Erythropoiesis-Stimulating Agents

Transcription

1 Erythropoiesis-Stimulating Agents Policy Number: Last Review: 8/2017 Origination: 8/2015 Next Review: 8/2018 Policy Blue Cross and Blue Shield of Kansas City (Blue KC) will provide coverage for erythropoiesisstimulating agents when the following medically necessary criteria are met. Prior authorization is recommended for prescription benefit coverage of Epogen, Procrit, and Aranesp. Throughout this Policy, unless otherwise stated: The term ESA refers to epoetin alfa (Epogen, Procrit ) and darbepoetin alfa (Aranesp ). Nonmyeloid malignancies include solid tumors and the nonmyeloid hematologic malignancies myeloma, lymphoma, and chronic lymphocytic leukemia. When Policy Topic is covered The use of epoetin alfa (Epogen/Procrit), darbepoetin (Aranesp), or pegylated (PEG) epoetin beta (Mircera) may be considered medically necessary for: treatment of anemia associated with chronic kidney disease. a,b,c The use of PEG-epoetin beta is investigational for all other indications. The use of epoetin alfa (Epogen/Procrit) or darbepoetin (Aranesp) may be considered medically necessary for: treatment of anemia in cancer patients with non myeloid malignancies where anemia is due to the effect of concomitantly administered chemotherapya, b ; treatment of anemia related to therapy with AZT (zidovudine) in HIV-infected patients a ; reduction of allogeneic blood transfusion in surgery patients a ; treatment of patients after allogeneic bone marrow transplantation; treatment of patients with myelodysplastic syndromes to reduce transfusion dependency; and treatment of patients with hepatitis C and anemia related to ribavirin treatment. In the medically necessary conditions noted above, the following criteria also apply: The lowest dose of erythropoiesis-stimulating agents (ESAs) should be used in order to avoid red blood cell transfusions; ESAs should not be used to raise the hemoglobin (Hb) level above 12 g/dl; and ESA therapy should not be administered without adequate iron stores. For medically necessary use in cancer patients, these additional U.S. Food and Drug Administration criteria also apply: Epoetin alfa or darbepoetin therapy should not be initiated at hemoglobin levels of 10 g/dl or higher; and

2 Epoetin alfa or darbepoetin treatment should be discontinued after completion of a myelosuppressive chemotherapy course. The use of epoetin alfa or darbepoetin is investigational for: treatment of patients after high-dose chemotherapy with autologous stem-cell support; treatment of noniatrogenic chronic anemia of cancer; and other cancer-associated anemia except as noted above. a FDA-approved label for epoetin alfa (Epogen, Procrit ). b FDA-approved label for darbepoetin alfa (Aranesp ). c FDA-approved label for PEG-epoetin beta (Mircera ). When Policy Topic is not covered The use of PEG-epoetin beta is investigational for all other indications The use of epoetin alfa or darbepoetin is investigational for: treatment of patients after high-dose chemotherapy with autologous stem-cell support; treatment of noniatrogenic chronic anemia of cancer; and other cancer-associated anemia excepted as noted above. Considerations Administration Erythropoiesis-stimulating agents (ESAs) and pegylated (PEG) epoetin beta are to be administered according to current Food and Drug Administration (FDA) approved labeling for each product, using recommended hemoglobin (Hb) levels for starting, stopping, and dose adjustment. This includes decreasing the dose of ESA as the Hb approaches the target level. Before commencing ESA or PEG-epoetin beta therapy, the patient s iron stores, blood ferritin, and transferrin saturation should be evaluated, adjusted, and maintained within normal physiological limits. ESA or PEG-epoetin beta therapy should not be administered without adequate iron stores. Blood Pressure Monitoring Blood pressure should be adequately controlled before initiation of ESA therapy and closely monitored and controlled during treatment. ESAs and PEG-epoetin beta are contraindicated in patients with uncontrolled hypertension. Discontinuation Erythropoiesis-Stimulating Agents Patients with myelodysplastic syndromes should be initially limited to a 3-month trial period with ESA. If no response to ESA is observed, ongoing therapy would be futile. ESAs and PEG-Epoetin Beta Patients with chronic kidney disease who do not respond adequately over a 12-week dose escalation period should not have their ESA or PEG-epoetin beta dose increased further. Increasing ESA or PEG-epoetin beta dose further is unlikely to improve response and may increase risks; the lowest ESA or PEG-epoetin beta dose that maintains adequate Hb to avoid recurrent red blood cell transfusions should be used. Other causes of anemia should be evaluated. If responsiveness does not improve, discontinue ESA or PEG-epoetin beta therapy. Risk Evaluation and Mitigation Strategy Epoetin alfa and darbepoetin must be prescribed and dispensed in accordance with a risk evaluation and mitigation strategy (REMS) drafted by the manufacturer and approved by FDA.

