Evaluating blood eosinophil levels to guide treatment
Together with clinical symptoms, high blood eosinophil counts can identify a severe asthma patient as having severe asthma with eosinophilic inflammation.
The DREAM study, which evaluated mepolizumab i.v. over 52 weeks, determined that the biomarker of blood eosinophil counts ≥150 cells/µL at screening, or ≥300 cells/µL in the past 12 months, predicted patients who would benefit most from mepolizumab therapy.1-2
Baseline FeNO was also measured, but was found to have a lower association than blood eosinophil count when identifying patients who would respond to mepolizumab.1
Screening eosinophils is predictive of the eosinophil count in the following year3
The solid blue line represents the predicted geometric mean eosinophil count after screening in patients receiving placebo only. The pink line represents the line of identity, which indicates where the values would fall if there was no change between eosinophil count at screening (X-axis) and at Week 56 (Y-axis). The graph is divided into quadrants, with the perpendicular line representing the cell counts ≥150 cells/µL or <150 cells/µL at screening. The horizontal line represents the geometric mean after screening of ≥150 cells/µL or <150 cells/µL.2
On the basis of data obtained from patients who received placebo in the DREAM clinical trial, a single blood eosinophil count of ≥150 cells/μL predicted subsequent average measurements of ≥150 cells/μL in 85% of patients in this population.3Using the average of two, three or four blood eosinophil counts of ≥150 cells/μL, the proportion of patients with post-screening averages was 85%, 90% and 92%, respectively. These data indicate that taking multiple measurements is unlikely to increase sensitivity compared with taking a single measurement, and therefore is not necessary.3
There are two different eosinophil thresholds to inform patient eligibility for mepolizumab. These were determined using a data-driven approach, based on the findings of the DREAM study where mepolizumab treatment was associated with clear clinical benefits in patients with an eosinophil count of ≥150 cells/μL at screening or ≥300 cells/μL in the prior 12 months.3The results of the Phase III clinical trials, MENSA and SIRIUS, have confirmed that this is an appropriate criterion for selecting a patient population most likely to respond to mepolizumab.45
Which patients are eligible for mepolizumab treatment?
Anti-IL-5 treatment can reduce the blood eosinophil count in patients with elevated blood eosinophil counts.1IL-5 plays a key role in eosinophil production and recruitment and is related to eosinophilic inflammation.6
Historically, the blood eosinophil threshold that defines severe asthma with eosinophilic inflammation has been a subject of debate within the asthma community. At present, no definitive threshold exists.
Data from the DREAM clinical trial have driven the blood eosinophil threshold for mepolizumab treatment eligibility.1These levels best predicted exacerbation reduction with mepolizumab treatment:
“Patients with severe asthma and two or more exacerbations in the past 12 months despite high-dose ICS and additional controllers, with or without OCS, and blood eosinophils ≥150 cells/μL at treatment initiation or ≥300 cells/μL in the past 12 months. 1”
“Mepolizumab is indicated as add-on treatment for severe refractory eosinophilic asthma in adult patients, at a dose of 100 mg s.c. 2”
These blood eosinophil count thresholds were prospectively tested in the MENSA and SIRIUS studies and confirmed the patients responding best to mepolizumab therapy.4-5
An early study with mepolizumab was conducted in a population of moderate asthma patients who received moderate doses of ICS (maximum dose of beclomethasone: 1000 µg/day).7Patients were required to have FEV1 50–80% predicted value and reversibility of ≥12% with salbutamol, but were not required to meet a predefined eosinophil threshold (blood or sputum) or have a history of frequent exacerbations.7Patients were excluded if they had used OCS in the previous 4 weeks or had poorly controlled asthma that meant they had been hospitalised or visited the emergency department in the previous 6 weeks.7
The primary endpoint was change from baseline in morning PEF at Weeks 12 and 20 with either mepolizumab or placebo.7Based on the results of this earlier study, the inclusion criteria for future studies were modified to require a history of exacerbations in order to determine whether exacerbation reduction may be the primary benefit of treatment with mepolizumab.7
This early study showed limited clinical benefits of mepolizumab on pulmonary function endpoints and symptoms. However, among the relatively small number of exacerbations recorded, a difference towards a reduction in the frequency of exacerbations was noted in the highest dose group (750 mg i.v.) compared with placebo. This observation was not statistically significant.7
The international ERS/ATS guidelines on definition, evaluation and treatment of severe asthma identifies sputum eosinophil count as one measure to identify patients with an eosinophilic phenotype.8
Both blood and sputum eosinophil levels were measured in the DREAM clinical trial.1However, only blood counts were used as an entry requirement in the MENSA and SIRIUS clinical trials with mepolizumab.1457Blood eosinophil counts were performed for all patients at all study visits, whilst sputum eosinophil counts were performed in a subset of 94 patients from sites with previous experience in sputum collection and processing, and only at baseline or screening, and Weeks 4,6and 52.1The figure below shows that baseline blood eosinophils correlated with response to mepolizumab in terms of exacerbation reduction; sputum eosinophils did not.13
Step 1: Convert eosinophils % into a decimal.
