The ABO Blood Group is an Independent Prognostic Factor in Patients With Resected Non-small Cell Lung Cancer

Background The ABO blood group is reported to be associated with the incidence and patient survival for several types of malignancies. We conducted a retrospective study to evaluate the prognostic significance of the ABO blood group in patients with resected non-small cell lung cancer (NSCLC). Methods A total of 333 patients (218 men and 115 women) with resected NSCLC were included in this study. In addition to age, sex, smoking status, preoperative serum carcinoembryonic antigen (CEA) level, operative procedure, histology of tumors, pathological stage (p-stage), and adjuvant therapy, the association between the ABO blood group and survival was explored. Results The 5-year overall and disease-free survival rates were 83.0% and 71.6% for blood group O, 67.2% and 62.3% for blood group A, 68.8% and 68.8% for blood group B and 69.2% and 65.3% for blood group AB, respectively. A multivariate analysis for overall survival showed the ABO blood group (group A vs. group O: HR 2.47, group AB vs. group O: HR 3.62) to be an independent significant prognostic factor, in addition to age, sex, smoking status, p-stage, and serum CEA level. A multivariate analysis for disease-free survival also showed the ABO blood group to be an independent significant prognostic factor. Conclusions The ABO blood group is an independent prognostic factor in patients with resected NSCLC. Studies of other larger cohorts are therefore needed to confirm the relationship between the ABO blood group and the prognosis among patients with resected NSCLC.


INTRODUCTION
Lung cancer is the leading cause of cancer death worldwide. In 2008, 1.6 million people received a new diagnosis of lung cancer, comprising 13% of all new cancer diagnoses, and 1.37 million people died of lung cancer, accounting for 18% of all cancer deaths in the world. 1,2 Patients with lung cancer, especially non-small cell lung cancer (NSCLC) without metastatic disease, are considered to be candidates for surgical resection. Although complete resection is often achieved in such patients, some patients experience relapse after surgery. In order to improve the outcomes of surgically managed patients, new prognostic factors must be explored, in addition to established factors such as a high preoperative or postoperative serum carcinoembryonic antigen (CEA) level, 3,4 positive results on pleural lavage cytology, 5 and a high standardized uptake value on positron emission tomography. 6 At the beginning of the 20th century, the Austrian scientist Karl Landsteiner identified the ABO blood group system. This discovery was the first detection of a genetic polymorphism in humans. Recently, an increasing number of studies have shown that the ABO blood group, in addition to its fundamental role in transfusion medicine, plays an important role in several human diseases, including venous thromboembolism (VTE), 7 coronary heart disease, 8 ischemic stroke, 9 and neoplastic disorders. Some reports have evaluated the association between the ABO blood group and the incidence of various types of malignancies, including gastric cancer, 10 pancreatic cancer, 11 and renal cell carcinoma. 12 In addition, there are two studies that evaluated the association between the ABO blood group and the prognosis of cancer patients, such as those with pancreatic cancer 13 and renal cell carcinoma. 14 Both studies found that the prognosis of blood group O patients is superior to that of non-blood group O patients. 13,14 However, few studies have assessed the relationship between the ABO blood group and the prognosis among patients with lung cancer. The aim of the present study was to clarify the prognostic significance of the ABO blood group in patients with resected NSCLC.

METHODS
Between January 2004 and December 2007, 337 patients with NSCLC underwent pulmonary resections at Nagoya University Hospital. In order to evaluate both overall survival (OS) and disease-free survival (DFS), 4 patients who had pleural dissemination at the thoracotomy (R2 resection: macroscopic residual tumor) were excluded from this study. All of the eligible patients underwent an ABO blood group examination prior to surgery. The ABO blood group evaluations were carried out via agglutination technology using the ® BioVue system (Ortho Clinical Diagnostics Japan, Tokyo, Japan). All clinical and pathological data were collected using a clinical database and charts. In addition to the ABO blood group, examined factors included age, sex, smoking status, preoperative serum CEA level, operative procedures, histology, postoperative adjuvant therapy, and pathological stage (p-stage).
Fisher's exact test and an analysis of variance were used to compare each variable between the blood groups, as appropriate. OS was calculated from the date of surgery to death. DFS was defined as the period from the date of surgery to the date of identification of recurrent disease or death from any cause. Two patients were excluded from DFS analysis because of missing data. The Kaplan-Meier method was used to calculate the survival rate with a 95% confidence interval (CI), and the log-rank test was used to compare the survival curves. A Cox proportional hazard model was used for the univariate and multivariate survival analyses. Reported P values were two-sided, and those less than 0.05 were considered statistically significant. The statistical analyses were performed using the computer software program STATA/ SE Ver.12.1 (State Corp., College Station, TX, USA). The Institutional Review Board of Nagoya University Hospital approved this retrospective study.

