Evaluation of anemia and iron status in Japanese patients with gynecological malignancy: retrospective cohort study

Anemia and iron deficiency (ID) are common complications frequently observed during cancer treatment, especially chemotherapy and radiotherapy [1]. Anemia is also seen in these patients at cancer diagnosis because of tumor bleeding, nutritional deficiencies, erythropoietin (EPO) deficiency due to renal disease, marrow involvement with tumor, and their chronic inflammatory state [2‐4]. These conditions can not only reduce health-related quality of life, but can also interrupt treatment [5, 6]. Anemia at diagnosis is one of the prognostic factors in cervical [7] and ovarian cancers [8]. A comprehensive understanding of anemia and iron status in patients with malignancies is critical.

ID anemia (IDA), which is the most globally prevalent type of anemia, is characterized by low ferritin, iron saturation, and increased transferrin total iron binding capacity (TIBC) levels [9]. Ferritin levels can be elevated by local or systemic inflammation, and consequently tend to correlate less with ID in patients with malignancies [2]. Additionally, increased inflammatory cytokines due to cancer or the toxicity of cancer therapy can induce dysfunction of erythropoietic activity and iron homeostasis [1]. In recent years, transferrin saturation (TSAT) has gained recognition as an important marker in patients with malignancies as it reflects functional ID, even when serum ferritin levels are elevated due to inflammation [1]. The National Comprehensive Cancer Network (NCCN) and European Society for Medical Oncology (ESMO) have published clinical practice guidelines for the management of anemia and ID evaluated with hemoglobin (Hb) levels, TSAT, and ferritin in patients with malignancies [1, 10]. Based on this background, anemia and iron status associated with cancer or cancer treatment have been investigated in several carcinomas [11, 12]. However, systemic iron metabolism and iron status vary between racial and ethnic groups, reflected by genetic variation, environmental factors, and dietary factors [13‐16]. The iron status at initial diagnosis and its changes during cancer treatments in Japanese patients with gynecological malignancies are not well understood. Accurate assessment of the iron status enables the appropriate management of anemia, thereby preventing interruptions in treatment. Moreover, it may facilitate the development of optimal therapeutic strategies and identification of novel biomarkers. This study aimed to evaluate anemia and iron status in Japanese patients with gynecological malignancies before treatment, along with its effect on Hb levels during treatment.

The present retrospective study was conducted at the University Hospital of Kyoto Prefectural University of Medicine and was approved by the institutional review board of Kyoto Prefectural University of Medicine (ERB-C-2969). Participation of patients was obtained through an opt-out methodology or written informed consent. Patients were enrolled if they were ≥ 18 years of age and diagnosed with endometrial cancer, cervical cancer, or ovarian, fallopian-tube, and primary peritoneal cancer. Patients who had other active malignancies, regardless of the treatment status, were excluded from the analysis. In addition, those with a documented history of hemorrhagic gastrointestinal disease were excluded. Patients with other complications were included in this study. Iron status of enrolled patients was evaluated based on TSAT and serum ferritin levels. Data on age, Hb level, mean corpuscular volume (MCV), primary cancer site, International Federation of Gynecology and Obstetrics (FIGO) stage, and treatment type. The difference between Hb levels before treatment and the lowest Hb levels within 6 months of administering chemotherapy or concurrent chemoradiotherapy was calculated. Blood transfusion attributed to surgical blood loss was excluded. FIGO stage was determined in accordance with the 2014 FIGO staging system for endometrial cancer, the 2018 FIGO staging system for cervical cancer, and the 2014 FIGO staging system for ovarian cancer. All data were obtained from electronic records of patients between January 2022 and July 2024. Anemia was defined as Hb levels ≤ 11.0 g/dL. ID status was divided into three categories in accordance with NCCN guidelines as follows: absolute ID (ferritin < 30 ng/mL and TSAT < 20%), functional ID (30 ng/mL ≤ ferritin ≤ 500 ng/mL and TSAT < 50%), possible functional ID (TSAT < 50% and 500 ng/mL < ferritin ≤ 800 ng/mL), and no ID (ferritin > 800 ng/mL or TSAT ≥ 50%) [10].

