Patients with anemia in the setting of end-stage renal disease commonly require treatment with an erythropoiesis-stimulating agent. Therapeutic options for the management of anemia have expanded with the development of new formulations of erythropoiesis-stimulating agents.

In 1967, a study of erythropoiesis in patients receiving hemodialysis (HD) found no measurable plasma erythropoietin activity in 14 of 16 patients, a mean hematocrit of 24% maintained by blood transfusions and a shortened red cell lifespan (half of the normal value).1 In 1985, prior to the availability of recombinant human erythropoietin (rhEPO), one study reported an average hemoglobin concentration of 8 g/dL and 10% of patients dependent on regular blood transfusions; 30% of patients receiving HD had hemoglobin concentrations less than or equal to 6 g/dL and 15% had hemoglobin concentrations greater than 10 g/dL.²

In 1993, 4 years after the introduction of rhEPO, more than 80% of patients at a large U.S. dialysis provider were receiving rhEPO, and the mean hemoglobin concentration had reached 9.5 g/dL.³ Among the 12% of patients with hemoglobin concentrations less than or equal to 10 g/dL at that time, the odds ratio of death was two-fold higher compared to patients with hemoglobin greater than 10 g/dL. The most recent United States Renal Data System (USRDS) data show that approximately 80% of patients receiving HD are treated with an erythropoiesis-stimulating agent (ESA) in any given month for management of anemia and approximately 20% of patients on HD receive one or more blood transfusions per year.^4-5

For more than 2 decades, ESAs have been standard treatment for anemia in patients receiving dialysis.^2-6 Anemia management remains a major focus in the care of patients with advanced kidney disease.^6-7 In the United States, patients with kidney disease commonly use epoetin alfa and darbepoetin ESA formulations for the treatment of anemia. Epoetin alfa has a half-life ranging from 4 to 13 hours and is generally administered greater than once per week.⁸ Darbepoetin has a mean half-life of 21 hours and is typically provided in the range of once per week to once every 2 weeks.^9-10

In 2014, methoxy polyethylene glycol-epoetin beta (Mircera, F. Hoffmann-La Roche) received FDA approval for use in patients with kidney disease in the United States. Mircera is a continuous erythropoietin receptor activator commonly (CERA), with a mean half-life of 134 hours and a dosing paradigm of once every 2 to 4 weeks.^10-11 Notably, despite their 10 plus-fold difference in half-life, epoetin alfa and CERA showed similar dose-dependent responses in clinical trials, with a hemoglobin rise within 2 weeks at high dose and within 6 weeks at low dose.^12-13 Due to of its longer half-life, CERA has longer-acting pharmacokinetics than shorter acting ESAs, allowing it to be administered less frequently.

PAGE BREAK

The efficacy and safety of CERA have been demonstrated in randomized and observational clinical trials worldwide.^14-17 Although CERA was originally approved by the FDA in 2007 for the treatment of anemia in the setting of kidney disease,¹⁸ it was not permitted for use until 2014 due to patent disputes.¹⁹ It has been established that short-, middle- and long-acting ESAs have equivalent efficacy for the treatment of anemia in kidney disease.²⁰

Other potential benefits of long-acting ESAs include reduced staff time for medication administration, improved maintenance of hemoglobin during hospitalizations and competitive pressure to reduce ESA prices. Several studies performed in Europe and the United States found monthly administration of long-acting ESAs resulted in a reduction of health care staff time for the management of anemia.^21-24 A potential disadvantage of long-acting ESAs is the decreased opportunity for titrating the ESA dose, reducing the physician’s ability to immediately address hemoglobin concentrations that are above the target range. However, the activity of shorter-acting ESAs in promoting the proliferation of progenitor cells leads to differentiation into red blood cells with effects that last weeks to months. Notably, the prescribing information provided as package inserts for shorter-acting ESAs indicate that dose changes should occur no more often than every 4 weeks except in response to an increase in hemoglobin of greater than 1 g/dL for 2 weeks.^9.10

In late 2014, Fresenius Kidney Care (FKC) initiated a rollout of CERA in U.S. dialysis facilities with concurrent analytics to assess the potential advantages of including this longer-acting ESA on the drug formulary. During the operational deployment and rollout, we assessed effectiveness of routine use, safety and logistical challenges.

