Cholesterol metabolism dysregulation in eyelid sebaceous gland carcinoma and its clinical significance

Abstract

Purpose:

Eyelid sebaceous gland carcinoma (SGC) is a rare and aggressive malignancy of the eyelids. Dysregulated cholesterol metabolism has been found to be involved in pathogenesis of various cancers including glioblastoma, breast, prostate, and pancreatic cancer. The role of cholesterol metabolism in SGC and its molecular characterization remains unexplored. This study evaluates the clinical significance of the expression of a key enzyme involved in cholesterol metabolism, acyl-CoA cholesterol acyltransferase 1 (ACAT1) in eyelid SGC.

Methods:

A cohort of 50 eyelid SGC cases diagnosed over a period of 12 years were selected. Cholesterol quantification, immunohistochemistry (IHC), and real-time PCR to determine the protein and mRNA expression of ACAT were performed.

Results:

Increased concentration of total cholesterol, free cholesterol, and cholesteryl esters was observed in SGC tumor tissues. Overexpression of ACAT1 at protein and mRNA levels was observed in 66% of the eyelid SGC patients. Significant association of high ACAT1 mRNA expression was observed in patients with lymph node metastasis (P = 0.03) and advanced (III/IV) tumor stage (P = 0.02). SGC patients with a high ACAT immunoexpression also had reduced disease-free survival.

Conclusion:

Overexpression of ACAT1 is a frequent event in the pathogenesis of eyelid SGC and is associated with high risk clinicopathological features. It could serve as a reliable poor prognostic biomarker.

Eyelid sebaceous gland carcinoma (SGC) originates from the specialized sebaceous glands found in and around the eyelids predominantly in the tarsus (meibomian glands) and cilia (Zeis glands) and rarely in the caruncle and eyebrow regions. It is a rare and aggressive eyelid carcinoma that could metastasize to lymph nodes or distant organs and has a risk of recurrence.[1,2,3,4,5] Various studies from the Caucasians have demonstrated that SGC is the second most common malignant eyelid tumor. It is the most common among the Asian population ranging from 8% to 60% as compared to the Western population, which accounts for 5% of the eyelid tumors.[6,7,8,9,10,11,12]

Cholesterol is an essential cellular component, playing a role in membrane structure and fluidity, signal transduction, and hormone production. It is acquired through the uptake of low-density lipoprotein (LDL) and high-density lipoprotein (HDL)-associated cholesterol from circulation or de novo synthesis from acyl-CoA present on the endoplasmic reticulum. Free cholesterol inside the cells is esterified and stored as cholesteryl esters (CEs) in lipid droplets (LDs). CE formation is catalyzed by a key enzyme called acyl-CoA cholesterol acyltransferase 1 (ACAT1), an isoform of Sterol O-acyltransferase 1 (SOAT1). It is frequently upregulated in cancer to support membrane biogenesis and lipid storage.[13]

Metabolic reprogramming is a common feature in cancer cells to meet energy demands. In various cancers, cholesterol metabolism reprogramming activates the oncogenic pathways leading to cancer progression. Intratumoral CE accumulation is known to alter cell signaling mechanisms, leading to increased tumor proliferation, invasiveness, and survival or chemotherapy resistance,[14] for instance, the Wnt/β-catenin pathway in pancreatic adenocarcinoma,[15] the sonic hedgehog pathway in prostate tumorigenesis[16] and pituitary adenomas,[17] caveolin-1/SREBP pathways in pancreatic cancer,[18,19] AKT/mTOR axis activation in renal cell carcinoma tissues,[20] and prostate cancer cells.[21] While ACAT1 is a known driver of metabolic reprogramming in many malignancies like breast, renal, and ovarian, its role in sebaceous gland carcinoma (SGC) of the eyelid remains unexplored.

Although occasional studies have reported dysregulation of cholesterol pathway genes in eyelid SGC, the clinical relevance of cholesterol metabolism in eyelid SGC patients remains unexplored. The present study aims to evaluate the significance of acyl-CoA: cholesterol acyltransferase (ACAT) expression, a key enzyme involved in cholesterol esterification and storage, in tissue samples from patients with eyelid SGC. This study bridges a critical gap in the literature by identifying ACAT1 as a distinct metabolic vulnerability unique to lipid-producing ocular malignancies.