Patient characteristics and treatments are presented as numbers and percentages. Quantitative data were evaluated and reported as medians with 75% interquartile ranges (IQRs). Statistical comparison of risk factors was performed with the Mann–Whitney U test, Student t-test, and one-way analysis of variance (ANOVA) with significance set at 0.05. The relationships between variables were evaluated with Spearman’s correlation coefficients. Statistical analyses were performed using IBM SPSS Statistics version 27 (IBM, Armonk, NY, USA).

A total of 103 patients were enrolled, and the iron status of all patients was evaluated before treatment (Table 1). The median age (IQR) of enrolled patients was 59 (52–68) years. The primary cancer diagnoses were endometrial cancer (34.0%), cervical cancer (45.6%), and ovarian, fallopian-tube, or primary peritoneal cancer (20.4%). Forty-six patients (44.6%) were diagnosed with stage III, IV, and recurrent cancer. Overall, 40.8% and 21.4% of patients received chemotherapy and concurrent chemoradiotherapy, respectively; only four patients (3.9%) received best supportive care.

Table 2 shows patient anemia and iron status before treatment. The median Hb level was 12.5 (10.6–13.2) g/dL, but Hb levels in 30 patients (29.1%) were < 11.0 g/dL. The median MCV was 91.8 (87.2–95.2) fL, with 14 patients (13.6%) having MCV < 85 fL. The median ferritin level was 93 (41–165) ng/mL, with levels < 30 and < 100 ng/mL observed in 16.0% and 53.1% of patients, respectively. The median TSAT was 20.7% (10.4–27.2), with TSAT < 20% and < 50% seen in 46.7% and 99.0% of patients, respectively. The classification of anemia in accordance with NCCN guidelines indicated eight patients (9.9%) had absolute ID, nine (11.1%) had functional ID, and two (2.5%) had possible ID.

Figure 1 shows the Spearman’s correlation between TSAT and Hb levels or serum ferritin. TSAT positively correlated with both Hb levels (r = 0.537, p < 0.01) and serum ferritin (r = 0.287, p = 0.01) in all patients before treatment. Patients were then classified based on primary site and evaluated for factors associated with anemia (Table 3). Although one-way ANOVA found no significant differences in Hb levels, MCV, and TSAT among primary sites, there was a statistically significant difference in serum ferritin levels (p = 0.01).

Changes in Hb levels during or after treatment were evaluated in patients who received chemotherapy or concurrent chemoradiotherapy (CCRT) (Table 4), with 90.3% of patients developing anemia within 6 months of treatment. The maximum change in Hb levels after treatment was 2.5 (1.2–3.5) g/dL, and the lowest Hb level during treatment was 9.6 (8.8–10.3) g/dL. In patients without anemia at the beginning of chemotherapy or CCRT, Hb levels in 84.8% decreased to < 11 g/dL at their lowest, and the maximum change in Hb levels after treatment was 3.1 (2.6–3.9) g/dL. Blood transfusion and iron supplementation were performed in 3.0% and 6.1% of patients, respectively. Although the maximum change in Hb levels after treatment was 1.1 (0.0–2.2) g/dL in patients with Hb ≤ 11 g/dL before treatment, blood transfusion and iron supplementation were performed in 17.2% and 44.8%, respectively.

To evaluate the importance of TSAT in anemia and iron status before treatment, patients were divided into TSAT < 20 and 20 ≤ TSAT < 50 (Table 5). Among all patients, there were significant differences between groups in Hb, MCV, and serum iron levels (p < 0.01). Patients who received chemotherapy or CCRT also showed significant differences between groups in Hb levels and lowest Hb levels (p < 0.01). No significant difference was observed in the magnitude of Hb decline during treatment. When the analysis was limited to patients with Hb < 12 g/dL at baseline, these significant differences disappeared.

