hoc analysis was performed also using the χ2 test (multiple comparisons). Kaplan-Meier survival analysis was performed to evaluate the probability of survival over time (events being lymph node involvement, distant metastasis, or death). The log-rank test was performed to test if the survival probabilities dif- fered among the tumor subtypes. A p value of less than 0.05 was considered statistically significant. Results A total of 536 patients with a histopathology-proven diagnosis of eyelid malignancy were included in this study. The mean age at presentation with eyelid malig-nancy was 58 years (median, 60 years; range, 4–100 years). The mean age at presentation was 58 years for SGC, 60 years for BCC, 55 years for SCC, and 50 years for miscellaneous tumors. Histopathology-proven diag- noses of these patients (Table 1; Fig. 1) included SGC (n = 285, 53%), BCC (n = 128, 24%), SCC (n = 99, 18%),malignant melanoma (n = 12, 2%), mucoepidermoid/ adenosquamous carcinoma (n = 7, 1%), lymphoma (n = 2, <1%), Merkel cell carcinoma (n = 1, <1%), adenoid cystic carcinoma (n = 1, <1%), and metastases from lung adenocarcinoma (n = 1, <1%). There was a female pre- ponderance for all malignant tumors of the eyelid (Table 2). History of prior surgical intervention was present in 189 (35%) cases including SGC (n = 118, 41%), BCC(n = 23, 18%), SCC (n = 42, 42%), and miscellaneous tu- mors (n = 16, 67%). Based on multiple comparisons and statistical analysis, statistically significant differences were noted among various tumors (Table 3). The statis- tically significant differences among the 4 groups (SGC, BCC, SCC, and miscellaneous tumors) are mentioned below.SGC versus BCCThe tumor epicenter was most often the upper eyelid in SGC (59%) and the lower eyelid in BCC (59%) (p < 0.001). Statistically significant differences between SGC and BCC included higher rates of tumor recurrence (21 vs. 3%; p < 0.001), regional lymph node metastasis (16 vs.<1%; p < 0.001), systemic metastasis (13 vs. 0%; p < 0.001), and death due to metastasis (9 vs. 0%; p = 0.001).* 18 patients were lost to follow-up after incision biopsy at presentation; ** Kruskal-Wallis test; *** χ2 test; na, not applicable. a Post hoc analysis with Bonferroni correction showed that only BCC and SCC were significantly different from each other (p = 0.003, Mann-Whitney test). b Post hoc analysis with Bonferroni correction showed that only BCC was significantly different from SCC (p = 0.004, χ2 test) and SGC (p < 0.001, χ2 test). c Post hoc analysis with Bonferroni correction showed that only BCC was significantly different from SGC (p < 0.001, χ2 test). d Post hoc analysis with Bonferroni correction showed that only BCC was significantly different from SGC (p = 0.006, χ2 test). e Post hoc analysis with Bonferroni correction showed that only BCC was significantly different from miscellaneous tumors (p = 0.003, χ2 test). f Post hoc analysis with Bonferroni correction showed that only BCC was significantly different from SCC (p = 0.004, χ2 test) and miscellaneous tumors (p = 0.003, χ2 test). g Post hoc analysis with Bonferroni correction showed that only SCC was significantly different from SGC (p = 0.001, χ2 test) and BCC (p = 0.005, χ2 test). Post hoc analysis with Bonferroni correction showed that only SGC was significantly different from BCC (p < 0.001, χ2 test) and SCC (p = 0.012, χ2 test). i Post hoc analysis with Bonferroni correction showed that only BCC was significantly different from SGC (p < 0.001, χ2 test), SCC (p = 0.007, χ2 test), and miscellaneous tumors (p < 0.001, χ2 test). j Post hoc analysis with Bonferroni correction showed that only BCC was significantly different from SGC (p < 0.001, χ2 test) and miscellaneous tumors (p < 0.001, χ2 test). k Post hoc analysis with Bonferroni correction showed that only BCC was significantly different from SGC (p = 0.001, χ2 test).The clinicopathological correlation was better with SGC compared to SCC (91 vs. 46%; p = 0.001). Tumor recurrence was more common with SGC compared to SCC (21 vs. 8%; p = 0.012).The patients with BCC were older than those with SCC (60 vs. 55 years; p = 0.003). The following factors were more common and statistically significant with SCCcompared to BCC: tumor epicenter location in the upper eyelid (40 vs. 17%; p = 0.004), the need for orbital exen- teration (19 vs. 5%; p = 0.004), and locoregional lymph node metastasis (8 vs. <1%; p = 0.007). The clinicopatho- logical correlation was better with BCC compared to SCC (86 vs. 46%; p = 0.005).BCC versus Miscellaneous TumorsThe following factors were more common with mis- cellaneous tumors compared to BCC: orbital tumor extension (29 vs. 6%; p = 