Methods

Patients and tissue samples

A retrospective observational study was conducted including 50 human eyelid SGCs diagnosed over a period of 2012–2024 at a tertiary care hospital in North India. Forty-eight patients underwent wide excision, and two were subjected to exenteration. Postoperative chemotherapy of (Carboplatin and 5-Fluorouracil) was given to seven patients and one patient was given both chemotherapy and radiotherapy (50–60 Gy). The patients were followed up every 6 months from 15 to 156 months (70.6 ± 49 months). The ethical approval was obtained from the Institute Ethics Committee ((Ref No IEC-345/8.05.2020, RP-59/2020), and the investigation was conducted according to the principles expressed in the Declaration of Helsinki. Written informed consent was obtained from the patients prior to tissue collection. Fresh tumor samples were collected from the main tumor mass and normal skin tissue of eyelid from exenteration samples of patients with SCC and melanoma (control) immediately after surgery and stored in RNA later (RNAprotect, Qiagen, Germany) at -20°C. The remaining tumor samples were fixed in formalin and embedded in paraffin. The clinicopathological features and radiological details of histopathologically confirmed cases of eyelid SGC were noted from the medical records. Staging was done according to the American Joint Committee on Cancer (AJCC, 8^th edition)[22,23] Histopathological slides were reviewed by two pathologists (SS and SK).

Immunohistochemistry

The avidin biotin peroxidase method was used for immunohistochemistry (IHC) with monoclonal antibodies against ACAT1 (ab168342; Abcam; rabbit; 1:150 dilution). Unstained slides were deparaffinized with xylene and rehydrated, followed by heat antigen retrieval at pH 6 (citrate buffer) at 100°C for 20 min. Blocking endogenous activity was done using hydrogen peroxide for 30 min, followed by overnight primary antibody incubation at 4°C. Subsequently, after the 30 min of biotinylated secondary antibody incubation, chromogenic substrate 3, 3’-diaminobenzidine (DAB) was applied (UltraVision Quanto Detection kit; Thermo Scientific, California, USA). Tissue sections were hematoxylin-counterstained and mounted. Appropriate positive controls (kidney) and negative controls (primary antibody incubation omitted) were carried out.

IHC slides were visualized under the light microscope and were scored semiquantitively for each marker (SR and SS). Intensity was scored as 0 (negative), 1 (weak positivity), 2 (moderate positivity), and 3 (strong positivity). The positive staining cells were scored as 1 (0–25%), 2 (26–50%), 3 (51–75%) and 4 (>75%). The percentage of positivity and intensity scores were multiplied to obtain a total immunoexpression for a particular marker (immunoreactivity score, IRS). The mean score of all samples served as the threshold to define high and low protein expression to minimize bias and balance group sizes.[24] ACAT1 immunoexpression score <9 was considered low expression, whereas expression score ≥9 was considered high expression.

RNA isolation and quantitative real-time PCR

Tissues (20–25 mg) were minced in mortar and pestle and grinded to fine powder using liquid nitrogen. The frozen tissue powder scooped into lysis buffer for 30 min and homogenized using a Dounce Homogenizer. Total RNA was isolated from the homogenized tissues as per the RNA extraction kit manufacturing protocol (RNeasy, Qiagen, Germany). RNA (1 µg) as a template was then reverse-transcribed to complementary DNA (cDNA) using a Verso cDNA synthesis kit (cat no# AB1453A, ThermoFisher Scientific, USA).

Quantitative real-time PCR was performed on 50 tumor tissues and 16 normal eyelid skin tissues of eyelid from exenteration samples of patients with SCC and melanoma (control) using SYBR green technology (cat no# A6001, GoTaq qPCR Mastermix, Promega, USA) on an Applied Biosystems (Thermo Scientific, Foster City, CA, USA) real-time PCR machine. Primers of ACAT1 and GAPDH (reference gene) were synthesized by IDT (Coralville, IA, USA) with sequences for ACAT1: 5’GCAGGCTTACCTATTTCTACTC3’ (forward primer) 5CAGTTAGCCCGTCTTTTACAATC3′ (reverse primer) and β-actin 5’TTGCCGACAGGATGCAGAA3’ (forward primer) 5’GCCGATCCACACGGAGTACT3’ (reverse primer). Data normalization was executed by using the 2-ΔΔCt method (Livak method). Ct values of tumor and normal samples were first normalized to the internal reference control (ΔCt). Further, the ΔCt of tumors was calibrated against the mean of the normal eyelid skin tissues to yield relative fold change values. Fold change expression >1.5 was taken as high expression.

