Preoperative Anemia Is an Independent Prognostic Factor for Esophageal Cancer Patients Who Receive Curative Treatment

Abstract

Background/Aim

We retrospectively evaluated the clinical impact of anemia during the perioperative period on both short- and long-term oncological outcomes in patients with resectable esophageal cancer who received curative treatment.

Patients and Methods

We retrospectively reviewed the medical records and collected data from consecutive patients with esophageal cancer who underwent curative resection at Yokohama City University from 2005 to 2022.

Results

A total of 198 patients were included in this study. According to previous studies and the 3- and 5- year overall survival rates, a hemoglobin level of 11 g/dl was selected as the optimal cutoff value in the present study; preoperative hemoglobin of <11 g/dl and >11 g/dl were observed in 34 patients (Hb-low group) and 164 patients (Hb-high group), respectively. The 3- and 5-year OS rates were 39.1% and 34.8%, respectively, in the Hb-low group and 70.3% and 62.1% in the Hb-high group. There were significant differences between the two groups (p=0.003). Univariate and multivariate analyses demonstrated that the preoperative Hb level was as an independent prognostic factor for OS [hazard ratio (HR)=1.809; 95% confidence interval (CI)=1.073-3.050, p=0.026]. Moreover, the 3- and 5-year recurrence-free survival (RFS) rates were 16.6% and 16.6%, respectively, in the Hb-low group and 51.8% and 45.1% in the Hb-high group (p<0.001). In the multivariate analysis, the preoperative Hb status was also selected as an independent prognostic factor for RFS (HR=1.977; 95%CI=1.240-3.151 p=0.004).

Conclusion

Preoperative anemia is an independent prognostic factor in patients with esophageal cancer. Our results suggest its potential significance in the treatment and management of these patients.

Introduction

Esophageal cancer (EC) was the eighth most common cancer and sixth leading cause of cancer-related death in 2020 (1,2). Perioperative adjuvant treatment and esophagectomy are standard treatments for resectable EC (3,4). In the perioperative period, patients with EC often experience anemia (5,6). EC patients have bleeding tumors and poor oral intake due to esophageal stenosis (7,8). These symptoms easily cause iron deficiency, resulting in severe anemia. Previous studies have demonstrated that 30-60% patients with EC develop anemia during the perioperative period.

Recently, perioperative anemia was shown to have negative impacts on both short- and long-term oncological outcomes in various malignancies (9-14). In terms of short-term oncological outcomes, patients with malignancies easily experience postoperative surgical complications and adverse events during perioperative adjuvant treatment. In terms of long-term oncological outcomes, patients with malignancies have factors that negatively impact the continuation of adjuvant treatment, the response to adjuvant treatment, and survival. Considering these findings, we hypothesized that perioperative anemia in patients with resectable EC has a negative impact on both short- and long-term oncological outcomes. However, few studies have evaluated the clinical impact of perioperative anemia in patients with resectable EC. The identification of a clinical relationship between perioperative anemia and oncological outcomes would assist patients and physicians in making optimal decisions and allow anemia to be treated during the perioperative period, enabling patients with EC to receive optimal treatment for EC.

To confirm our hypothesis, we retrospectively evaluated the clinical impact of anemia during the perioperative period on both short- and long-term oncological outcomes in patients with resectable EC who received curative treatment.

Patients and Methods

Patients. We conducted a retrospective review of medical records, gathering data on patients with EC who underwent curative resection at Yokohama City University from 2005 to 2022. The inclusion criteria were as follows: histologically diagnosed primary esophageal adeno-carcinoma or squamous cell carcinoma, stage IB to III disease (according to the 7^th edition of the Tumor-Node-Metastasis classification by the Union for International Cancer Control), and successful resection (R0).

Surgery and adjuvant therapy. In subtotal esophagectomy, the standard procedure involves a right thoracotomy followed by reconstruction using a gastric tube. For middle- to lower-thoracic tumors, two-field lymph node dissection was performed, whereas three-field lymph node dissection was conducted for upper-thoracic tumors. Patients with squamous cell carcinoma received two courses of preoperative chemotherapy, consisting of 5-FU (800 mg/m², days 1-5) and CDDP (80 mg/m², day 1), admi-nistered every 3 weeks. In contrast, patients with adenocarcinoma did not receive preoperative chemotherapy.

Postoperative complications. The Clavien-Dindo classification system was used to define postoperative complications. Complications (grade 2-5) that occurred either during the hospital stay or within 30 days after surgery were identified through a retrospective review of the patients’ medical records (15).

Follow-up. Patient follow-up was conducted at outpatient clinics. Physical examinations and hematological tests, including the measurement of tumor markers (Carcinoembryonic antigen and carbohydrate antigen 19-9), were conducted every three months for five years post-surgery. Computed tomography was performed every three months for the first three years post-surgery, then every six months until five years post-surgery.