0.003), need for orbital exenteration (26 vs. 5%; p = 0.003), locoregional lymph node me- tastasis (26 vs. <1%; p < 0.001), and systemic metastasis (13 vs. 0%; p < 0.001). Overall, SGC (n = 38) and miscellaneous tumors (ma- lignant melanoma [n = 3], mucoepidermoid/adenosqua- mous carcinoma [n = 2], lymphoma [n = 1], and metas- tasis [n = 1]) were more commonly associated with locoregional tumor invasion into the orbit/paranasal si- nuses/brain compared to SCC (n = 16) and BCC (n = 9). Locoregional lymph node metastasis was more common in patients with miscellaneous tumors (n = 6, including malignant melanoma [n = 2], Merkel cell carcinoma [n = 1], lymphoma [n = 1], eyelid metastasis [n = 1], and mu-coepidermoid/adenosquamous carcinoma [n = 1]). Over a mean follow-up period of 19 months (median, 7 months; range, 1 week to 273 months), tumor recurrence (21%), systemic metastasis (13%), and death due to metastasis (9%) occurred more commonly in patients with SGC compared to other eyelid malignancies.The 5-year Kaplan-Meier estimates for locoregional lymph node metastasis for SGC, BCC, SCC, and miscel- laneous tumors were 43, 1, 22, and 47% (p < 0.0001), re-spectively (Fig. 2; Table 4), for distant metastasis 36, 0, 11,and 18% (p = 0.0002), respectively (Fig. 3; Table 4), andfor metastasis-related death 25, 0, 11, and 16% (p = 0.003),respectively (Fig. 4; Table 4). Discussion BCC is the most common malignant eyelid tumor in the West [1–5]. In our series, SGC was the most common malignant eyelid tumor and this finding is consistent with some studies from other Asian countries including India, Japan, and Nepal [8, 15, 18–21, 23–26]. In our study, SGC (53%) was twice as common as BCC (24%) and 3 times more common than SCC (18%). Other studies from Chi- na, Korea, Singapore, Taiwan, and Thailand have shown BCC as the most common malignant eyelid tumor [7, 9, 11, 14, 16, 24, 27–33]. However, these studies have also shown a lower proportion of BCC and a higher propor- tion of SGC compared to the West [1–5, 7, 9, 11, 14, 16, 27–33] (Table 5). A study from Japan has shown SCC as the most common malignant eyelid tumor, constituting 48% of all eyelid malignancies [17], while in our study SCC was less common compared to SGC and BCC, con- stituting only 24% of all eyelid malignancies. Similar to our study, the majority of studies on malignant eyelid tu- mors in Asian Indians have shown that SGC is the most common eyelid malignancy (32–56%) in the Asian Indi- an population [15, 18–21, 23, 25, 26]. Referral bias to the tertiary care referral institution may also play a role in this difference of eyelid tumor proportions. The referral of BCC would be much lower compared to other tumors since these tumors may have been more easily managed by referring physicians. In our study, the history of prior intervention was highest for miscellaneous tumors (67%), SCC (42%), and SGC (41%), suggestive of difficulty in managing these cases compared to BCC (18%). Malignant eyelid tumors occur more commonly in the elderly [1–21, 23–32, 34–36]. In our series, the mean age at presentation of all malignant eyelid tumors was >50 years. The patients presenting with miscellaneous tumors (50 years) and SCC (55 years) were younger compared to those with SGC (58 years) and BCC (60 years). Female predominance of malignant eyelid tumors was noted in some studies, while male predominance or no predilec- tion has also been reported [34]. In our study, though there was a slight female predilection, it was not statisti- cally significant. The reason for female preponderance in our study is unknown, though it may be related to in- creased cosmetic concerns of eyelid disfigurement in fe- males, resulting in a higher rate of doctor consult in fe- males compared to males. Periocular SGC most commonly arises in the upper eyelid, accounting for half to two thirds of cases due to a predominance of meibomian glands in the upper eyelid [1]. In our study, 59% of tumors arose from the upper eyelid. Female preponderance of SGC has been reported in previous studies [33, 37, 38], while few studies have shown a male preponderance [39]. In our study, there was a female predilection with a male:female ratio of 1:1.5. A Chinese-American collaborative study on malignant eye- lid tumors including 525 cases from China and 1,543 cas- es from the USA revealed a disparity between the fre- quencies of SGC in both populations, accounting for 33% in China and 2% in the USA [33]. Based on these findings, the authors proposed that the incidence of SGC is higher in the Asian population compared to Caucasians and this could be related to genetics and racial predisposition for SGC in Asians [33].