Cholesterol/cholesteryl ester assay

Total cholesterol and free cholesterol were estimated using a Cholesterol/Cholesteryl Ester Assay Kit (Abcam ab65359). Fresh or frozen tissues of representative eyelid SGC and normal eyelid skin (10–20 mg) were homogenized in chloroform, isopropanol, and Triton X-100 solution (7:11:0.1 v/v/v). Homogenates were centrifuged at 13,200 rpm for 10 min. Supernatants were dried at 50°C, followed by vacuum drying for 30–40 min. Dried pellets were resuspended in 200 μl cholesterol buffer. Total cholesterol and free cholesterol were measured according to the manufacturer’s instructions.

Statistical analysis

Statistical analysis was performed using Graph Pad Prism 8 and MedCalc. The IHC and real-time PCR results obtained were correlated with clinicopathological features using Fisher’s Exact or χ2 test, and the statistical significance was defined as P < 0.05. Kaplan–Meier analysis was done to calculate disease-free survival of the patients, and the difference between the groups was analyzed by the log-rank test. Cox proportional hazards model (univariate and multivariate) analysis was performed to determine the prognostic significance.[25,26]

Results

Clinical and histopathological features

Fifty eyelid SGC patients were recruited for the study. The mean duration of the disease was 70.6 ± 49 months (range: 15–156 months), and the mean age of the patients was 60.84 ± 12.68 years (range: 39 to 84 years). Thirty-six patients (72%) had an age of >50 years. A slight male preponderance was noted (1:1.1). The tumor was located in the right eyelid in 27 (54%) and left in 23 (46%) cases. Two cases (4%) involved both upper and lower eyelids, 36 (72%) involved the upper lid, and 12 cases (24%) involved the lower eyelid. LTD (large tumor diameter) >20 mm was observed in 20 cases (40%). Lymph node metastasis was present in 8 (16%) and recurrence in 9 cases (18%). Death and distant metastasis were observed in 2 cases (4%) each. The TNM categories, that is, T₁/T₂, were observed in 29 cases (58%), whereas T₃/T₄ in 21 cases (52%). Higher AJCC stage (III/IV) was present in 9 patients (18%). On histopathological analysis, there was poor differentiation in 34 cases (68%) and 16 (32%) were well differentiated. Pagetoid spread was present in 27 cases (54%) [Table 1].

Table 1.

Clinicopathological features in eyelid sebaceous gland carcinoma

Clinical Features				Number (n=50)		Percentage (%)
Age		≤50 years		14		28
		> 50 years		36		72
Sex		Male		26		52
		Female		24		48
Laterality		Left Eye		23		46
		Right Eye		27		54
Location		Upper Lid		36		72
		Lower Lid		12		24
		Both Upper and Lower lids		2		4
LTD		≤20 mm		30		60
		>20 mm		20		40
Surgery		Frozen section guided excision Exenteration		48 2		96 4
Pagetoid Spread		Present		27		54
		Absent		23		46
AJCC* Staging		IA-II		41		82
		IIIA-IV		9		18
TNM		T1+T2		29		58
		T3+T4		21		42
Nodes		Positive		8		16
Metastasis		Distant		2		4
Tumor Differentiation		Poor Differentiation		34		68
		Well Differentiation		16		32
Death				2		4
Recurrence				9		18

*AJCC: American Joint Committee on Cancer (8th edition)

Cholesterol and cholesteryl esters in eyelid sebaceous glands carcinoma

The level of total cholesterol was elevated significantly in representative SGC tumor samples as compared to normal eyelid skin [Fig. 1]. A significant increase in CE levels in SGC suggests enhanced cholesterol esterification activity, potentially mediated by ACAT1.

