ETM Experimental and Therapeutic Medicine 1792-0981 1792-1015 D.A. Spandidos ETM-28-5-12710 10.3892/etm.2024.12710 Articles Cancer antigen 153: A risk factor for ocular metastases in patients with breast cancer GeQian-Min * HuQiao-Hao * LiuJia-Xiang * MinYou-Lan LiangRong-Bin LiBiao ShiWen-Qing LinQi YuanQing LiQiu-Yu ShaoYi Department of Ophthalmology, The First Affiliated Hospital, Jiangxi Medical College, Nanchang University, Nanchang, Jiangxi 330006, P.R. China Correspondence to: Dr Yi Shao, Department of Ophthalmology, The First Affiliated Hospital, Jiangxi Medical College, Nanchang University, 17 Yongwaizheng Street, Donghu, Nanchang, Jiangxi 330006, P.R. China drsadia@uitm.edu.my freebee99@163.com

*Contributed equally

 11 2024 10 09 2024 28 5 421 27 09 2019 19 12 2022 Copyright: © 2024 Ge et al. 2024 This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

Ocular metastasis (OM) in breast cancer (BC) always predicts poor prognosis. The present study explored differences in tumor markers in patients with BC with and without OM, and attempted to determine risk factors for OM in patients with BC. This study involved 629 patients with BC. Patients' clinical features were tested using χ2 test, unpaired Student's t-test and Mann-Whitney U. These parameters were analyzed using binary logistic regression to obtain risk factors for OM. A receiver operating characteristic curve was then established to determine the diagnostic value for OM. There were no significant differences in age, sex, menopausal state, and pathological type between the two groups. Significantly more axillary lymph node metastases were observed in the OM group compared with the non-ocular metastases group. Cancer antigen 153 (CA153) was revealed to be a significant independent risk factor for OM in patients with BC. The cutoff CA153 value for diagnosis of OM was 43.00 u/ml, the sensitivity was 96.15% and the specificity was 96.02%. In conclusion, CA153 was demonstrated to be a risk factor for OM in patients with BC. High levels of CA153 were associated with OM in patients with BC.

 breast cancer tumor markers diagnostic value ocular metastases Funding: This work was supported by the National Natural Science Foundation of China (grant no. 82160195), Jiangxi Double-Thousand Plan High-Level Talent Project of Science and Technology Innovation (grant no. jxsq2023201036) and the Key R & D Program of Jiangxi Province (grant no. 20223BBH80014). Introduction

Breast cancer (BC) is the most commonly diagnosed cancer in women globally in addition to being the leading cause of cancer mortality in women in >100 countries, with a continuously increasing incidence (1). Moreover, >1/3 of patients with BC will develop distant metastases such as lung, liver, bone and brain metastases (2-5).

Ocular metastasis (OM), an uncommon distant metastasis, is easily neglected because of its obscure clinical symptoms in the early stage (6). When it develops to an advanced stage, OM causes ocular pain, foreign body sensation, vision loss, visual field defects and other symptoms, thus seriously affecting patients' quality of life (7). Currently, positron emission computed tomography/computed tomography (PET/CT), magnetic resonance imaging (MRI) and ultrasonic testing are often used in clinical practice for diagnosis of OM (8,9). However, these approaches have obvious disadvantages, including their expense and damage from high-dose radiation (10). Thus, it is important to explore improved methods for predicting OM in breast cancer. Serum testing for clinical parameters is a practical method to assess the possibility of distant metastases, as it can shed light on the progress of tumors. Among the various parameters that have been used in clinical practice, tumor markers are considered to be reliable indices to predict distant metastases in patients with cancer (11).

A cancer biomarker is a substance or process that indicates the presence of cancer in the body (12). Measured either in the tumor or in blood, tumor biomarkers can be used to evaluate tumor condition and thus predict the possibility of developing distant metastases (11). Traditional cancer biomarkers include embryonic antigens, as well as protein, carbohydrate, enzyme and hormonal markers (13).

Cancer antigen 153 (CA153) is used to detect Mucin 1 (MUC-1), a transmembrane protein consisting of two subunits (14). MUC-1 is expressed at the apical plasma membrane in normal secretory epithelial cells (15). However, it is released into the serum when metastatic BC occurs (16). Numerous other advanced types of cancer, including ovarian, pancreatic, gastric and lung cancer, also result in elevated CA153 levels (17-20). High levels of C153 can also be observed in a number of benign diseases, including chronic active hepatitis, liver cirrhosis, sarcoidosis and metaplastic anemia (17,21,22). Owing to its low specificity, the use of CA153 in diagnosing early BC has been limited, although it provides useful prognostic information. Numerous studies have demonstrated that high preoperative levels of CA153 are associated with shorter disease-free and overall survival times in patients with BC (23,24).

