Editor’s note: This is the first part in a three part series. Part two covers trends in iron therapy, and part three will cover an emerging class of agents for anemia therapy.


Erythropoietic stimulating agents have revolutionized the treatment of anemia in patients with chronic kidney disease and end-stage renal disease. ESAs have become such an integral part of the treatment of patients with ESRD that it is almost impossible to envision nephrology practice without them. This series will review the history of ESAs in nephrology practice, their contribution to improving patient care, and what lies ahead in the development of new treatments for anemia.

The pre-ESA era

Over half the nephrologists currently in practice had not yet completed their training in 1989 when recombinant human erythropoietin (rHuEPO) was approved by the U.S. Food and Drug Administration. These nephrologists do not remember dialysis patients who had characteristic pale yellow skin that was due to a combination of severe anemia and uremic pigment deposition. Such patients were so fatigued they could not carry out the activities of daily living and were dependent on monthly transfusions and virilizing androgen therapy—both variably effective—to raise their hematocrit (Hct) levels from the low to mid 20s.

Read also
Trends in iron therapy: Emerging treatments for anemia in patients with CKD, ESRD, Part 2

Although the FDA has removed all quality of life claims from the ESA product information, those practitioners who witnessed the dramatic improvement in patient well-being following the introduction of ESA therapy know the FDA action is misleading. For the patients who benefited, ESAs have been a godsend and, although ESAs may not have lived up to their unrealistic promise of being a panacea, it is unlikely that many informed patients would abandon ESA therapy and return to the life of their pre-1989 counterparts.

Randomized-controlled trials

Results from a number of observational trials published in the 1990s suggested that higher Hcts (>36%) or hemoglobin (Hb) levels (>12 g/dL) were associated with improved outcomes in dialysis patients. The NKF Dialysis Outcomes Quality Initiative (DOQI) anemia guidelines released in 1997 recommended a target Hct level of 33-36% and the NKF Kidney Disease Outcomes Quality Inititative (KDOQI) anemia guidelines released in 2001 recommended a target Hb level of 11-12 g/dL in anemic patients with CKD. By the mid-2000s, the mean Hb level among hemodialysis patients in the US was around 12 g/dL, despite the FDA product information for ESAs recommending a target Hb in the 10-12 g/dL range.

There was concern regarding the cost of ESA overuse, and the Centers for Medicare & Medicaid Services instituted a policy whereby ESA reimbursement would be withheld if the 3-month rolling average Hct exceeded a certain level in dialysis patients.

Additional concerns regarding the safety of ESAs were raised by the findings from a number of randomized clinical trials (RCTs), including:

The Normal Hematocrit Cardiovascular Trial, published in 1998, involving 51 sites in the US examining whether further benefit could be gained from normalizing Hct in 1,265 adult dialysis patients versus a target Hct of 30%. The study group involved patients with high cardiac risk, i.e., patients with ischemic heart disease or congestive heart failure, but the Monitoring Committee terminated the study early when the hemodialysis patients treated with rHuEPO to the Hct target of 42% were found to have a 1.3 relative risk (CI 0.9-1.9) for death and non-fatal myocardial infarction compared to those in the control arm with the target Hct of 30%.

The Correction of Hemoglobin Outcomes in Renal Insufficiency (CHOIR) study, published in 2006, demonstrated composite cardiovascular end-points (including mortality) to be significantly higher among non-dialysis CKD patients with target Hb of 13.5 g/dL treated with rHuEPO compared to those with target Hb 11.3 g/dL treated with rHuEPO.

The Trial to Reduce Cardiovascular Events with Aranesp Therapy (TREAT) study, published in 2009, examined type 2 diabetes anemic CKD patients treated with darbepoetin, a longer-acting ESA. Patients were randomized to a target Hb of 13 g/dL or a placebo control arm with rescue darbepoetin therapy if the Hb fell below 9 g/dL.

Patients in the high Hb target arm had no differences in cardiovascular end-points except for a higher incidence of stroke and a lower incidence of coronary revascularization procedures. However, there was an increased rate of cancer deaths in patients with prior cancer who were randomized to the high target Hb arm. Following the publication of the TREAT study, the FDA revised the product information for rHuEPO and darbepoetin to emphasize the adverse events noted in the RCTs, and to recommend a target Hb for ESA therapy that is the lowest possible to avoid transfusions. The FDA also removed all quality of life claims for ESA therapy from the product information.

