RESEARCH Significance of ABO-Rh Blood Groups in Response and Prognosis in Breast Cancer Patients Treated with Radiotherapy and Chemotherapy

Background: To evaluate whether ABO-Rh blood groups have significance in the treatment response and prognosis in patients with non-metastatic breast cancer. Materials and Methods: We retrospectively evaluated files of 335 patients with breast cancer who were treated between 2005 and 2010. Demographic data, clinic-pathological findings, treatments employed, treatment response, and overall and disease-free survivals were reviewed. Relationships between clinic-pathological findings and blood groups were evaluated. Results: 329 women and 6 men were included to the study. Mean age at diagnosis was 55.2 years (range: 26-86). Of the cases, 95% received chemotherapy while 70% were given radiotherapy and 60.9% adjuvant hormone therapy after surgery. Some 63.0% were A blood group, 17.6% O, 14.3% B and 5.1% AB. In addition, 82.0% of the cases were Rh-positive. Mean follow-up was 24.5 months. Median overall and progression-free survival times were 83.9 and 79.5 months, respectively. Overall and disease-free survival times were found to be higher in patients with A and O blood groups (p<0.05). However rates did not differ with the Rh-positive group (p=0.226). In univariate and multivariate analyses, ABO blood groups were identified as factors that had significant effects on overall and disease-survival times (p=0.011 and p=0.002). Conclusions: It was seen that overall and disease-free survival times were higher in breast cancer patients with A and O blood groups when compared to those with other blood groups. It was seen that A and O blood groups had good prognostic value in patients with breast cancer.


Significance of ABO-Rh Blood Groups in Response and Prognosis in Breast Cancer Patients Treated with Radiotherapy and Chemotherapy
Yasemin Benderli Cihan (Klimant et al., 2011, Xing et al., 2014. Many blood group systems have been identified by using this antigenic structure. Environmental factors don't have any effect on blood group development and it is a quantitative characteristic that exclusively represents genetic basis. ABO blood group system was discovered by Landsteiner at 19 th century and it is the first discovered and most widely used blood group system. There is another blood group system based on presence or absence of Rh antigen. Today, ABO and Rh blood group systems are most widely used systems. In the literature, there are several studies on blood groups. These studies can be classified into two groups: studies directly addressing blood groups and its genetic, and those addressing relationship between genetics of blood group and benign or malign diseases (Holdsworth et al., 1985;Klimant et al., 2011;Miao et al., 2013).
ABO blood group genes have a differential distribution in the population. It is known that this is a risk factor for development of diseases. It has been reported that some cancer types are most frequently observed in subtypes of ABO blood groups. In the literature, many studies defined relationship between ABO blood groups and metastasis, prognosis, stage and histopathological diagnosis in several cancer types (Holdsworth et al., 1985;Iodice et al., 2010;Kos et al., 2014), although others showed no role (Urün et al., 2012;Unal et al., 2013;Utkan et al., 2013). In recent years, although there are published studies suggesting that blood groups are important in treatment response and prognosis in breast cancer, there is paucity in data in this field (Holdsworth et al., 1985;Klimant et al., 2011;Gates et al., 2012;Miao et al., 2013).
Primarily, it was aimed to investigate whether blood groups have prognostic value in breast cancer by using ABO blood groups that can be readily detected and represent genetic structure exclusively.

Materials and Methods
We retrospectively reviewed clinical data of 335 cases with non-metastatic breast cancer confirmed by histopathology and received chemotherapy, radiotherapy, and hormone therapy (according to receptor status) between 2005 and 2010. The following data were extracted from patient files: age, gender, menopausal status, stage, surgery type, adjuvant therapies (chemotherapy, radiotherapy, hormone therapy), histopathological data (histological subtypes, tumor size, axillary lymph node involvement, grade according to Scharf-Blood-Richardson grading system, hormone receptor status, and HER2/neu expression) and blood groups.
Staging was performed based on AJCC 2002 staging system. Presence of recurrence, date of recurrence, localization of recurrence and overall and disease-free survivals were extracted from patient files. Patients who didn't attend follow-up visit within prior 6 months were contacted by telephone. This study was planned in accordance to Helsinki Declaration, Patient Rights regulation and Ethic principles.

Radiotherapy
Radiotherapy was given to the patients with tumor diameter >5 cm and axillary lymph node≥3-positive. In these cases, radiotherapy was delivered to whole breast/ thorax wall and axillary and supraclavicular regions with gamma beam by using Co-60 device and 6 MV X-beam by using Linax device. Additional electron doses of 10 or 16 Gy were delivered to tumor bed and incision in patients at risk. Radiotherapy was delivered at a total dose of 50-66 Gy with 25-33 Gy fractions.