Other cancer biomarkers have also been used in diagnosing tumors and metastases. In patients with colorectal cancer and liver metastasis, higher carcinoembryonic antigen (CEA) level is associated with shorter median progression-free survival and median liver progression-free survival (25). In Zhao's study (26), the sensitivity and specificity of cancer antigen 125 (CA125) for diagnosing ovarian cancer were 88.2 and 67.4%, respectively. Cancer antigen 199 (CA199) level >300 mg/ml is an independent prognostic factor for postoperative survival in Zheng's study (hazard ratio, 3.76; 95% CI, 2.18-6.49) (27). Alkaline phosphatase (ALP) is used to diagnose bone metastases in BC with a high specificity (93.96%) but a low sensitivity (65.14%) (28).

A number of studies have reported altered levels of CA153 in patients with cancer with metastases at various sites and investigated their diagnostic value. However, whether CA153 levels could be used to detect OM in patients with BC remains unknown. The present retrospective study analyzed levels of CA153 and other common tumor biomarkers of patients with BC to determine their value in diagnosing OM.

 Materials and methods <sec> <title>Study design

This study was designed to determine the associations between levels of cancer biomarkers and OM in patients with BC. The data range of sample collection is Jan 2005-Jan 2019. All BC were diagnosed pathologically, and the pathologist was independent from the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of the First Affiliated Hospital of Nanchang University (approval no. CDYFY-2015-112; date, 2015-05-06; Nanchang, China). All the methods complied with relevant guidelines and regulations. Breast tissues from patients were resected during surgery and the diagnosis of BC was made on the basis of pathological biopsy. CT and MRI were used to make the OM diagnosis. Inclusion criteria were all patients who were diagnosed with BC based on the pathologically. Exclusion criteria were primary ocular malignant tumor, primary ocular benign tumor and secondary BC. As this is a retrospective study, informed consent of the patients was waived, which was approved by the Ethics Committee at The First Affiliated Hospital of Nanchang University.

 Data collection

Age, sex, histopathological subtype and menopausal status were recorded as basic information and analyzed for differences among patients (Table I). As well as CA153, axillary lymph node metastasis (ALNM), CEA, CA125, CA199, ALP and calcium were analyzed as these parameters are frequently used in clinical practice and reflect the condition of tumors in patients with cancer. All clinical indices were collected from medical records when patients were first diagnosed with BC. Consecutive data are represented as mean ± standard deviation.

 Statistical analysis

First, basic clinical features including age, sex, histopathological subtype, menopausal status and ALNM number were compared using Fisher's exact test, χ2 test and unpaired Student's t-test. Then, clinical parameters of the ocular metastases (OM) group and non-ocular metastases (NOM) group were compared across different subgroups using the Mann-Whitney U test. Binary logistic regression was carried out to determine independent risk factors for OM. A receiver operating characteristic (ROC) curve was constructed with MedCalc 18.2.1 (MedCalc Software, Ltd.), and the cutoff value, area under the curve (AUC), sensitivity and specificity were obtained. Measurement data are presented as mean ± standard deviation (SD). P<0.05 was considered to indicate a statistically significant difference.

 Results <sec> <title>Clinical features of patients with BC

There were 629 patients with BC in total (627 female and two male), of which 26 patients had OM and 603 had NOM. There were no differences in age (P=0.241), sex (P=1.000), menopausal status (P=0.248), or histopathology (P=0.656) between the OM and NOM groups. The average age of patients in the OM group was 45.85±7.76 years, compared with 48.31±10.58 years in the NOM group. In the OM group, 19 patients had premenopausal status and seven were postmenopausal, whereas in the NOM group the numbers were 373 and 230, respectively. Regarding histopathology, the OM group contained 20 cases of invasive ductal carcinoma and six cases of other types. In the NOM group, there were 440 cases of invasive ductal carcinoma and 163 cases of other types. A statistical difference was revealed for ALNM (P<0.001); the OM group contained four cases without ALNM and 22 cases with ALNM, whereas the NOM group contained 259 and 344 cases, respectively. Details are presented in Table I.