3 REMS for epoetin alfa and darbepoetin alfa each comprise elements to assure safe use and an implementation system. ESA manufacturers must ensure that all hospitals and healthcare professionals who prescribe and/or dispense ESAs to patients with cancer have enrolled and completed training in the ESA APPRISE (Assisting Providers and Cancer Patients with Risk Information for the Safe use of ESAs) Oncology Program. The ESA APPRISE program began on March 24, 2010 after FDA s initial approval of separate but similar REMS for epoetin alfa and darbepoetin alfa on February 16, Both REMS were subsequently modified, most recently on December 31, Healthcare providers and hospitals that prescribe and/or dispense an ESA for chronic kidney disease (CKD) must provide each patient with a copy of the REMS Medication Guide included in the product label and ensure that patients are adequately informed of the risks associated with ESA treatment. However, prescribers are not required to enroll in and complete the ESA APPRISE program. PEG-epoetin beta does not have a REMS. On March 27, 2012, FDA approved a REMS for peginesatide with a communication plan as its only component. The plan s goal was to inform all healthcare professionals who might prescribe the drug that peginesatide is indicated only for adult patients with CKD on dialysis, and of potentially fatal risks associated with its use in CKD patients not on dialysis. Peginesatide is currently discontinued. This Blue Cross and Blue Shield of Kansas City policy Statement was developed using available resources such as, but not limited to: Hayes Medical Technology Directory, Food and Drug Administration (FDA) approvals, Facts and Comparisons, National specialty guidelines, Local medical policies of other health plans, Medicare (CMS), Local providers. Description of Procedure or Service Endogenous erythropoietin is a glycoprotein hematopoietic growth factor that regulates hemoglobin levels in response to changes in the blood oxygen concentration. Erythropoiesis-stimulating agents (ESAs) are produced using recombinant DNA technologies and have pharmacologic properties similar to endogenous erythropoietin. The primary clinical use of ESAs is in patients with chronic anemia. The evidence for ESAs (eg, epoetin alfa, pegylated epoetin beta, darbepoetin) in individuals who have chronic kidney disease and anemia includes randomized controlled trials (RCTs) and systematic reviews of RCTs. Relevant outcomes are symptoms, morbid events, medication use, and treatmentrelated mortality and morbidity. All 3 ESAs have been studied and approved for this use. Most of the evidence has demonstrated a decrease in the need for blood transfusions but has failed to demonstrate any significant improvement in other clinical outcomes, including mortality, morbidity, functional status, or quality of life. Many studies have, in fact, demonstrated increased mortality risk and increased risk for venous access thrombosis and stroke, prompting U.S. Food and Drug Administration (FDA) warnings and a risk evaluation and mitigation strategy. The evidence is sufficient to determine qualitatively that the technology results in a meaningful improvement in the net health outcome. The evidence for ESAs (eg, epoetin alfa, darbepoetin) in individuals who have cancer-related anemia includes RCTs and systematic reviews of RCTs. Relevant outcomes are symptoms, morbid events, medication use, and treatment-related mortality and morbidity. The available trials have demonstrated an increase in hemoglobin concentration and a decrease in the need for blood transfusions. However, the evidence has also demonstrated increased mortality rates and possible tumor promotion, as well as increased risk of thromboembolic events, when target hemoglobin levels were above 12 g/dl. Epoetin alfa and darbepoetin are the ESAs approved for use in treatment of cancer-related anemia; pegylated epoetin beta is not FDA approved for this indication, as studies have demonstrated increased mortality

4 and no significant improvement in clinical outcomes. The evidence is sufficient to determine qualitatively that the technology results in a meaningful improvement in the net health outcome. The evidence for ESAs (eg, epoetin alfa, darbepoetin) in individuals who have hepatitis C infection treated with ribavirin includes RCTs. Relevant outcomes are quality of life and medication use. Evidence from RCTs demonstrates that treatment with ESAs improves the ability to maintain full-dosing of ribavirin, as anemia is often a limiting effect for treatment. There may also be a positive effect on quality of life, although this is less certain. Epoetin alfa and darbepoetin are the ESAs approved for this use. The evidence is sufficient to determine qualitatively that the technology results in a meaningful improvement in the net health outcome. Endogenous erythropoietin is a glycoprotein hematopoietic growth factor synthesized by cells near the renal tubules in response to changes in the blood oxygen concentration. When a patient is anemic, the ability of the blood to carry oxygen is decreased. An oxygen-sensing protein in the kidney detects the decrease in blood oxygen concentration and induces the production of endogenous erythropoietin, which then acts on the erythroid cell line in the bone marrow to stimulate hematopoiesis, thereby effectively increasing blood hemoglobin (Hb) concentrations. Suppression of erythropoietin production or suppression of the bone marrow response to erythropoietin results in anemia in several disease processes, including chronic kidney disease (CKD), many types of cancer treatment, other chronic diseases, and use of certain drugs. The severity of anemia is defined by blood Hb concentration. Normal ranges are 12 to 16 g/dl in women and 14 to 18 g/dl in men. Mild anemia is defined as Hb from 10 g/dl to the lower limit of normal ranges, moderate anemia is 8 to 10 g/dl, and severe anemia is 8 g/dl or less. Erythropoiesis-stimulating agents (ESAs) are produced using recombinant DNA technologies. They were initially developed as replacement therapy to treat anemia due to endogenous erythropoietin deficiency that commonly occurs in patients with chronic renal failure (CRF) secondary to CKD. Patients with CRF will become severely anemic and experience severe fatigue and reduced exercise tolerance unless treated with blood transfusions or an ESA. Partial correction of anemia by ESA treatment of patients with CRF reduces the need for red blood cell (RBC) transfusions and enhances physical functioning. In cancer, anemia occurs with varying degrees of frequency and severity. It occurs most commonly in genitourinary, gynecologic, lung, and hematologic