LF-PRLR Expression In Human Breast Cancer:Its Influences On Survival And Correlations With Clinical Factors

Backgrounds and purposesWorldwide,the incidence rate of breast cancer has increase to be in the first among women carcinoma.The incidence rate of breast carcinoma in China is increasing year by year.Patients age,menstruation,tumor tissue histological grading,stage of progress(indicators including tumor size、metastasis of axillary lymph node and pTNM stage),the expression of ER/PR/HER-2 are important clinical factors for breast cancer patients.These clinical factors are widely used in the classification of risk degree and molecular typing and in the fields of treatment and prognosis.The biological function of Prolactin(PRL)is very wide and scholars have found out more than 300 kinds of biological functions,which covering almost all major areas of physiological functions.At the same time,PRL play a positive promoting role in genesis and progression of breast cancer.This biological functions is induced after PRLR on breast cancer cell surface has binded by PRL,namely PRLR expression on surface of breast cancer cell is one of the key factors that promoting the biological effects of PRL in breast cancer.About 70%of human breast cancer cells or tissues expresses PRLR and the positive site is mainly located in the cytoplasm.Studies have found two important sub-type of PRLR in breast cancer cell:the long form prolactin receptor(LF-PRLR)and the short form prolactin receptor(SF-PRLR)(LF-PRLR is much more than SF-PRLR).LF-PRLR take positive role in the occurrence and progression of breast cancer,contrastly the biological effect of SF-PRLR is to inhibite and regulate the biological effects of LF-PRLR.So far results from studies on PRLR in breast cancer about its influence on prognosis and and its relation with clinical factors are much different,so that scholars have difficult to draw consistent conclusions on these questions.Reasons for difference in these results maybe contribute to differences in the distribution of different sub-type of PRLR and not controling some possible influence or interference factors.Only deep into the most LF-PRLR sub-type which mostly relate to occurrence and progression of breast cancer and controling possible influence or interference factors through stratified analysis,the prognosis value of PRLR expression and correlation between PRLR expression and clinical factors can be revealed exactly and truly in breast cancer.In this study,we deeped into the LF sub-type(LF-PRLR)and explored its prognosis value and the correlation between LF-PRLR expression and various clinical factors for breast cancer patients;At the same time,we tried to find out possible influence or interference factors through stratified analysis.Through controling these influence or interference factors,we hope to further revealed exactly and truly the prognosis value of LF-PRLR and correlations between LF-PRLR expression and clinical factors of breast cancer patients,which will provide new clues and ideas for the researches of anti-PRLR treatment combined by other kinds of treatment and the individual pattern of anti-PRLR treatment.Materials and methodsEntry criteria:(1)Initial treatment of Chinese female breast cancer patients whose ages ranging from 18 to 75 years old;(2)Breast cancer was confirmed by the primary lesions biopsy of pathology,and there was no distant metastasis taking place;(3)Having received surgical treatment that including breast conserving surgery and radical(or modified radical)surgery requiring the dissectional number of axillary lymph node was more than 10 in order to stage accurately;(4)Before surgery no anti-tumor neoadjuvant treatment was used,and necessary adjuvant therapy after surgery was received if patients had corresponding indications;(5)After surgery,specimen slide and wax block was well preserved and there was enough tissue for detection and the expression of ER/PR and HER-2 were detected;[6]Excluding cases:patients whose clinical data were not complete,or whose pathological datas were not completely preserved,or whose treatment is not standard or complete.One hundred and twenty nine Chinese female patients with breast cancer having received therapy in the Cancer Center of Sun Yet-san University from January 2000 to June 2001 Preoperative chest X-ray and abdominal ultrasound and whole body bone ECT scan(CT or MRI added when necessary)were received to exclude the possibility of distant metastases.The patients aged from 26 to 75 years old,with a median age of 46 years.Among 129 patients,premenopause and postmenopause patients were 80 and 49 cases respectively.According to the pathologic classification of breast cancer by WHO,118 invasive ductal carcinoma,5 microinvasive ductal carcinoma,and 6 other types of carcinoma.According to the pTNM stagestandard of breast cancer by AJCC/UICC in 2002,20 patients were classified to phase 1,62 patients were classified to phase Ⅱ and 47 patients were classified to phase III.The expression of LF-PRLR in postoperative wax samples was detected by immunohistochemistry with LSAB kit according to the manufacturers’ instructions,with the positive contrast of well known positive cells and the negative contrast of PBS instead of primary monoclonal antibodies.Cells were determined LF-PRLR expression positive cells when yellow or dark brown granules were seen in the place of cytoplasm.The entire tissue slice was observed under an optical microscope by pathologists.When the percentage of positive cells were ≤10%of the total number of cells,the tissue was categorized as "LF-PRLR negative expression" and when the percentage was>10%,the tissue was categorized as "LF-PRLR positive expression."Survival table method was used to calculate the 3、5 and 10 years DMFS,RFS,PFS and DSS,the survival of different groups patients were