“Salmeterol and Fluticasone Propionate and Survival in Chronic Obstructive Pulmonary Disease”
by the Towards a Revolution in COPD Health (TORCH) investigators
N Engl J Med. 2007 Feb 22;356(8):775-89. [free full text]
When the TORCH study was published in 2007, no prospective study to date had demonstrated a mortality benefit of inhaled corticosteroids (ICS) in COPD. Pulmonary inflammation occurs in COPD, and it had been hypothesized that ICS would improve COPD in multiple measures. Previously, ICS had been shown to reduce the frequency of COPD exacerbations, and retrospective data suggested that ICS reduced mortality, particularly when used in combination with a long-acting beta-agonist (LABA). TORCH was designed to evaluate prospectively the potential mortality benefit of combined ICS/LABA vs. ICS vs. LABA vs. placebo.
Population: COPD patients age 40-80, current or former smokers with ≥ 10-pack-year smoking hx, FEV1 < 60% predicted value and increase in FEV1 < 10% with albuterol administration, and prebronchodilator FEV1/FVC ratio of ≤ 70%
Intervention: combination salmeterol 50 µg and fluticasone propionate 500 µg BID
- placebo BID
- salmeterol 50 µg BID
- fluticasone 500 µg BID
Note: all patients underwent a two-week run-in period during which the use of all corticosteroids and long-acting bronchodilators was stopped. Other classes COPD medications were allowed throughout the study.
Primary – time to all-cause mortality by 3 years, per log-rank test
- time to all-cause mortality, per Cox proportional hazards model
- time to all-cause mortality, per log-rank test stratified by smoking status and country of residency
- frequency of COPD exacerbations
- quality of life per the St. George’s Respiratory Questionnaire
- lung function, per postbronchodilator spirometry
- incidence of pneumonia
6184 patients were randomized, but only 6112 were included in the final analyses (several sites excluded for not adhering to quality standards). The four groups were similar in all baseline characteristics (see Table 1).
All-cause mortality at 3 years was 12.6% in the combination-therapy group, 15.2% in the placebo group, 13.5% in the salmeterol group, and 16.0% in the fluticasone group. The hazard ratio for the comparison between combination-therapy and placebo was 0.825 (95% CI 0.681–1.002, p = 0.052, per log-rank test). See Figure 2B. This comparison was repeated in a pre-specified secondary analysis, using the Cox proportional hazards model, which yielded a HR of 0.811 (95% CI 0.670-0.982, p = 0.03), and in another pre-specified secondary analysis, using the log-rank test stratified according to smoking status and country of residency, which yielded a HR of 0.815 (95% CI 0.673-0.987, p = 0.04). In the primary analysis, the mortality risk did not differ among the salmeterol or fluticasone groups relative to the placebo group (see Table 2). Mortality risk in the combination-therapy group was less than that of the fluticasone group (HR 0.774, 95% CI 0.641-0.934, p = 0.007).
COPD exacerbations occurred at an annual rate of 0.85 in the combination therapy group and 1.13 in the placebo group, thus the rate ratio for exacerbations was 0.75 (95% CI 0.69-0.81, p < 0.001, NNT = 4). Exacerbation rates were also lower in the salmeterol and fluticasone groups (see Table 2).
The adjusted mean quality of life score per the St. George’s Respiratory Questionnaire improved in the combination-therapy, salmeterol, and fluticasone groups, and worsened slightly in the placebo group (see Table 3). All groups initially demonstrated an improvement in quality of life. In pairwise comparisons, combination therapy was superior to placebo, salmeterol, and fluticasone (p ranging from < 0.001 to 0.02).
Mean postbronchodilator FEV1 averaged over 3 years improved in the combination therapy group and decreased in the other groups. In all groups, the overall trend was a decrease in FEV1 following an initial improvement (see Figure 2E). In pairwise comparisons, combination therapy was superior to the other groups with respect to change in FEV1 (see Table 3).
The incidence of pneumonia was increased in groups receiving an ICS. The probability of developing pneumonia within the 3 year period was 19.6% in the combination-therapy group, 12.3% in the placebo group, 13.3% in the salmeterol group, and 18.3% in the fluticasone group (p < 0.001 for comparison between both combination-therapy versus placebo and fluticasone versus placebo).
44% of patients in the placebo group withdrew from the study. Only 34% of the combination-therapy group withdrew.
In this large, international, double-blind, placebo-controlled, randomized, parallel-group trial of patients with COPD, combination therapy with ICS/LABA did not improve mortality when compared to a placebo. However, combination therapy improved the frequency of COPD exacerbations, improved quality of life, and slowed the decline in FEV1 relative to placebo.
It is notable that, according to this study’s pre-specified secondary analyses of mortality per Cox proportional hazards and log-rank test stratified by smoking status and location, there was a mortality benefit of combination therapy.
The authors suspect that there is indeed a mortality benefit, but that the trial was underpowered to detect it. Furthermore, the higher rate of treatment-group withdrawal among placebo patients may have biased the study toward a null result, given the intention-to-treat analysis.
In the years since TORCH, meta-analyses that included TORCH have concluded that ICS therapy in COPD slows the rate of decline in FEV1 and decreases the rate of COPD exacerbations when compared with placebo, but it does not reduce mortality.
Today, inhaled corticosteroids remain an integral component of our management of moderate to very severe COPD. See the Global Initiative for Chronic Obstructive Lung Disease (GOLD) Pocket Guide to COPD Diagnosis, Management, and Prevention (2017) pages 14-16.
1. Wiki Journal Club
2. 2 Minute Medicine
3. UpToDate “Role of inhaled glucocorticoid therapy in stable COPD”
4. “Inhaled corticosteroids for stable chronic obstructive pulmonary disease.” Cochrane Database Syst Rev (2012).
5. Global Initiative for Chronic Obstructive Lung Disease (GOLD) Pocket Guide to COPD Diagnosis, Management, and Prevention (2017)
Summary by Duncan F. Moore, MD