 Distributions of patients with BC and condition of their distant organ metastases

The top four cities patients came from were Nanchang, Shangrao, Fuzhou, and Jian (marked in red in Fig. 1A). The majority of the patients were from Jiangxi province (marked in red and yellow in Fig. 1A), but there were also 11 patients from other provinces (marked in blue in Fig. 1A). Details of the patient distribution are presented in Fig. 1A.

Numbers of distant organ metastases (bone, liver, lung and brain) in the two groups were counted. Bone was the most common metastatic site for BC, accounting for 25 and 62 cases in the OM and NOM group, respectively (Fig. 1B).

 Subgroup analysis and binary logistic regression analysis

To exclude potential confounding by other distant metastases and clarify the diagnostic value of CA153, data were further divided into subgroups. Differences in CA153 levels between the OM group and NOM group were revealed to be statistically significant in each subgroup (Table II). The P-values are summarized in Table III. The binary logistic regression analysis identified CA153 as a risk factor for OM in patients with BC (P<0.001; Table IV).

 Sensitivity, specificity, AUC and cutoff value for CA153

A ROC curve for CA153 was established to evaluate its diagnostic value. The AUC for CA153 was 0.977, with a 95% CI of 0.958-0.995. The cutoff value was 43.3 µ/ml, with a corresponding sensitivity of 96.15% and specificity of 96.02% (Fig. 2). Details are presented in Table V.

 Discussion

BC has a high incidence of distant metastases and has increased the medical cost burden of our society. Although the widespread use of mammography, PET/CT and MRI has increased the detection rate, there are still some disadvantages (29). Various risk factors and clinical parameters associated with distant metastases of BC have been taken into consideration to make up for the limitations of these techniques (28,30-32) (Table VI).

Tumor marker serum tests are widely used in patients with cancer to evaluate tumor occurrence and condition. Several studies have already investigated the use of tumor markers in a number of types of cancer. Brand et al (33) revealed higher levels of CA199 in patients with pancreatic ductal carcinoma (PDC) compared with healthy controls, indicating that CA199 can be used to diagnose PDC. In Stojkovic's study (34), higher CEA levels are associated with advanced stage in patients with colorectal carcinoma (CRC); the study also demonstrates the use of CEA as a diagnostic factor to predict the severity of CRC. In another study (35), CA125 has been used to diagnose ovarian cancer; the sensitivity and specificity of CA125 according to the ROC curve were 79.6 and 82.5%, respectively. Also, in work by Tang et al (36), CA153 has been used in the diagnosis of BC with a sensitivity of 63% and a specificity of 82%.

Notably, tumor markers are also helpful to assess a patient with cancer's risk of distant metastasis. Yuan et al (37) observed that CA125 can increase A2780 and OVCAR-3 cell migration, indicating that CA125 may play an important part in tumor metastasis. Moreover, Zhou et al (38) revealed CA125 to be an independent risk factor of bone metastases in patients with lung cancer. Cao et al (31) revealed that CA153 is useful to predict liver metastasis in patients with BC.

OM is often associated with poor prognosis and low quality of life (39). The incidence of OM in BC has varied among different studies, with rates between 5-30% (40,41). The choroid is the most common site of OM owing to its rich blood supply, followed by the orbit, iris, ciliary body, optic nerve, conjunctiva and eyelid (42). When tumor cells grow in these locations, the substances they produce and release enter the blood and can be tested (43). Thus, tumor markers can well reflect the stage of an OM tumor. However, the diagnostic value of tumor markers in OM of patients with BC remains unclear.

The present study analyzed the differences in basic clinical features and common tumor markers between the OM group and NOM group. Among the basic features, ALNM demonstrated significant differences between the two groups. There were also statistically significant differences in CA153 levels in all subgroups and in the whole group, indicating its diagnostic value as a risk factor for OM in patients with BC.

The present study was the first to evaluate the diagnostic value of CA153 in patients with BC with OM. Compared with patients with NOM, patients with OM had higher CA153 levels. The cutoff value was 43.3 µ/ml, suggesting that patients with BC with CA153 levels higher than this value were at risk of OM. Moreover, the sensitivity and specificity for diagnosing OM in patients with BC were 0.96 and 0.96, respectively. Given the high AUC value (0.977), CA153 represents a relatively high predictive accuracy. As an important specific marker of BC, it is widely used in diagnosis of BC and in assessing prognosis and metastasis (44). In a meta-analysis performed by Tang et al (36), the sensitivity and specificity of CA153 in breast secretions for predicting BC were 0.63 and 0.82, respectively. Another meta-analysis conducted by Li et al (45), demonstrated that elevated CA153 is associated with shorter disease-free survival times in patients with BC. In a retrospective study performed by Chen et al (28), CA153 was revealed to be an independent risk factor for BC bone metastases, with sensitivity and specificity of 0.77 and 0.87, respectively.