compared by Kaplan-Meier analysis and Log-lank test;and COX regression model was used to conduct multivariate survival analysis.The Chi square test(categorical datas)or Spearsman rank correlation analysis(rank datas)was used to probe the correlation between LF-PRLR expression and clinical factors in breast cancer patients.P<0.05 was defined as statistically significant.ResultsIn 129 cases,LF-PRLR expression was negative in 41 cases,and positive in 88 cases(the positive rate was 68.2%).In 41 cases of LF-PRLR negative patients,there happened recurrence in 2 cases,distant metastases in 6 cases;tumor free survival in 34 cases and 4 cases died from breast cancer;In 89 cases of LF-PRLR positive patients,there happened recurrence in 10 cases,distant metastases in 30 cases,tumor free survival in 55 cases and 23 cases died from breast cancer.Comparison by Log-Lank test found that 3,5 and 10 years of DMFS,PFS and DSS in LF-PRLR negative patients were higher than that in the LF-PRLR positive patients(P value were 0.032,0.026 and 0.047 respectively,all<0.05).There was no statistically significant difference in 3,5 and 10 years of RFS between LF-PRLR negative patients and LF-PRLR positive patients(P=0.205>0.05);But the survival curves in both groups distinctly separated.In this study,we also used COX risk regression model to conduct multiple factors survival analysis.Factors taken probably into COX risk regression model included age,menopause or not,tumor quadrant,tumor tissue grade,T stage,N stage,the expression of ER、PR、HER-2 and LF-PRLR.Factor was final taken into COX risk regression when its P value from single factor Log-Lank test was lower than 0.20.Our results showed that the expression of LF-PRLR finally stayed in DMFS,PFS and DSS multivariate Cox model(all P value<0.05),and was not finally stayed in RFS multivariate Cox model(P value>0.05).Our results on correlations between LF-PRLR expression and clinical factors in patients with breast cancer were shown as followed:(1).LF-PRLR positive rate in"≤55 years old" patients was 74.5%,higher than that in">56 years old" patients(48.4%)(Chi square value=7.401,P=0.007<0.05),namely that correlation exists between LF-PRLR expression and ages of patients.Moreover,when stratified by ER or PR expression status,correlation between LF-PRLR expression and ages of patients was only showed in ER or PR negative sub-layer patients(P value 0.009 and 0.006 respectively.both<0.05).In addition,we also used "50 years old"as a age breakpoint(50 years old used widely for prognosis in BC)and separated 129 patients into two groups:"≤50 years old"patients and "50 years old" patients,as a result,LF-PRLR positive rate in two groups showed no significant statistical difference(P>0.05).(2).In ER or PR positive expression sub-layers patients,LF-LF-PRLR positive rate showed an increasing trend when the menarche age changed from "12～13 years old" to "14 years old"(P value 0.098 in ER positive subgroup and 0.030 in PR positive subgroup,both nearly or lower than 0.05);and showed an discreasing trend when the menarche age changed from "14 years old" to "14-16 years old" and to "≥17 years old"(P value 0.006 in ER positive subgroup and 0.053 in PR positive subgroup,both nearly or lower than 0.05).On the contrary,these correlations between LF-PRLR expression and the menarche age of patients was not seen in ER or PR negative sub-layer(both P value<0.05).In the PR positive and age higer than 35 years old sub-layer patients,LF-PRLR positive rate in premenopausal patients was higher than in postmenopausal patients(P=0.05);on the contrary,there were no difference in LF-PRLR positive rate between premenopausal patients and postmenopausal patients in PR negative sub-layer patients(P>0.05).As to the age of menopause,we divided fifty postmenopause patients into two groups according to their age at menopause:"≤50 years old menopause" and ">50 years old menopause",as a result that LF-PRLR positive rate in "≤50 year old menopausal" patients was higher than in ">50 years old menopause "patients(P value=0.006<0.05),namely that LF-PRLR expression was correlational with the age of menopause for breast cancer patients.(3).LF-PRLR positive rate in tumor histological grade Ⅰ,Ⅱ and Ⅲ patients were 85.7%,72.7%and 44.4%respectively,Spearman rank correlation analysis showed that there had negative correlation between LF-PRLR expression and histological grading for breast cancer(P value=0.002<0.05).When stratificated by ER expression,we found that the negative correlation between LF-PRLR expression and histological grading existed only in ER negative sub-layer patients.(4).The Spearman correlation test for all 129 patients showed that LF-PRLR positive rate was not associated indicators used for appraising sooner or later for breast cancer(tumor size,axillary lymph node metastasis and pTNM stage)(r value was 0.121,0.086 and 0.091 respectively,all P value>0.05).However,will the three indexes(tumor size,axillary lymph metastasis and pTNM staging)were divided into three step-up levels:first level:"primary tumors ≤2 cm","no axillary lymph node metastasis" and "stage Ⅰ";the second level:"primary tumor>2cm and≤5cm","1～3 axillary lymph node metastasis" and "stage Ⅱ";the third level:"primary tumor>5cm","≥4 axillary lymph node metastasis" and "stage Ⅲ".It was found that regardless of tumor size,axillary lymph node metastasis and pTNM stage,LF-PRLR positive rates in the second level patients were higher than that in the first level patients(all P value<0.05);and no statistically significant differences in LF-PRLR positive rate were seen between the second level and the third level patients(all P value>0.05).