You are here
Home > Blog > Oncology > Risk Factors for Ocular Metastases in Invasive Ductal Carcinoma

Risk Factors for Ocular Metastases in Invasive Ductal Carcinoma

Risk Factors for Ocular Metastases in Invasive Ductal Carcinoma

Invasive ductal carcinoma (IDC), also known as infiltrating ductal carcinoma, is the most common type of breast cancer, accounting for about 72% of all cases. IDC can occur at any age, but many people are over the age of 55 at the time of diagnosis. IDC starts from the mammary ducts, and grows through the duct wall into the breast stoma. Because of its invasiveness, over time, IDC spreads along the lymphatic and circulatory channels.

Breast Cancer & Ocular Metastasis

In women, ocular metastasis (OM), also termed uveal metastasis is usually from breast cancer. Several studies have concluded that a vast majority of ocular metastatic sites arise from hematogenous spread. The uveal tract (iris, ciliary body, and choroid) is the primary ocular site of breast cancer metastasis. The choroid has the highest risk for ocular metastasis (81%), followed by the iris (9%), optic disc (5%), and ciliary body (2%).

The extensive anastomoses between the choroidal vessels have been proposed as a possible reason for increased metastasis in the choroid. The most common clinical symptoms include limited ocular mobility, proptosis, diplopia, blurred or decreased vision, or ocular pain. In the majority of the cases, clinicians fail to notice signs of ocular metastasis at an early stage. Ophthalmectomy is often recommended for patients at an advanced stage.

Slit-lamp biomicroscopy (SLB) is the gold standard for identifying metastatic sites, alongside the history of breast cancer. Additional imaging methods such as fluorescein angiography, fundus autofluorescence (FAF), and optical coherence tomography (OCT) are also used in the diagnostic process. Although magnetic resonance imaging (MRI), and positron emission tomography-computed tomography scan (PET-CT) are used to detect distant metastases, they have the disadvantages of high expenditure and radiation doses.

Recently, serum indices and clinical features of patients have been widely analyzed to detect risk factors and to predict distant metastases. Therefore, this study by Liang et al. aimed to find out risk factors and their diagnostic values for ocular metastases in IDC patients.

Materials and Methods

Patient Population & Data Collection

This retrospective study was conducted in The First Affiliated Hospital of Nanchang University. A cohort of 1192 patients who had IDC from 2008 to 2017 were analyzed retrospectively. An informed consent form was obtained from all included participants.

  • Inclusion criteria were: (1) Primary breast cancer patients diagnosed at The First Affiliated Hospital of Nanchang University; (2) OM (Ocular metastasis) patients with intraocular space‐occupying lesions; (3) The pathological diagnosis of IDC confirmed by the Pathology Department of The First Affiliated Hospital of Nanchang University.
  • The exclusion criteria: Patients with metastatic breast cancer or primary ocular malignant tumor were not included in this study.

Age, menopausal status, and axillary lymph node metastases (ALNM) of the cohort were compiled from medical records. Levels of triglycerides (TG), total cholesterol (TC), high‐density lipoprotein (HDL), low‐density lipoprotein (LDL), apolipoprotein A1 (ApoA1), apolipoprotein B (ApoB), lipoprotein A (LipA), and alkaline phosphatase (ALP) were also documented. Furthermore, tumor markers such as carcinoembryonic antigen (CEA), carbohydrate antigen 125 (CA125), carbohydrate antigen 153 (CA153), and carbohydrate antigen 199 (CA199) were also compiled for data analysis.

The researchers divided the cohort into the Ocular Metastases (OM) and the non-Ocular Metastases (NOM) groups. A biopsy was done to diagnose OM and IDC. The clinical features of both the study groups were then compared. Binary logistic regression was employed to determine the risk factors of OM in IDC patients.

All procedures performed in this study were in accordance with the ethical standards of The First Affiliated Hospital of Nanchang University and with the Helsinki declaration.

Statistical Analysis

To figure out the risk factors, the complied clinical parameters were further evaluated statistically.