malignancies. Anemia may be directly related to cancer type or to its treatment. Oncologic anemia occurs by a variety of mechanisms: (1) Poor oral intake or altered metabolism may reduce nutrients (folate, iron, vitamin B12) essential for RBC production. (2) Antibodies and/or immunoregulatory abnormalities associated with certain tumor types (most commonly, B cell malignancies) may cause increased erythrocyte destruction (hemolysis). (3) Tumors may cause blood loss via tissue invasion (eg, gastrointestinal bleeding from colon cancer). (4) Other neoplasms, particularly hematologic malignancies (leukemia, lymphoma, and multiple myeloma) can invade the bone marrow and disrupt the erythropoietic microenvironment. (5) In more advanced cases, there may be marrow replacement with tumor or amyloid. (6) However, marrow dysfunction can occur even in the absence of frank invasion. (7) Inflammatory proteins from interactions between the immune system and tumor cells are thought to cause inappropriately low erythropoietin production and poor iron utilization, as well as a direct suppression of RBC production. Cancer treatments also may cause anemia: (1) radical cancer surgery can result in acute blood loss; and (2) radiotherapy and many cytotoxic chemotherapeutic agents suppress marrow to varying degrees. Damage is due to a variety of mechanisms. For example, alkylating agents cause cumulative DNA damage; antimetabolites damage DNA indirectly; and platinum-containing agents appear to damage erythropoietin-producing renal tubule cells. RBC transfusion is the traditional approach to quickly ameliorate anemia symptoms. However, this approach carries risk for several potential adverse events. The highest adverse event risk (1 per 432 whole blood units transfused) is for transfusion-related acute lung injury (TRALI). Adverse events due to errors in transfusion (eg, type mismatch) are estimated to occur at a rate of 1 per 5000 to 10,000

5 units of blood transfused. Current transfusion medicine and blood bank practices have significantly reduced the risk of transmissible infections, primarily due to better donor selection and screening for infectious diseases. Estimated risks per unit of blood transfused for transmission of hepatitis B virus (<1 in 400,000), hepatitis C virus (<1 in 1,000,000), HIV (<1 in 1,000,000), and bacterial contaminants (1 per 10,000 to 100,000) have fallen dramatically since the early 1990s. Therefore, although the initial impetus to commercialize erythropoietin replacement products was based on reduction in the risks associated with blood transfusion, current practices have mitigated many of those risks. Nonetheless, blood shortages, transfusion errors, and risks of alloimmunization and TRALI provide sufficient rationale for the use of ESA therapy in appropriately indicated patients. Four ESA products have been licensed in the United States: Epoetin alfa is manufactured, distributed, and marketed by Amgen, under the proprietary name, Epogen. The same epoetin alfa product manufactured by Amgen is also marketed and distributed by Janssen Products, a subsidiary of Johnson and Johnson, under the proprietary name, Procrit. Under a contractual agreement with Amgen, Janssen Products has rights to develop and market Procrit for any indication other than for treatment of anemia associated with CRF in patients on dialysis or use in diagnostic test kits. Epogen and Procrit have identical labeling information for all FDA-approved indications. A second ESA, darbepoetin alfa, is marketed solely by Amgen, under the proprietary name, Aranesp. The third ESA product, peginesatide, was codeveloped and commercialized by Affymax and Takeda Pharmaceuticals, who market it under the proprietary name, Omontys. In February 2013, Affymax, Takeda, and FDA announced a voluntary recall of all lots of peginesatide due to post-marketing reports of serious hypersensitivity reactions, including anaphylaxis. FDA currently lists peginesatide (Omontys) as discontinued. Epoetin beta is currently unavailable in the U.S. However, a methoxy pegylated (PEG) form of epoetin beta, called continuous erythropoietin receptor activator or CERA, has a prolonged half-life that permits once monthly dosing. PEG-epoetin beta was FDA approved in 2007 under the brand name Mircera. Mircera sales in the United States were prohibited from 2009 until 2015 due to a copyright infringement lawsuit, however, Hoffmann-La Roche is now manufacturing and supplying the drug to Galencia, and it is currently available. Epoetin alfa and epoetin beta have the same amino acid sequence as endogenous erythropoietin but differ from each other in glycosylation; clinical effects are considered interchangeable. Darbepoetin alfa is similar to endogenous erythropoietin but has 2 additional oligosaccharide chains. In contrast, peginesatide lacks any amino acid sequence homology to erythropoietin. It is a synthetic dimer of identical 21-amino acid peptides bound to a linker and to polyethylene glycol, with a total molecular weight of approximately 45,000 Da. (The molecular weight of endogenous erythropoietin is approximately 34,000 Da.) However, the epoetins, darbepoetin, and peginesatide all have pharmacologic actions similar to those of the endogenous hormone. Each binds to and activates the human erythropoietin receptor and thus increases the number of RBCs and the blood concentration of Hb, when given to individuals with functioning erythropoiesis. Both brands of epoetin alfas, PEGepoetin beta, and darbepoetin alfa are FDA-approved to treat anemia in patients with CKD who are on dialysis or not on dialysis. Peginesatide is approved only for adult patients with anemia from CKD who are on dialysis. Epoetin alfa and darbepoetin alfa also are approved for other indications. Boxed Warning Epogen/Procrit and Aranesp has a boxed warning that ESAs increase the risk of death, myocardial infarction (MI), stroke, venous thromboembolism, thrombosis of vascular access and tumor progression or recurrence. 1-2 For patients with CKD, controlled trials have demonstrated that patients experienced greater risks for death, serious adverse cardiovascular (CV) reactions, and stroke when given ESAs to target a hemoglobin (Hb) level > 11.0 g/dl. No trial has identified a Hb target level, ESA dose, or dosing strategy that negates such risks. Use the lowest Epogen/Procrit dose necessary to reduce the need for RBC transfusions. For use in cancer, ESAs shorten overall survival and/or increase the risk of