Figure 1.

Quantification of cholesterol and cholesteryl esters in normal eyelid skin and sebaceous gland carcinoma (SGC) tissues. Graphs represent mean concentrations of Total Cholesterol, Free Cholesterol, and Cholesteryl Esters (*P < 0.05)

Immunohistochemical and mRNA expression of ACAT1 and correlation with clinical and histopathological features

High ACAT1 immunoexpression was noted in 66% (33/50) of eyelid SGC cases [Fig. 2]. Patients aged over 50 years (66.6%) and those with tumors classified as TNM stage T3/T4 (62%) demonstrated high ACAT1 immunoexpression, respectively. In addition, elevated ACAT1 expression was observed in approximately 50–55% of cases with aggressive clinical features, including LTD >20 mm, presence of lymph node metastasis, advanced tumor stage (III/IV), and recurrence. Furthermore, all cases with systemic metastasis and death showed high ACAT expression. Histologically, 66.6% cases with pagetoid spread exhibited ACAT1 overexpression [Supplementary Table S1].

Figure 2.

(a) Light microscopy to show a case of SGC (H and E x200); (b) Cytoplasmic overexpression of ACAT1 in a case of poorly differentiated SGC (Scale = 100 μm, Avidin Biotin × 200)

Table S1.

Clinicopathological features and association with Acyl-CoA cholesterol acyltransferase 1 (ACAT1) expression in eyelid sebaceous gland carcinoma

Clinical Features (n, %)		ACAT1 immunoexpression						ACAT1 mRNA expression
		High (33, 66%)		Low (17, 34%)		P value		High (32, 64%)		Low (18, 36%)		P value
Age
≤50 years (14, 28%)		9 (64.2%)		5 (35.7%)		>0.99		8 (57.1%)		6 (42.8%)		0.53
> 50 years (36, 72%)		24 (66.6%)		12 (33.3%)				24 (66.6%)		12 (33.3%)
Sex
Male (26. 52%)		17(65.3%)		9 (34.6%)		>0.99		17(65.3%)		9 (34.6%)		>0.99
Female (24, 48%)		16 (66.6%)		8 (33.3)				15 (62.5%)		9 (37.5%)
Laterality
Left Eye (23, 46%)		17 (73.9%)		6 (26%)		0.37		14 (60.8%)		9 (39.1%)		0.77
Right Eye (27, 54%)		16 (59.2%)		11 (40.7%)				18 (66.6%)		9 (33.3%)
Location
Upper Lid (36, 72%)		23 (63.8%)		13 (36.1%)		0.57		23 (63.8%)		13 (36.1%)		0.52
Lower Lid (12, 24%)		8 (66.3%)		4 (33.3%)				7 (58.3%)		5 (41.6%)
Both lids (2, 4%)		2 (100%)		0				2 (100%)		0
LTD
≤20 mm (30, 60%)		22 (73.3%)		8 (26.6%)		0.23		18 (60%)		12 (40%)		0.55
>20 mm (20, 40%)		11 (55%)		9 (45%)				14 (70%)		6 (30%)
Pagetoid Spread (27, 54%)		18 (66.6%)		9 (33.3%))		>0.99		18 (66.6%)		9 (33.3%)		>0.99
AJCC Staging
IA-II (41, 82%)		28 (68.2%)		13 (31.7%)		0.46		23 (56%)		18 (44%)		0.02*
IIIA-IV (9, 18%)		5 (55.5%)		4 (44.4%)				9 (100%)		0
TNM
T1+T2 (29, 58%)		20 (68.9%)		9 (31.1%)		0.76		16 (55.1%)		13 (44.9%)		0.14
T3+T4 (21, 42%)		13 (62%)		8 (38%)				16 (76.2%)		5 (23.8%)
Positive Nodes (8, 16%)		4 (50%)		4 (50%)		0.41		8 (100%)		0		0.03*
Metastasis (2, 4%)		2 (100%)		0		0.54		2 (100%)		0		0.52
Tumor Differentiation
Poor (34, 68%)		21 (61.7%)		13 (38.2%)		0.52		24 (70.5%)		10 (29.4%)		0.21
Well (16, 32%)		12 (75%)		4 (25%)				8 (50%)		8 (50%)
Death (2, 4%)		2 (100%)		0		0.54		1 (50%)		1 (50%)		>0.99
Recurrence (9, 18%)		5 (55.5%)		4 (44.4%)		0.46		8 (88.8%)		1 (11.1%)		0.13

