Return to American College of Clinical Pharmacy                                     Print This
Oncology-Associated Anemia: Optimizing the Value of Erythropoietic Growth Factors 

Introduction

Oncology-associated anemia affects more than half of all treated patients. Some data indicate this figure is as high as 90%. Two clinical practice guidelines regarding the use of erythropoietic growth factors for this indication are available from the American Society of Hematology (ASH)/American Society of Clinical Oncology (ASCO), and the National Comprehensive Cancer Network (NCCN). They indicate that hemoglobin (Hb) levels of 10 g/dL or less should trigger the use of this therapy, and that “individual clinical circumstances” should guide treatment of patients with Hb levels between 10 and 12 g/dL (Rizzo et al. J Clin Oncol 2002;20:4083-4107).  

Over 80% of oncologists believe this condition is undertreated by their peers. Program Chair Thomas L. Rihn, PharmD, Associate Professor of Clinical Pharmacy at the Duquesne University School of Pharmacy, noted that patients with mild-to-moderate forms of anemia are especially neglected. “We have a tendency to focus on the more severe forms of anemia,” he commented, “even though the treatment of lower grades of anemia has been shown to improve fatigue, cognitive functioning, and other quality of life factors rather dramatically.”  

Among a number of anemia-related symptoms, fatigue is one of the most significant, according to data from the NCCN. Anywhere from 70% to 100% of cancer patients receiving therapy experience fatigue to an extent that it compromises their quality of life and restricts their activities. Treatment for patients with anemia begins with identification and correction of any underlying causes, such as nutritional deficiencies or bleeding. Before the availability of erythropoietic growth factors, red blood cell (RBC) transfusion was the only treatment available. Although RBC transfusion remains an option, significant risks and concerns, including adverse events (e.g. fever, urticaria, volume overload, blood-borne infections, and blood shortages) make pharmacologic intervention a preferred therapy.  

The two available erythropoietic growth factors, epoetin alfa (Procrit® ) and darbepoetin alfa (Aranesp®), have similar indications and modes of action. Epoetin alfa is characterized by 3 N-linked carbohydrate (CHO) chains, whereas darbepoetin alfa has an increased carbohydrate content, with 5 N-linked carbohydrate chains. This increased carbohydrate content extends darbepoetin’s half-life three-fold. This increased half-life provides the potential to maintain minimum effective concentrations over longer periods of time and further extend dosing intervals. Darbepoetin has less erythropoietin receptor binding than epoetin, but greater overall biological activity. The clinical significance of these differences is not known. But Dr. Rihn commented that “...the
predominant determinant of response, and ultimately the increase in hemoglobin, is really the half-life of the specific growth factor.” 

How Erythropoietin Growth Factors are Studied Affects Results
Data emerging from clinical trials with erythropoietic growth factors vary in ways that can confound a direct comparison of results. Many trials with these agents exclude a large percentage of enrolled patients from final analyses of results. The preferred type of analysis is an intention-to-treat (ITT) analysis, in which all patients in the original cohort of enrolled patients are included in the final analysis. “Completer,” or treated, analyses include only patients who finish the study. Because this type of analysis includes only a proportion of the original group, the clinical response (in this case Hb response), is
likely to be higher.
 
Another variable in data analysis
affecting the interpretation of results is the timing of the clinical surrogate response — whether early (4-6 weeks) or late (at study end). The effect of RBC transfusion on Hb must also be considered in the data analysis. Dr. Rihn explained. “Significant numbers of patients with moderate to severe forms of anemia will receive a transfusion. The transfusion will boost hemoglobin concentrations for up to 28 days—the life of the red cells.” This must be factored into the analysis, “...otherwise, some of the early rise in hemoglobin may be attributed to the action of the drug, when it is in fact due to the transfusion.” The FDA issued a letter in June of 2003 warning investigators and clinicians to use caution when interpreting and comparing study results of erythropoietic growth factors. The letter stated that an early 1 g/dL increase in Hb at 28 days could be misleading if the transfusion effect is not adjusted for, and the FDA further advised authors to disclose whether data are generated via ITT or completer analyses.