We describe our experience in the first year of transitioning from the use of short- or middle-acting ESAs to CERA in a large national network of outpatient in-center HD facilities.

Operational plan and deployment

We deployed educational resources in a phased rollout to convert from use of shorter-acting ESAs to a long-acting ESA at FKC clinics starting in December 2014. To roll out this new ESA consistently and swiftly, we provided an electronic algorithm for physicians to consider. If ordered by the physicians, this algorithm performed daily calculations of the changes in dose and frequency required by a physician-signed algorithm, rather than necessitating manual calculation.

Prior to initiating the rollout, workgroups set the project timeline, milestones, vigilance monitoring, analytical plans and prepared systems to accommodate documentation of the new medication. An introductory meeting with clinics included information on CERA’s mode of action, results from clinical development and international pragmatic experiences, safety profiles, operational benefits and the deployment plan for CERA.

PAGE BREAK

Training at facilities was led by the deployment education team in the 8 weeks prior to the CERA rollout date, and formal open forum follow-up calls were offered in the 8 weeks following the rollout date to allow clinicians to ask questions or share concerns that arose. The training consisted of physician and staff education on the following topics: overview of the medication (mode of action and pharmacokinetics); contraindications; treatment logistics; medication storage and handling; drug order entry; pre-filled syringe training; adverse event- and product complaint-reporting; and case studies. Additionally, we provided a medical science liaison consultant to each clinic to offer additional clinical and operational support.

Moreover, we provided dedicated staff for additional vigilance monitoring of adverse events and complaints.

The transition from shorter-acting ESAs to a long-acting ESA required logistical planning and operational changes to accommodate the conversion to the new drug. CERA is supplied in single-use pre-filled syringes and, like all ESAs, requires refrigeration at 2°C to 8°C to ensure the stability of the exogenous hormone. As single- use syringes require more space than vials, all facilities were provided with additional medical-grade refrigerators to house the increased volume. To ensure the stability of CERA in transit, single-use syringes were packaged and shipped with a cold chain monitoring system that would indicate if any temperature excursion occurred during shipment.

The ESA conversion required a shift from the typical thrice weekly dosing with a short-acting ESA to every 2- to 4-week administration with the long-acting ESA. To make this transition easier for physicians and clinicians, FKC developed an electronic algorithm which calculated dose changes based on hemoglobin values and other clinical information such as recent hospitalizations, missed doses and iron parameters for new patients on dialysis. Even after signing the algorithm, physicians hold discretion as to whether to override individual dose calculations based upon their own medical judgment, and the electronic algorithm recognizes such overrides. The algorithm was also designed to be flexible enough to accommodate facility variations in collection of labs and preferred dosing schedules.

PAGE BREAK

During the first 8 weeks of deployment, education was provided to physicians and clinicians to reinforce the basic principles of anemia management and ESA pharmacokinetics, as well as instructions for clinical staff on how to use the algorithm, including practical details such as how to select appropriate syringes for administering the prescribed CERA dose. This education also reminded clinicians to communicate with hospitals to prevent premature dosing with a short-acting agent. This education was provided as a combination of webinars, written materials and support by medical science liaisons available to answer specific clinical questions.

The number of new clinics de-ployed per month increased during the first 12 months with 18 clinics in month 1, 41 new clinics in month 2 and 161 new clinics by month 12; the greatest number of new clinics deployed per month was 410 in month 7 (Figure 1).

Clinical parameters

We examined data from the first year of the rollout and transition (December 2014 to December 2015). During the rollout period, 1,966 clinics were included in the rollout of CERA, and by December 2015, 155,115 patients had received their first dose of CERA (Figure 1). To understand changes in clinical parameters with the conversion, we examined all in-center HD patients who first received CERA during the rollout period and did not use any other kind of ESA (epoetin alfa or darbepoetin alfa) during their follow-up period. Patients who received any home dialysis treatments (HD or peritoneal dialysis) during the baseline or follow-up periods were also excluded from the analyses.

Figure 1. During 12 months of the CERA rollout (from December 2014 to December 2015), a total of 1,966 clinics were transitioned to CERA (A). A total of 155,115 patients were converted and had received a first dose of CERA (B).

