Background

Patients with cancer commonly experience anemia, with estimates suggesting as high as 90% of patients with solid tumor malignancies may be affected.^1–3 Although the causes of anemia in cancer are often multifactorial, including both cancer treatment and tumor-related factors, the presence of anemia has been demonstrated to be an independent predictor of survival among a variety of solid tumors and is associated with worse quality of life.^1–6 Given the important role that anemia may play in the survival and well-being of patients with cancer, the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Hematopoietic Growth Factors recommend evaluation for possible causes of anemia among patients with hemoglobin (Hgb) levels ≤11 g/dL.⁴ In a prior retrospective study, we demonstrated that rates of compliance with NCCN Guidelines were low among patients with gynecologic cancer, with only 33% of patients with anemia undergoing NCCN-recommended evaluation.⁷ Our objectives in this study were to determine the prevalence of anemia among a larger cohort of patients diagnosed with solid cancers at Vanderbilt University Medical Center (VUMC) and to evaluate compliance with NCCN Guidelines for the evaluation and management of anemia.

Methods

Results

Baseline Characteristics

A total of 25,018 patients met the inclusion criteria (Table 1). Median age at diagnosis was 60 years (IQR, 51–68 years), and most patients were White (88%) and of non-Hispanic ethnicity (95%). The most common malignancies were respiratory tract, prostate, and nonprostate urologic cancers (11% each). Approximately one-fourth of patients (n=6,502; 26%) presented with stage I disease.

Table 1.

Demographic and Oncologic Characteristics

View Table

Anemia Characteristics

Of the 25,018 patients, 8,695 had Hgb values available before diagnosis, with 1,484 (17%) noted to be anemic before diagnosis. A total of 11,019 (44%) were anemic within 6 months of diagnosis (Table 2). Among patients who were anemic before diagnosis, median Hgb level was 9.6 g/dL (IQR, 8.9–10.5), and most had normocytic anemia (n=1,073; 72%). Most patients were noted to have either grade 1 (n=642; 43%) or grade 2 anemia (n=616; 42%). Among patients who were anemic within 6 months of diagnosis, median Hgb level was 9.0 g/dL (IQR, 7.7–10.1). Of these patients, almost half (n=4,686; 43%) had grade 2 anemia, and most (n=9,623; 87%) had normocytic anemia. Patients with retroperitoneal/peritoneal cancers had the highest prevalence of anemia (83/110; 75%), whereas patients with prostate cancer had the lowest prevalence of anemia (171/2,841; 6%).

Table 2.

Anemia Characteristics Among Patients With Solid Tumors

View Table

Compliance With NCCN Guidelines

Among all patients found to be anemic within 6 months of diagnosis, a total of 4,125 (37%) underwent any evaluation of their anemia, of whom 1,742 (16%) had iron studies performed and 1,528 (14%) had vitamin B₁₂ or folate studies performed (Table 3). Of those with iron studies performed, 197 (11%) patients had absolute iron deficiency, 103 (6%) had possible functional iron deficiency, and 686 (39%) had functional iron deficiency based on laboratory criteria. Of those with other nutritional evaluation, 74 (5%) had vitamin B₁₂ deficiency and 69 (5%) had folate deficiency. Fewer than half of patients with anemia (n=4,318; 39%) received treatment for anemia, including blood transfusion (n=3,528; 32%), oral iron supplementation (n=1,279; 12%), or intravenous iron supplementation (n=97; 1%). Details on types of treatment by grade are listed in Table 4. Treatment of anemia was significantly more frequent as grade of anemia increased (any treatment among mild, 12%; moderate, 31%; severe, 77%; χ² [2, n=11,019]=3,020.6; P<.001). Patients with penile or testicular cancers had the highest prevalence of anemia evaluation (n=57; 79%). In contrast, patients with breast cancer had the lowest prevalence of anemia evaluation (n=175; 17%).

Table 3.

Compliance With NCCN Guidelines: Evaluation and Treatment of Anemia Among Patients With Solid Tumors

View Table

Table 4.

Treatment of Anemia Among Patients With Solid Tumors, by Grade

View Table

Discussion

The etiology of anemia in patients with cancer is frequently multifactorial, influenced by nutritional deficiency, tumor-driven chronic inflammation, treatment-driven bone marrow suppression, nephrotoxicity, and surgical blood loss.⁴ Accordingly, the NCCN Guidelines recommend evaluation of possible causes of anemia among patients with cancer to guide further management. We previously demonstrated low compliance (33%) with NCCN Guidelines for anemia evaluation among a smaller cohort of gynecologic oncology patients at our institution.⁷ In this study, we aimed to expand on our previous findings by assessing the prevalence of anemia among patients with solid tumor malignancies at our institution, and compliance with NCCN Guidelines for evaluation of anemia in patients with a Hgb level ≤11 g/dL. In this large study population (n=25,018), approximately 17% of patients with Hgb values available before diagnosis (n=8,695) were anemic before diagnosis, whereas almost half (n=11,019) had anemia within 6 months of diagnosis. Despite the high prevalence of anemia, only 37% of patients with anemia had NCCN-recommended evaluation of anemia, including iron, other nutritional, or hemolysis laboratory studies, and only 39% of patients received treatment for anemia, including blood transfusions and iron supplementation. More than half (53%) of patients who underwent anemia evaluation ultimately received any treatment. These findings are in line with those of our prior work, which showed that 36% of patients with gynecologic malignancies and anemia received any anemia evaluation and 42% received treatment for anemia. Rates of absolute and possible functional iron deficiency among patients with anemia in our cohort were overall low (11% and 6%, respectively), as were rates of other nutritional deficiency (vitamin B₁₂, 5%; folate, 5%). Estimated rates of functional iron deficiency based on laboratory criteria were higher, at 39%; however, this may be overestimated because we did not have information on treatment with erythropoietin-stimulating agents among our cohort. Regardless, this is in line with prior studies demonstrating high rates of functional iron deficiency among patients with cancer.¹⁰ At both initial presentation and within 6 months of diagnosis, most anemia was normocytic, suggesting anemia of chronic disease or treatment-induced myelosuppression rather than chronic blood loss or iron deficiency. Anemia of chronic disease itself is a multifactorial process, frequently involving functional iron deficiency, wherein inflammatory cytokines increase levels of hepcidin, which impairs normal functioning of ferroportin, a critical iron transport protein.¹¹