Although CA153 is an independent risk factor for both bone metastases and OM, there are differences between the two conditions. First, in the present study, bone metastases were more common, whereas OM was relatively rare. Second, OM was more likely to occur at an advanced stage, indicating the progression of the disease, whereas bone metastases could be found in early BC. Third, when bone metastases occurred, elevated CA153 and ALP levels were detected, whereas OM was not associated with high levels of ALP. Regarding the diagnostic value of CA153, high levels of CA153 were detected in both metastases, but with some differences. On the one hand, the cutoff values were different. For diagnosing bone metastases, the cutoff value was 25.42 µ/ml, whereas that for OM was 43.3 µ/ml. On the other hand, the sensitivity and specificity for OM were both >95%. Given these differences and clinical symptoms, it is not difficult to distinguish bone metastases from OM.

However, the present study had some limitations. First, as OM is rare, the sample size was relatively small. Second, the majority of participants were from the same province; more studies of patients from different places are needed. Third, although the present study indicated an association between CA153 and OM, it could not identify their causal relationship. Hence, tests from more institutions are necessary to validate the conclusions of the present study. It would also be appropriate to explore the diagnostic value of CA153 in other metastases. Furthermore, associated molecular experiments could be carried out to explore the specific correlations.

The present study clarified CA153 as an independent risk factor for OM in patients with BC. Patients with BC with CA153 >43.3 µ/ml were more likely to have OM. Although CA153 levels had high predictive value for OM according to the present study, CA153 alone was insufficient to diagnose OM in patients with BC. Combining relevant clinical parameters with clinical symptoms would be an improved strategy. However, the present study indicated that physicians should be vigilant when CA153 levels are over the cutoff value. Treatment should be performed in time to avoid severe metastases. These results will help physicians to improve prediction of OM in patients with BC and provide some insight into the specific mechanisms of tumor markers in cancer metastases.

 Acknowledgements

Not applicable.

 Data availability and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

 Authors' contributions

YS designed the study. QG, QH and JL analyzed and interpreted the patient data. YS and QG confirm the authenticity of all the raw data. QG, QH, JL, QYL and YS made major contributions writing the manuscript. YM, BL, and RL performed the statistical analyses. WS, QL, QY, and QYL collected the clinical data. All authors read and approved the final manuscript.

 Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of the First Affiliated Hospital of Nanchang University (approval no. CDYFY-2015-112). All patients provided written consent to participate.

 Consent for publication

Not applicable.

 Conflict of interest

The authors declare that they have no competing interests.

