Ajay K Singh
MBBS FRCP MBA
Brigham and Women’s Hospital/Harvard Medical School
Reducing erythropoiesis stimulating agent (ESA) dosage could be metaphorically likened to kicking over a beehive: in other words, who needs it? It makes treating anaemia more complicated and it makes less money for all concerned. On the other hand, several trials demonstrate that treating anaemia with ESAs in patients with chronic kidney disease (CKD) is associated with increased risk. Exposure to high ESA dose appears the likeliest reason for this increase in risk. Pharmacists have a key role in ensuring that ESAs are used rationally and safely. Here I make the case for individualising anaemia treatment with a goal towards reducing ESA dose to reduce potential risks associated with therapy.
In the USA, from 24 March 2010 onwards, the Food and Drug Administration (FDA) has begun requiring a ‘Risk Evaluation and Mitigation Strategy’ (REMS) for ESAs in the treatment of cancer-induced anaemia.1 However, the FDA is also asking nephrologists and dialysis providers to educate patients receiving ESAs about safety concerns with these agents. The FDA has asked the ESA manufacturers in the USA, Amgen and Centocor-Ortho Biotech, to require oncologists to register and participate in a programme called Assisting Providers and Cancer Patients with Risk Information for the Safe Use of ESAs (APPRISE). Oncologists need to undergo training on the risks and benefits of ESAs in order to continue prescribing them. This involves discussion of the risks and benefits of ESAs with patients who have cancer before beginning a course of treatment. Doctors have to document that this discussion took place.
Why is reducing ESA dose in managing the anaemia so important? The TREAT study,2 as well as the randomised controlled trials preceding it such as CHOIR,3 CREATE4 and the Normal Haematocrit study,5 and a recent meta-analysis6 have demonstrated that targeting a higher haemoglobin (Hb) concentration with high ESA dosage is associated with increased risk. Observational studies and secondary analyses of the anaemia trials have raised the possibility of several factors accounting for this increased risk observed in the trials (Table 1). However, accumulating evidence suggests that exposure to high ESA dose is the most likely explanation for this increased risk.7
The approach to treating anaemia in CKD patients has been one-dimensional, especially among CKD patients not on dialysis. Both European and US guidelines have advocated aiming for a haemoglobin level within a target range, regardless of patient characteristics and the clinical situation. Recently, an individualised approach has been strongly recommended.8 In Figure 1, I propose such an approach for non-dialysis CKD patients. (In dialysis patients, while individualising therapy should be the objective, most patients will require ESA because of severe kidney disease and failure of adequate endogenous erythropoietin production; aiming for lower doses of ESA should be a priority but withholding ESA is usually not possible). Randomised clinical trials in anaemia have unequivocally demonstrated that Hb is not a particularly good surrogate marker for clinical outcomes. Raising the Hb from 11 to 13g/dL is associated, at best, with an inconsistent and modest improvement in quality of life: indeed, no improvement in cardiovascular outcomes or mortality is observed with partial or complete correction of anaemia.
Since Hb is an unreliable intermediate clinical outcome, an alternative specific clinical outcome needs to be identified for individual patients. This clinical outcome could be reducing fatigue or preventing the need for a blood transfusion. Some patients – particularly younger patients – may be well compensated at a haemoglobin level of 9g/dL, functioning relatively normally and without complaints of fatigue or tiredness. For these patients, there will be no need for ESA therapy. In the TREAT study,2 approximately one-half of the patients in the placebo rescue arm of the study required no ESA therapy, whereas the other one-half of patients required very low dosages of ESA.
After individualising the goals of anaemia therapy, it is important to identify the ESA hyporesponsive patient. Most definitions of ESA hyporesponsiveness focus on the ESA dose (the commonest definition of ESA resistance is the continued use of >300IU/kg/week of epoetin-alpha or its equivalent). However, considering the clinical context is also important. ESA hyporesponsive patients tend to be inflamed. For example, a failed kidney allograft may still be in place, or the patient may have low-level activity of his or her lupus (SLE) or rheumatoid arthritis (RA). Alternatively, the patient may have evidence of an indolent infection. For example, an infected skin ulcer or a low-grade infection of an indwelling dialysis catheter. The patient may have an underlying malignancy. Laboratory abnormalities suggestive of infection and/or inflammation may be helpful. These include a chronically elevated white blood count, an increased ferritin with a low transferrin saturation (Tsat), a low serum albumin or an elevated high sensitivity C-reactive protein (hsCRP).
