Carbohydrate antigen 125, carbohydrate antigen 15–3 and low-density lipoprotein as risk factors for intraocular metastases in postmenopausal breast cancer

Abstract The prognosis of patients with postmenopausal breast cancer (PBC) could be improved by the early detection of intraocular metastases (IOMs). However, serum biomarkers for IOMs in PBC remain elusive. In the current study, we investigated patients with PBC, and compared serum parameters in an IOM and a non-IOM group, and then differentiated the risk factors related to IOMs. A comparison between an IOM and a non-IOM (NIOM) group was performed using Student t-test and a Chi-Squared test. After constructing a Poisson regression model to identify risk factors, we plotted receiver operating characteristic curves to evaluate the predictive value of significant risk factors in detecting IOMs. The incidence of IOMs in PBC was 1.16%. The histopathology results were not significantly different between the 2 groups. The levels of serum carbohydrate antigen 125 (CA-125), carbohydrate antigen 15–3 (CA15–3) and alkaline phosphatase were significantly elevated in IOMs compared with NIOMs (P = .082, P < .001, and P < .001, respectively). Compared with NIOMs, age, carbohydrate antigen 19 to 9, hemoglobin, calcium, total cholesterol, low-density lipoprotein (LDL) and apolipoprotein A1 were remarkably lower in IOMs (P = .038, P < .001, P < .001, P = .032, P = .041, P < .001, and P = .001, respectively). Poisson regression suggested that CA-125, CA15–3 and LDL were contributing to IOMs in PBC as risk factors (OR = 1.003, 95% CI: 1.001–1.005; OR = 1.025, 95% CI: 1.019–1.033; OR = 0.238, 95% CI: 0.112–0.505, respectively). A receiver operating characteristic curve revealed that the cut-off values for CA-125, CA15–3 and LDL were 16.78 0 U/mL, 63.175 U/mL, and 2.415 mmol/L, respectively. The combination of CA-125 and CA15–3 showed significant diagnostic value (area under the curve [AUC] = 0.982, P < .001). Our investigation suggests that CA-125, CA15–3 and LDL remarkably predict IOMs in PBC as risk factors, and the combination of CA-125 and CA15–3 shows considerable diagnostic value.


Introduction
Breast cancer (BC) is the most common cancer in females worldwide, and BC is the main source of cancer-related death in women. [1] Metastases play an important role in cancer-related death, and improvements in therapies for metastatic cancer have been slow to emerge. [2] Currently, to facilitate postmenopausal BC (PBC) prognosis, identifying early metastases is necessary.
Intraocular metastases (IOMs) are regarded as a leading cause of intraocular malignancy. [3] The choroid is the most common site of IOMs. [4] It was reported in a cohort study involving 1111 participants, that BC demonstrated the highest proportion of IOMs, at 37%, and the average age at which patients experienced IOMs was 57. [5] In other words, despite the low incidence of IOMs in BC, BC is the main primary lesion for IOMs.
Currently, imaging techniques are indispensable in detecting ocular tumors, and these include computed tomography (CT), high-frequency ultrasound, fluorescein angiography optical coherence tomography, and magnetic resonance imaging (MRI). [6] However, the cost and radiation dose associated with such techniques are too high for long-term follow-ups. Moreover, the incidence of IOMs is somewhat low. Consequently, although these imaging tests have considerable diagnostic potential in this context, they are not suitable as a global-scale tool to regularly screen PBC patients during long-term follow-up. However, serological monitoring is non-invasive, reproducible, and costeffective, which make it a better option for detecting IOMs. [7] Recently, it has been confirmed that serum monitoring has considerable potential in BC diagnosis and screening; for instance, monitoring of exosomal miRNA [8] and serum 25hydroxyvitamin D [9] has been shown to be useful. To promote the ability of early detecting IOMs in PBC, the identification of serum risk factors is necessary and is becoming more feasible.
With the goal of improving PBC patient prognosis, a retrospective study was conducted by our research group to identify serum risk factors for IOMs in PBC by evaluating the link between clinical serum pathological parameters and IOMs.

Ethics statement
The Medical Research Ethics Committee of the First Affiliated Hospital of Nanchang University approved this study (CDYFY-LL2006023). Because of the retrospective nature of the study, informed consent of participants was waived. The methodology used in this study is in accordance with approved guidelines and related regulations.

Study design
Participants with BC involved in this retrospective observational study from April 1995 to July 2017 were diagnosed via histopathological sections obtained through surgical resection or biopsy. The IOMs were confirmed through local CT and MRI.
The key inclusion criteria were female BC patients with postmenopausal status, and BC was the primary lesion. The key exclusion criteria were patients with primary ocular malignancies or benign tumors without pathology reports.