To assess the impact of the conversion to CERA, we retrospectively examined the percentages of weekly hemoglobin concentrations in the range of 10 g/dL and 12 g/dL, 3-month-average transferrin saturation (TSAT) in the range of 30% to 50%, as well as IV iron dosing patterns during the 6-week baseline period prior to the first dose of CERA and the 24-week follow-up period after the first dose of CERA. In addition, we examined the average CERA dose at 8 through 12 weeks in patients with central venous catheters (CVCs) as compared to patients with arteriovenous fistulas (AVFs), and a further sub-analysis examined the average CERA dose at the time of CVC removal compared to the dose 8 to 12 weeks later. We also compared the hospitalization rate in the 24 weeks prior to conversion to the 24 weeks after conversion among the patients who converted.

PAGE BREAK

A total of 95,024 patients distributed across 1,811 clinics were included in the analyses. The cohort had the following characteristics: median age 67 years (25th to 75th percentile range of 57 to 77 years); median vintage 5 years (25th to 75th range of 3 to 7 years); 55% were men; 56% Caucasian, and 35% African-American. Almost two-thirds (64%) of the cohort had a documented diagnosis of diabetes mellitus, 23% had a central venous catheter (CVC) for vascular access, and the median serum albumin concentration was 3.8 g/dL.

Figure 2. Analysis of hemoglobin levels in patients who converted to CERA during the 6 weeks prior to initiation up through 24 weeks after the first dose is shown (A). Analysis of the percent of patients meeting the TSAT goal (30% to 50%) 3 months before and 3 months after the initiation of CERA (B). Mean IV iron dosing in patients that started receiving CERA during the 6-week baseline period up through 24-week follow-up period (C). Mean IV iron dosing trend separated into ESA naïve patients and patients who had received ESA previously (D).

The management of anemia was evaluated using the weekly percent of patients within the hemoglobin range of 10 g/dL to 12 g/dL during the baseline and follow-up periods. The percent of patients within the hemoglobin range of 10 g/dL to 12 g/dL was relatively unchanged (Figure 2A). The decline from 70% to 65% in the percent of patients with hemoglobin concentrations in range observed 1 week prior to conversion may have been secondary to patients who happened to be on hold from their prior ESA due to higher hemoglobin concentrations, as the conversion algorithm required that patients with a hemoglobin of greater than 11.5 g/dL wait until the hemoglobin decreased to at least 11.1 g/dL before initiating CERA. The distribution of patients with 3-month-average transferrin saturation (TSAT) within range (30% to 50%) was similar in the baseline and follow-up periods (see Figure 2B). The mean IV iron dose per HD was generally stable during the baseline and follow-up period (see Figure 2C) with the minor increase at the time of initiation of CERA appearing to be due to ESA-naïve patients receiving iron repletion at the time of starting CERA (see Figure 2D).

The management of anemia was further examined with respect to vascular access type. Patients were categorized by their vascular access type at the time of conversion CERA and only patients with CVCs (n=22,147) or AVFs (n=54,613) were examined in these secondary analyses. Eight to 12 weeks after conversion the mean CERA dose in patients with CVCs was 11.2 µg/HD treatment compared to a mean CERA dose of 9.7 µg/HD treatment in patients with AVFs. A similar difference in CERA dose was observed in a subset of more than 5,000 patients whose CVC was removed when the CERA dose immediately prior to CVC removal was compared to the CERA dose 8 to 12 weeks after CVC removal (results not shown).

Moreover, we assessed the changes in the frequency of ESA administrations at HD treatments in patients 12 weeks before and after conversion to CERA. During the twelve-week period before conversion, we observed that patients received a short- or middle-acting ESA at 80% and 29% of HD treatments respectively. Twelve weeks after conversion to CERA, patients received the longer-acting ESA less frequently at 12% of HD treatments.

PAGE BREAK

We also examined hospitalizations recorded in our EHR in the period before using CERA and after CERA initiation. The hospitalization records in the EHR have specific sections to indicate whether the patient received a transfusion or an ESA dose while in the hospital. The baseline period for hospitalization was defined as the 24 weeks before an individual’s first CERA dose and the follow-up period consisted of the 24 weeks after initiation. We calculated the decrease in hemoglobin over the course of each hospitalization and compared results between the baseline period and the follow-up period.