Approximately 30% of patients with gynecological cancers had Hb ≤ 11 g/dL, and 9.9% of patients had absolute ID before treatment. In patients with a pre-treatment Hb ≥ 12 g/dL, 84.8% developed treatment-related anemia (Hb ≤ 11 g/dL) within 6 months of starting chemotherapy or CCRT. TSAT < 20% before treatment was a predictive marker for lower Hb, MCV, and serum iron levels; patients with TSAT < 20% developed more severe anemia during treatment than those with higher TSAT. Hb levels in patients with pre-treatment anemia decreased less during treatment than those in patients without anemia due to frequent therapeutic intervention with blood transfusions and iron administration.

Iron plays a critical role in DNA synthesis and repair, genomic integrity, and erythropoiesis [17, 18]. It is also involved in oxidative stress-associated carcinogenesis through the generation of reactive oxygen species, as well as ferroptosis, a novel iron-dependent form of cell death [17, 19]. Regarding erythrocyte differentiation, the burst-forming unit–erythroid cells to colony-forming unit–erythroid cells transition during erythropoiesis is regulated by EPO, while later stages of erythroid maturation rely on iron availability [20]. Hypoxia resulting from anemia promotes EPO transcription through a HIF-2α–dependent mechanism. Erythroferrone, a hormone secreted by erythroblasts stimulated by EPO, suppresses hepcidin production in hepatocytes. Subsequently, the inhibitory effect of hepcidin on ferroportin is reduced, thereby facilitating the mobilization of intracellular iron into the circulation and enhancing systemic iron metabolism.

On the other hand, iron metabolism is altered in patients with cancer. In the tumor microenvironment (TME), which mainly consists of not only tumor cells but also stromal cells and a wide range of immune cells—including T helper cell, regulatory T cells, cytotoxic CD8 + T cells, macrophages, neutrophils, myeloid-derived suppressor cells, natural killer cells, and dendritic cells—active inflammatory signaling is constantly generated [21]. These key inflammatory cells in the TME secrete inflammatory mediators, including interferon (IFN), interleukin-1 (IL-1), IL-2, and tumor necrosis factor (TNF), thereby creating a highly active and complex network of intercellular crosstalk [21, 22]. These cytokines, such as IL-1-1 and − 6 and TNF, stimulate hepcidin synthesis, which prevents iron absorption, storage, and utilization for erythropoiesis by binding to ferroportin [2, 4, 23]; as a result, increased hepcidin activity causes low serum iron and functional ID [2, 24].

Although serum ferritin and TSAT both reflect total body iron stores under normal circumstances [9], ferritin synthesis is markedly upregulated during inflammatory conditions, obscuring true ID [23]. On the other hand, TSAT, calculated by dividing serum iron by TIBC, serves as an important biochemical marker reflecting iron status, as it takes into account both serum iron and its main transport protein, transferrin [25]. In the present study, the serum ferritin levels at diagnosis were significantly elevated in patients with ovarian cancer. Higher serum ferritin levels have been associated with the stage, tumor grade, and tumor size in patients with cancer [26]. Moreover, advanced ovarian cancer frequently presents with peritoneal carcinomatosis and a chronic inflammatory state within the TME [27]. Therefore, this finding may be attributable to the higher proportion of advanced-stage cases in patients with ovarian cancer than those with endometrial or cervical cancer.

A previous report showed that 20% of patients with gynecological cancer had anemia at the time of cancer diagnosis, and 67% developed anemia within 6 months of diagnosis [11]. Other studies demonstrated that 74%–90% of patients with gynecological malignancies receiving chemotherapy developed chemotherapy-induced anemia (CIA) [6, 28]; the incidence and severity of treatment-related anemia depended on the type of cancer and chemotherapy regimen [29]. We found that 29.1% of Japanese patients with gynecological cancers had anemia before cancer treatment, and > 90% of patients developed anemia within 6 months of starting chemotherapy or CCRT. This might be due to the low frequency of intervention of therapeutic administration for anemia in this study, compared with a previous report where frequent blood transfusions were performed [11]. In the present study, all cases that received transfusions as therapeutic interventions had Hb levels < 8.0 g/dL. Furthermore, all cases had a TSAT of less than 20% prior to treatment. In addition, although anemia treatment type and its effects on CIA were not evaluated, the present study also demonstrated that therapeutic interventions could prevent Hb decline, even if Hb was < 12 g/dL before cancer treatment.