Clinical Evidence and Practice Standards

Clinical practice standards have been developed from both large randomized, controlled and open-label community-based studies of erythropoietic growth factors in cancer care. Three open-label studies which together enrolled over 7,700 patients demonstrated the safety and efficacy of epotein alfa, according to speaker Jill M. Kolesar, PharmD, FCCP, BCPS, Associate Professor of Pharmacy at the University of Wisconsin. Glaspy et al, with the Procrit Study Group (J Clin Oncol 1997;15(3): 1218-34), evaluated the impact of therapy with epoetin alfa on clinical outcomes in patients with non-myeloid malignancies during cancer chemotherapy in community oncology practice. Demetri et al (J Clin Oncol 1998;16(10):3412-25), presented results confirming a quality of life benefit independent of disease response or tumor-type with epoetin alfa. Gabrilove and colleagues (J Clin Oncol 2001;19(11): 2875-82) established that once-weekly epoetin alfa given to anemic cancer patients provided improvements in hemoglobin and quality of life similar to three-times weekly dosing. Each study resulted in a statistically significant mean increase in Hb from baseline as well as decreases in transfusion requirements. It is important to note, however, that in the Gabrilove study and other epoetin trials, transfusion effect was not adjusted for and completer analyses were used in assessing Hb response and study-end hematopoietic response. Somewhat similar results were reported in non-US studies with rHuEPO (Quirt
et al. J Clin Oncol 2001;19(2):4126-34. Littlewood et al. J Clin Oncol 2001; 19(11):2865-74).  

In a systematic review and meta-analysis of 22 trials (n = 1927) to evaluate epoetin treatment of anemia associated with cancer therapy, Seidenfeld et al (J Natl Cancer Inst 2001;93(16):1204-14), concluded that epoetin treatment reduces the odds of transfusion in anemic cancer patients undergoing therapy, but that evidence was insufficient to determine whether earlier initiation of epoetin, when Hb is higher, spares more patients from transfusion or results in a better quality of life versus waiting until Hb concentrations have declined to nearly 10 g/dl.  

Dr. Kolesar summarized what is known about epoetin alfa; namely, thatit improves Hb, decreases transfusion requirements, improves quality of life,
and is well tolerated. What is not known, she said, is its impact on survival, the ideal target Hb, the optimal treatment schedule (weekly, biweekly, Q3 weeks, or monthly), and possible additional effects such as neuroprotection. 

Darbepoetin Alfa
Darbepoetin alfa was approved by the FDA in July of 2002 for the treatment of anemia in patients with non-myeloid malignancies, where the anemia is due to the effect of concomitantly administered chemotherapy. Five phase I/II studies established the dosing range of between 2.5 mcg/kg/wk to 9 mcg/kg/wk, and clinical responses from darbepoetin alfa treatment in anemic patients with solid tumors receiving chemotherapy, and in patients with chronic anemia (Hedenus M, et al. Br J Haematol. 2003;122(3): 394-403. Smith RE Jr, et al. Br J Cancer. 2003;88(12):1851-8. Hedenus M, et al. Br J Haematol. 2002; 119(1):79-86. Kotasek D, et al. Eur J Cancer. 2003;39(14): 2026-34.) 

In a phase III study, Vansteenkiste et al, for the Ananesp 980297 Study Group, compared the effectiveness of once-weekly darbepoetin alfa versus placebo in the treatment of anemia in subjects with lung cancer receiving multi-cycle, platinum-based chemotherapy (J Natl Cancer Inst 2003;21(94):1211-20). In this study, which enrolled 320 anemic patients, those with a Hb of 11 g/dl or less were administered weekly darbepoetin injections (2.25 mcg/kg/wk) for 12 weeks, or placebo. The 297 patients who completed at least the first 28 days of study were assessed for red blood cell transfusions during weeks 5-12, the primary endpoint. Patients were also evaluated for hematopoietic response, adverse events, antibody formation to darbepoetin alfa, hospitalizations, Functional Assessment of Cancer Therapy (FACT)-Fatigue score, and disease outcome.  

Transfusion effect was adjusted for and an ITT data analysis was performed. Results from this study demonstrated that darbepoetin alfa-treated patients required statistically fewer transfusions (27% versus 52%; p <.001), more hematopoietic responses (66% versus 24%; p <.001) and better improvement in FACT-Fatigue scores (56% versus 44% overall improvement, p = .019), versus placebo. Patients receiving darbepoetin alfa did not develop antibodies to the drug. Adverse events were similar between the treatment and placebo groups.