Subsequent studies from Asian coun- tries have supported this theory. However, this theory of racial predilection may not be true. In a retrospective study of 1,349 cases from a US-based population registry, the incidence of SGC was 2.03 cases per 1000,000 popula- tion in Whites versus 1.07 cases per 1,000,000 population in Asian/Pacific Islanders versus 0.48 per 1,000,000 popu- lation in Blacks, suggestive of a lack of racial predilection of SGC [39]. In Asians, SGC accounts for a higher propor- tion of eyelid malignancies, similar to our study, but this is not due to a higher incidence of SGC but to a relative lack of other malignant eyelid tumors [39]. Asians with malignant eyelid tumors are 6.21 times (range, 3.8–10.1) more likely to have SGC compared to non-Asians [39]. Overall, BCC is the most common malignant eyelid tumor in the world [1, 40], with a male predilection [41]. However, it has been reported that eyelid BCC is more common in young females (<50 years) [39]. In our study, BCC was more common in females across all age groups, with a male:female ratio of 1:1.5. BCC is more common in the lower eyelid (50–66%), followed by medial canthus (13–30%), upper eyelid (15–16%), and lateral canthus (3– 5%) [42, 43]. In our study, the most common tumor loca- tion was the lower eyelid (59%) and the tumor epicenter location in the medial canthus was less frequent at 10%. Pigmented BCC of the eyelid has been noted in 1–50% pa- tients and was more common in patients with darker skin color [44–46]. Haye and Dufier [45] noted that pigmented BCC of the eyelid was 1% in a Parisian population and in- creased to 45% in a Mediterranean population, suggestive of increased incidence in a darker-skinned population. In our study, pigmented BCC was noted in 55% of cases. Eyelid SCC is a disease of the elderly, with a mean age at presentation of 60 years, and has a male predilection [47]. However, in our study, the mean age at presentation of SCC was 55 years and the disease had a slight female preponderance with a male:female ratio of 1:1.1. There was no preference for upper or lower eyelids. The occur- rence of malignant melanoma and other eyelid tumors was rare in our series. However, younger patients (<20 years of age) had a preponderance for non-SGC/BCC/ SCC eyelid malignancies. The recommended modality of treatment for all ma- lignant eyelid tumors is wide excision biopsy under fro- zen section or Moh’s micrographic surgery control. In our study, wide excision biopsy under frozen section was performed in all cases when the tumor was limited to the eyelid. Clinical misdiagnosis was more common with SCC and miscellaneous malignant eyelid tumors with a clinicopathological correlation of 46% for SCC and 38% for miscellaneous tumors. Eyelid SCC (85%) was com- monly misdiagnosed as SGC, and the most common mis- diagnosis of miscellaneous tumors was eyelid metastasis (29%). Locoregional tumor invasion (13%), locoregional lymph node metastasis (16%), tumor recurrence (21%), systemic metastasis (13%), and disease-related death (9%) were more common with SGC compared to SCC and BCC. These findings are comparable with other stud- ies [1, 4, 37–39]. Locoregional lymph node metastasis was also more common with miscellaneous tumors (25%), including malignant melanoma (n = 2), Merkel cell carci- noma (n = 1), lymphoma (n = 1), metastasis (n = 1), and mucoepidermoid/adenosquamous carcinoma (n = 1). In summary, the proportion of SGC is higher in Asian Indians compared to other malignant eyelid tumors. This finding differs from the Western population. SGC, malignant melanoma, and Merkel cell Lificiguat carcinoma are associated with poorer prognosis, with higher chances of locoregional lymph node and systemic metastasis compared to BCC and SCC. This finding is similar to studies from the Western population.