Statistical analysis. The association between the AFR and clinicopathological parameters was analyzed using the chi-square test. The Kaplan-Meier method was used to generate overall survival (OS) and recurrence-free survival (RFS) curves. A Cox proportional hazards model was used to conduct univariate and multivariate survival analyses. Statistical significance was defined as p<0.05. All statistical analyses were performed with SPSS (v27.0 J Win; IBM Corp., Armonk, NY, USA).

Ethical approval. The Institutional Review Board of Yokohama City University approved this study (IRB number: F220500064).

Results

Patient background. A total of 198 patients were included in this study. Among them, 168 patients were male and 30 were female, 104 patients were younger than 70 years of age and 94 were older than 70 years of age, and the median preoperative hemoglobin level was 12.5 g/dl (range=6.8-16.2 g/dl). According to previous studies and the 3- and 5- year overall survival rates, a hemoglobin level of 11 g/dl was determined to be the optimal cutoff value in the present study (Table I). In the present study, 34 patients had preoperative hemoglobin levels of <11 g/dl (Hb-low group) and 164 patients had preoperative hemoglobin levels of >11 g/dl (Hb-high group). When comparing patient backgrounds between the Hb-low and Hb-high groups, there were significant differences in T status. The Hb-low group had a more aggressive tumor status (≥T2) than the Hb-high group (88.2% vs. 51.8%, p=0.001). However, the percentages of patients with lymph node metastasis (47.1% vs. 46.3%, p=0.939), age >70 years (38.2% vs. 49.4%, p=0.236), and male patients (79.4% vs. 86.0%, p=0.331) were similar in the Hb-low and Hb-high groups.

Table I. Comparison of survival rates stratified by patient characteristics.

OS: Overall survival.

Survival analysis. The 3- and 5-year OS rates were 65.1% and 57.5%, respectively, in the entire cohort, 39.1% and 34.8% in the Hb-low group, and 70.3% and 62.1% in the Hb-high group (Figure 1). There were significant differences between the two groups (p=0.003). In the univariate analysis of OS, T status, lymph node metastasis status, vascular invasion, lymphatic invasion, and preoperative Hb status were selected as significant prognostic factors (Table II). In addition, the preoperative Hb status was identified as an independent prognostic factor for OS [hazard ratio (HR)=1.809; 95% confidence interval (CI)=1.073-3.050, p=0.026]. Moreover, the 3- and 5-year recurrence free survival (RFS) rates were 16.6% and 16.6%, respectively, in the Hb-low group and 51.8% and 45.1% in the Hb-high group (Figure 2). There were significant differences between the two groups (p<0.001). In the univariate analysis of RFS, T status, lymph node metastasis status, vascular invasion, lymphatic invasion, and preoperative Hb status were identified as significant prognostic factors (Table III). In addition, the preoperative Hb status was identified as an independent prognostic factor for RFS (HR=1.977; 95%CI=1.240-3.151, p=0.004).

Figure 1.

Overall survival of patients with esophageal cancer in the hemoglobin-high and hemoglobin-low groups.

Table II. Uni and multi-variate Cox proportional hazards analysis of clinicopathological factors for overall survival.

Figure 2.

Recurrence-free survival in esophageal cancer patients in the hemoglobin-high and hemoglobin-low groups.

Table III. Uni and multi-variate Cox proportional hazards analysis of clinicopathological factors for recurrence-free survival.

The perioperative clinical course in the Hb-low and Hb-high groups. When comparing the perioperative clinical course between the Hb-low and Hb-high groups, significant differences were observed in the rate of perioperative blood transfusion. The rate of perioperative blood transfusion was 52.9% in the Hb-low group and 16.4% in the Hb-high group (p<0.001). However, there were no significant differences in postoperative surgical complications (64.7% vs. 68.3%, p=0.684), pneumonia (23.5% vs. 24.4%, p=0.915), or anastomotic leakage (32.3% vs. 30.5%, p=0.830) between the two groups. Table IV shows the recurrence patterns between the Hb-low and Hb-high groups. The patterns of recurrence showed no significant differences between the Hb-low and Hb-high groups.

Table IV. Recurrence patterns in patients with esophageal cancer based on hemoglobin levels.

Discussion

This study aimed to evaluate the clinical impact of preoperative anemia in patients with EC who received curative treatment. The major finding of our study was that preoperative anemia was an independent prognostic factor for patients with EC. Our results suggest that preoperative anemia is a promising prognostic factor for the management and treatment of patients with EC.