The SOAR Trial evaluated the use of darbepoetin alfa in anemic cancer patients with non-myeloid malignancies and Hb levels of 11 g/dl or less (Vadhan-Raj S, et al. J Support Oncol 2003;1:131-8.), Patients (n = 1,173) were started on darbepoetin alfa 3 mcg/kg every 2 weeks. If at week 7, the patients did not experience a hemoglobin increase of at least 1 g/dl, the dose was increased to 5 mcg/kg every 2 weeks. If the Hb increase met this threshold at week 7, the original 3 mcg/kg dose was continued. The study was concluded with a final dose at week 15, and patients were seen for a follow-up at week 17, two weeks after their final dose of darbepoetin alfa. Transfusion effect was adjusted for in this trial and an ITT data analysis was used. At week 17, 43% of patients underwent the mandated dose-escalation; 71% achieved a hemoglobin response of 2 g/dl or greater; and 84% had a hematopoietic response rate of 2 g/dl or greater and/or Hb of 12 g/dl or higher.  

In a comparison and summary of data supporting the use of epoetin alfa versus darbepoetin alfa for the treatment of anemia in cancer patients, Dr. Kolesar commented that darbepoetin alfa is comparable to epoetin alfa in terms of increasing Hb, decreasing transfusion requirements, and improving quality of life. Both agents are well tolerated to a comparable degree. Dosing differs, however, with weekly dosing of epoetin alfa and Q2 or 3 week dosing with darbepoetin alfa, a factor that may influence the choice of agents due to the greater convenience and coordination with commonly used chemotherapy regimens.

Novel Dosing Regimens and Emerging Clinical Outcomes

Another way of assessing the usage and effectiveness of erythropoietic growth factors is to conduct a medical use evaluation (MUE) and assess other large clinical trial-based MUEs, according to speaker Philip D. Hall, PharmD, FCCP, BCPS, BCOP, Associate Professor from the College of Pharmacy and Hollings Cancer Center at the Medical University of South Carolina in Charleston. Compared with the standard design of a clinical trial, which involves a restricted population, the MUE is a “real world” look at how the agents are being used and the associated outcomes. A MUE is observational, and provides outcomes that are more generalizable than those resulting from a clinical trial. A MUE minimizes bias via consecutive patients, whereas a clinical trial minimizes bias through randomization. Critical to maximizing the strength and power of data emerging from a MUE study, Dr. Hall emphasized, are prospectively defined data elements and data analysis to ensure data integrity. 

Dr. Hall described a large multicenter MUE involving 16 sites throughout the USA, including 2 university hospitals, 1 community-based health system, 2 large community hospitals, and 11 community oncology clinics. Enrolled patients were adults with cancer receiving concurrent chemotherapy. Physicians determined which erythropoietic agent to use as well as the dose. Patient groups (those treated with epoetin alfa and those treated with darbepoetin alfa) were balanced in terms of tumor type, with approximately equal representation of breast, lung, GI, lymphoma, and other cancers. An interesting finding in this MUE was a dose escalation rate of 14% for epoetin and 11% for darbepoetin. These “real world” dose escalation rates are significantly lower than those reported in clinical trials. Other data from this study emphasize the similarity of responses between the two drugs. 

Looking at how the benefits of these agents could be improved, Dr. Hall suggested that the long-half-life of darbepoetin alfa may be exploited to permit administration concurrently with the most commonly used Q3 week chemotherapy administration schedules.

This possibility was investigated in an international, randomized, double-blind, placebo-controlled, dose-finding study of darbepoetin alfa administered SC every 3 weeks (Eur J Cancer 2003;39:2026-34). Patients had solid tumors, were receiving cyclic chemotherapy, and had Hb of 11 g/dl or less, transferrin saturation greater than 15% and ferritin higher than 10 mcg/L. Darbepoetin alfa was administered on a dose-escalation schedule of 4.5, 6.75, 9, 12, and 15 mcg/kg every 3 weeks for 12 weeks, or placebo administered Q3-weeks.  

Results showed a 51% hematopoietic response rate at a darbepoetin alfa dose of 4.5 mcg every 3 weeks. A slight dose-response relationship for Hb endpoints between the 4.5 and 12 mg/kg dose levels was seen, Dr. Hall reported. For transfusion-based endpoints, no dose-response relationship was observed. Mean change in FACT-Fatigue scores from baseline to end of the treatment showed a statistically significant improvement in patients
receiving darbepoetin alfa, especially notable when patients’ Hb level rose by more than 2 g/dl. Side effects were minimal even at higher doses. An analysis of serum concentrations of drug revealed no accumulation over the course of the study. 