 References BrayFFerlayJSoerjomataramISiegelRLTorreLAJemalAGlobal cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countriesCA Cancer J Clin6839442420183020759310.3322/caac.21492 MinnAJGuptaGPSiegelPMBosPDShuWGiriDDVialeAOlshenABGeraldWLMassaguéJGenes that mediate breast cancer metastasis to lungNature43651852420051604948010.1038/nature03799 BacalbasaNBalescuIIlieVFloreaRSoropABrasoveanuVBrezeanIVilcuMDimaSPopescuIThe impact on the long-term outcomes of hormonal status after hepatic resection for breast cancer liver metastasesIn Vivo321247125320183015045210.21873/invivo.11372 OruçZKaplanMAArslanÇAn update on the currently available and future chemotherapy for treating bone metastases in breast cancer patientsExpert Opin Pharmacother191305131620183012937310.1080/14656566.2018.1504922 BosPDZhangXHNadalCShuWGomisRRNguyenDXMinnAJvan de VijverMJGeraldWLFoekensJAMassaguéJGenes that mediate breast cancer metastasis to the brainNature4591005100920091942119310.1038/nature08021 LiangRBYuKWuJLLiuJXLinQLiBZhangYQGeQMLiQYShuHYShaoYRisk factors and their diagnostic values for ocular metastases in invasive ductal carcinomaCancer Med1082483220213333693210.1002/cam4.3656 GeorgalasIParaskevopoulosTKoutsandreaCKardaraEMalamosPLadasDPapaconstantinouDOphthalmic metastasis of breast cancer and ocular side effects from breast cancer treatment and management: Mini reviewBiomed Res Int201557408620152607895610.1155/2015/574086 ReddySKurliMTenaLBFingerPTPET/CT imaging: Detection of choroidal melanomaBr J Ophthalmol891265126920051617011310.1136/bjo.2005.066399 RazekAAElkhamarySMRI of retinoblastomaBr J Radiol8477578420112184936310.1259/bjr/32022497 ShaoYHTsaiKKimSWuYJDemissieKExposure to tomographic scans and cancer risksJNCI Cancer Spectr4pkz07220193233749010.1093/jncics/pkz072 MartinTAYeLSandersAJLaneJJiangWGCancer Invasion and Metastasis: Molecular and Cellular Perspective. In: Madame Curie Bioscience Database [Internet]. Landes Bioscience, Austin, TX, 2000-2013. WagnerPDSrivastavaSNew paradigms in translational science research in cancer biomarkersTransl Res15934335320122242443610.1016/j.trsl.2012.01.015 ShimetaniNStatus quo of tumor markers and important points in medical practiceRinsho Byori5112161220200314743745(In Japanese) KufeDWMucins in cancer: Function, prognosis and therapyNat Rev Cancer987488520091993567610.1038/nrc2761 DuffyMJEvoyDMcDermottEWCA 15-3: Uses and limitation as a biomarker for breast cancerClin Chim Acta4111869187420102081694810.1016/j.cca.2010.08.039 RahnJJDabbaghLPasdarMHughJCThe importance of MUC1 cellular localization in patients with breast carcinoma: An immunohistologic study of 71 patients and review of the literatureCancer911973198220011139157510.1002/1097-0142(20010601)91:11<1973::aid-cncr1222>3.0.co;2-a HayesDFZurawskiVRKufeDWComparison of circulating CA15-3 and carcinoembryonic antigen levels in patients with breast cancerJ Clin Oncol4154215501986242894910.1200/JCO.1986.4.10.1542 StieberPMolinaRChanDWFritscheHABeyrauRBonfrerJMFilellaXGornetTGHoffTJägerWEvaluation of the analytical and clinical performance of the Elecsys CA 15-3 immunoassayClin Chem4721622164200111719486 StieberPMolinaRChanDWFritscheHABeyrauRBonfrerJMFilellaXGornetTGHoffTJägerWClinical evaluation of the Elecsys CA 15-3 test in breast cancer patientsClin Lab491524200312593471 BonGGvon Mensdorff-PouillySKenemansPvan KampGJVerstraetenRAHilgersJMeijerSVermorkenJBClinical and technical evaluation of ACS BR serum assay of MUC1 gene-derived glycoprotein in breast cancer, and comparison with CA 15-3 assaysClin Chem4358559319979105258 ColomerRRuibalAGenolláJRubioDDel CampoJMBodiRSalvadorLCirculating CA 15-3 levels in the postsurgical follow-up of breast cancer patients and in non-malignant diseasesBreast Cancer Res Treat131231331989273096010.1007/BF01806524 SymeonidisAKouraklis-SymeonidisAApostolopoulosDArvanitopoulouEGiannakoulasNVassilakosPZoumbosNIncreased serum CA-15.3 levels in patients with megaloblastic anemia due to vitamin B12 deficiencyOncology6735936720041571399110.1159/000082919 MolinaRAugeJMFarrusBZanónGPahisaJMuñozMTorneAFilellaXEscuderoJMFernandezPVelascoMProspective evaluation of carcinoembryonic antigen (CEA) and carbohydrate antigen 15.3 (CA 15.3) in patients with primary locoregional breast cancerClin Chem561148115720102047282510.1373/clinchem.2009.135566 