(5).The Chi quare test for all 129 patients showed that LF-PRLR positive rate in ER positive patients was higher than that in ER negative patients(P value 0.025<0.05),namely that there had a positive correlation between LF-PRLR expression and ER expression in breast cancer patients.When stratificated by HER-2 expression,age,menarche age and axillary lymph node metastasis status,this positive correlation only existed in ER negative,>55 years old,menarche age<17 years old and no axillary lymph node metastasis sub-layers patients.The Chi quare test for all 129 patients showed that there had no statistical difference in LF-PRLR positive rate between PR positive patients and PR negative patients(P>0.05),namely that there had no correlation between LF-PRLR expression and PR expression in breast cancer patients.But when stratificated by HER-2 expression,this positive correlation(LF-PRLR positive rate of PR positive patients was higher than that of PR negative patients)was seen in the HER-2 positive sub-layer patients(P value 0.038<0.05),on the contrary there was no similar correlation in the HER-2 negative sub-layer patients(P value>0.05).The Chi quare test for all patients showed that there had no statistical difference in LF-PRLR positive rate between HER-2 positive patients and HER-2 negative patients(P>0.05),namely that there had no correlation between LF-PRLR expression and HER-2 expression in breast cancer.But when stratificated by ER/PR expression,positive correlations(LF-PRLR positive rate of HER-2 positive patients was higher than that of HER-2 negative patients)were seen in the ER positive or PR positive sub-layer groups(P value 0.007 and 0.030 respectively both<0.05),on the contrary there was no similar correlation in the ER negative or PR negative sub-layer groups(all P value>0.05).Conclusions and ProspectsLF-PRLR expression affects distant metastasis,tumor-free survival and disease specific survival after surgery in patients with breast cancer:positive expression of LF-PRLR was an independent pore prognostic factor.The influence of LF-PRLR expression on the prognosis of local recurrence was perhaps more weak,and expanding samples may be able to achieve positive results.LF-PRLR expression related with some clinical factors include age,menstrual status,tumor histological grade,disease stage indicators(including tumor size,axillary lymph metastasis stutas and pTNM staging)and the expression of ER/PR/HER-2.Further more,when LF-PRLR expression is associated with one of these clinical factors,another one or more than one clinical factors may affect or interfere this association,whose details are as follows:(1)Correlation between LF-PRLR expression and age performs that LF-PRLR positive rate was significantly decreased after 55 years old,and this correlation was interfied by ER or PR expression.(2)Correlation between LF-PRLR expression and menarche age performs as a single peak curve with a peak of LF-PRLR expression at "14 years old menarche",and this correlation depends on the positive expression of ER or PR.Correlation between LF-PRLR expression and "menopause or not" depends on the positive expression of PR,which showes statistical difference when ≤35 years old of patients were excluded.The expression of LF-PRLR is associated with "menopausal age":the positive rate of LF-PRLR was higher in earlier menopause patients.(3)LF-PRLR expression was correlated with the degree of histological grading,but the correlation was interfered by the positive expression of ER.(4)There was a complicated correlation between LF-PRLR expression and the phase of disease:regardless of tumor size,axillary lymph node metastasis statue or pTNM staging indexs,from the "the first level" to "the second level" LF-PRLR positive rates increased significantly(P<0.05),and from the "the second level" to "the third level"LF-PRLR positive rates all inversely showed a decline trend(although P>0.05).That is to say that " the second level" of the phase of disease had the peak lever of LF-PRLR expression.The expression of ER/PR affacted or interfered this relationship between LF-PRLR expression and the phase of disease.(5)There was a positive correlation between LF-PRLR expression and ER expression,and HER-2 expression status,patient age,menarche age,axillary lymph metastasis status are influencial or interferent factors as to the correlation between LF-PRLR expression and ER expression.The positive correlation between LF-PRLR expression and PR expression depends on the positive expression of HER-2.The positive correlation between LF-PRLR expression and HER-2 expression depends on the positive expression of ER or PR.In the event of showing positive correlation with the expression of LF-PRLR,ER/PR expression and HER-2 expression presence complex relationship in which not only inter-dependence but also mutual promotion are seen.Our study aimed at the LF sub-type that was most closely related to the occurrence and progression of breast cancer and found out that LF-PRLR has some relevances with age,menstrual status,tumor histological grade,disease stage and and the expression of ER/PR/HER-2;moreover,we also have found that these relevances were affected or interfered by one or more than one clinical factor.In this study we reveals deeper these relevances which may provide some new clues for laboratory research and promote some new ideas for the combined use of anti-PRLR therapy with other types of treatment and the individual research on anti-PRLR therapy.However,in addition to the clinical factors involved in this study,whether are there other clinical factors such as VGFR and P53 that also have correlation with the expression of LF-PRLR?Is there other factors that will affect or interfere the correlation between LF-PRLR expression and clinical factors in breast cancer?What biological mechanisms exist behind the correlations and corresponding influences or interferences?In addition to the LF sub-type of PRLR,what results from another major sub-types PRLR—SF1b?All these questions need to be further replied.