6 tumor progression or recurrence in clinical studies involving patients with breast, non-small cell lung, head and neck, lymphoid, and cervical cancers. Prescribers and hospitals are required to enroll and comply with the ESA APPRISE Oncology Program to prescribe and/or dispense Epogen/Procrit to those with cancer. Use the lowest dose needed to avoid RBC transfusions. Use ESAs only for anemia due to myelosuppressive chemotherapy. ESAs are not indicated for patients receiving myelosuppressive chemotherapy when the anticipated outcome is cure. Discontinue ESAs after the completion of a chemotherapy course. In surgery patients, deep vein thrombosis (DVT) prophylaxis is recommended due to the increased risk of DVT. 1-2 Rationale This evidence review was developed in 1995 and updated periodically with a literature review of the MEDILNE database. The most recent literature review was performed for the period through November 13, Primary data sources for oncology included a 2006 comparative meta-analysis on the outcomes of epoetin or darbepoetin for managing anemia in patients undergoing cancer treatment, which was prepared for the Agency for Healthcare Research and Quality (AHRQ), and the 2005 AHRQ report, updated in ; a meta-analysis using individual patient data for outcomes of erythropoiesisstimulating agent (ESA) therapy in patients with cancer, 5,6 with additional outcomes reported in ; American Society of Clinical Oncology/American Society of Hematology (ASCO/ASH) 2010 clinical practice guidelines on the use of epoetin and darbepoetin to treat chemotherapy-associated anemia 8 ; 2007 briefing documents available from the U.S. Food and Drug Administration (FDA) Oncologic Drugs Advisory Committee (ODAC); and a 2007 Decision Memorandum from the Centers for Medicare and Medicaid Services on the use of ESAs for nonrenal disease indications. Information on the use of ESAs in chronic renal failure (CRF) was obtained from several sources including 2007 briefing documents from a joint meeting of FDA s Cardiovascular and Renal Drugs Advisory Committee (CRDAC) and Drug Safety and Risk Management Advisory Committee (DSRMAC) to reassess ESA risks; and, a meta-analysis of blood hemoglobin (Hb) targets for patients with CRFassociated anemia. 11 FDA-approved labels for ESAs available in the United States comprised additional data sources for this evidence review, in particular, recommended dosing information for the different clinical settings covered. The 2010 ASCO/ASH clinical practice guideline for the use of ESAs considers epoetin and darbepoetin, used at dosages recommended in current FDA-approved package inserts, to be equivalent with respect to effectiveness and safety. Epoetin and darbepoetin are identical with respect to: (1) indications for use in chemotherapy-induced anemia, (2) Hb limits for adjusting doses, initiating or discontinuing treatment, (3) warnings and cautions to consider, and (4) increased rates of thromboembolic events in the experimental arms of separate trials on each product versus controls/placebo. 8 Chronic Kidney Disease and Anemia Epoetin Alfa, Epoetin Beta, and Darbepoetin At initial approval of epoetin in 1989, the primary objective of treatment was to raise Hb concentration sufficiently to avoid transfusion, with a target range of 9 to 10 g/dl in anemic patients with chronic kidney disease (CKD). The first National Kidney Foundation Kidney Disease Outcomes Quality Initiative (NKF-KDOQI) guidelines in 1997 recommended an Hb concentration of 11 g/dl, a level that was increased by subsequent NKF-KDOQI anemia guidelines, to 11 to 13 g/dl in With increased experience in the use of ESAs, it became unclear whether higher Hb target concentrations, including normalization, would yield additional benefits, in particular, physical function and improved cardiovascular outcomes. Clinical doubts increased with publication of the first large randomized controlled trial (RCT) of Hb normalization using epoetin alfa in hemodialysis patients (Normal Hematocrit Cardiac Trial [NHCT]). 17 NHCT showed a trend toward increased mortality risk and significantly increased risk for vascular access thrombosis with ESA treatment to a hematocrit (Hct) target of 42%. Subsequently, 4 published RCTs in hemodialysis patients with end-stage renal disease

7 (ESRD) and 8 in nondialysis patients with CKD found improved physical function at higher Hb targets, but none demonstrated significant improvements in cardiovascular end points or mortality. 18 The Epogen/Procrit labeling was modified in 1996 to include results of the NHCT study that showed a higher mortality rate for anemic dialysis patients randomized to an Hct of 42%, compared with an Hct of 30%. In 2006, the CHOIR study reported worse cardiovascular outcomes for anemic CRF patients who were not undergoing dialysis and who were randomized to a target Hb of 13.5 g/dl, compared with an Hb of 11.3 g/dl. 19 Subsequent analyses of outcomes in CHOIR showed shorter times to progression of kidney disease and higher rates of renal replacement therapy and death among patients randomized to the higher Hb target. 20 The CREATE study, also reported in 2006, was similar to CHOIR but enrolled fewer patients. 21 CREATE did not demonstrate statistically significant differences in adverse cardiovascular outcomes for the higher Hb group, but the general trend of major cardiovascular outcomes was similar to the CHOIR findings. The 2009 TREAT study randomized 4038 patients with type 2 diabetes mellitus, Hb of 11 g/dl or less, and CKD not on dialysis. 22 Patients in 1 arm were treated with darbepoetin to a target Hb of 13 g/dl, and those in the other arm received darbepoetin only if Hb fell below 9 g/dl. Risks for 2 end points were not significantly different between arms: death or a cardiovascular event (hazard ratio [HR], 1.05; 95% confidence interval [CI], 0.94 to 1.17; p=0.41) and death or ESRD (HR=1.06; 95% CI, 0.95 to 1.19; p=0.29). However, fatal or nonfatal stroke was significantly increased among patients randomized to the higher Hb target (HR=1.92; 95% CI, 1.38 to 2.68; p<0.001). Multivariate analysis found no statistically significant relationship of increased stroke risk to any baseline characteristic; to effects on blood pressure, Hb, or platelet count; or to darbepoetin dose. 23 A 2012 meta-analysis by Vinhas et al included only large RCTs (N>500) with a minimum duration of 1 year. 24 Outcomes of interest were vascular access thrombosis, stroke, progression to ESRD, and allcause mortality. Five trials (7902 patients), including the CHOIR, CREATE, NHCT, and TREAT trials, were identified. Mean or median duration of follow-up ranged from 14 to 36 months. In 2012, the American Society of Nephrology released its evidence-based recommendations for the Choosing Wisely campaign to improve patient care and resource use. 25 Citing the evidence reviewed, the society included the following among its top 