DuffyMJDugganCKeaneRHillADMcDermottECrownJO'HigginsNHigh preoperative CA 15-3 concentrations predict adverse outcome in node-negative and node-positive breast cancer: Study of 600 patients with histologically confirmed breast cancerClin Chem5055956320041472646710.1373/clinchem.2003.025288 PengSHuangPYuHWenYLuoYWangXZhouJQinSLiTChenYPrognostic value of carcinoembryonic antigen level in patients with colorectal cancer liver metastasis treated with percutaneous microwave ablation under ultrasound guidanceMedicine (Baltimore)97e004420182951766110.1097/MD.0000000000010044 ZhaoTHuWCA125 and HE4: Measurement tools for ovarian cancerGynecol Obstet Invest8143043520162716072610.1159/000442288 ZhengBHYangLXSunQMFanHKDuanMShiJYWangXYZhouJFanJMaZYGaoQA new preoperative prognostic system combining CRP and CA199 for patients with intrahepatic cholangiocarcinomaClin Transl Gastroenterol8e11820172898108210.1038/ctg.2017.45 ChenWZShenJFZhouYChenXYLiuJMLiuZLClinical characteristics and risk factors for developing bone metastases in patients with breast cancerSci Rep71132520172890028510.1038/s41598-017-11700-4 KashyapDPalDSharmaRGargVKGoelNKoundalDZaguiaAKoundalSBelayAGlobal increase in breast cancer incidence: Risk factors and preventive measuresBiomed Res Int2022960543920223548013910.1155/2022/9605439 SlimaneKAndreFDelalogeSDunantAPerezAGrenierJMassardCSpielmannMRisk factors for brain relapse in patients with metastatic breast cancerAnn Oncol151640164420041552006510.1093/annonc/mdh432 CaoRWangLPSerological diagnosis of liver metastasis in patients with breast cancerCancer Biol Med9576220122369145710.3969/j.issn.2095-3941.2012.01.011 HuJLiGZhangPZhuangXHuGA CD44v subpopulation of breast cancer stem-like cells with enhanced lung metastasis capacityCell Death Dis8e267920172830083710.1038/cddis.2017.72 BrandRENolenBMZehHJAllenPJEloubeidiMAGoldbergMEltonEArnolettiJPChristeinJDVickersSMSerum biomarker panels for the detection of pancreatic cancerClin Cancer Res1780581620112132529810.1158/1078-0432.CCR-10-0248 Stojkovic LalosevicMStankovicSStojkovicMMarkovicVDimitrijevicILalosevicJPetrovicJBrankovicMPavlovic MarkovicAKrivokapicZCan preoperative CEA and CA19-9 serum concentrations suggest metastatic disease in colorectal cancer patients?Hell J Nucl Med20414520172831590710.1967/s002449910505 DayyaniFUhligSColsonBSimonKRolnyVMorgensternDSchlumbrechtMDiagnostic performance of risk of ovarian malignancy algorithm against CA125 and HE4 in connection with ovarian cancer: A meta-analysisInt J Gynecol Cancer261586159320162754069110.1097/IGC.0000000000000804 TangSWeiLSunYZhouFZhuSYangRHuangYZhangHXuHYangJCA153 in breast secretions as a potential molecular marker for diagnosing breast cancer: A meta analysisPLoS One11e016303020162763655210.1371/journal.pone.0163030 YuanQSongJYangWWangHHuoQYangJYuXLiuYXuCBaoHThe effect of CA125 on metastasis of ovarian cancer: Old marker new functionOncotarget8500155002220172863700610.18632/oncotarget.18388 ZhouYYuQFPengAFTongWLLiuJMLiuZLThe risk factors of bone metastases in patients with lung cancerSci Rep7897020172882771910.1038/s41598-017-09650-y LinQChenXYLiuWFZhuPWShiWQLiBYuanQMinYLLiuJMShaoYDiagnostic value of CA-153 and CYFRA 21-1 in predicting intraocular metastasis in patients with metastatic lung cancerCancer Med91279128620203121884910.1002/cam4.2354 KreuselKMWiegelTStangeMBornfeldNFoersterMHIntraocular metastases of metastatic breast carcinoma in the woman. Incidence, risk factors and therapyOphthalmologe9734232620001089227810.1007/s003470050534(In German) NelsonCCHertzbergBSKlintworthGKA histopathologic study of 716 unselected eyes in patients with cancer at the time of deathAm J Ophthalmol957887931983685918810.1016/0002-9394(83)90066-1 DemirciHShieldsCLChaoANShieldsJAUveal metastasis from breast cancer in 264 patientsAm J Ophthalmol13626427120031288804810.1016/s0002-9394(03)00192-2 TarroGPernaAEspositoCEarly diagnosis of lung cancer by detection of tumor liberated proteinJ Cell Physiol2031520051538963710.1002/jcp.20195 LuFPanSQiYLiXWangJThe clinical application value of RDW, CA153, and MPV in breast cancerClin Lab6720213361634410.7754/Clin.Lab.2020.200507 LiXDaiDChenBTangHXieXWeiWClinicopathological and prognostic significance of cancer antigen 15-3 and carcinoembryonic antigen in breast cancer: A Meta-analysis including 12,993 patientsDis Markers2018986309220182985402810.1155/2018/9863092