Modulating inflammation by either removing an inflammatory focus (such as failed kidney allograft)9 or by treating an indolent infection in a tunnelled line,10 can lead to reduced ESA dose. Drug therapy has also been effective in reducing inflammation. In a study by Cooper et al,11 pentoxyfylline inhibited the pro-inflammatory cytokines TNF-alpha and interferon-gamma and resulted in greater ESA responsiveness (as measured by a higher haemoglobin level on stable ESA dose). Modulating the inflammatory response to improve ESA hyporesponsiveness could entail treating the underlying source of the inflammation, such as an underlying SLE or rheumatoid arthritis flare. Other strategies include the use of ultrapure water and biocompatible membranes in dialysis patients, statins and vitamin E supplementation.12
While the serum markers for iron deficiency are not perfect, using the serum ferritin and Tsat to ensure that iron stores are fully replete and that iron is available for erythropoiesis is very important. The National Kidney Foundation Kidney Disease Outcome Quality Initiative (NKF-KDOQI)13 recommends initially using oral iron in non-dialysis CKD patients and intravenous iron in dialysis patients aiming for a serum ferritin >200ng/mL and a Tsat >20%.
While all non-dialysis patients are being treated with subcutaneously administered ESA, this is not necessarily the case with ESA administration in dialysis patients. Many receive ESA intravenously three times each week with their dialysis treatment. Switching these patients to subcutaneous epoetin is a powerful way to reduce ESA dosage. The KDOQI guidelines state:13 ‘In the opinion of the Work Group, convenience favors intravenous administration in HD-CKD patients.’ A trial by Kaufman et al14 demonstrates that EPO given subcutaneously three times a week will maintain the haematocrit at the same level and at a one-third lower dose than intravenous epoetin. In fact, in almost all published studies, subcutaneous administration is associated with a 25–50% reduction in epoetin dose. Survey data from the American Association of Kidney Patients (AAKP)15 that we have analysed indicates that patients are willing to accept subcutaneously injected epoetin if it is explained to them that it is important for safety reasons.
Optimisation of metabolic parameters results in reduced ESA utilisation. Evidence demonstrates that folic acid deficiency16 can result in ESA hyporesponsiveness especially in elderly patients with poor dietary folate intake without regular oral supplementation. There is good evidence that hyperparathyroidism is an important factor in ESA hyporesponsiveness.17 L-carnitine therapy may also lower ESA utilisation, however its use in dialysis patients remains controversial.18 In dialysis patients, optimising the dose of dialysis19 and/or reverting to nocturnal dialysis20 or peritoneal dialysis21 is effective in lowering ESA utilisation.
In conclusion, pharmacists play a central role in advising and scrutinising therapeutic managements in patients and could play an important role in changing how clinicians approach the use of ESAs in treating anaemia in CKD patients. While we are not certain that exposure to high ESA dose is risky and this question continues to be debated,21 it makes both economic and clinical sense to use the lowest possible dose to achieve the desired outcome. The desired outcome should be individualisation of anaemia care and not the one-dimensional approach that has been used so far.
1. Available online at: www.fda.gov/Drugs/DrugSafety/forPatientsandProviders/ucm200297.htm. (Accessed 20 April 2010).
2. Pfeffer M et al. N Engl J Med 30 Oct 2009 [Epub ahead of print].
3. Singh A et al. N Engl J Med 2006;355:2085–98.
4. Drueke T et al. N Engl J Med 2006;355:2071–84.
5. Besarab A et al. N Engl J Med 1998;339(9):584–90.
6. Palmer S et al. Ann Intern Med 3 May 2010 [Epub ahead of print].
7. Singh A. J Am Soc Nephrol 2009;20:1436–41.
8. Agarwal R. Clin J Am Soc Nephrol 6 May 2010 [Epub ahead of print].
9. López-Gómez J et al. J Am Soc Nephrol 2004;15(9):2494–501.
10. Hung A and Ikizler T. Semin Dial 2008;21(5):401–4.
11. Cooper A et al. J Am Soc Nephrol 2004;15(7):1877–82.
12. Johnson D et al. Nephrology (Carlton) 2007;12(4):321–30.
13. KDOQI. Am J Kidney Dis 2007;50(3):471–530.
14. Kaufman J et al. N Engl J Med 1998;339(9):578–83.
15. Patel T et al. Nephrol News Issues 2007;21(11):57,59,63–64.
16. Schiffl H and Lang S. Nephrol Dial Transplant 2006;21(1):133–37.
17. Drüeke T and Eckardt K. Nephrol Dial Transplant 2002;17(Suppl 5):28–31.
18. Hedayati S. Semin Dial 2006;19(4):32.
19. Ayli D et al. J Nephrol 2004;17(5):701–6.
20. Schwartz D et al. Clin Nephrol 2005;63(3):202–8.
21. Coronel F et al. J Nephrol 2003;16(5):697–702.