Data collection
The patient medical records were the source of the study data. All data, such as demographic characteristics and clinical pathology parameters, were collected before patients received anti-tumor therapy (radiotherapy, chemotherapy, or surgery). These mainly included: the age of the tumor being diagnosed, histopathology and the condition of metastases, and laboratory results such as alkaline phosphate, hemoglobin, triglycerides, low-density lipoprotein (LDL), apolipoprotein A, and serum calcium levels. The concentrations of common tumor markers in serum, such as carcinoembryonic antigen, cancer antigen 125 (CA-125), cancer antigen 153 (CA15-3) and cancer antigen 19-9, were also recorded. As the link between different serological pathological parameters and IOMs was analyzed, we estimated the predictive value of significant risk factors in detecting IOMs.

Statistical analyzes
SPSS software (version 19.00; SPSS, Chicago, IL) and GraphPad Prism (8.0.1.244) were used for statistical analysis, and Excel 2010 (Excel, Microsoft Corp, USA) was used for data collation. Initially, a comparison between an IOM and a non-IOM (NIOM) group was performed using Student t test and the Chi-Squared test to find clinical pathology parameters with significant differences. Quantitative variables are reported as mean ± standard deviation. The occurrence of IOMs in PBC is a rare event (10/865; ≈0.0116). Consequently, a multivariate Poisson regression analysis model for rare events was constructed to identify statistically independent risk factors and determine the odds ratio (OR) and 95% confidence interval (95% CI). Finally, receiver operating characteristic (ROC) curves were plotted to evaluate the predictive value of significant risk factors in detecting IOMs. The area under the curve (AUC) of the different factors, and the sensitivity and specificity of the optimal cut-off point were calculated to estimate the capability of the risk factors in predicting IOMs. P values ofless than 0.05 were regarded as statistically significant, and all reported P values were bilateral. When adding variables into the Poisson regression analysis model, we used P < .20 as a standard to ensure that low-impact factors were not being missed. The test level (a) was set to 0.05.

Demographics and clinical characteristics
A total of 2373 PBC patients participated in this investigation, and 865 patients were finally involved based on the inclusion and exclusion criteria. Table 1 lists baseline data such as the demographic and clinical characteristics of the IOM group (10 patients) and the NIOM group (855 patients). As for age, we observed that patients in the IOMs were much older than that in the NIOMs (P = .038). The histopathology results were not significantly different between the 2 groups. Besides, most IOMs were 2 to 5 cm and TNM 2-3 Stage. Detailed clinical data for all elderly patients participating in the study are listed in Table 1.

Differences in the clinical features and the risk factors associated with intraocular metastases
The results showed that triglyceride, high density lipoprotein, apolipoprotein B, lipoprotein a and carcino-embryonic antigen were not significantly different between the IOM and NIOM group (P > .05). However, serum CA-125, CA15-3 and alkaline phosphatase levels were significantly elevated in the IOM group compared with the NIOM group (P = .082, P < .001, and P < .001, respectively). Compared with the NIOM group, age, CA-19-9, hemoglobin, calcium, total cholesterol, LDL and apolipoprotein A1 were lower in the IOM group (P = .038, P < .001, P < .001, P = .032, P = .041, P < .001 and P = .001, respectively) (  Table 4 and Fig. 1).