- % Eosinophil ÷ 100
Step 2: Convert white blood cell (WBC) count into cells/µL, if not already in these units.
- WBC count (in K/µL or GI/L or x 103 cells/µL) x 1000 = WBC in cells/µL
Step 3: Multiply eosinophils by WBC count in cells/µL.
- % Eosinophil ÷ 100
There are a number of conditions and medications that can affect blood eosinophil counts, including medications used to treat severe asthma.6It is vital to consider the patient’s history when determining whether they have severe asthma associated with eosinophilic inflammation, so that both the longitudinal blood eosinophil response to treatment and exacerbation history are taken into account.6
Corticosteroids are a strong eosinophil suppressant, for both their activation and their survival.6They prevent production of many inflammatory mediators, including IL-3, IL-4, IL-5 and GM-CSF, which are vital to eosinophil production and survival.6
Montelukast is a leukotriene receptor antagonist indicated for the treatment of asthma, exercise-induced bronchoconstriction and allergic rhinitis.9Clinical trials have shown a suppressant effect of montelukast treatment on blood eosinophil counts, comparable to increasing budesonide dose.10Montelukast inhibits leukotriene receptors, which are expressed on eosinophils and other inflammatory and airway cells, and block the induction of oedema, smooth muscle contraction and inflammatory cell modulation by leukotriene D4.9
This indicates that montelukast therapy should be taken into account when assessing blood eosinophil counts.
Omalizumab is an anti-IgE antibody used for the treatment of severe allergic asthma and does not have a MoA directly related to eosinophil production.11However, a pooled analysis found an association between the reduction of IgE and significant decreases in blood eosinophil counts in patients with moderate-to-severe persistent allergic asthma.12
Anti-IL-5 agents are a new class of asthma treatment. They are biologics targeted against IL-5, which is a key cytokine responsible for eosinophil proliferation and maturation in the bone marrow.6
Anti-IL-5 agents have been shown to reduce the number of eosinophils in peripheral blood.
Eosinophils are primarily responsible for the body’s antiparasitic defences. Parasitic infection can elevate blood eosinophil count and present a confounding factor when diagnosing severe asthma.6
In countries where parasitic diseases are prevalent, parasite infection should always be considered as a differential diagnosis when using blood eosinophils as a diagnostic or treatment biomarker.6
Severe asthma should not be diagnosed on the basis of blood eosinophil counts alone. Other differential diagnoses for high blood eosinophil counts should include allergy, symptoms of cancer, Churg-Strauss syndrome, hypereosinophilic syndrome and a hypersensitivity reaction resulting from drug ingestion.61314The table below shows examples of diseases associated with high blood eosinophil levels.
Diseases associated with high blood eosinophil levels
|Type of disease
||Potential disease cause
||Invasive helminth infection
||Eosinophilic pneumonitis; asthma; allergic bronchopulmonary aspergillosis; chronic eosinophilic pneumonia; subset of COPD patients
||Inflammatory bowel disease; eosinophilic gastroenteritis; allergic colitis|
||Allergic rhinoconjunctivitis; asthma; eczema; atopic dermatitis
|Systemic||Idiopathic hypereosinophilic syndrome; vasculitis; Churg-Strauss syndrome/eosinophilic granulomatosis with polyangitis
|Iatrogenic||Drug reaction; cytokine infusions|
|Malignant||Lymphoma; colonic carcinoma; Langerhans cell histiocytosis; myeloid or stem cell neoplasms
COPD, chronic obstructive pulmonary disease.
Adapted from Rothenberg 1998 and Valent, et al. 2012.14
- Pavord ID, et al. Lancet 2012;380:651–9.
- Mepolizumab SmPC; GlaxoSmithKline 2016
- Katz LE, et al. Ann Am Thorac Soc 2014;11:531–6.
- Ortega HG, et al. N Engl J Med 2014;371,1198–207.
- Bel EH, et al. N Engl J Med, 2014;371:1189–97.
- Rothenberg M. N Engl J Med 1998;338:1592–600.
- Flood-Page P, et al. Am J Respir Crit Care Med 2007;176:1062–71.
- Chung KF, et al. Eur Respir J 2014;43:343–73.
- Montelukast SmPC; Accord Healthcare 2016
- Price DB, et al. Thorax 2003;58:211–6.
- Omalizumab SmPC; Novartis 2016
- Massanari M, et al. Resp Med 2010;104:188–96.
- Bochner B, et al. J Allergy Clin Immunol 2012;130:587–96.
- Valent P, et al. Expert Rev Hematol 2012;5:157–76.