In the present study, we found that preoperative anemia was a prognostic factor for both OS and RFS in patients with EC who received curative treatment. Previous studies have demonstrated that anemia during EC treatment is a prognostic factor. Zhang et al. evaluated the clinical impact of perioperative anemia in 79 patients with EC who received radiotherapy (16). All patients received curative radiation therapy alone, preoperative or postoperative radiotherapy, or radiochemotherapy. Radiotherapy was a total radiation dose of 66 to 72 Gy for those managed without surgery and a total radiation dose of 50 Gy for those managed with surgery. The patients were divided into anemia (n=23) and no-anemia (n=56) groups according to a cutoff value of Hb 12 g/dl for men and 11 g/dl for women. When comparing the patient background factors between the Hb-low and Hb-high groups, there were no significant differences in age, tumor stage, and daily life habits. The 2-year OS and disease-free survival (DFS) rates were 36% and 25%, respectively, in the non-anemia group and 17% and 13% in the anemia group. In the prognostic factor analysis, anemia was identified as an independent prognostic factor for both OS (HR=2.125, 95%CI=1.190-3.796, p=0.011) and DFS (HR=1.898, 95%CI=1.085-3.320, p=0.025). In addition, Zhang clarified the prognostic impact of perioperative anemia in 103 patients with EC who received primary radiotherapy (17). They divided the patients into anemic and non-anemic groups according to the cutoff value of Hb 12 g/dl for men and 11 g/dl for women. There were no significant differences between the anemic and non-anemic groups. The 3- and 5-year OS rates were 43% and 37%, respectively, in the non-anemia group and 20% and 17% in the anemia group. Moreover, the 3- and 5-year DFS rates were 37% and 26%, respectively, in the non-anemia group and 13% and 10%, respectively, in the anemia group. In the prognostic factor analysis, Hb was selected as an independent prognostic factor for both OS (HR=1.916, 95%CI=1.155-3.180, p=0.012) and DFS (HR=1.973, 95%CI=1.204-3.233, p=0.007). Accordingly, although the treatment modality was different, the Hb level during EC treatment might be a promising prognostic factor for patients with EC.

There are several possible reasons why anemia during treatment may affect long-term oncological outcomes. The most likely reason is that perioperative anemia is associated with blood transfusions during EC treatment. In the present study, we demonstrated that perioperative anemia was significantly associated with blood transfusion during treatment. Previously, blood transfusion was associated with poor survival in patients with EC (18-20). Kosumi et al. evaluated the clinical impact of blood transfusion in 260 patients with EC (18). They divided the patients into those who received blood transfusion during treatment (n=91) and those who did not (n=169). They demonstrated that patients who received blood transfusion had a significantly worse prognosis than those who did not. Blood transfusion was an independent prognostic factor (HR=2.07, 95%CI=1.09-3.91, p=0.026). Thus, perioperative anemia necessitates blood transfusion, resulting in a poor prognosis. Another possible reason is that perioperative anemia is associated with postoperative surgical complications or adverse events associated with the treatment of EC. Masaoomi et al. evaluated the clinical relationship between anemia and acute respiratory failure after esophagectomy in 6352 patients with EC (21). The overall rate of acute respiratory failure after esophagectomy was 27.08%. They demonstrated that anemia was an independent risk factor for acute respiratory failure after esophagectomy (odds ratio=1.45; 95%CI=1.20-1.75). Although we did not demonstrate a clinical relationship between preoperative anemia and postoperative surgical complications and pneumonia after esophagectomy, some studies have suggested that preoperative anemia is associated with the occurrence of postoperative surgical complications, resulting in a poor prognosis.

Study limitations. First, it is a retrospective study conducted at a single institution. Therefore, there may have been selection bias in the present study. In addition, there may have been a time bias in the present study. The present study included patients with EC who were treated between 2005 and 2022. During this period, perioperative treatment, especially adjuvant treatment, has dramatically changed. Preoperative neoadjuvant chemotherapy and postoperative immune checkpoint inhibitors have been introduced for resectable EC (22,23). Second, the present study did not evaluate the clinical impact of anemia during the preoperative adjuvant chemotherapy period. Previously, Voelter et al. evaluated anemia during preoperative adjuvant chemotherapy in 29 patients with EC (24). They demonstrated that the median decline in Hb was 2.9 g/dl. Moreover, patients who received preoperative blood transfusions had a significantly higher risk of developing postoperative complications. Therefore, it is necessary to evaluate the clinical impact of anemia due to neoadjuvant chemotherapy. Accordingly, our study needs to be validated in a larger cohort in the future.

In conclusion, preoperative anemia is an independent prognostic factor in patients with EC. Our results suggest that preoperative anemia is a promising prognostic factor for the treatment and management of patients with EC.

Conflicts of Interest

The Authors declare no conflicts of interest in association with the present study.

Authors’ Contributions

KS and TA contributed substantially to concept and design. TA, IH, YM, NK, NO, MJ, MF, RE, SY, MU, NN, KK, and NY made substantial contributions to the data acquisition, analysis, and interpretation. TA, AS, AT, and YR were involved in drafting and critically revising the manuscript for important intellectual content. TA and KS approved the final version of the manuscript.

Acknowledgements

This study was supported in part by the nonprofit organization of the Yokoyama Surgical Research Group (YSRG).