Our results are in line with previously published studies showing a high prevalence of anemia in oncology patients with rates ranging from 30% to 90%.^10,12,13 Contrary to previously published studies, but similar to our prior published cohort, we identified a much lower prevalence of patients with iron deficiency anemia.^5,7,13–15 This may be a result of high variability in definitions of iron deficiency among patients with cancer.¹⁵ When broader definitions of absolute iron deficiency (TSAT <20% and ferritin level <100 ng/mL) and functional iron deficiency (TSAT <20% and serum ferritin >100 ng/mL), such as those used by ESMO, were applied to patients in our cohort who had both of these iron studies available, these definitions increased the number of patients with absolute and functional iron deficiency anemia to 29% and 35%, respectively.¹⁵ Additionally, due to the small percentage of our patients having iron studies performed, we may be underreporting the true prevalence of iron deficiency anemia in this cohort. We hypothesize that the poor uptake of NCCN Guideline recommendations is likely multifactorial, reflecting lack of awareness, lack of coordination, and possibly lack of documentation. Our cohort is a heterogeneous cohort of patients who are often cared for by a combination of surgeons and medical oncologists. Additional research is needed to understand the rationale behind non–guideline-concordant care.

Strengths of our study include a large cohort of patients treated at an NCI-designated Cancer Center over a period of 10 years. Few studies have assessed compliance with NCCN Guidelines, and our study contributes to the literature showing that compliance with NCCN Guidelines for evaluation and treatment of anemia is suboptimal. Limitations of our study include the retrospective nature, as well as the use of deidentified data from the electronic medical record, which may have been incomplete. Due to the size and heterogeneity of our cohort, we did not explore associations of anemia with cancer treatment, which may influence the prevalence of anemia among patients with cancer. Moreover, we were unable to obtain data regarding vitamin B₁₂ or folate supplementation within this cohort, therefore potentially underestimating the number of patients who may have received treatment for nutrition-associated anemia. We acknowledge that LDH is a nonspecific marker and may reflect assessment of disease status, and its use as a surrogate marker of hemolysis evaluation may result in overestimation of hemolysis work-up. Additionally, we did not include direct antiglobulin testing (Coombs test) among the laboratory parameters that could signify assessment for hemolysis. We did not evaluate for the presence of bleeding diatheses or preexisting oral iron intake, which may have confounded our results. We only included patients who had Hgb laboratory tests completed at our institution; therefore, information regarding further evaluation, treatment, or care from other facilities may have been missing from our records and thereby underreported. Lastly, AJCC cancer staging has changed over time for numerous malignancies, and we abstracted stage based on the AJCC stage in use at the time of diagnosis; therefore, staging was not standardized among malignancies in which the AJCC cancer staging has changed.

Our data suggest there are opportunities to better characterize the prevalence of anemia and compliance with NCCN Guidelines among patients with cancer. NCCN Guidelines currently recommend that “any cause of anemia that is independent of cancer or chemotherapy should be treated as indicated.” Therefore, in oncology patients, we should consider the use of intravenous iron for treatment of absolute iron and functional iron deficiency (in patients receiving erythropoietin-stimulating agents), and in select patients with possible functional iron deficiency to avoid allogeneic transfusion. Although rates of absolute iron deficiency were low in our cohort when using NCCN definitions, treatment of both absolute and functional iron deficiency with intravenous iron may be an effective treatment strategy in the cancer setting. Available limited evidence suggests that intravenous administration of iron may overcome functional resistance to iron and may even be an effective prophylactic to prevent anemia among patients undergoing cancer-directed therapy.^10,16–19 We have embarked on several quality improvement initiatives to improve compliance with NCCN Guidelines based on the findings of our research. This includes increased perioperative evaluation and optimization for oncology patients undergoing surgery through utilization of our High-Risk Surgical Encounter (Hi-RISE) Clinic, and increased awareness and evaluation among patients undergoing chemotherapy. Future studies should consider large-scale prospective evaluations of anemia among oncology patients as well as implementation strategies such as electronic medical record integration for clinical decision support to optimize evaluation and treatment.

Conclusions

Anemia is common in patients with solid tumors; yet, compliance with NCCN Guidelines for evaluation and treatment of anemia remains low. Opportunities exist to improve compliance with NCCN Guidelines for management of anemia across the spectrum of cancer care.