Distributions of patients with BC and conditions of distant organs metastases. (A) Distributions of patients with BC were counted. (B) Other distant metastases of patients with BC were counted. OM, ocular metastases; NOM, non-ocular metastases.

Diagnostic of ocular metastases based on CA153 value. AUC was 0.977 (P<0.001; 95% confidence interval, 0.958-0.995). The cutoff value was 43.3 µ/ml, with a sensitivity of 96.15% and a specificity of 96.02%. AUC, areas under the curve.

Clinical features of patients with breast cancer.

Characteristic OM NOM Whole P-value
Age, years±SDa 45.85±7.76 48.31±10.58 48.21±10.48 0.241
Sex, nb        
     Women 26 601 627 1.000
     Men 0 2 2  
Menopausal status, nc       0.248
     Premenopausal 19 373 392  
     Postmenopausal 7 230 237  
Histopathology, nc       0.656
     Invasive ductal carcinoma 20 440 460  
     Other types 6 163 169  
Axillary lymph node metastases, nc       <0.001
     0 4 259 263  
     1-4 7 210 217  
     >4 15 134 149  

aUnpaired Student's t-test was used;

bFisher's exact test;

cχ2 test was used. OM, ocular metastases; NOM, non-ocular metastases.

Subgroup analysis.

A, Premenopausal
Clinical features OM (n=19) NOM (n=373) P-value
CEA (ng/ml) 39.59±107.83 2.93±9.99 <0.001
CA125 (µ/ml) 57.23±103.74 22.86±67.94 <0.001
CA153 (µ/ml) 158.59±129.37 17.16±26.55 <0.001
CA199 (µ/ml) 30.89±50.44 13.28±13.76 0.375
ALP (µ/l) 134.32±101.45 59.60±24.64 <0.001
Calcium (mmol/l) 2.30±0.42 2.32±0.55 0.688
B, Postmenopausal
Clinical features OM (n=7) NOM (n=230) P-value
CEA (ng/ml) 43.56±72.41 39.45±531.00 0.009
CA125 (µ/ml) 138.19±203.30 24.63±174.47 0.002
CA153 (µ/ml) 147.14±72.26 19.55±32.91 <0.001
CA199 (µ/ml) 8.37±4.55 27.50±216.90 0.350
ALP (µ/l) 107.29±59.49 82.15±42.56 0.391
Calcium (mmol/l) 2.12±0.60 2.32±0.15 0.622
C, NALNM
Clinical features OM (n=4) NOM (n=259) P-value
CEA (ng/ml) 53.01±101.07 2.39±4.27 0.079
CA125 (µ/ml) 118.43±209.20 14.84±31.73 0.213
CA153 (µ/ml) 217.99±219.97 13.13±9.72 <0.001
CA199 (µ/ml) 11.62±6.95 12.90±11.85 0.920
ALP (µ/l) 97.50±43.61 63.67±22.47 0.053
Calcium (mmol/l) 2.65±0.16 2.31±0.13 <0.001
D, ALNM
Clinical features OM (n=22) NOM (n=344) P-value
CEA (ng/ml) 38.41±99.88 27.75±434.30 <0.001
CA125 (µ/ml) 71.87±126.60 30.08±156.44 <0.001
CA153 (µ/ml) 144.15±89.88 21.79±37.24 <0.001
CA199 (µ/ml) 27.22±47.59 23.08±177.62 0.646
ALP (µ/l) 132.41±97.68 71.61±40.89 <0.001
Calcium (mmol/l) 2.18±0.47 2.33±0.57 0.235
E, Bone metastases
Clinical features OM (n=25) NOM (n=62) P-value
CEA (ng/ml) 42.14±99.88 141.30±1021.78 0.095
CA125 (µ/ml) 81.67±140.05 105.98±365.58 0.019
CA153 (µ/ml) 156.81±117.64 54.64±77.02 <0.001
CA199 (µ/ml) 25.33±44.89 72.86±416.87 0.704
ALP (µ/l) 130.00±92.31 102.05±77.10 0.222
Calcium (mmol/l) 2.26±0.48 2.49±1.30 0.980
F, Liver metastases
Clinical features OM (n=14) NOM (n=33) P-value
CEA (ng/ml) 36.01±56.65 252.70±1400.68 0.009
CA125 (µ/ml) 53.37±109.41 122.59±462.49 0.061
CA153 (µ/ml) 140.13±127.64 36.24±53.15 <0.001
CA199 (µ/ml) 31.53±55.19 117.97±570.94 0.735
ALP (µ/l) 140.36±101.37 93.33±81.60 0.036
Calcium (mmol/l) 2.32±0.42 2.30±0.25 0.761
G, Lung metastases
Clinical features OM (n=14) NOM (n=39) P-value
CEA (ng/ml) 53.26±131.52 8.04±29.49 0.047
CA125 (µ/ml) 78.84±148.35 35.70±90.15 0.023
CA153 (µ/ml) 131.96±125.65 20.37±21.18 <0.001
CA199 (µ/ml) 37.96±57.54 14.28±17.47 0.175
ALP (µ/l) 90.50±42.53 70.69±20.57 0.102
Calcium (mmol/l) 2.45±0.26 2.31±0.15 0.068
H, Brain metastases
Clinical features OM (n=12) NOM (n=12) P-value
CEA (ng/ml) 20.99±31.48 6.06±8.68 0.160
CA125 (µ/ml) 77.45±131.97 16.19±12.78 0.012
CA153 (µ/ml) 185.05±151.17 28.40±21.97 <0.001
CA199 (µ/ml) 44.28±60.21 10.18±7.11 0.052
ALP (µ/l) 137.25±105.89 64.42±22.92 0.045
Calcium (mmol/l) 2.41±0.29 2.24±0.12 0.155
I, Whole
Clinical features OM (n=26) NOM (n=603) P-value
CEA (ng/ml) 40.66±98.15 16.86±328.08 <0.001
CA125 (µ/ml) 79.03±137.87 23.53±120.14 <0.001
CA153 (µ/ml) 155.51±115.46 18.07±29.14 <0.001
CA199 (µ/ml) 24.82±44.05 18.71±134.39 0.792
ALP (µ/l) 127.04±91.70 68.20±34.42 <0.001
Calcium (mmol/l) 2.25±0.47 2.32±0.44 0.958