The decrease in hemoglobin was calculated using the difference between the last hemoglobin result prior to the start of the hospitalization and the first hemoglobin result after the hospital discharge. We examined a 90,337 hospitalizations among a subset of 52,684 patients who started using CERA during the rollout period and experienced at least one hospitalization in either the baseline or follow-up periods.

During the baseline period, 50,291 hospitalizations were observed and the mean hemoglobin decrease was 0.27 g/dL. In the follow-up period, 40,046 hospitalizations were observed and the mean decrease in hemoglobin was 0.18 g/dL. In both the baseline and follow-up periods, the rate of patients who had documented blood transfusions in the EHR during hospitalization was 7%. Our records also indicated 28% of patients received an ESA dose during their hospital admission in the baseline period as compared to 14% of patients received an ESA dose during their hospital admission in the follow-up period.

Figure 3. Counts of SAEs (A) and other non-serious AEs (B) by month, except for the first 3 months, which were grouped together due to low numbers. Hashed bars indicate numbers of adverse events occurring on the first dose of CERA. Solid bars indicate events occurring on later doses. Dashed and solid lines represent total counts by month (with and without adverse events) of first doses and later doses, respectively.

We examined adverse event data from all CERA administrations given between December 2014 and December 2015 (see Figure 3). This analysis included all patients who received one or more doses of CERA. Figure 3 shows counts of serious adverse events (SAEs) and other non-serious adverse events (AEs) by month. Due to small patient numbers, data for the first 3 months were grouped together. During the 13-month period, 139,588 first doses and 735,628 later doses were given. A total of 18 SAEs, including 13 first-dose events and five later events, and 185 other AEs, including 133 first-dose events and 52 later events, occurred. During that time, the rates of SAE were 0.93 SAE per 10,000 first doses and 0.068 SAE per 10,000 later doses (see Figure 3A). The corresponding rates of other non-serious AEs were 9.5 other AEs per 10,000 first doses and 0.71 other AEs per 10,000 later doses (see Figure 3B). There was one death, which occurred in the hospital 2 days after the patient exhibited seizure-like behavior after receiving a second dose of CERA. It should be noted that these rates of AEs are lower than the rates reported in clinical trials, which collect symptoms that may be considered normal consequences of renal failure or dialysis treatment, such as pruritis or low blood pressure, and include hospitalizations and deaths regardless of attributed cause. For example, in registrational clinical trials with 6 and 12 months of follow-up, AE were reported in 38% of CERA patients and in 42% of patients on other ESAs, which corresponds to more than 200 AEs per 10,000 doses.¹¹

PAGE BREAK

Discussion

In our experience transitioning to using CERA for management of anemia related to ESRD, we found the widespread conversion during 12 months in a population of more than 155,000 patients on dialysis at more than 1,900 clinics was feasible with incremental rollout periods. Relatively stable hemoglobin ranges of 10 g/dL to 12 g/dL were observed throughout the transition and a small number of observed AE relative to the number of doses administered were observed.

When examining decreases in hemoglobin during the course of documented hospitalizations among patients who converted to CERA, we observed a smaller average hemoglobin decrease during the follow-up when patients were receiving CERA compared to the average decline in hemoglobin during the baseline when patients had been receiving shorter acting ESAs. Although we think this is an important observation, the limitations should be noted, including the challenge of receiving transition-of-care records that include transfusions and medications administered during hospitalization, and the need for dialysis staff to manually enter hospitalization-related data into the FKC EHR. Based on our observations using the data available, it appears hemoglobin stability with different ESAs during the course of hospitalizations in patients with dialysis-dependent ESRD is a topic for further study.

Another major benefit of transitioning to a long-acting ESA may be the reduction in resource utilization for anemia management. In our experience, we found the frequency of ESA administrations per HD treatment was more than 65% lower after conversion to CERA for patients who were previously treated with a short-acting ESA. A recent analysis performed in Australia on patients receiving HD found switching from a short-acting to a long-acting ESA reduced up to 80% of the time needed for health care professionals to administer the ESAs, which further reduced costs associated with labor.25 Additional studies performed in Europe and the United States have also concluded that monthly administration of long-acting ESAs results in a reduction of time employed by health care staff to manage anemia.^21-24

Conclusion

We observed that CERA showed comparable clinical performance to shorter-acting ESAs. Our clinical experience demonstrates the utility of a disciplined operational deployment of a long-acting ESA agent into the treatment of patients with anemia related to ESRD.