We also evaluated the clinical importance of TSAT for treatment-related anemia. Patients were divided into two groups based on a TSAT cutoff following the NCCN guidelines for anemia. Patients with TSAT < 20 had significantly lower Hb levels and lowest Hb values than those with higher TSAT among the patients receiving chemotherapy or CCRT. Focusing on patients with Hb < 12 g/dL before cancer treatment, significant differences in Hb levels, lowest Hb, and decrease in Hb were not found, which was attributed to the intervention for anemia in patients with TSAT < 20%. These results indicate that TSAT could be a predictive marker for anemia and iron status, and that treatment-related anemia could be prevented through appropriate treatment based on TSAT.

We also evaluated whether MCV could be an indicator of iron status in patients with gynecological malignancies. Low MCV is a crucial factor of microcytic hypochromic anemia, of which IDA is one common type [30]. Although MCV is a strong indicator of cancer, its effectiveness varies among ethnic groups [31, 32], suggesting ethnic differences in patients with cancer [15]. Therefore, it is important to clarify the relationship between MCV and iron status in Japanese patients with gynecological malignancies. In the present study, only 13.6% of patients had MCV < 85 fL before treatment, and no significant differences in MCV values were observed between patients with different primary sites. When patients were classified using TSAT, MCV was significantly lower in patients with TSAT < 20 than in those with TSAT ≥ 20, indicating that it could be a biomarker of iron status in Japanese patients with gynecological malignancies.

Clinical guidelines for ID have been published by several committees and academic organizations, including the NCCN and ESMO. However, these guidelines provided recommendations for therapeutic strategies for anemia with various ferritin level and TSAT criteria [33]; whereas intravenous or oral iron supplementation are recommended in the NCCN guidelines for patients with Hb < 11 g/dL or a > 2 g/dL decrease from baseline and absolute ID, the ESMO guidelines recommend only intravenous iron supplementation for patients with Hb 10–11 g/dL and TSAT < 20% or serum ferritin < 100 ng/mL. Although the ESMO guidelines clearly state the iron dosage, the NCCN guidelines do not, indicating that the ESMO guidelines may provide more clinically relevant information.

The present study had some limitations. It was a single-center retrospective study with few cases. Accordingly, performing stratified or fully adjusted analyses by FIGO stage or baseline laboratory data was not feasible. Additionally, it did not evaluate the efficacy of iron supplementation on the prevention of treatment-related anemia and continuation of cancer treatment. Interventions addressing treatment-related anemia may have been affected by selection bias, potentially impacting the outcomes observed in this study. Further study should be conducted to emphasize our results and their effect on the oncological outcomes. A retrospective study revealed that compliance with the NCCN guidelines for evaluation and treatment of anemia was low in patients with gynecological cancers [11]. In the present study, the treatment decision for anemia was made by the attending physicians; as such, the efficacy of therapeutic interventions for treatment-related anemia was not evaluated, and it remains unclear which guidelines are most effective for treatment-related anemia. However, to prevent treatment interruption due to anemia, clinicians need to understand its management during cancer treatment and provide appropriate support for patients in compliance with guidelines. To clarify guideline efficacy on treatment-related anemia, a randomized controlled trial that is prospectively designed to evaluate the safety and efficacy of the blood management program in patients with gynecological cancer has been initiated [34].

Anemia and iron deficiency were common in Japanese patients with gynecological cancers, and low TSAT (< 20%) before treatment was associated with more severe anemia during therapy. TSAT may serve a useful biomarker for predicting and managing treatment-related anemia. Early intervention based on TSAT levels could help maintain Hb levels and prevent treatment interruption. Further studies are needed to standardize blood management for patients with gynecological cancers.