Mann-Whitney U test was used. OM, ocular metastases; NOM, non-ocular metastases; ALNM, axillary lymph node metastases; NALNM, non-axillary lymph node metastases; CA, Cancer antigen.

Summary of P-values.

Clinical features PRE POST NALNM ALNM BM LIM LUM BRM Whole
CEA <0.001 0.009 0.079 <0.001 0.095 0.009 0.047 0.160 <0.001
CA125 <0.001 0.002 0.213 <0.001 0.019 0.061 0.023 0.012 <0.001
CA153 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
CA199 0.375 0.350 0.920 0.646 0.704 0.735 0.175 0.052 0.792
ALP <0.001 0.391 0.053 0.000 0.222 0.036 0.102 0.045 <0.001
Calcium 0.688 0.622 <0.001 0.235 0.980 0.761 0.068 0.155 0.958

These were the P-values of all subgroup analyses. PRE, premenopausal; POST, postmenopausal; NALNM, non-axillary lymph node metastases; ALNM, axillary lymph node metastases; BM, bone metastases; LIM, liver metastases; LUM, lung metastases; BRM, brain metastases; CEA, carcinoembryonic antigen; CA, cancer antigen; ALP, alkaline phosphatase.

Binary logistic regression analysis.

Factors B Exp(B) Exp(B) 95% CI P-value
CEA -0.020 0.998 0.992-1.003 0.458
CA125 0.001 1.001 0.998-1.004 0.464
CA153 0.028 1.029 1.018-1.039 <0.001
CA199 0.002 1.002 0.989-1.015 0.759
ALP 0.001 1.001 0.991-1.010 0.892
Calcium -0.064 0.938 0.266-3.311 0.921

Enter method was used in the binary logistic regression analysis. CI, confidence interval; CEA, carcinoembryonic antigen; CA, cancer antigen; ALP, alkaline phosphatase.

Receiver operating characteristic curve analysis.

Variable Cutoff value Sensitivity (%) Specificity (%) AUC 95% CI P-value
CA153 43.3 96.15 96.02 0.977 0.958-0.995 <0.001

Sensitivity and specificity were obtained at the point of cutoff value. AUC, area under the curve; CI, confidence interval.

Literature summary of risk factors for common distant metastases in patients with BC.

Author (Refs.) Year Distant metastasis Risk factors
Slimane et al (30) 2004 Brain Negative hormone receptor status
Cao et al (31) 2012 Liver Lactate dehydrogenase + γ- glutamyltransferase + CA153
Chen et al (28) 2017 Bone Axillary lymph node metastases + CA125 + CA153 + alkaline phosphatase + hemoglobin
Hu et al (32) 2017 Lung CD44v

CA153, cancer antigen 153; CA125, cancer antigen 125.