responses. In vitro PDE- 4 inhibition suppresses T-lymphocyte cytokine production and proliferation,
decreases neutrophil chemotaxis as well as oxidative burst and cytokine production, reduces TNF-α
production from dendritic cells and macrophages,
and decreases mast cell degranulation and eosinophil leukotriene production.114 In clinical studies,
PDE- 4 inhibition reduced sputum neutrophils and
eosinophils in patients with COPD after treatment
for 4 weeks.115
Theophylline is a nonspecific oral PDE inhibitor
that has been used as a bronchodilator in the treatment of airway diseases for over 70 years.116 Nevertheless, enthusiasm for its use has waned given
the high serum concentrations that are required for
clinical efficacy, which lead to significant side effects (ie, a narrow therapeutic window). Recently,
theophylline has been used at lower doses to
promote anti-inflammatory action independent of
its PDE inhibition. Histonedeacetylase 2 (HDAC2)
is a nuclear enzyme that inhibits expression of
inflammatory genes and is reduced in the lungs
of patients with COPD, most likely secondary to
increased oxidative stress. 13 As alluded to earlier,
HDAC2 appears essential for anti-inflammatory
action of corticosteroids. 13,117 At lower plasma concentrations than would be required for bronchodilation, theophylline activates HDAC2 and reverses
corticosteroid resistance in patients with COPD.116
Ford et al randomized 30 patients with COPD
to treatment with theophylline and either inhaled
fluticasone propionate (500 µg twice per day) or
inhaled placebo for 4 weeks.85 Theophylline and
fluticasone combination reduced total sputum eosinophils, percentage of sputum neutrophils, and
sputum chemokine (C-X-C motif ligand 8/IL- 8).
Combination treatment was also associated with
improvements in FEV1 predicted and mid expiratory flow rates.85
Roflumilast, the only compound approved for
clinical use in its class, is an oral PDE- 4 inhibitor
with prominent anti-inflammatory properties and
no acute bronchodilator effects.118 Several phase
III placebo-controlled efficacy trials demonstrated
beneficial effects of the compound. Calverley
and colleagues reported the results of 2 placebo-controlled double-blind randomized trials (AURA
and HERMES) which recruited participants with a
clinical diagnosis of COPD from an outpatient setting, if they had chronic cough, sputum production,
at least 1 recorded COPD exacerbation requiring
corticosteroids and/or hospitalization, and FEV1
≤50%.119 Patients were randomized to receive
roflumilast 500 µg once a day (n = 1537) versus
placebo (n = 1554) for 1 year. Importantly, ICS and
long-acting anticholinergic drugs were not allowed.
The first primary endpoint, prebronchodilator FEV1,
increased by 48 mL compared with placebo in the
intervention group (P < 0.001). The second primary endpoint, rate of moderate to severe exacerbations, was also improved in patients who received
roflumilast ( 1.14 versus 1.37/year, 17% reduction;
95% CI, 8 to 25; P < 0.001). Adverse events were
significantly more common with roflumilast and
were related to the known side effects of the drug,
namely, diarrhea, weight loss, decreased appetite,
and nausea. Dropouts from the study, however,
occurred at a similar rate (roflumilast 35% and
placebo 31%). In another publication combining 2
placebo-controlled double-blind multicenter trials
(EOS, n = 933, and HELIOS, n = 743), Fabbri et
al studied the combinations of roflumilast with salmeterol and roflumilast with tiotropium compared
with placebo controls in patients with COPD who
had postbronchodilator FEV1 of 40% to 70% of
predicted.120 The patients in the HELIOS trial also
had to have chronic cough and sputum production
as well as frequent use of rescue medications de-