RESULTS
The patient characteristics are shown in Table 1. This study included 218 men and 115 women, ranging in age from 31 to 85 years (median: 68 years). The median observation period in the survivors was 73 months (range: 1-107 months). The pathological characteristics were as follows: 210 tumors were adenocarcinomas (ADs), 93 tumors were squamous cell carcinomas (SQs), and 30 tumors were other NSCLCs (others). Meanwhile, 227 patients had p-stage I disease, 49 had p-stage II disease, and 57 had p-stage III disease. Sixty-eight patients (20.4%) received adjuvant therapy (chemotherapy and/or radiation therapy) after surgery. There were 108 (32.4%) patients with blood group O, 140 (42.1%) with blood group A, 59 (17.7%) with blood group B, and 26 (7.8%) with blood group AB. The distribution of each blood group was similar to that of the general population in Japan. 15 There were significantly more advanced-stage patients in blood group O than in any other group (P = 0.037). The overall survival curves of each blood group are shown in Figure 1. The five-year overall survival (OS) rate was 83.6% (95% CI, 75.0%-89.5%) for blood group O, 68.5% (95% CI, 59.8%-75.6%) for blood group A, 68.8% (95% CI, 55.2%-79.1%) for blood group B, and 69.2% (95% CI, 47.8%-83.3%) for blood group AB. The patients in blood group O showed significantly better survival than the patients in non-O blood groups (P = 0.016). Stratified analysis by pstage revealed that the association between the ABO blood group and overall survival was similar in each p-stage group.
A were independent significant prognostic factors (Table 2). Blood groups A and AB, which express the blood group A antigen on erythrocytes, were independently associated with poor OS among patients with resected NSCLC.
Univariate analysis for DFS showed that age, sex, preoperative serum CEA level, histology, and p-stage were significant prognostic factors (

DISCUSSION
Recently, several studies have suggested important roles for the ABO blood group in the development of hemostasis and neoplastic disease, as ABO antigens are highly expressed on the surface of a variety of human cells and tissues. 16 As listed in Table 4, there are a number of reports regarding the relationship between the ABO blood group and the incidence of several types of cancers. Affirmation of this relationship has   27 Meanwhile, no affirmation of the relationship has been reported for colorectal cancer 28 and cervical/endometrial cancer. 20 Trends in the relationship between blood group and incidence of various types of cancers have been noted; namely, blood groups O or non-A show a low incidence of cancer, while blood groups non-O or A demonstrate a higher incidence of cancer. Among these studies, the relationship between the ABO blood group and the incidence of gastric and/or pancreatic cancer is considered to be reliable and convincing, as the studies were large-scale meta-analyses. 10  There are also a few studies regarding the relationship between the ABO blood group and the prognosis in patients with malignant tumors. Kaffenberger et al reported that the non-O blood type was found to be associated with a significantly decreased OS among 900 surgically managed patients with renal cell carcinoma according to a multivariate survival analysis (HR 1.68; 95% CI, 1.18-2.39). 14 The authors also reported that the non-O blood type was associated with marginally decreased DFS in the same cohort (HR 1.53; 95% CI, 0.97-2.41). Rahbari et al analyzed a total of 627 patients who underwent resection for pancreatic ductal adenocarcinoma and revealed a favorable and independent impact of blood group O (vs. non-O) on OS according to a multivariate survival analysis (HR 0.78; 95% CI, 0.62-0.99). 13 Furthermore, OuYang et al recently demonstrated the prognostic value of the ABO blood group in patients with nasopharyngeal carcinoma treated with intensitymodulated radiotherapy (IMRT) or conventional radiotherapy (CRT). 29 In their multivariate survival analysis, patients with blood type A had a significantly lower OS and distant metastasis-free survival than those in the non-A group, among both the IMRT group (n = 924) and CRT group (n = 1193). These results are consistent with our observations, in which we found that the OS and DFS of the resected NSCLC  patients in blood group A or AB are significantly poorer than that of patients in blood group O, despite the higher incidence of advanced cancer among blood group O individuals. The mechanisms by which the ABO blood group influences the prognosis of cancer patients have not been fully investigated. The A and B antigens are expressed on the surface of red blood cells as well as numerous other tissues throughout the body, 30 including lung cancer tissues. 31 One hypothesis is that ABO antigens in tumor cells play an important role in intracellular adhesion and membrane signaling, both of which are critical to the progression and spread of malignant cells. 32 Lee et al reported that the expression of blood group antigen A on lung cancer tissue is an important favorable prognostic factor in blood group A patients. 31 We speculate that the A and B antibodies in the plasma of blood group O patients have protective effects against tumor cell progression.
Recently, two genome-wide association studies have suggested that single nucleotide polymorphisms at the ABO gene locus are associated with two serum markers, namely tumor necrosis factor-α (TNF-α) and soluble intracellular adhesion molecule-1 (sICAM-1). 33,34 TNF-α is an inflammatory cytokine that affects tumor progression. In addition, the levels of sICAM-1 are known to be elevated in several types of malignancies and may play a role in escape from immune surveillance by tumor cells. 35 Therefore, we also suspect that the ABO gene locus influences the prognosis of cancer patients via the effects of these serum proteins.
There are some limitations to our retrospective analysis. First, the number of study subjects was small; however, to our knowledge, this is the first report to show the prognostic significance of the ABO blood type in surgically managed NSCLC patients. Secondly, our data regarding the ABO blood groups were obtained using a serological technique based on the phenotype, not the genotype, of the blood group. Nakao et al reported that the number of non-O alleles was found to be associated with an increased risk of pancreatic cancer in a Japanese population. 26 The number of non-O alleles may therefore have an additive effect on the prognosis of patients with NSCLC.

Conclusion
Our multivariate survival analysis showed the ABO blood group to be an independent prognostic factor in addition to age, sex, smoking status, p-stage, and serum CEA level. The blood group A antigen may have a negative effect on the prognosis of surgically managed patients with NSCLC. Studies using other larger cohorts are needed to confirm a robust relationship between ABO blood group and the prognosis of patients with resected NSCLC.

ONLINE ONLY MATERIAL
Abstract in Japanese.