In 1986 recombinant erythropoietin is the product of years of research

Anaemia was a major cause of morbidity in patients with end stage renal disease until recombinant erythropoietin became available at the end of  the 1980s. 
Before long-term dialysis was available, haemoglobin (Hb) levels fell to a mean of about 8 g/dl (80 g/l) at a glomerular filtration rate (GFR) of 10ml/min, and about 7 g/dl at a GFR of 5, with a significant minority of patients running levels below 5 g/dl. 

The anaemia of renal failure in a pre-dialysis and pre-EPO study.  Mishra and Kerr,  with permission.

Dialysis alone elevated Hb a little, but many patients were severely symptomatic and disabled by their anaemia.  Androgens and other treatments were in use but had only slight impact.  Blood transfusions raised Hb to relieve the worst symptoms or life-threatening crises, but were problematic for long term use because:

• Weekly or alternate weekly treatment was commonly required
• Iron overload was a serious consequence of regular transfusions 
• Antibody formation made blood increasingly difficult to match
• Sensitization to transplant antigens could preclude later transplantation
• Risks of transmission of viral infections (hepatitis devastated renal units 1964-72)

For these reasons, transfusion to near-normal haemoglobin values was not feasible and many patients lived with chronic severe anaemia. 

The full explanation for renal anaemia was controversial. It was known that red cells had shortened survival in renal failure. Regular blood loss was an additional factor for haemodialysis patients. However the more severe anaemia of patients who had had bilateral nephrectomies suggested that something produced by the kidneys was important.
Many pieces of scientific work led to the discovery of Erythropoietin (EPO), which was purified from urine in 1977. Assays to measure EPO were developed in the 1980s, and showed that although renal patients produced some erythropoietin, EPO levels failed to rise as high as they did in patients with anaemia of other causes.

EPO was a revolutionary treatment for these relatively young patients. Recombinant EPO became the first blockbuster of a new generation of genetically engineered protein drugs. But it was extremely expensive, adding about £5000 per annum, over 30% of the cost of dialysis in 1990. Subcutaneous administration and increased iron therapy were subsequently found to be ways to reduce the amounts needed and minimize cost.

Recommendations from the UK Renal Association in 1990 were that EPO should be given to patients with Hb below 8g/dl, or who were transfusion dependent or had their lives threatened by anaemia, or were at risk of sensitization to transplant antigens because of their transfusion requirement. It was hoped that this would note extend to more than 20% of dialysis patients. 

Today’s patients are older and have more comorbid disease than the patients treated 15-20 years ago.  They tolerate low haemoglobins less well and arguably have more to gain from successful treatment of anaemia. This led to an edging up of target haemoglobins, supported by mostly pharma-sponsored research that showed improved activity and well-being in patients whose Hb was lifted over 10. ‘Minimum acceptable’ haemoglobin levels were now identified in guidelines. The proportion of haemodialysis patients receiving EPO or similar agents rose to 90%.

Testing of whether these higher targets were safe came late, and wasn’t so encouraging. Beginning with the Normal Haematocrit Study in 1998, and culminating in the CHOIR and CREATE studies in 2006, it became apparent that although having normal haemoglobins seemed to make patients feel better, it might be at the cost of increased thrombotic events, death, and cancer. No safe dose of EPO, or safe Hb limit that can completely prevent these adverse effects, has been established. 

ESAs (meaning Erythropoiesis Stimulating Agents of all types, including recombinant EPO) are now a core part of the management of advanced CKD. They have been hugely beneficial, and national and international guidelines assume their use. But how to balance the benefits with the potential long-term risks is still not solved. 

Further reading

Cotes PM 1982 Immunoreactive erythropoietin in serum. I. Evidence for the validity of the assay method and the physiological relevance of estimates  Br J Haematol 50:427-38
Eschbach JW, Egrie JC, Downing MR, Browne JK, Adamson JW. 1987. Correction of the anemia of end-stage renal disease with recombinant human erythropoietin. Results of a combined phase I and II clinical trial. N Engl J Med. 316:73-8.
Winearls CG, Oliver DO, Pippard MJ, Reid C, Downing MR, Cotes PM. 1986. Effect of human erythropoietin derived from recombinant DNA on the anaemia of patients maintained by chronic haemodialysis. Lancet. 1986 328:1175-8.
Goldsmith DJA, CG Winearls.  2008.  EPO – taken for granted now, but 20 years ago a new era was dawning.  UK Renal Association.  (not available online)
Jhaveri KD, 2012. A history of erythropoietin development – nice timeline.