Recombinant human erythropoietin (rHuEPO) was developed in the 1980s and was noted to have a surprisingly high efficacy in correcting anemia of CKD [13] (Table 5.1). The extraordinary success of rHuEPO led to the spawning of several derivatives of the original molecule that can be collectively called epoetins or erythropoiesis-stimulating agents (ESAs). rHuEPO and its derivatives are glycosylated proteins and hence are produced from mammalian cell lines and in vivo action is dependent on the degree and nature of glycosylation. The epoetins α and β were the first molecules manufactured and became widely used after supportive evidence from multiple clinical trials. The typical dosing schedule of rHuEPO in routine clinical practice is three times a week, with the medication being administered along with the HD treatment, typically by the dialysis nurse. This schedule also assures a close to perfect adherence with the prescribed regimen, enhancing the effectiveness of the therapy.

Intravenous (IV) administration continues to be the preferred route for rHuEPO in patients receiving maintenance HD, even though the bioavailability of the IV route is inferior compared to the subcutaneous (SQ) route [14]. The half-life of rHuEPO in vivo is only about 4–11 h [15]. SQ administration leads to a more sustained increase in serum rHuEPO levels; the bioavailability is 20–25% and measurable rHuEPO levels can be detected even 4 days after administration [15, 16]. The SQ dose is also ~25% lower for the same clinical effect compared to the IV dose, thus providing cost benefits [17, 18]. One complication possibly linked to SQ use of certain products is pure red cell aplasia (PRCA) [19]. As a result of this possibly idiosyncratic reaction, the SQ route has become less popular in spite of its multiple other benefits.

Following the successful adoption of rHuEPO, there have been several attempts to pharmacologically manipulate its structure in order to achieve a more favorable therapeutic profile. Darbepoetin-α was the first such agent introduced; due to enhanced glycosylation darbepoetin has a larger size and a much longer half-life compared to rHuEPO [20, 21]. In the case of darbepoetin the route of administration (IV vs. SQ) does not seem to influence efficacy, and less frequent dosing makes it ideal for certain patients such as those on home dialysis therapy. Adoption has been slow due to costs but has been favored once incorporated into center-specific protocols. A conversion factor of 200 units rHuEPO to 1 μg of darbepoetin is typically used for initial dosing. Another newer agent developed to further increase the half-life of rHuEPO was the continuous EPO receptor activator (CERA) , created by binding EPO to polyethylene glycol. CERA has a half-life of 5 days irrespective of route of administration [22, 23]. Studies have proven the non-inferiority of CERA [24], which has been used for many years now in other countries, and also recently started being adopted by dialysis providers in the US for anemia management. Introduction of CERA (Mircera^®, Roche) into the market was supported by its non-inferiority vs. darbepoetin in a randomized trial in CKD patients [25]. With a very long half-life and extended administration intervals, CERA offered a convenient alternative in non-dialysis CKD [26] and peritoneal dialysis patients whose contact with the dialysis providers is not on a daily basis [27]. IV CERA was pitted against darbepoetin in dialysis patients in a randomized trial and was found to be clinically as effective in maintaining Hgb levels in the target range as darbepoetin [28], thus paving the way for its mainstream use. Though cost-effectiveness is still not proven [29], CERA is a promising addition to the quiver of agents that are currently at our disposal.

Medicines that mimic biologic molecules but are produced artificially, that are clinically similar in action but having minor differences in clinically inactive metabolites are called biosimilars. Biosimilar peptides have been around for several years now and have been successful when used in many different diseases and conditions. A wide variety of drugs are available in this class including hormones, vaccines, growth factors, monoclonal antibody-based agents, etc. In the strictest sense, the advent of biosimilars in the management of anemia occurred with the introduction of darbepoetin. That said, the term “biosimilar” is currently applied to the generic formulations of the patented molecules epoetins-α and β in the US. In Europe, the first few biosimilars licensed were generic manipulations of epoetins-α and ζ (zeta) after their patent expirations [30]. The complex structure of these proteins necessitates an equally complicated manufacturing process, which results in subtle differences in the end product of different manufacturers. These end products also differ in their pharmacologic actions [31]. Therefore, an identical replication of epoetins-α or ζ is not possible, even though their main biologic actions are compatible with the mechanism of action of rHuEPO; hence the term “biosimilar” is applied to them. Since they are produced in different cell lines, it is possible that there are isoforms in the same batch of products that can alter clinical pharmacology; possibly contributing to serious side effects such as PRCA [32, 33]. One deterrent to the manufacturing of biosimilars was the discovery of several cases of PRCA with one of the epoetin-α biosimilar brands [34]. Another was the ever-decreasing price of the novel ESAs and hence the loss of financial impetus to continue to make newer biosimilars. With patent expiration of the original ESAs, the introduction of several biosimilars of rHuEPO onto the US market is imminent.