Hormone therapy and follow-up
Hormone therapy was initiated in the patients with positive estrogen and progesterone receptors. Tamoxifen and/or LHRH analogs were given to premenopausal patients while tamoxifen or aromatase inhibitors were given to postmenopausal patients. Follow-up visits were scheduled by 3-months interval within first year; biannually until end of year 5; and annually thereafter. Complete blood count, biochemical parameters, Ca 15-3 and CEA levels were measured biannually, while chest radiographs, mammography, abdominal sonography and bone scintigrapy were obtained annually.

Statistical analysis
Data were analyzed by using SPSS for Windows version 15.0 (SPSS Inc., Chicago, Illinos, USA). Normality was tested by using Kolmogrov-Simirnov method. Numeric variables were expressed as median and minimum-maximum values. Categorical variables were expressed as percents. Correlations between categorical variables were tested by using Pearson Chi-square and Monte Carlo tests. Overall survival was calculated as the time from diagnosis and death due to any reason, while disease free survival was calculated as the time from diagnosis to recurrence. Survival analysis was performed by using Kaplan-Meier curves. Univariate analysis was performed by using log-rank test, while multivariate analysis was performed by using Cox regression test. p<0.05 was considered as statistically significant. Table 1 presents demographic and clinic-pathological characteristics of the patients. Local recurrence (14 cases) or distant metastasis (45 cases) was detected during median follow-up of 19.5 months (mean: 24.5; range: 10 days-112 months). Forty-five cases died due diseaserelated or irrelevant causes. Multiple-organ metastasis was seen in 17 cases (38.0%), bone metastasis in 15 cases (33.0%), lung metastasis in 7 cases (15.0%), brain metastasis in 2 cases (4.4%), and other organ metastasis in 8 cases (8.8%). Median overall and progression-free survival times were 83.9 and 79.5 months. Five-and 7-years overall survival rates were 76% and 54%, while 5-and 7-years disease-free survival rates were 61% and 57%, respectively. Table 2 presents distribution of ABO-Rh blood groups according to clinic-pathological features and results of analyses. A significant difference was found in histological grade among ABO blood groups (p<0.008). No significant difference was found among ABO-Rh blood groups regarding age, gender, menopausal status, body mass index (BMI), localization, stage, tumor diameter, lymph node involvement, pathology, histological grade, perivascular and lymphovascular invasion, hormone receptor status, HER2 positivity and adjuvant therapies used. Table 3 presents disease-free and overall survival times according to ABO-Rh blood groups. Overall survival was found as 80.6 months in A blood group, 47.4 months in AB blood group, 59.0 months in B blood group and 84.7

Figure 1. A) Overall and B) Disease-Free Survival According to ABO Blood Groups
A B months in O blood group. When overall survival time was assessed according to ABO blood groups, there was significant difference among groups (p<0.05; Figure 1a). Disease-free survival time was found as 79.6 months in A blood group, 51.9 months in AB blood group, 43.5 months in B blood group and 86.0 months in O blood group. When overall survival time was assessed according to ABO blood groups, there was significant difference among groups (p<0.05; Figure 1b). When considered according to Rh blood groups, it was found that overall and disease-free survival times were found to be higher in Rh-positive group when compared to Rh-negative group, but the difference didn't statistical significance (p=0.262 and p=0.226). Tables 4 and 5 present results of univariate and multivariate analysis overall and disease-free survival. In univariate analysis, ABO blood groups and pathologic subtype were identified as factors that had significant effect on overall survival; ABO blood groups and hormonotherapy had significant effect on disease-survival (p<0.05). These factors (ABO blood, pathologic subtype and hormonotherapy) remained to be significant in multivariate analysis.

