Mortality and VTE Associated with ESA Administration to Anemic Cancer Patients

Erythropoietin and darbepoetin were FDA-approved in 1993 and 2002, respectively, to treat chemotherapy-associated anemia in patients with non-myeloid malignancies. At that time of approval for erythropoietin, concerns were raised about venous thromboembolism (VTE) and tumor progression that might occur because of ESA treatment in these patients. Expression of erythropoietin and erythropoietin receptors has been shown in several human cancers. In certain cancer types, in vitro studies have shown that activation of erythropoietin/erythropoietin receptors leads to proliferation, antiapoptosis, and invasion. While systematic studies have confirmed VTE risks in humans, studies have not identified an association between treatment with ESAs and increased mortality risk.

The study described below evaluates VTE and mortality rates of ESA use in phase 3 trials comparing ESAs with placebo or standard of care treatment to treat chemotherapy-associated anemia in cancer patients. Data were extracted from trials reviewed in the Cochrane Collaboration (January 1, 1985 - April 1, 2005) and MEDLINE and EMBASE databases (April 1, 2005 - January 17, 2008).

Mortality risk was evaluated for 13,611 patients with cancer from 51 phase 3 trials and VTE risk was evaluated for 8,172 patients with cancer from 38 phase 3 trials. Trials differed with respect to study drug, patient numbers, treatment duration, concomitant treatments, and cancer diagnoses.

Once again, a significantly increased risk of VTE was found in patients treated with ESA (334 events / 4,610 patients) vs. control patients (173 events / 3562 control patients) (RR, 1.57; 95% CI, 1.31-1.87). Overall, the risk of mortality was also greater in ESA treated patients (HR 1.10; 95% CI 1.01-1.20). Although there was a trend towards increased mortality in each subset (ESA treatment in patients with the anemia of cancer or those with chemotherapy-induced anemia), in neither group did this reach a level of statistical significance (anemia of cancer; HR, 1.29; 95% CI, 1.00-1.67;P=.05), (chemotherapy-associated anemia; HR, 1.09;95%CI, 0.99-1.19).

Please reference the source article:
Venous thromboembolism and mortality associated with recombinant erythropoietin and darbepoetin administration for the treatment of cancer-associated anemia. Bennett CL, Silver SM, Djulbegovic B, Samaras AT, Blau CA, Gleason KJ, Barnato SE, Elverman KM, Courtney DM, McKoy JM, Edwards BJ, Tigue CC, Raisch DW, Yarnold PR, Dorr DA, Kuzel TM, Tallman MS, Trifilio SM, West DP, Lai SY, Henke M. JAMA. 2008 Feb 27;299(8):914-24.

NAAC Expert Commentary:
Although the explanation for higher complications (including mortality) among cancer patients treated with ESAs remains unclear, the accumulated data, including that from this report warrant both strict compliance to newly established guidelines and additional study.

One concern regarding prior meta-analyses is that their findings are based primarily on aggregated data from a variety of different clinical situations. Data is commonly obtained from divergent studies that could involve chemotherapy, radiation therapy, or no treatment, and the use of various ESAs with different doses and target hemoglobin levels.

The current report takes a step forward from prior studies, since the investigators were able to discriminate between patients receiving ESAs for cancer-associated anemia and patients receiving ESAs for chemotherapy-associated anemia. The distinction is important in light of the analysis, which identified a trend toward higher mortality in both subsets; although a trend was observed, statistical significance was not reached for either subset.

The question remains as to whether strict compliance to current recommendations regarding ESA use will be associated with increased VTE and mortality. This question will be addressed in ongoing prospective randomized trials. Although the increased risk of VTE is of important consequence, it is unlikely to explain the higher rate of mortality experienced by ESA-treated cancer patients. One explanation might be that many cancers express erythropoietin receptors, which might be stimulated towards proliferation by exogenous ESA. As intuitive as this may seem, the data, including in vitro and animal studies, has been far from conclusive.

Thus, a number of unresolved issues remain. Yet, most clinicians would agree that the existing quality of life data, which is quite favorable with regard to ESA use in anemia that is associated with cancer treatment, must be balanced with those concerns raised by analyses that demonstrate increased risks, including mortality. For the short term, oncologists must rely on their clinical judgment as to which patients with anemia should be treated, and by what means (i.e., transfusion, ESA). Adherence to current guidelines, such as those forwarded by ASH/ASCO or the EORTC, is highly recommended.