5 recommendations: Do not administer erythropoiesisstimulating agents to CKD patients with hemoglobin levels 10 g/dl without symptoms of anemia. A 2014 Cochrane review included 8 trials (total N=2051) that compared darbepoetin with epoetin (alfa or beta) in adults with anemia due to CKD. 26 No statistically significant differences were observed in random effects meta-analyses of final Hb or mean change in Hb level, overall mortality, cardiovascular events or cardiovascular mortality, blood transfusions, or adverse events due to hypertension or vascular access thrombosis. Risk of bias was rated as moderate to high, and statistical heterogeneity was minimal (I2=0%) for all outcomes. Pegylated Epoetin Beta FDA s 2007 approval of PEG-epoetin beta (Mircera) was based on 6 phase 3, international, open-label, RCTs in patients with anemia due to CKD (see Table 2). In 2 trials (total N=505), patients were not receiving ESA therapy (correction trials), and in 6 trials (total N=1894), Hb was stable on maintenance ESA therapy (maintenance trials). All but1 trial (ARCTOS) enrolled dialysis-dependent patients. The primary efficacy outcome in all trials was maintenance of Hb levels over 24 to 52 weeks, adjusted for baseline Hb and center, in the intention-to-treat and per-protocol patient samples. For this outcome, the trials demonstrated noninferiority of PEG-epoetin beta once or twice monthly to epoetin (alfa or beta) 1 to 3 times weekly (AMICUS, MAXIMA, PROTOS, RUBRA) and to darbepoetin weekly or twice monthly (ARCTOS, STRIATA). In the correction trials (ARCTOS, AMICUS), median time to response was longer in the PEG-epoetin beta groups (43 days and 57 days, respectively) compared with the darbepoetin (29 days) and epoetin (31 days) groups. Although target Hb ranges in these trials included levels that have since been associated with increased mortality in CKD (ie, >11 g/dl), 27 FDA s summary review of safety (based on 1789 PEG-

8 epoetin beta treated patients [64% for >1 year] and 948 ESA-treated patients) reported that mortality was similar between the 2 groups (10% vs 11%, respectively). Incidence of serious adverse events also was similar between groups (37% vs 40%, respectively), although serious bleeding events (5.2% vs 4%), serious gastrointestinal bleeding events (1.2% vs 0.2%), and thrombocytopenia less than platelets/l (7.5% vs 4.4%) occurred more commonly in PEG-epoetin beta treated patients. FDA reviewers attributed these imbalances to the greater proportion of patients on hemodialysis in the PEG-epoetin beta group (84% vs 80%), and considered the risks of hemorrhage and thrombocytopenia similar to or slightly increased above that for other ESAs. Trials excluded patients with poorly controlled hypertension; 27% of enrolled patients required increases in antihypertensive therapy. Since FDA approval, subsequent short-term trials (24-40 weeks; total N=841) have replicated the findings of the pivotal correction trials in patients on hemodialysis 37 and not on hemodialysis,38,39 and of the pivotal maintenance trials in patients on hemodialysis. 40,41 Of 324 nondialysis patients in the ARCTOS correction trial, 296 (91%) entered a 24-week extension study. 42 Patients who responded to PEG-epoetin beta biweekly (n=145) were rerandomized 1:1 to biweekly or monthly dosing to maintain Hb between 11 to 13 g/dl. Mean (SD) Hb levels were 11.9 (0.9) g/dl, 11.7 (0.9) g/dl, and 11.9 (1.0) g/dl in the PEG-epoetin biweekly, PEG-epoetin monthly, and darbepoetin (weekly or biweekly) groups (n=151), respectively. Within-patient variation in Hb levels was similar across groups. Peginesatide FDA approved Omontys (peginesatide) to treat anemia in CKD patients on dialysis in March 2012 based on 2 randomized active-controlled noninferiority trials, which were summarized in a TEC Specialty Pharmacy Report. The first trial, EMERALD-1, enrolled 803 patients in the United States, and controls received epoetin alfa (Epogen, Procrit). The second trial, EMERALD-2, enrolled 823 patients in the United States and Europe, and controls received epoetin alfa or epoetin beta (not available in the United States). Adults on dialysis for at least 3 months with stable Hb concentrations (between 10.0 g/dl and 12.0 g/dl) on ESA therapy for at least 8 weeks were eligible for randomization to peginesatide once monthly or continued epoetin 1 to 3 times weekly for 36 weeks. Results for the primary efficacy outcome (between-arm difference of the change from baseline Hb to the mean value during weeks 29 to 36 [the evaluation period], with a noninferiority margin of -1.0 g/dl) demonstrated the noninferiority of peginesatide in each trial (-0.15 g/dl in EMERALD-1, in EMERALD-2). The relative risk (RR) for red blood cell (RBC) transfusion also did not differ significantly between arms (RR=1.21; 95% CI, 0.76 to 1.92 in EMERALD 1; RR=0.79; 95% CI, 0.50 to 1.24 in EMERALD-2). Two other trials (PEARL-1 and PEARL-2; total N=656 randomized to peginesatide, N=327 to darbepoetin) were conducted for patients with CKD who were not on dialysis.43 The trials prospectively evaluated cardiovascular risk of ESAs. The 4 trials together were powered for a primary safety outcome, which was to rule out an increase of 30% or more in the risk of the composite safety end point, based on a 2-sided 90% CI. The composite safety end point comprised death, stroke, myocardial infarction, and hospitalization for congestive heart failure, unstable angina or arrhythmia. Incidence of the composite safety outcome did not differ significantly between groups randomized to peginesatide or active comparator (HR=1.06; 95% CI, 0.89 to 1.26). In an analysis limited to the 2 trials of patients on dialysis (EMERALD-1 and -2), the 2 groups again did not differ with respect to incidence of the composite safety end point (HR=0.95; 90% CI, 0.79 to 1.13). However, peginesatide significantly increased the incidence of this composite safety end point in a pooled analysis of the 2 trials for patients not on dialysis (PEARL-1 and -2; HR=-1.32; 90% CI, 1.02 to 1.72). Cardiovascular harms in the nondialysis population were considered unacceptably high and the indication was abandoned. The risk evaluation and mitigation strategy (REMS) communication plan developed by the manufacturer and approved by FDA was designed to inform health care providers who might prescribe peginesatide of these findings. Note also that thus far, no data are available on safety or efficacy of peginesatide for any other ESA indications (eg, patients with a non myeloid malignancy who are anemic while receiving palliative chemotherapy, HIV patients who are anemic while receiving zidovudine, or perisurgical patients unable to donate autologous blood).