Discussion
Several aspects of clinic-pathological correlations of ABO-Rh blood group with many diseases, particularly cancers, have been subject of investigation from beginning of the century (Iodice et al., 2010;Gates et al., 2012;Mortazavi et al., 2014). The fact that there were discrepancies in the temporal data of blood group investigations regarding material, character, genetic and experimental applications indicates that blood groups remain to be mysterious. One can suggest that further comprehensive studies with different designs are needed to obtain consistent and concrete results. Thus, a different perspective was used to investigate the relationship between ABO blood groups and breast cancer in the present study. It was aimed to detect value of blood groups in treatment response and prognosis in breast cancer.
In breast cancer, demographic characteristics such as age, menopausal status and ethnicity, tumor characteristics such as tumor size, axillary lymph node status, and histopathological subtype, and biomarkers such as oncogene, tumor suppressor genes, growth factors and proliferation measures are known to be potential prognostic factors (Iodice et al., 2010;Klimant et al., 2011;Sozen and Benderli Cihan, 2012;Miao et al., 2013;Xing et al., 2014). Unfortunately, based on current understanding, it is impossible to precisely determine patients who would recover by local treatment or those who would die due to recurrence despite treatment; thus, there is an ongoing effort to identify novel prognostic factors. It has been thought that ABO-Rh blood groups, a genetic feature, have prognostic value in patients with breast cancer. However, data are scarce in this field. In the present study, it was aimed to clarify relationship between known blood group phenotypes and above-mentioned prognostic factors and to enable formerly detection of risk groups thought to be relevant.
In our study, it was seen that the highest percent distribution was in A blood group, while lowest percent distribution in AB blood group. The relationship between breast cancer and blood groups were first described by Aird et al. In that study, it was reported that there was no association between breast cancer and blood groups (Aird et al., 1954). Miao et al. evaluated blood group distribution in 9665 patients with breast cancer and compared with 244,768 healthy controls. Authors reported that breast cancer incidence was similar among all ABO blood groups (Miao et al., 2013). In a study by Hems and Anderson, it was reported that A blood group was more commonly seen in patients with breast cancer (Hems, 1970;Anderson et al., 1985). In the study by Tryggvadottir et al., it was reported that B blood group type was 2 fold more common among cases with familial breast cancer when compared to sporadic cases (Tryggvadottir et al., 1988). Iodice et al. evaluated blood group distribution in 15,359 cancer patients and found that breast cancer incidence was higher among patients with O blood group but the difference didn't reach statistical significant (p=0.60) (Iodice et al., 2010). In another study by Cihan et al, the distribution of ABO-Rh blood groups in 255 patients with skin cancer was compared to those obtained from 25,071 healthy blood donors. In the control group, the most frequent blood group was A blood group (44.3%); followed by O blood group (31.5%), B blood group (16.1%) and AB blood group (8.1%). In the patient group, the most frequent blood group was A blood group (50.2%); followed by O blood group (26.3%), B blood group (16.1%) and AB blood group (7.5%). There was significant difference between patient and control groups regarding distribution of ABO-Rh blood groups. Skin cancer was more commonly observed in A blood group (Cihan et al., 2013). When our study was compared to control group of our previous study, it was seen that distribution of blood groups in the breast cancer were in line with control group. In addition, our results were consistent with those reported by Guleria et al., Anderson et al. (1985) and Hems in study.
In our study, overall and disease-free survival was highest in O blood group; followed by A blood group in breast cancer. This difference was found to be statistically significant (p<0.05). Although there are discrepant results in the literature, Guleria et al. demonstrated that breast cancer was more prevalent and associated with poor prognosis in women with A blood group (Guleria et al., 2005). In a study by Gates et al., it was reported that there was no association between ABO blood groups and breast cancer risk or survival (Gates et al., 2012). In a study on 426 patients with breast cancer by Klimant et al., A blood group was identified in 198 patients (46.5%), while O blood group in 163 (38.3%), B blood group in 43 (10.1%) and AB blood group in 22 patients (5.2%). Blood group distribution in patients with breast cancer didn't differ from that in general population (p=0.08). However, there was a trend towards higher rates of B blood group and lower rates of AB blood group. In that study, 5-years overall and disease free survivals were 93.0 months in AB blood group, 80.6 months in A blood group, 79.6 months in O group and 74.5 months in B blood group. Authors concluded that there was no significant difference in overall and disease-free survivals among blood groups. Also, no correlation was reported between blood group type and HER2/neu, ER and PR status (Klimant et al., 2011). In our study, it was seen that survival was higher in patients with A and O blood groups when compared to other blood groups. In agreement with the results of Klimant et al. (2011); Gates et al. (2012) no correlation was observed between ABO-Rh blood groups and age, menopausal status, BMI, localization, stage, tumor diameter, lymph node involvement, perivascular and lymphovascular invasion, hormone receptor status and HER2 positivity (Klimant et al., 2011;Gates et al., 2012). In our study, it was found that overall and disease free survivals were higher in Rh-positive patients when compared to Rh-negative patients, but the difference didn't reach statistical significance. There is limited number of studies about relationship of Rh factor to cancer, reporting inconsistent results. In a study by Ronco et al., it was reported that there was higher risk for breast cancer in Rh-negative population (Ronco et al., 2009), while no such relationship was found in the studies by Dede et al. and Stamatakos et al. Ronco et al. reported breast cancer risk was higher by 50% in Rh-negative patients when compared to Rh-positive patients (Ronco et al., 2009;Stamatakos et al., 2009;Dede et al., 2010). In the study by Stamatakos et al., it was reported that metastasis risk was 4.2 fold higher in Rh-negative patients with breast cancer when compared to Rh-positive patients with breast cancer (Stamatakos et al., 2009).
In conclusion, in our study, it was seen that breast cancer is more common in patients with A and Rh-positive blood groups. It was seen that overall and disease-free survival times were higher in breast cancer patients with A and O blood groups when compared to those with other blood groups. Further comprehensive studies are needed to elucidate relationship between ABO-Rh blood groups and breast cancer. It can be suggested that our study will provide a basis for future studies in this topic.