Background: While broadly utilized for treatment of anemia in cancer patients, treatment with erythropoietin stimulating agents (ESA) is not recommended in patients who fail to respond within 6-8 weeks.
Objective: This study examined the rate of adverse events (AE) and costs of continued exposure to ESAs in the absence of clinical response compared to those responding to ESA treatment.
Methods: Electronic medical records of breast, head & neck, cervical, non-small cell lung cancer, and multiple myeloma patients initiating ESAs during 1/1/2007 - 7/1/2011 were extracted (ACORN, Memphis, TN). Patients with ESA use ≥8 weeks, valid Hemoglobin (Hb) values ≤28 days before and 43-70 days after ESA start were retained. Following the ASCO/ASH guidelines, “responders” had Hb increase ≥1 g/dL from start to between days 43 to 70. Non-responders were then divided into “continuing” or “discontinuing” based on presence of ESA treatment after the 43-70 day window. Days on ESA, anemia-related costs (ESA, laboratory, visits), and AE rates were analyzed. ESA costs were estimated by 2011 Wholesale Acquisition Cost (WAC) dose billed and other anemia-related costs by applying Current Procedural Terminology (CPT) pay proxies, derived from the 2011 MarketScan Databases.
Results: Compared to responders, continuing non-responders had significantly longer mean ESA treatment (+17 days; P = 0.023) and significantly higher median anemia-related costs (+$3,385; P <.001) during post-response period and for entire ESA treatment period (+$2,927; P <.001). Discontinuing non-responders incurred significantly lower median anemia-related costs than responders during ESA treatment (–$2,368; P <.001). Additionally, continuing non-responders experienced a higher number of AEs compared to discontinuing non-responders (9.4% vs 3.4%, respectively).
Conclusion: For every one non-responding cancer patient who discontinued treatment according to recommendations, there were three non-responders who continued treatment. The treatment duration for non-responding patients who continued ESAs was longer even than for those responding to ESA treatment. Treatment costs were significantly higher for non-responders continuing ESA treatment than the other two cohorts. It was not clear from these data whether the increased AE rate in non-responders who continued ESAs, compared to those who discontinued, was offset by other positive outcomes for the patients.