References:
1. Eschbach JW, et al. N Engl J Med. 1967;doi:10.1056/NEJM196703232761 202.
2. Winearls CG. Nephrol Dial Transplant. 1995;10Suppl2:3-9.
3. Madore F, et al. J Am Soc Nephrol. 1997;8(12):1921-1929.
4. Saran R, et al. Am J Kidney Dis. 2018;doi:10.1053/j.ajkd.2018.01.002.
5. Pisoni RL, et al. Am J Kidney Dis. 2004;doi:10.1053/j.ajkd.2004.03.023.
6. Wetmore JB, et al. Am J Nephrol. 2015;doi:10.1159/000431335.
7. DOQI. Am J Kidney Dis. 2007;doi:10.1053/j.ajkd.2006.12.005.
8. Epogen [package insert]. 1989. Available from: www.pi.amgen.com/~/media/amgen/repositorysites/pi-amgen-com/epogen/ epogen_pi_hcp_english.pdf. Accessed Oct. 27, 2017.
9. Aranesp [package insert]. 2001; Available from: www.accessdata.fda.gov/drugsatfda_docs/label/2011/103951Orig1s5173_103951Orig1s5258lbl.pdf. Accessed Oct. 27, 2017.
10. Biggar P, et al. Kidney Res Clin Pract. 2017;doi:10.23876/j.krcp.2017.36.3.209.
11. Mircera [package insert]. 2015; Available from: www.accessdata.fda.gov/drugsatfda_docs/label/2016/125164s071s072s073lbl.pdf Accessed April 26, 2018.
12. Eschbach JW, et al. N Engl J Med. 1987;doi:10.1056/NEJM198701083160 203.
13. Provenzano R, et al. Clin Nephrol. 2007;doi:10.5414/CNP67306.
14. Fliser D, et al. Curr Med Res Opin. 2010;doi:10.1185/03007991003666652.
15. Locatelli F, et al. Adv Ther. 2016;doi:10.1007/s12325-016-0309-6.
16. Locatelli F, et al. Clin Nephrol. 2010;doi:10.5414/CNP73094.
17. Aggarwal S. Nat Biotechnol. 2008;doi:10. 1038/nbt1108-1227.
18. FDA Administration. 2007. Available from: www.accessdata.fda.gov/drugsatfda_docs/nda/2007/125164TOC.cfm. Accessed April 26, 2018.
19. Reuters. 2009; Available from: www.reuters.com/article/amgen-roche/update-1-amgen-wins-anemia-drug-patent-dispute-with-roche-idUSN2212529420091222. Accessed April 26, 2018.
20. Palmer Suetonia C, et al. Cochrane Database Syst Rev. 2014;doi:10.1002/14651858.CD010590.pub2.
21. Saueressig U, et al. Blood Purif. 2008;doi:10.1159/000171873.
22. Schiller B, et al. Hemodial Int. 2008;doi:10.1111/j.1542-4758.2008.00308.x.
23. De Cock E, et al. Blood Purif. 2015;doi:10.1159/000437133
24. Stephens JM, et al. Curr Med Res Opin. 2016;doi:10.1185/03007995.2015.1123144.
25. Johnson DW, et al. J Ren Care. 2017;doi:10.1111/jorc.12213.
The authors are employed by Fresenius Medical Care North America in Waltham, Mass. Jeffrey Hymes, MD, is the chief medical officer and senior vice president of Fresenius Kidney Care. Marta M. Reviriego-Mendoza, MD, is a senior scientific liaison. Norma J. Ofsthun, PhD, is vice president of data analytics. Amanda K. Stennett, PhD, is director of electronic algorithms and kinetic modeling. Patricia McCarley, MSN, RN, ACNP-BC CNN, is vice president of strategic quality operations. Adam J. Davey, PMP, is director of operational excellence. Carol Lee, BSN, RN, CDN, is manager of patient safety. Tanya Taft, RN, CNOR, CEEP, CPPS, is vice president of patient safety. John W. Larkin, MS, is director of publications and research. Franklin W. Maddux, MD, FACP, is chief medical officer and executive vice president for clinical and scientific affairs. Disclosures: The authors report no financial disclosures.