9 On February 23, 2013, Affymax, Takeda, and FDA announced a voluntary recall of all lots of peginesatide due to post-marketing reports of serious hypersensitivity reactions, including anaphylaxis. Serious reactions occurred within 30 minutes of the first dose of peginesatide; serious reactions after subsequent doses have not been reported. Among approximately 25,000 patients who received peginesatide postapproval, the estimated overall incidence of hypersensitivity reactions was 2 per 1000, and the estimated incidence of fatal reactions was 2 per 10,000. FDA currently lists peginesatide (Omontys) as Discontinued on its website (Drugs@FDA). Comparative Efficacy of Different ESAs While there are RCTs that directly compare different ESAs, the evidence base on the comparative efficacy of different ESAs is incomplete. A Cochrane network meta-analysis was published in 2014 that attempted to use indirect comparisons via network meta-analysis to evaluate comparative efficacy. 45 This analysis included RCTs that compared one ESA with placebo, no treatment, or another ESA for the treatment of CKD. A total of 56 studies involving 15,596 patients were included in the review, the majority of which were judged to have high or uncertain risk of bias. The network meta-analysis did not allow conclusions on the comparative efficacy of different agents in preventing blood transfusions. There were very few studies that included patient-reported outcomes (eg, quality of life) and as a result, the evidence base was also insufficient to make conclusions on these other outcomes. Section Summary: Chronic Kidney Disease and Anemia Three ESAs are FDA-approved for use in patients with CRF: epoetin alfa, PEG-epoetin beta, and darbepoetin alfa. Placebo-controlled clinical trials have established that epoetin alfa and darbepoetin alfa effectively increase Hb concentrations and decrease the need for blood transfusions. Evidence does not support an improvement in other clinical outcomes such as mortality, morbidity, functional status, or quality of life (QOL). Some trials and a meta-analysis published in 2012 have reported increased cardiovascular events and/or increased mortality in patients treated with ESAs. These trials generally have treated to an Hb of 12 g/dl or higher. The optimal target Hb is unclear, and it is uncertain whether treating to lower Hb levels avoids the increase in adverse events. Both peginesatide and PEG-epoetin beta have been compared with other ESAs in randomized trials. Peginesatide has shown noninferiority to epoetin for adult patients with CRF on dialysis. There are no trials reporting benefit for peginesatide for other indications or in pediatric patients with kidney disease. Currently, peginesatide is unavailable and should not be used. PEG-epoetin beta has shown noninferiority to epoetin and darbepoetin for correcting or maintaining Hb levels in RCTs of patients on dialysis or not on dialysis. In meta-analyses of trials involving dialysis patients, no statistical differences were reported in overall mortality, blood transfusions, or adverse events due to hypertension or venous access thrombosis. Evidence on comparative effectiveness of the different agents is lacking. A Cochrane network meta-analysis was not able to make any conclusions on comparative effectiveness due to limited comparative evidence. Cancer-Related Anemia Epoetin Alfa, Epoetin Beta, and Darbepoetin In 1993, FDA approved Procrit/Epogen (epoetin alfa) to treat anemia in patients receiving cancer chemotherapy based on data from 2 multicenter randomized placebo-controlled, double-blind clinical trials (1 enrolled 344 adult patients and the second enrolled 222 pediatric patients), and an additional pooled analysis of 6 smaller double-blind RCTs enrolled a total of 131 patients. Patients in all 3 studies received at least 12 weeks of concurrent chemotherapy and were randomized (1:1) to receive Procrit/Epogen or placebo subcutaneously for 12 weeks. Overall, the data showed a reduction in the proportion of patients requiring blood transfusion during the second and third months of epoetin treatment. The approval of Aranesp (darbepoetin alfa) in 2002 for the treatment of anemia associated with cancer chemotherapy was based on demonstration of a significant reduction in the proportion of patients transfused during chemotherapy from week 5 through the end of treatment. Study , a phase 3, double-blind, placebo-controlled randomized (1:1) multicenter, multinational trial of darbepoetin alfa enrolled 314 anemic patients with previously untreated non-small-cell or small cell lung cancer receiving at least 12 weeks of platinum-containing chemotherapy.