The same extended dosing interval, but with a front-loading dose, was studied with epoetin alfa in an open-label pilot trial involving 20 patients (Proc ASCO 2003;22:754 Abstract 3033). Chemother-apy-treated patients with non-myeloid malignancies and anemia received epoetin 60,000 units SC weekly for up to 8 weeks. If Hb increased to greater than 1.3 g/dl in any 2-week period, the dose was reduced to 40,000 units. The maintenance dose was 120,000 units every 3 weeks. Because of a greater than 1.3 g/dl increase in hemoglobin, 35% of patients had the loading dose of 60,000 units SC reduced. At week 13 and at week 24 (at early withdrawal measurements), patients on the maintenance dose of epoetin alfa had 13.1 and 13.3 g/dl hemoglobin levels. The investigators of this small study concluded that larger-scale trials to evaluate this dosing regimen are needed. 

In reviewing the dose-escalation trial of weekly darbepoetin alfa, a dose-response relationship was evident in patients receiving weekly darbepoetin alfa up to doses of 4.5 mcg/kg per week. At this dose, the Hb response rate was 76% and median time-to-response was 7 weeks, versus10 weeks for patients receiving 2.25 mcg/week.  

In a pilot trial of front-loading darbepoetin alfa (N = 127), published in 2003 (Cancer 2003;97:1312-21), four patient groups were randomized to one of three darbepoetin alfa doses, or epoetin alfa.

Efficacy was measured by mean change in Hb, proportion of patients who achieved a hemoglobin response, the time to response, and the mean change in FACT-Fatigue scores. After 4 weeks of treatment, the mean change in Hb concentration was 0.53 g/dl, 0.70 g/dl, and 0.90 g/dl in darbepoetin alfa groups 1-3 respectively, and 0.39 g/dl in the epoetin alfa group. According to the author, Dr. Glaspy, the early erythropoietic response in darbepoetin-treated patients was associated with an early and maintained reduction in patient-reported fatigue. The adverse event profile was comparable with all darbepoetin doses and epoetin alfa. The authors concluded that when given in a front-loaded dose, followed by lower and/or less frequent doses, darbepoetin alfa appears to be efficacious “and may decrease the time to response relative to treatment with epoetin alfa.” 

Why do some patients fail to respond?
In a review of several case studies, Dr. Hall addressed some reasons why patients may not respond to erythropoietic growth factors.  

Functional Iron Deficiency: A 54-year-old woman with breast cancer receiving chemotherapy and reporting fatigue is found to have a white blood cell count of 7.22 x 103 cells/mm3, Hb 10.0 g/dl, Hct 30.8%, MCV 87.7, platelets 680 x 103/mm3 and albumin of 2.4 g/dl. Her iron studies revealed a ferritin of 778.8 ng/ml, a serum iron of 16 mcg/dl, a TIBC of 195 mcg/dl, and a transferrin saturation of 8%. Her labs also revealed a markedly elevated CA 27.29 breast cancer marker. Ferritin is an acute phase reactant, Dr. Hall remarked, whose levels vary according to the presence and severity of inflammation. Dr. Hall recommended treatment with 325 mg of ferrous sulfate taken three times a day. It worked for this patient. “Her hemoglobin went to 11.1 g/dl within 2 weeks, and was normal within 4 weeks. He noted that screening for his study of erythropoietin growth factors in cancer-treatment-induced anemia uncovers “about 20%- 40%” of patients with functional iron deficiency, with low transferrin saturation and normal to elevated ferritin concentrations, indicating inadequate iron for utilization. 

Positive versus negative signals: A 55-year-old woman with 3 lymph node positive adenocarcinoma of the breast, receiving adjuvant chemotherapy with epirubicin and cyclophosphamide, complains of feeling tired after her first cycle of treatment. She is found to be neutropenic and anemic, with an EPO concentration of 161.5 mU/ml; TSH of 1.48; and normal folate and B12. Her baseline hemoglobin, at initiation of treatment with 200 mcg of darbepoetin, was 8.4 g/dl; trans saturation 22.4%, ferritin 359 ng/mL, and albumin 4.3 g/dL. At week 2, her Hb was 7.1, and 6.5 at week 4. “At that time she refused a RBC transfusion, so we escalated her dose to 300 mcg every 2 weeks. At week 6, Hb was 8.5, and 10.5 at week 8. What accounts for lack of response? Dr. Hall explained that erythropoiesis is a multi-step process involving erythroid proliferation and differentiation, which are primarily controlled by a “balance of positive and negative signals.” Positive signals are stem cell factor, interleukin-3 and erythropoietin. Production of negative signals of tumor necrosis factor-alpha, transforming growth factor-beta, and interferon-gamma may be responsible for a non-response in some patients. Animal models of anemia of chronic disease demonstrated that neutralizing tumor necrosis factor alfa with monoclonal antibodies improves the animal’s response to recombinant erythropoietin. This may be a strategy in the future. Patience may be called for, Dr. Hall said, but he asked rhetorically whether it is safe to remain patient if there are alternative strategies such as front-loading and dose-escalation?