  • Chi‐squared test: to detect differences in age, menopausal status, and ALNM levels.
  • Normality tests: to determine the normal distribution of the collected samples.
  • Rank‐sum test: failure to observe normal distribution led the researchers to perform a rank-sum test between OM and NOM groups.
  • Binary logistic regression: for risk factor determination of OM in IDC patients.
  • ROC curves: to determine the best diagnostic modality of OM in IDC patients, ROC curves of both single risk factors and combined risk factors were established.


  • Of the 1192 IDC patients selected for this study, 19 were with OM and 1173 without OM.
  • The mean age of the OM group was 44.79 ± 7.91 years, whereas that of the NOM group was 48.06 ± 10.32 years.
  • 15 patients in the OM group were premenopausal and 4 were postmenopausal. On the other hand, in the NOM group, 732 patients were premenopausal and 441 were postmenopausal (p = 0.139).
  • Of 728 patients with axillary lymph node metastases (ALNM), 16 were from the OM group and 712 were from the NOM group. Of the 464 patients without ALNM, 3 belonged to the OM group and 461 to the NOM group.
  • Though there was no significant difference in age or menopausal status, the incidence of OM rate was directly proportional to lymph node metastasis. Also, OM incidence showed an increasing trend.
  • When the binary logistic regression results were analyzed, it was observed that IDC patients with high levels of CA153 (>43.3 u/mL), low levels of ApoA1 (<1.11 g/L), and low Hb (<112g/L) were more susceptible to developing OM.


The identification of risk factors is crucial in designing timely treatment and surveillance plans after acute treatment. Factors that may increase the risk of ocular metastases in invasive ductal carcinoma include:


CA15-3 is a high molecular weight mucinous glycoprotein, a product of the MUC-1gene. MUC-1 gene is present in almost all epithelial cells, and its overexpression is often associated with colon, breast, ovarian, lung, and pancreatic cancers. Previous studies showed that elevated levels of CA15-3 are usually detected in patients with metastatic breast carcinoma. However, CA15-3’s role as a tumor marker in early-stage or breast cancer subtypes is still controversial. Population studies have shown that in individual patients, CA15-3 measurement may help to determine whether the symptoms are due to a benign or malignant tumor, as elevated levels are frequently associated with metastatic disease. Uehara et al. and Atoum et al. reported significant differences in CA15-3 levels in different stages of breast cancer. Some studies also found that CA15-3 may be useful as an indicator to determine treatment efficacy. In this study by Rong‐Bin Liang et al., CA15-3 showed a sensitivity of about 94.74% and enabled OM diagnosis in IDC patients. Furthermore, the cutoff value of 43.3 u/mL in IDC patients indicated that they are more prone to develop OM.


Apolipoprotein A-I (ApoA1), the major structural component of high-density lipoprotein (HDL), plays a vital role in lipid metabolism and transport. ApoA1 also displays anti-inflammatory and antioxidant properties, meaning it may be an antitumor agent. Evidence from numerous studies has shown that ApoA1 can inhibit tumor cell proliferation in vitro. A previous study documented ApoA1’s inhibitory action on activated neutrophils, thereby reducing tumor cell proliferation.

Many recent reports have linked ApoA1 with malignant tumors. For instance, elevated ApoA1 levels were seen in small cell lung carcinoma (SCLC), hepatocellular carcinoma (HCC), and bladder cancer. However, in some kinds of cancers, the expression of ApoA1 was reduced. Therefore, reduced serum ApoA1 levels are used as an independent predictor for the prognosis of ovarian cancer, non-small cell lung carcinoma (NSCLC), lymphoma, prostate cancer, and acute lymphoblastic leukemia.

Farias‐Eisner et al. reported that a biomarker panel of ApoA1, prealbumin (TTR), and transferrin (TF) can detect endometrial cancer with a sensitivity of 71%. According to the Tuft et al., higher APoA1 mRNA levels can be used to differentiate ovarian from breast carcinoma. They found a direct relation between APoA1 mRNA levels and survival rate in patients with ovarian serous adenocarcinoma.