Prevalence of Fibromyalgia in Iron Deficiency Anemia and Thalassemia Minor

Fibromyalgia (FM) is a chronic pain condition that shares many symptoms with iron deficiency anemia (IDA) and thalassemia minor (TM). According to criteria established by the American College of Rheumatology, FM is a chronic pain condition characterized by a history of widespread pain in all four body quadrants and the presence of 11 of 18 tender points on physical examination. The prevalence of FM in developed nations varies between 0.5 - 4%, and the disease has an increased prevalence in patients with rheumatic diseases like rheumatoid arthritis (RA), lupus erythematosus, Sjogren’s syndrome, osteoarthritis, and Behcet’s disease; the prevalence of the disease is increased in chronic diseases like diabetes mellitus.

Prior to the publication of this study, the prevalence of FM in patients with IDA or patients with TM was not known. The study described below was conducted to (1) determine the prevalences of FM in patients with IDA or TM, (2) identify any factors associated with the presence of FM, and (3) determine the prevalence of IDA in FM and evaluate IDA’s affects on the clinical findings of FM.

The study population consisted of 205 consecutive patients with IDA and 40 patients with TM who were admitted to the Internal Medicine Outpatient Clinic located in Edirne, Turkey. Also included were 196 patients with FM diagnosed at the Rheumatology Outpatient Clinic. The control group consisted of 100 hospital employees (89 females and 11 males). All patients and healthy subjects were obtained for the study during the period of 2003 to 2005.

Fibromyalgia was identified using the ACR 1990 diagnosis criteria. Blood samples were taken for various biochemical factors identifying IDA and TM. IDA was defined as a hemoglobin (Hb) level of less than 13.5 g/dl in men (normal male adult 13.5–17.5 g/dl) or less than 12 g/dl in women (normal female adult 12.0–15.5 g/dl) with serum ferritin less than12 ng/ml, low serum iron, raised total iron binding capacity, and transferrin saturation less than 16%. Interviews were conducted to obtain demographic information. Function and pain parameters associated with FM were determined using a wall analog scale and the FM Impact Questionnaire score.

Fibromyalgia was diagnosed in 36 patients (all females) with IDA (17.6%), 8 patients (seven females, one male) with TM (20%) and 6 patients (all females) in the control group (6%).The prevalences of FM in patients with IDA compared to patients with TM was significantly higher than in the control group ( p=0.006 and p=0.025, respectively). The ratio of females in the IDA group was higher than that in the TM group (22.5% vs. 9.3%, p=0.027). There was no difference in Hb and iron parameters between IDA patients with or without FM. Fibromyalgia diagnosed in patients with IDA was associated with female sex, marriage, and history of pica (p<0.05). The results of this study indicate that FM has an increased frequency in patients with IDA and patients with TM. In patients with IDA and TM, nonspecific symptoms may not be explained by only anemia, and FM may be present.

Please reference the source article:
An increased prevalence of fibromyalgia in iron deficiency anemia and thalassemia minor and associated factors. Pamuk GE, Pamuk ON, Set T, Harmandar O, Yesil N. Clin Rheumatol. 2008 Apr 11.

NAAC Expert Commentary:
Fibromyalgia (FM) is a very common chronic pain syndrome. The most current hypothesis for its pathogenesis is that FM patients have central sensitization, a situation in which their processing of afferent sensory input results in these patients experiencing greater distress than normal individuals. Functional MRI scanning during painful stimuli has demonstrated that FM patients display increased blood flow to different parts of the brain compared to normal controls. Other central sensitization syndromes include low back pain, temporo-mandibular pain, and muscle contraction headaches. Fibromyalgia occurs as a primary syndrome, and commonly accompanies systemic disorders such as rheumatoid arthritis and lupus.

The article by Pamuk et al from Turkey evaluates the prevalence of FM in iron deficiency anemia (IDA) and thalassemia minor (TM). This analysis is difficult, given that the diagnosis of FM requires the absence of comorbid conditions that might account for the symptoms attributed to FM (such as fatigue, diffuse pain, etc.). Furthermore, it is hard to imagine how the presence of anemia would affect central processing of pain signals, although there might be a connection which is unrecognized.

Physicians must avoid the trap of attributing symptoms of diffuse pain and fatigue to FM, when there is an unrecognized medical problem present, such as anemia, malignancy, congestive heart failure, etc. Pamuk’s association of FM with IDA and TM probably falls under the category of “True and Unrelated.”

Effects of Oral Iron Treatment in Children with ADHD

Significantly lower serum ferritin levels have been observed in children with attention deficit hyperactivity disorder (ADHD). In fact, in an earlier study conducted by the authors of the study described below, 84% of children n with ADHD had serum ferritin levels < 30 ng/mL, compared with 18% of controls (P < 0.001).