In the study on the efficacy of biosimilars , the first trial compared the efficacy of the molecule HX575, the first biosimilar approved in Europe in 2007, to original rHuEPO molecules. Based on the structure of epoetin-α, HX575 was pharmacokinetically equivalent to epoetin-α but patient exposure was different as seen in the area under the curve (AUC) levels [35]. This finding is suggestive of different clinical response to equivalent dosing with biosimilars and hence close monitoring to hematologic indices was recommended [36]. When given to patients as maintenance therapy for anemia, HX575 was deemed clinically equivalent to epoetin-α [37] and as safe as epoetin-α [38]. However, with some evidence of differing clinical efficacy based on manufacturing locations [39], it was not deemed feasible to reliably incorporate HX575 in anemia protocols. SB309 was the second biosimilar agent to be developed based on epoetin-ζ. This molecule’s lower bioavailability compared to epoetin-α was attributed to the higher protein content of epoetin-α [40]. SB309 was found to be clinically equivalent to epoetin-α in the initial correction phase study and in two maintenance trials [41–43]. This was further corroborated in a crossover trial with a similar safety profile as epoetin-α [44]. Immunogenicity is a legitimate concern with biosimilars and this was also studied in ESA-naïve CKD patients. Of 337 patients randomized to receive either HX575 or SB309 via the SQ route, two patients developed antibodies and one had PRCA [45]; this was attributed to tungsten exposure during manufacturing [46]. None of the biosimilar studies have reported such antibodies when given IV [47]. Due to the rare nature of this adverse effect, only long-term clinical use will provide robust information about the magnitude of PRCA with biosimilars.

In the US, the first biosimilar agent for any indication was approved under the biosimilar approval pathway Biologics Price Competition and Innovation Act (BPCIA ) in 2014 [48]. Under this act, protection for the reference epoetin lasts 4 years from a biosimilar agent [49]. Specifically for management of anemia, there are at least six trials registered using biosimilar agents by the two companies aiming to introduce biosimilars in the US [50, 51] in both dialysis and non-dialysis-dependent CKD patients in different stages of completion. The Food and Drug Administration has a stringent and time-consuming process for biosimilar agents and this may slow down introducing more biosimilars into the US market. This cautious approach is supported by an increase in the prevalence of PRCA in countries that have used biosimilars for several years [52]. Though production and use have not been as controlled as the original epoetins even in Europe, reassurance can be had from the European experience of almost a decade of use, especially in a highly regulated and stringent environment. That and with a regimented approach paralleling that of the original biologic agents, there is not a strong argument to keep biosimilars out of reach of US nephrologists’ use in the near future.

For several years now and in several countries, epoetin-like peptides have been developed without licensing and stringent production standards. They differ dramatically in their structure and hence also in their clinical efficacy and side-effect profile [53]. There is a concern for increased incidence of PRCA with these products due to potentially lower production standards and for various other adverse effects due to unregulated or clandestine use, but due to a mismatch between demand and affordability, these are probably the only means of access to ESAs in many countries [54]. More data are awaited on these “copy” epoetins.

The efficacy of rHuEPO in correcting anemia in CKD and ESRD has been successfully established shortly after the manufacturing of the molecule for therapeutic utilization. After rHuEPO administration, a measurable increase in reticulocytes is seen in 7–10 days with an increase in hemoglobin following in 2–3 weeks. A concomitant drop in serum iron indices is evident unless iron is replenished simultaneously (more about the role of iron below). The widespread availability of rHuEPO in dialysis patients has resulted in a rapid uptick in its clinical application, and a broadening of its recipient base from dialysis patients to the much larger group of patients with non-dialysis-dependent CKD, and separately to patients with anemia caused by chemotherapy. The initial enthusiasm surrounding rHuEPO led to a belief that complete therapeutic correction of anemia of CKD and normalization of blood hemoglobin concentrations using this agent was possible and beneficial; hence a series of randomized controlled clinical trials were designed to test this hypothesis.