10 After the first approval of an ESA for treatment of chemotherapy-associated anemia in 1993, additional data became available regarding increased risks of mortality and possible tumor promotion from the use of ESAs. Increased mortality has been observed in patients with cancer (BEST, ENHANCE, , EPO-CAN-20 studies) when ESA treatment strategies were designed to achieve and maintain Hb levels above 12 g/dl. 8 In addition, ESA treatment strategies intended to achieve and maintain Hb levels above 12 g/dl have demonstrated poorer tumor outcomes (BEST, ENHANCE, DAHANCA studies). More recently, a 2009 meta-analysis using individual patient data on 13,933 subjects from 53 RCTs reported significantly greater on-study mortality (HR=1.17; 95% CI, 1.06 to 1.30) and poorer survival to end of follow-up (HR=1.06; 95% CI, 1.00 to 1.12), with little heterogeneity between trials. 5,6 Results were qualitatively similar when the analysis was limited to 10,441 patients receiving concurrent chemotherapy in 38 trials, and there was little evidence for a difference between trials of patients receiving different chemotherapy regimens. Data from multiple trials, consistent with data presented to ODAC in May 2004, led to revised product labeling with broader and more detailed warnings against ESA treatment strategies targeting Hb levels above 12 g/dl. More recent data, including the individual patient data meta-analysis summarized earlier, 5,6 suggested that factors such as the planned Hb ceiling for stopping ESA therapy had little influence on increased mortality resulting from ESA treatment. Although risks of Hb targets greater than needed to avoid transfusions are now well-established, data from adequate, well-controlled studies employing recommended ESA doses and Hb targets are as yet insufficient to assess effects on survival or tumor promotion. The only data provided to FDA which used the recommended dose and Hb target was from Amgen Study , which demonstrated significantly shorter survival in cancer patients receiving ESAs compared with those supported by transfusion alone. However, this study was not adequately designed to assess effects on tumor promotion or on thrombotic risks. Despite these caveats, data from available studies were sufficient for FDA to reassess the safety of ESAs in patients with cancer and to re-evaluate the net clinical benefit of ESAs in this setting. Results of the updated AHRQ comparative effectiveness review (2013) were consistent with those reported in Among patients receiving chemotherapy and/or radiotherapy for malignancy, use of ESAs to treat anemia reduced the risk of transfusion and increased the risk of thromboembolic events and on-study mortality. Both thromboembolic events and on-study mortality were reduced (but not eliminated) when ESA treatment was initiated at Hb less than 10 g/dl. Although the reviewed evidence incorporated higher baseline and target Hb levels than those currently recommended, sensitivity analyses suggested that these findings were robust. QOL, as assessed by the Functional Assessment of Cancer Therapy (FACT) fatigue scale, was improved in patients receiving ESAs, but the magnitude of improvement was less than the minimal clinically important difference of 3 points. Fifteen included trials did not support an association between ESA use and tumor response or progression; metaanalysis was not possible due to varying outcome definitions. A 2014 meta-analysis examined the incidence of thromboembolic events in patients with solid and hematologic cancers who received ESAs, and found a similar result. 46 Gao et al pooled 51 RCTs (12,115 patients) and reported a 75% increased odds of thromboembolic events among patients receiving ESAs (pooled odds ratio, 1.75; 95% CI, 1.50 to 2.05; I2=0%). Pegylated Epoetin Beta PEG-epoetin beta is not FDA-approved for anemia due to cancer chemotherapy, and Hoffmann- LaRoche, manufacturer of PEG-epoetin beta, has not sought this indication. A 2010 phase 2, openlabel RCT by Gascon et al compared 3 doses of subcutaneous PEG-epoetin beta with subcutaneous darbepoetin in 153 patients who were receiving first-line chemotherapy for stage 3B or 4 non-small-cell lung cancer. 47 Baseline Hb at screening was 11 g/dl or less. PEG-epoetin beta was administered every 3 weeks, and darbepoetin was administered weekly or every 3 weeks. The primary efficacy outcome, mean change from baseline Hb during weeks 5 to 13, did not differ between groups and indicated inadequate treatment responses in all groups (0.17 g/dl and 0.26 g/dl in the PEG-epoetin beta and

11 darbepoetin groups, respectively). At week 12, the trial was terminated due to more deaths in the 3 PEG-epoetin beta groups compared with the darbepoetin group (29 [25%] of 114 patients vs 4 [10%] of 39 patients, respectively). Post hoc analyses did not convincingly demonstrate that baseline imbalances accounted for the mortality difference. Section Summary: Cancer-Related Anemia Epoetin alfa and darbepoetin alfa are approved for patients with anemia associated with concurrent cancer chemotherapy. These ESAs effectively increase Hb concentrations and decrease the need for blood transfusions in patients with anemia caused by cancer chemotherapy. The evidence does not support an improvement in other clinical outcomes such as mortality, morbidity, functional status, or QOL. Some trials have reported higher thromboembolic events and/or mortality in cancer patients treated with ESAs, and 2 meta-analyses published in 2012 and 2013 also reported increases in mortality and thromboembolic events. Trials that reported increased adverse events have generally treated to an Hb of 12 g/dl or higher, and adverse events appear to be correlated with higher treatment targets. However, it is unclear whether treating to a lower Hb reduces or eliminates these adverse events. These concerns over potential harm from ESAs have led FDA to reassess the risk/benefit ratio and to modify the labeled indications. Current FDA labeling recommends against starting ESA therapy in a cancer patient whose Hb exceeds 10 g/dl. PEG-epoetin beta is not FDA approved for patients with anemia due to cancer chemotherapy. A phase 2 RCT demonstrated increased mortality among patients with advanced non-small-cell lung cancer who received PEG-epoetin beta compared with those who received darbepoetin. Hepatitis C Infection Treated With Ribavirin Related Anemia Standard treatment for hepatitis C infection includes ribavirin. Anemia related to ribavirin use often is the limiting step in treatment. Options for treatment of ribavirin-related anemia are reduction in the dose of ribavirin and use of ESAs and/or blood transfusions as needed. However, a reduction in ribavirin dose has been associated with less favorable response rates, and therefore, some experts prefer using ESAs to maintain full-dose ribavirin. Evidence on the benefit of using ESAs for this purpose comprises several RCTs, some of which are reviewed next. At least 2 controlled trials randomized patients with hepatitis C virus and ribavirin-related anemia to epoetin alfa or usual care. The larger of these was conducted by Afdhal et al (2004). 