*Significant association, LTD: Large Tumor Diameter, AJCC: American Joint Committee on Cancer (8th edition)

High mRNA expression of ACAT1 was observed in SGC compared to normal skin of eyelid [Fig. 3]. More than 66% of the patients with high ACAT 1 had age >50 years, LTD >20 mm (70%), and TNM stage T3/T4 (76.2%). All the patients with systemic metastasis, recurrence (88.8%), and death (50%) demonstrated high mRNA expression. A statistically significant association of high ACAT1 mRNA expression was observed with lymph node metastasis (P = 0.03) and advanced III/IV stage tumors (P = 0.02). Increased ACAT1 mRNA was also noted in tumors with pagetoid spread (66.6%) and poor differentiation (70.5%) [Supplementary Table S1]. No significant correlation was observed between ACAT1 mRNA and immunoexpression, suggesting that post-transcriptional regulation could influence ACAT1 protein levels independently of mRNA levels.

Figure 3.

ACAT1 mRNA overexpression in eyelid sebaceous gland carcinoma cases as compared to normal eyelid tissues

Disease-free survival and Cox regression analysis

On Kaplan–Meier survival curve analysis, a higher number of events in the high-expression group was observed compared to the low-expression group. ACAT1 immunoexpression (IHC) demonstrated statistically significant reduced disease-free survival in the high-expression group, suggesting its role as a prognostic marker [Fig. 4a] (P = 0.013). The consistent pattern of high expression of ACAT1 across both IHC and real-time PCR [Fig. 4b] demonstrates that elevated ACAT1 expression is associated with adverse clinical outcomes. Univariate and multivariate Cox regression analysis demonstrated distant metastasis and ACAT immunoexpression (IHC) as significant poor prognostic indicators [Table 2].

Figure 4.

Kaplan–Meier disease-free survival analysis in eyelid sebaceous gland carcinoma patients: (a) ACAT1 protein overexpression (IHC) shows worse disease-free survival (P = 0.013); (b) ACAT1 mRNA expression did not show statistically significant reduced disease-free survival (P = 0.7)

Table 2.

Univariate and multivariate Cox regression analysis for disease-free survival in eyelid sebaceous gland carcinoma patients

Variables		Univariate				Multivariate
		HR (95% CI)		P		HR (95% CI)		P
Age		1.51 (0.4617 to 4.9830)		0.49		-
Sex		2.2 (0.7869 to 6.2077)		0.1323		-
LTD (>20 mm)		0.58 (0.1427 to 2.3818)		0.4524		-
Stage		1.08 (0.3825 to 3.0633)		0.8813		-
TNM		0.88 (0.2153 to 3.6465)		0.86		-
Lymph Node metastasis		1.24 (0.4358 to 3.5305)		0.68		-
Distant Metastasis		13.4 (1.8438 to 97.7591)		0.01*		15.89 (1.865 to 135.37)		0.011*
Recurrence		1.0 (0.3287 to 3.0491)		0.9985		-
Death		4.9 (0.9542 to 25.9531)		0.056		5.88 (0.9631 to 35.9266)		0.055
Pagetoid Spread		0.68 (0.2403 to 1.9293)		0.49		-
Poor differentiation		0.98 (0.3335 to 2.9160)		0.97		-
Acyl-CoA cholesterol acyltransferase 1 (ACAT1) expression (IHC)		6.42 (1.2720 to 32.4768)		0.02*		5.34 (1.0126 to 29.0398)		0.05*
Acyl-CoA cholesterol acyltransferase 1 (ACAT1) mRNA expression		1.53 (0.1837 to 12.8001)		0.69		-

*Significant association, HR: Hazard Ratio, CI: Confidence Interval, LTD: Large Tumor Diameter