Although the pathophysiology of ADHD is complex and not completely understood, several lines of evidence suggest that ADHD is the result of an imbalance in the dopaminergic and noradrenergic systems. Since iron is an important modulator of these systems, it is believed that subnormal brain iron stores may negatively influence these systems, and contribute to the development of ADHD. All children in the previous study had normal hemoglobin (Hb) levels, suggesting that ADHD was more likely to be caused by low serum ferritin levels, compared to anemia. However, the study was not designed to infer causality between iron deficiency and ADHD. In order to provide strong evidence of causality, children with ADHD would need to demonstrate a clear benefit from iron supplementation. Therefore, the following double-blind, placebo-controlled, randomized pilot study tested the effects of iron supplementation on iron deficient, nonanemic children with ADHD.

The study population consisted of outpatient children with ADHD (confirmed by DSM-IV criteria) aged 5-8 years old who had normal Hb levels and serum ferritin levels < 30 ng/mL (the definition of iron deficiency in this study). Iron status was based on a variety of biochemical measurements, in particular - the normalization of serum ferritin levels in these children. Patients with ADHD were randomly assigned (3:1) to receive 80 mg ferrous sulfate tablets for a period of 12 weeks. At baseline and week 12, children were tested using the Conners’ Parent Rating Scale (primary endpoint), Conner's Teachers Rating Scale, Attention Deficit Hyperactive Disorder rating scale, DSM-IV criteria, and the Clinical Global Impression Severity-Scale. Patients were also tested for restless legs syndrome using the International Restless Legs Syndrome Study Group criteria for children.

Twenty three children (18 boys and 5 girls) met the inclusion criteria and were randomly assigned to receive oral iron therapy (n=18) or placebo (n = 5). Two patients discontinued iron therapy due to constipation and one patient was lost to follow-up.

Serum ferritin levels increased in the iron treatment group, from 29.1 +/- 17.6 ng/mL at baseline to 55.7 +/- 20.4 ng/mL at week 12 (P = 0.000). In contrast, the iron status of children in the placebo group did not improve. A progressive significant decrease in the ADHD Rating Scale was observed in children after 12 weeks of iron therapy (-11.0 +/- 13.9; P < 0.008), compared with children receiving placebo (3.0 +/- 5.7; P = 0.308). However, children receiving iron therapy did not show improvement that reached statistical significance as measured by the Conners’ Parent Rating Scale (P = 0.055) and Conners’ Teacher Rating Scale (P = 0.076). Symptoms of restless leg syndrome were improved in children receiving iron therapy compared to placebo. However, no correlation was found between serum ferritin level and occurrence of restless leg symptoms. A major limitation of this study, according to the authors, was the small sample size. However, the authors concluded that the results provide the rationale for larger multisite studies.

Please reference the source article:
Effects of iron supplementation on attention deficit hyperactivity disorder in children. Konofal E, Lecendreux M, Deron J, Marchand M, Cortese S, Zaïm M, Mouren MC, Arnulf I. Pediatr Neurol. 2008 Jan;38(1):20-6.

NAAC Expert Commentary:
Although the pathophysiology of ADHD is complex and not completely understood, there is increasing evidence that brain iron stores influence the monoamine-dependent functions that are altered in ADHD. Previously, low serum ferritin levels have indicated decreased iron stores even before the development of anemia. In addition, a relationship between iron deficiency, cognitive impairment and motor instability has been recognized in children even in the absence of anemia.(1)

This study presents evidence that iron deficiency may contribute to the overall pathophysiology of ADHD, even though a causal link has not been established. Important to consider is that the use of the CPRS total score as the primary outcome may be problematic because non-ADHD factors could mask or dilute the effect of iron supplementation. As commented on by the authors, this was only a small pilot study, and a larger multisite trial will be required before changes in clinical practice can be recommended. Future studies should address normal serum ferritin values and specific iron doses, especially because the long-term effects of iron therapy in this pediatric population is not known.

Normal serum ferritin values vary by age in young children, and indicate total body iron stores in the absence of liver disease, chronic inflammation or infection. Children with chronic medical conditions including malnutrition were excluded from the current study, but other possible cofounders of serum ferritin were not addressed.

When caring for young children it is also important to assess dietary history and screen for iron deficiency in those patients at risk, including:

• toddlers and adolescent females
• young children poorly absorbing or not consuming iron-rich foods
• infants born prematurely
• children who consume excessive cow’s milk
• adolescents with poor diet, rapid growth or menstrual blood loss

References in the commentary:
(1) Attention-deficit/hyperactivity disorder: a selective overview. Biederman J. Biol Psychiatry. 2005 Jun 1;57(11):1215-20.