48 This trial included 185 patients with an Hb level of 12 g/dl or less who received 8 weeks of epoetin alfa at a dose of 40,000 units weekly. Outcomes included the proportion of patients who were able to maintain fulldose treatment with ribavirin, mean Hb level, and QOL as measured by 36-item Short-Form Health Survey. More patients in the epoetin group (88%) than in the usual care group (60%, p<0.001) were able to maintain full-dose ribavirin. Increase in mean Hb level also was higher in the epoetin group (2.2 g/dl) than in the usual care group (0.1 g/dl, p<0.001). Improvement in QOL was significantly greater for the epoetin group on 7 of 8 domains, with incremental improvement ranging from 1.3 to 10.0 for patients on epoetin. A second RCT by Dieterich et al (2003) 49 was similar to the Afdhal trial. Dieterich et al enrolled 64 patients with hepatitis C and ribavirin-related anemia, as defined by Hb less than 12 g/dl. Patients were followed for 16 weeks and treated with epoetin alfa 40,000 units weekly. Primary end points were ribavirin dose and Hb level. Mean ribavirin dose decreased less in the epoetin group (-34 mg/d) than in the usual care group (-146 mg/d), but this difference was not statistically significant (p=0.06). More patients in the epoetin group (83%) than in the usual care group (54%, p=0.02) were able to maintain full-dose ribavirin. Mean Hb level was higher in the epoetin group (13.8 g/dl) than in the usual care group (11.4 g/dl, p<0.001). A third RCT by Shiffman et al (2007) evaluated ESAs for anemia in patients with hepatitis C who were treated with ribavirin. 50 This trial randomized 150 patients to 3 groups at the onset of treatment: (1) ribavirin at standard dose; (2) ribavirin at standard dose plus epoetin alfa; and (3) ribavirin at higher dose plus epoetin alfa. Primary end points were reduction in ribavirin dose and the proportion of patients with a sustained virologic response (SVR). Fewer patients treated with epoetin required dose

12 reduction (10%) compared with patients not treated with epoetin (40%, p<0.05), but the proportion of patients with SVR did not differ between groups. Section Summary: Hepatitis C Infection Treated With Ribavirin Related Anemia RCTs of ESAs versus placebo for patients with hepatitis C and ribavirin-related anemia have demonstrated that use of ESAs can improve Hb levels and allow more patients to maintain treatment at full ribavirin doses. One RCT also reported improvement in QOL for patients treated with ESAs. Improvements in these parameters may lead to health outcome benefits, although no study has reported an improvement in clinical outcomes such as SVR or survival. Postapproval FDA Regulatory Actions In November 2006, FDA issued a Public Health Advisory regarding the serious cardiovascular risks from ESA therapy in patients with CKD, as evidenced in the CHOIR and NHCT studies. Subsequently, FDA received reports of increased risks associated with ESAs used to treat anemia in cancer patients who were receiving or not receiving chemotherapy, as well as a report of thrombotic risks in patients receiving ESAs in the perisurgical setting. These data prompted reassessment of the safety information contained in the labeling for Aranesp, Epogen, and Procrit and culminated in the approval of revised labels on March 9, Product labels have been revised and updated subsequently, most recently in December Regarding dosage information, periodic reassessment of ESA safety has determined that clinical data do not support a therapeutic Hb target free of risk for mortality. Consequently, revised Dosage and Administration sections of the product label deleted any specific therapeutic Hb or Hct "target" range for ESAs. Instead, revised labels recommended that prescribers use the lowest ESA dose that will gradually increase Hb concentration to the lowest level sufficient to avoid the need for RBC transfusion. For anemic CRF patients, this recommendation was primarily based on the NHCT and CHOIR study findings, as well as the lack of data for any specific Hb or Hct threshold or range. Clinical data did not identify specific Hb or Hct levels that directly correlated with a " reduction in the need for red blood cell transfusion," the main treatment benefit supporting ESA efficacy. The March 2007 label revision allowed prescribers to use their clinical judgment in determining the " lowest level sufficient to avoid the need for red blood cell transfusion." On November 8, 2007, FDA revised the product labeling for epoetin alfa and darbepoetin alfa These revisions clarified the evidence for safety and effectiveness of these products and provided more explicit directions and recommendations for their use. These recommendations were consistent with those made during the May 10, 2007 ODAC and the September 11, 2007 CRDAC and DSRMAC meetings. Revisions included strengthened boxed warnings and Warnings and Precautions sections, and changes to the Indications and Usage, Clinical Trials Experience, and Dosage and Administration sections of the product labels. Product labels for Epogen/Procrit and Aranesp have been revised many times since then. The revised black box warnings and limitations of use shown next reflect current labeling for these ESAs. Although the Mircera product label has not been updated since 2007, Warnings for use in CRF are similar to those listed next. Cancer ESAs shortened overall survival and/or increased the risk of tumor progression or recurrence in clinical studies of patients with breast, non small cell lung, head and neck, lymphoid, and cervical cancers. Because of these risks, prescribers and hospitals must enroll in and comply with the ESA APPRISE Oncology Program to prescribe and/or dispense an ESA to patients with cancer. To decrease these risks, as well as the risk of serious cardiovascular and thromboembolic reactions, use the lowest dose needed to avoid RBC transfusions. Use ESAs only for anemia from myelosuppressive chemotherapy. ESAs are not indicated for patients receiving myelosuppressive chemotherapy when the anticipated outcome is cure. Discontinue use after the completion of a chemotherapy course.