Discussion

In normal tissues, ACAT coverts the toxic-free cholesterol into less harmful CEs, and in malignancies, cancer cells hijack this homeostatic mechanism to fuel rapid proliferation and evade the immune system.[27] The resulting elevated ACAT-1 expression may serve a dual purpose, safeguarding tumor cells against free cholesterol-induced lipotoxicity and establishing a dense CE storage reservoir crucial for rapid membrane biogenesis. Various studies have shown that high cholesterol level or increased de novo cholesterol biosynthesis promotes cancer cell survival, proliferation, and metastasis.[28,29] Eyelid SGC is a rare and aggressive malignancy, and only a few bioinformatics studies have demonstrated dysregulation of cholesterol metabolism genes. We observed high levels of total cholesterol, free cholesterol, and cholesteryl esters in eyelid SGC as compared to normal eyelid skin tissues.

Increased ACAT and CE have been reported in various malignancies including breast cancer, glioblastoma, pancreatic cancer, renal cancer, leukemia, and prostate cancer.[30,31,32,33,34] Similarly, in the present study, significantly high levels of cholesteryl esters as well as protein and mRNA overexpression of ACAT1 in eyelid SGC patients were observed. Investigating ACAT1 expression in SGC offers a novel paradigm: determining how a tumor reprograms its native and specialized cholesterol homeostasis pathways. The pronounced lipid phenotype of eyelid SGC aligns with metabolic reprogramming strategies seen in other lipid-rich malignancies like clear cell renal cell carcinoma, epithelial ovarian cancer, and triple-negative breast cancer. In these tumors, ACAT-1 upregulation serves as a vital metabolic rheostat rather than a passive byproduct of fat accumulation. It actively channels lipotoxic-free cholesterol into inert cholesteryl esters for storage. By identifying a similar mechanism in eyelid SGC, our findings underscore a conserved oncogenic mechanism in which lipogenic tumors exploit cholesterol esterification to sustain rapid tumor prolifereation while evading lipid-induced cell death.[14,27,35,36] In the few previous published studies on eyelid SGC, whole-exome and transcriptome sequencing data showed several cholesterol metabolic genes to be mutated and differentially expressed.[37] In another bioinformatics study, differential expression of miRNAs and mRNAs associated with the enrichment in cholesterol biosynthesis pathways in SGC has been reported.[38] Another key regulator of cholesterol and lipid synthesis, Sterol regulatory element binding protein (SREBP-1) overexpression, has also been reported in eyelid SGC.[39]

SGC patients with high-risk clinicopathological features demonstrated overexpression of ACAT1 mRNA and protein. A significant correlation with lymph node metastasis and advanced stage suggests ACAT1 could play a critical role in promoting eyelid SGC progression, metastasis, and recurrence. In prostate cancer, high cholesterol level and increased mevalonate pathway have been reported to be involved and it correlated with bone metastases.[40]

Aberrant cholesterol metabolism has been shown to be associated with poor survival outcome, cell migration, and metastasis.[34] In the present study, a significantly reduced disease-free survival and poor prognosis was observed in patients with high ACAT1 immunoexpression. Similarly, in human pancreatic and breast cancer, aberrant accumulation of cholesteryl ester and high expression of ACAT-1 showed a correlation with poor patient survival.[31,41]

Although elevated ACAT1 transcription is associated with aggressive forms of SGC and several other cancers, contrasting evidence has shown that ACAT1 expression is downregulated in advanced stages of gastric cancer, nasopharyngeal carcinoma, glioblastoma, and renal cell carcinoma.[42,43,44,45]

Conclusion

In conclusion, our study demonstrates that ACAT1 is significantly overexpressed in eyelid SGC and correlates with aggressive clinicopathological features. Mechanistically, ACAT1 serves as a metabolic shield against lipotoxicity while providing energy reservoirs via cholesterol ester storage. Clinically, high expression of ACAT1 in SGC suggests that it could be a poor prognostic biomarker and potential novel metabolic target for therapeutic intervention. Larger multicentric studies are necessary to validate its therapeutic potential in eyelid SGC. The limitations of the study are a relatively small sample size and a low number of cases with adverse outcome.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Funding support for the study was provided by Indian Council of Medical Research (grant 5/4/6/15/OPH/2020-NCD-II).