Discussion
The incidence of IOMs in BC is low, at approximately 0.07% to 12%. [10] In the current study, the incidence of IOMs in PBC was 1.16%, within the range of 0.07% to 12%. We also found that patients in the IOMs were much older than that in the NIOMs. However, the result of poisson regression showed that p value of      The most frequent symptoms of IOMs are blurred vision and sight loss, which are non-specific symptoms, and the prognosis for IOMs is poor. [11] Moreover, according to a case report, after 34 years of BC remission, IOMs occurred. [12] Therefore, there is a necessity for long-term screening for IOMs. However, adherence to follow-up visits is low in PBC. [13] Consequently, the management of IOMs in PBC is confronted with an enormous challenge, because of non-specific IOM symptoms, unfavorable adherence, excessive follow-up times and the low incidence of IOMs. Moreover, imaging tests (Fig. 2) are likely not suitable to regularly screen PBC patients during long-term follow-up. However, recently, it has been revealed that serum biomarkers are related to prognosis in many cancers, suggesting that serological monitoring will play an increasingly important role in the detection of cancer during follow-up (Table 5). Therefore, to permit the early detection of IOMs in PBC, the identification of serum risk factors is necessary and is becoming more feasible.
CA-125, known as Mucin 16, is the largest mucin, and its overexpression occurs in numerous cancers. [14] It was revealed by a multicenter study that CA-125 has considerable diagnostic value in epithelial ovarian cancers (EOCs). [15] CA-125 exerts an important role in EOC cell proliferation and metastases. [16] Additionally, CA-125 has been linked to lung cance, pancreatic cancer, gastric cancer, colorectal cancer, endometrial cancer, and uterine papillary serous carcinoma. [17][18][19][20][21][22] It was also reported that CA-125 shows diagnostic value in identifying BC recurrence. [23] Moreover, it was confirmed that CA-125 has considerable value in diagnosing metastatic BC. [24] A study involving 2133 BC patients revealed that CA-125 was related to bone metastases in BC. [25] In fact, the P value of CA-125, (IOM vs NIOM) was 0.082 (Table 2), which was more than 0.05 and www.md-journal.com not statistically significant in the Student t test. In the Poisson regression analysis model, we used P < .20 as a standard to ensure that low-impact factors were not being missed and the OR of CA-125 turned out to be significant (P = .016, Table 3). Therefore, we hypothesized that CA-125 is a risk factor for IOMs in PBC. The current study revealed that the occurrence of IOMs is more likely in PBC patients with CA-125 >16.78 0 U/mL. CA15-3, also known as Mucin 1, is a membrane tethered glycoprotein, which contributes to cancer progression. [26] It was reported in a meta-analysis that CA15-3 could be a cancer biomarker. [27] Moreover, numerous studies have reported that CA15-3 is relevant in lung cancer, cervical cancer and colorectal cancer, and it is also useful in the diagnosis of malignant pleural effusion. [28][29][30][31] It was demonstrated by a study involving 1681 participants that CA15-3 in high levels plays a crucial role in the BC tumor burden. [32] Serum levels of CA15-3 are recommended for use in detecting metastatic BC by the American Society of Clinical Oncology Clinical Practice Guidelines. [33] It was reported by Wu SG et al [34] that CA15-3 is a risk factor for axillary lymph node metastasis in BC. Similarly, this link was also observed for liver metastasis in BC. [35] In a recent case report it was observed that CA15-3 increased in a 57-year-old BC patient with IOMs that arose 28 years after the initial BC diagnosis. [36] Consequently, we identified CA15-3 as a risk factor for IOMs in PBC. Patients with CA15-3 >63.175 U/mL are more likely to suffer IOMs.
LDL, a lipoprotein granule abundant in cholesterol, transports cholesterol to peripheral tissues from the liver. [37] LDL has been correlated with ocular diseases such as meibomian gland dysfunction and age-related cataracts. [38,39] Recently, it has been demonstrated that metabolic dysregulation contributes to an elevation in cancer mortality. [40] It was reported by Guang et al that LDL has an adverse link with the total risk of cancer. [41] Serum LDL has been linked to colorectal cancer, prostate cancer, esophageal squamous cell carcinoma and chronic lymphocytic leukemia. [42][43][44][45] LDL has been reported as a risk factor in the context of cancer progression and metastases. [46] Moreover, it was reported by Kumar et al. that high levels of LDL are linked to BC. [47] Increased BC proliferation and metastasis was observed following LDL stimulation, as the levels of tumor LDL receptors increased. [48] However, it was demonstrated that BC patients with metastases had a significantly lower level of LDL compared with a control group. [49] This may be because of the large consumption of LDL during rapid proliferation and metastases in BC. Therefore, we hypothesized that LDL at a low level was a risk factor for IOMs in PBC. The current study demonstrated that patients with LDL <2.415 mmol/L are at greater risk for IOMs. However, the ROC results indicated that the accuracy of LDL in predicting IOMs is not sufficient to detect IOMs alone.
Although the results are promising, the current study still has some limitations. Firstly, this is a retrospective observational study. As the study period is substantial, some correlative data are not adequate, such as patient survival time. Moreover, there may be differences in the methodological and technical aspects within the assay of tumor markers and biochemical parameters, which may exert a negative effect on the sensitivity and specificity of test parameters. Besides, the sample size of IOMs is limited and it may not be sufficiently large to enable the extrapolation of the results to the entire BC population. Thirdly, some confounding bias may be present, because all of the data in the current study derived from a single medical institution. Therefore, the results of the current study need to be validated through a multicenter study.

Conclusion
To summarize, this investigation revealed that CA-125, CA15-3, and LDL have considerable predictive value for IOMs in PBC as risk factors. These markers will facilitate the early detection of IOMs in PBC during long-term follow-up. Moreover, these results could inspire novel insights into the molecular mechanisms of CA-125, CA15-3, and LDL for antineoplastic applications.  The serum tumor biomarkers as risk factors of different types of cancer.