The role of ApoA1 levels in breast cancer is still a matter of debate. Many studies showed an inverse association between APOA1 expression and breast cancer development but in a few studies, elevated APOA1 expression was found to promote breast carcinomas.

The present study concluded that IDC patients with ApoA1 levels lower than 1.11 g/L were prone to developing OM. It can be used to diagnose OM in IDC patients, with the sensitivity and specificity of 84.21% and 94.88%, respectively.


Hemoglobin (Hb), the oxygen-binding protein in erythrocytes, acts as the two-way respiratory carrier in the circulation. Anemia is a common yet potentially detrimental complication in cancer patients, especially in patients at an advanced stage or those receiving treatment. The causes are multifactorial, including tumor-related hemorrhage, chemotherapy, marrow infiltration, surgery, nutritional deficiencies, and cytokine-mediated anemia.

Low Hb levels decrease the partial pressure of oxygen in tumor cells which translates into hypoxia. Hypoxia can negatively impact therapeutic outcomes by reducing the effectiveness of radiotherapy and some oxygen-dependent cytotoxic agents. Additionally, hypoxia can change the genetic makeup of tumor cells, which will favor tumor progression and metastasis.

Literature studies suggest that there is more than one reason to pay attention to Hb levels in cancer patients. Nagaraju et al found that tumor hypoxia can boost the expression of hypoxia‐inducible factor‐1α (HIF‐1α). HIF‐1α, a transcription factor, can mediate angiogenesis of tumor cells and accelerate tumor infiltration. This could ultimately lead to ocular metastasis. Furthermore, it is reported that by altering PI3 K/AKT/mTOR, MAPK, and NF‐ĸB pathways, hypoxia can induce tumor metastasis.

Currently, hemoglobin level is used as a prognostic and predictive factor in hematological malignancies (acute myeloid leukemia, non-Hodgkin’s lymphoma, and Hodgkin’s disease) and solid tumors (ovarian, breast, cervical, head, and neck cancers as well as renal carcinoma).

This study concluded Hb as an independent risk factor of OM in IDC patients; IDC patients with Hb levels < 112 g/L were more likely to develop OM. Therefore, Hb could be considered a positive predictive factor for OM in patients with IDC, with the sensitivity and specificity of 78.95% and 77.92%, respectively.

Combined diagnostic values

For the best predictive accuracy, diagnostic values of combined risk factors were analyzed. The results are as follows:

  • Among the two combined risk factor combinations, CA15-3 and ApoA1 duo showed better accuracy, with a sensitivity of 89.47% and specificity of 97.78%.
  • Among the three combined risk factors, CA15-3, ApoA1, and Hb trio gave the best accuracy. The sensitivity and specificity recorded were 89.47% and 99.32% respectively.

Axillary lymph node metastases

In most cancer types, lymph node status is not only a prognosis marker but also an important factor in deciding on surgery as well as postoperative therapy. Collective results of many clinical studies have indicated that the number of lymph node metastases indirectly reflects tumor evasiveness. One study reported that the hazard ratios for breast cancer patients with 1-3 and 4 or more lymph node metastases were 1.2 and 2.5, respectively. This indicated that the number of lymph nodes is a key factor in the prognosis after relapse.

Bones are the most frequent sites where metastatic breast cancer cells tend to invade. Chen et al reported that the incidence of breast cancer bone metastasis is higher in patients with four or more axillary lymph node metastases, elevated serum levels of CA12-5, CA15-3, and ALP and low Hb levels.


The incidence for ocular metastasis is higher among breast cancer patients. Therefore, identifying more risk factors and their molecular and genetic basis will enable clinicians to develop more reliable, non-invasive and less expensive diagnostic and treatment modalities. 


There are some limitations to this study. First, it is a retrospective review that only enrolled patients from First Affiliated Hospital of Nanchang University; thus, the generalizability of the findings is limited. Second, some IDC patients of this study might have other distant metastases that elevate CA15-3 levels, which might affect accuracy of the findings. Third, the sample size is relatively small. Fourth, though the statistical analysis was successful in demonstrating a connection between risk factors and OM in patients with IDC, the specific mechanisms involved need further study.

Similar Articles

Leave a Reply