It should be noted that the stratification of patients and analysis of primary outcomes in the RCTs was by target Hb level, not by ESA dose, so any link between higher ESA doses and adverse outcomes is an association, not cause and effect. It should also be noted that patients who achieved higher target Hb levels with small ESA doses had better outcomes than those who required large ESA doses to achieve lower target Hb, further strengthening the link between ESA doses and adverse outcomes.

Nonetheless, in 2017, ESAs remain the only class of agents to treat anemia in CKD and ESRD patients who are not iron deficient. During the 1990s and 2000s the liberal use of intravenous (IV) iron was promoted as a necessary therapy for treatment of anemia in hemodialysis patients following the recognition of ongoing iron losses and the phenomenon of functional iron deficiency induced by accelerated red blood cell (RBC) production from ESA therapy. Iron therapy has also come under increased scrutiny for possible adverse outcomes, so there exists an unmet need for effective but safer agents to stimulate RBC production and provide iron so that the supply of iron to the erythroid marrow meets demand.

Trends in ESA therapy

Modified epoetin

Branded rHuEPO (epoetin alfa, Procrit, Epogen) and darbepoetin (Aranesp) remain the mainstays of ESA therapy in the US for patients with non-dialysis CKD. Epoetin is generally administered every 1-2 weeks in non-hemodialysis (HD) patients and darbepoetin is generally administered every 2-4 weeks in non-HD patients. The choice of ESA in non-dialysis CKD patients is usually driven by the insurer based on contracting, tiering, and pricing. For patient convenience, darbepoetin is often preferred due to its less frequent dosing.

In the dialysis market, aggressive pricing by Amgen for darbepoetin has increased its market share from 5.1% in Aug. 2010 to 14.7% in Oct. 2016 (see DOPPS data below).


Pegylated epoetin beta (Mircera) is a longer acting ESA which is effective when administered once monthly. It was approved by the FDA in 2008, but its use in the US was delayed until 2015 by patent infringement issues. During that interim period, it was used in Europe where the patent infringement issues did not apply. Based on a favorable European experience, Fresenius began a pilot program to use pegylated epoetin beta in its US dialysis facilities in 2015, and the use was expanded over the following 18 months. By Oct. 2016 the dialysis market share of pegylated epoetin beta was 35.9% in the US, exclusively due to its use in Fresenius facilities.

In November 2016, the chief medical officer (CMO) of Fresenius issued a memo that pegylated epoetin beta would be the preferred ESA within Fresenius facilities and that epoetin alfa and darbepoetin would be placed in a restricted formulary category. The CMO cited the excellent clinical outcomes and safety record of pegylated epoetin beta, the reduced staff time devoted to ESA administration, and the significant reductions in cost as the basis for this decision. As of that time, pegylated epoetin beta was not being marketed in the US outside of Fresenius dialysis facilities.

In late 2016, Roche Pharma-ceuticals, the manufacturer of pegylated epoetin beta, entered an agreement with Vifor (the maker of iron sucrose [Venofer] and sucroferric oxyhydride [Velphoro]) to market pegylated epoetin beta in the US outside of Fresenius dialysis facilities. It is worth noting that DaVita Kidney Care recently signed a 6-year supply agreement with Amgen to provide a minimum of 90% of its ESA requirements, replacing prior agreements that were to expire in 2018. Anticipated competition from lower-priced biosimilar ESAs has likely accelerated the signing of such agreements.

Biosimilar epoetin

Biosimilar agents are non-identical “copies” of pharmacologic agents that are derived from living cells, which are termed biologic agents.  rHuEPO and its derivatives darbepoetin and pegylated epoetin beta are biologic agents. Like other pharmacologic proteins such as polypeptide hormones (e.g., insulin, human growth factor), monoclonal antibodies (e.g., infliximab, adalimumab), growth factors (e.g., filgrastim) and some vaccines, rHuEPO is made by transfecting a non-human cell with a human gene for the human target protein. In the case of rHuEPO, the cell is a Chinese hamster ovary cell. The pharmacologic protein produced has an inherent heterogeneity in its higher order structure and post-translation modifications (e.g. carbohydrate side-chains). Therefore, it is impossible to produce an exact copy of a substance which isn’t identical to itself, unlike the production of identical generic copies of small molecule drugs.

The FDA defines biosimilar agents as being “highly similar to the US licensed biological product (also referred to as the “reference” or “originator” product) notwithstanding minor differences in clinically active components.” There are no clinically meaningful differences between the biologic product and the reference product in terms of the safety, purity, and potency of the product. Biosimilar ESAs have been in use in Europe since 2007 with a favorable experience in terms of comparable safety and efficacy to reference ESAs. More robust patent protection laws in the US have prevented biosimilar ESAs from entering the US market until 2015.