Economics and Formulary Considerations

In the complex world of pharmacoeconomics, where costs, reimbursement, and perceptions come together in decision-making, the goal in simple terms is to assess “value.” Value, according to Gordon J. Vanscoy, PharmD, CACP, MBA, Assistant Dean for Managed Care and Associate Professor of Pharmaceutical Sciences at the University of Pittsburgh School of Pharmacy, takes into account not only cost and reimbursement issues, but also outcomes, with the overall goal of achieving “optimal outcomes for the most reasonable utilization of resources.” He provided an example. The average wholesale price (AWP) for 40,000 units of epoetin alfa is about $534, versus $499 for darbepoetin alfa. However, institutions do not pay the AWP; instead they benefit from list- or wholesale acquisition-price discounts and rebates. After these price adjustments are taken into account, some institutions pay as little as 71% of the AWP for epoetin and 56% of AWP for darbepoetin. Also, Dr. Vanscoy continued, there is an approximately 15% across-the-board reduction in the acquisition price for epoetin, regardless of market share, but as much as a 30% reduction from the AWP of darbepoetin, which is calculated based on market share increments.  

A Cost Minimization Example
In a cost-minimization (CM) example Dr. Vanscoy illustrated why dose escalation probabilities be taken into consideration, in addition to labeled dosing, or recommended dosing per standard of practice evidence. As he explained: “...according to darbepoetin clinical trial data, at week 6, 43% of patients received a dose escalation. Thirty-three percent of epoetin alfa patients received an escalation of dose at week 4. But in the “real world” MUE data that were presented, dose escalation was as low as 11% for darbepoetin alfa and 14% epoetin alfa. 

Dr. Vanscoy performed a sample calculation, using 100 patients. Applying the rates of dose escalation seen in clinical studies, and assuming that both drugs are obtained at the best available pricing, the net cost savings with darbepoetin is $195,540. This model can be utilized and adapted to any number of clinical treatment and institution-specific circumstances.” Dr. Vanscoy said, “It is this type of evaluation that caused institutions like the University of Pittsburgh Medical Center to implement a therapeutic interchange program for darbepoetin.” Dr. Vanscoy noted that a cost calculation tool is available at www.upa-llc.com to facilitate an individualized and unbiased evaluation of cost and basic reimbursement. 

Reimbursement adds another factor to the equation. Dr. Vanscoy noted that on the inpatient side, there is a fixed reimbursement per DRG, whether patients are on Medicare or enrolled in a managed care plan. In the setting of hospital outpatient clinics and physician offices, APCs and pass-through reimbursement codes, Medicare fee schedules, and variations of insurance plans (e.g. HMO, PPO) must be taken into consideration. In the case of home care, a DMERC fee schedule must be considered, along with variations of insurance plans. A recent ruling from the Centers for Medicare and Medicaid Services (CMS), effective January 4, 2004, includes the following:

• Epoetin alfa and darbepoetin alfa are still considered to be functionally equivalent
• The economic conversion rate has been adjusted to 330 units of epoetin to 1 mcg of darbepoetin (formerly 260 units to 1).
• The corresponding reimbursement rates are $ 11.76 per 1000 units of epoetin (a 29% increase) and $3.88 per 1 mcg of darbepoetin (a 64% increase).

“In terms of standard of practice dosing,” Dr. Vanscoy said, “most business is cost driven for most health care systems.” Pricing is dynamic. The bottom line for most physician offices is that that reimbursement has been decreased from 95% to 85% of AWP, effective January 4, 2004. This reimbursement includes a 20% patient co-pay. In both instances, darbepoetin alfa is most cost-effective, has a lower AWP and patient co-pay than epoetin alfa. 

Formulary Considerations
“As we move toward the future,” Dr. Vanscoy advised, “we encourage a synchronized approach to care,” which involves administration of the erythropoietic growth factor on a schedule that is similar to the chemotherapy or other treatment regimen. Optimal formulary management includes not just a product overview, with supporting clinical evidence, but also comprehensive economic information (i.e., prospective studies based on pivotal trials, plus retrospective and economic modeling studies), and an impact model report including comparisons to current best practice scenarios.



Return to American College of Clinical Pharmacy                                     Print This

All contents Copyright © 1999 - 2004 Medical Association Communications