The approval of ESAs was based on small clinical trials and other observational studies showing clinical benefit along with avoidance of repeated transfusions [55]. Encouraged by these data, the first prospective randomized trial to assess the possibility of benefits in normalizing blood hemoglobin concentration was the US Normal Hematocrit (Hct) trial [56]. The goal of the study was to ascertain if a higher target Hct achieved with ESA therapy was indeed better. Dialysis patients were randomized into two groups with different target Hct (30% vs. 42%), and followed until the development of the composite primary end point of death or first nonfatal myocardial infarction. The study was prematurely halted due to a higher proportion of patients in the higher Hct target group reaching the primary end point, thus obviating the possibility of any benefit in “normalizing” the Hct and suggesting that treatment with rHuEPO to a normal Hct target may in fact be detrimental. In addition to the primary end point, other clinical end points such as vascular access thromboses also suggested detrimental effects from the normal Hct treatment paradigm . Another study evaluating the change in the left ventricular volume index in dialysis patients with different Hgb targets (9.5–11.5 g/dL vs. 13.5–14.5 g/dL) detected no difference in outcome between the two groups, suggesting a lack of benefit from normalizing Hgb levels to another hypothetical treatment target [57]. Along with the clinical trials conducted in dialysis populations, several trials were designed to test similar hypotheses in anemic patients with non-dialysis-dependent CKD. The Cardiovascular Risk Reduction by Early Anemia Treatment with Epoetin Beta (CREATE) trial was an international randomized controlled clinical trial of patients with advanced non-dialysis-dependent CKD, which studied the incidence of cardiovascular events in two groups assigned to different Hgb targets and different timings of epoetin-β initiation [58]. No differences in primary end points were detected and, interestingly, the group initiating epoetin-β earlier experienced a shorter time to dialysis. The Correction of Hemoglobin and Outcomes in Renal Insufficiency (CHOIR) trial was another large randomized controlled clinical trial designed to address the effect of different Hgb treatment targets (11 g/dL vs. 13 g/dL) on clinical events [59]. The group assigned to the higher Hgb target experienced a significantly higher rate of the composite primary end point (congestive heart failure, hospitalizations, stroke, or myocardial infarction) compared to the low target group. Also, there was not any improvement in the quality of life as perceived by the participants in the higher Hgb target. Similar results were reported in the Trial to Reduce Cardiovascular Events with Aranesp Therapy (TREAT) trial using darbepoetin, in that the darbepoetin-treated arm experienced a significantly higher incidence of stroke compared to the placebo arm, and without a benefit in the other study end points [60]. Benefits from using ESAs were also examined in trials conducted in non-CKD patients, especially those with heart failure. In the Reduction of Events by Darbepoetin Alfa in Heart Failure (RED-HF) trial , patients with systolic heart failure were administered darbepoetin in an effort to study its effect on a composite primary end point of death or hospitalizations related to heart failure [61]. Interestingly, no differences were seen in the treatment group compared to placebo and a higher incidence of thromboembolic events in the darbepoetin group led to a halt in design of large trials using ESAs in heart failure. This is in contrast to the trials conducted with intravenous iron (IVI) in heart failure (discussed below).

From these clinical trials, it has become clear that normalization of Hgb using treatment with rHuEPO is still fraught with difficulties: it does not result in the expected clinical benefits, it could be detrimental, and last but not least, it is very expensive. Hence, current guidelines suggest maintenance of Hgb in the 10–12 g/dL range with focus on avoiding transfusion, recognizing the benefits of rHuEPO therapy in sparing blood transfusions in patients with severe anemia of CKD, but also acknowledging a lack of high-quality evidence on the ideal Hgb level for these interventions. It also remains unclear if the Hgb treatment targets established in the wake of the rHuEPO trials are valid end points for therapeutic strategies employing agents with distinctly different mechanisms of action (vide infra). Post-hoc analyses have suggested that the adverse effects seen in the high-Hgb target arms of the rHuEPO trials may have been attributable to the higher doses of rHuEPO and/or the relative iron deficiency induced by this strategy [62]; hence the ideal therapeutic targets for other anemia therapies will have to be evaluated in dedicated clinical trials.

Iron replenishment is an essential aspect in the management of anemia of various etiologies; mostly when iron deficiency is the main cause of the anemia, but also in the context of other treatments, such as with rHuEPO. In addition to the “usual” iron losses, CKD patients may also lose iron as a result of more frequent gastrointestinal or other bleeding, frequent blood testing or blood losses related to hemodialysis [94]. Coupled with ineffective iron absorption through the gastrointestinal tract in dialysis patients [95], this loss of iron can amount to a negative iron balance of 1.5–3 g per year [96]. Unless at least a similar amount of iron is supplemented, ESA treatment is expected to be ineffective. Also, with rHuEPO administration and enhanced erythropoiesis, there is significant decrease in iron stores and a lack of replenishment leads to suboptimal and eventually arrested response to rHuEPO or its derivatives. The traditional route of administration for iron has been oral. Ferrous sulfate in oral preparations has been effective in patients with iron deficiency who are devoid of chronic inflammation and liver disease, but less so in patients in whom hepcidin prevents effective restoration of iron stores. In such patients intravenous iron (IVI) preparations have gained popularity (Table 5.3), especially in cases where this route is also convenient and practical, such as chronic HD patients and this is supported by evidence of clinical benefit and cost savings [97–99].