There are two biosimilar ESAs from Pfizer and Sandoz that have undergone clinical trials in the US and are in the process of regulatory approval. A concern regarding biologic products in general and biosimilar products in particular is whether the manufacturing, packaging, and distribution process is sufficiently rigid to avoid alteration or denaturation of the protein and render it antigenic.

In the case of ESAs, the production by the recipient of an antibody against the drug could lead to cross-reaction with native erythropoietin and the development of pure red cell aplasia (PRCA). This has been a problem in non-industrialized countries where the drug approval and monitoring process is not as rigid as that in the US and EU, so poor quality ESAs appear on the market that may induce PRCA.


The few clusters of PRCA that have occurred in the EU have been aggressively pursued and their root causes eliminated quickly. Nonetheless, it is possible that concern regarding PRCA may retard the uptake of biosimilar ESAs despite their lower cost vs. the reference agent (epoetin alfa). Since the FDA does not require biosimilar agents to reproduce the extensive clinical trials of the originator agents, bringing a biosimilar agent to the market can be done at a fraction of the cost of bringing an originator biologic to the market. A number of issues related to the use of biosimilar therapeutics into nephrology practice were addressed in a report based on a scientific workshop sponsored by the National Kidney Foundation.1

ESA dosing trends

The use of ESAs in ESRD patients in the US declined 38% between 2007 and 2012 (per the US Government Accounting Office) and it has declined around another 10% between 2012 and 2016, according to DOPPS data. The major driving forces for this reduction have been

  • the concerns regarding the impact of high ESA doses, per the RCTs
  • the FDA’s change in ESA labeling in 2011

instituting a bundled payment for dialysis treatment in the US in 2011, which changed ESAs from a profit center to a cost center for providers. The mean Hb level in the DOPPS sample decreased from 11.3 g/dL in Aug. 2010 to 10.7 in Jan. 2013, but has remained remarkably stable through Oct. 2016. There has been a small increase in the transfusion rate among ESRD patients with the shift of the Hb distribution curve to the left.

The onset of the bundled payment provided a natural experiment to determine if the decrease in Hb levels and ESA doses are associated with a decrease in the major adverse cardiovascular events (MACE) attributed to ESA therapy in the RCTs. A 2016 report in the Journal of the American Society of Nephrology comparing various hard outcomes among dialysis patients in 2011 (prior to the effects of the bundled payment and a change in ESA labeling) and 2012 (after the effects of the bundled payment and the ESA labeling change would be expected to manifest) noted that all-cause mortality, cardiovascular mortality, and myocardial infarction in both years to be at the expected rates.

However, in 2012 there were lower-than-expected rates of venous thromboembolism, stroke and heart failure, although non-ESA related changes in practice may have confounded the results.2  A 2016 report in JAMA Internal Medicine compared a pre-bundling cohort (Jan. 2008-Dec. 2009) with a matched post-bundling cohort (July 2011-June 2013), all patients 66 years or older. With about a 40% reduction in mean ESA dose, the post-bundling cohort demonstrated a 10% increase in transfusions (HR 1.09, 95% CI 1.07-1.12, p<0.001) but no significant increase in the risk of MACE, death, hospitalized congestive heart failure, or venous thromboembolism. However, the risk of stroke decreased in the post-bundling cohort (HR 0.77, 95% CI 0.64-0.93, p<.01) and black patients in the post-bundling cohort had a significant reduction in MACE (HR 0.82, 95% CI 0.73-0.92, p<.001) and all cause mortality (HR 0.82, 95% CI 0.73-0-93, p<.002) as well.3


  1. Wish JB, Charytan CC, Chertow GM, Kalantar-Zadeh K, Kliger AS, Rubin RJ, et al. Introduction of biosimilar therapeutics into nephrology practice in the United States: report of a scientific workshop sponsored by the National Kidney Foundation. Am J Kidney Dis. 2016 Dec:68(6):843-852.
  2. Chertow GH, Liu J, Monda KL, Glibertson DT, Brookhart MA, Beaubrun AC, et al. Epoetin alfa and outcomes in dialysis amid regulatory and payment reform. J Am Soc Nephrol. 2016 Oct:27(10):3129-3138.
  3. Wang C, Kane R, Levenson M, Kelman J, Wernecke M, Lee J-Y, et al. Association between changes in CMS reimbursement policy and drug labels for erythrocyte-stimulating agents with outcomes for older patients undergoing maintenance hemodialysis covered by fee-for-service Medicare. JAMA Intern Med. 2016 Dec:176(12):1818-1825.