You see a 50-year-old woman for the first time in your office. She tells you she’s still smoking a pack per day. She’s trying to cut down, but it’s hard. For the past few years she’s had a cough each winter and can’t walk the stairs to her second floor apartment without getting winded. For these symptoms, she underwent a battery of cardiac testing with a prior doctor, but was never sent for pulmonary function testing. So you write the referral and sure enough, she returns to your office a few weeks later with a report and a new diagnosis: chronic obstructive pulmonary disease (COPD).
“What can I expect?” she asks.
You pause. Sure, you’ve learned that COPD is characterized by progressive decline in lung function, but at what rate? What factors determine the progression of disease? And are there any tests you can send to see if she’ll do better, or worse, than others with her diagnosis?
It’s an area of uncertainty, with a paucity of long-term data that examine the rate of forced expiratory volume in 1 second (FEV1) loss and the factors that contribute to this decline.
To deal with this information gap, the Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) study steps in with over 2,000 patients followed over three years with spirometry and blood tests. Patients with moderate COPD between the ages of 40 and 75, with a smoking history of 10 pack-years or greater were enrolled. All had a ratio of FEV1 to forced vital capacity of less than 70% and their FEV1 was less than 80% predicted. Patients underwent a chest CT to evaluate severity of emphysema at the start of the study, and investigators collected serum samples to identify disease biomarkers. Over the next 3 years they reported the number of COPD exacerbations they had experienced and underwent spirometry testing before and after use of a bronchodilator.
On average, FEV1 dropped 33 ml yearly with a standard deviation of 59 ml. While the majority of patients had a decline in FEV1 each year, for a minority (about 15% of those in the study), FEV1 actually increased. It’s not clear whether this was due to occult illness at the time that the enrollment FEV1 was measured (that is, the baseline was falsely low), to new onset of bronchodilator treatment, to the disease’s natural history, or to other factors.
There were highly variable outcomes, so which groups did worse? Patients with moderate disease had a faster decline than those with worse disease at outset; this may reflect the fact that those with moderate disease had more FEV1 to lose than those with severe disease. Exacerbations during follow-up resulted in a hit to FEV1. Those who continued to smoke also lost FEV1 faster: of note, however, the rate of decline didn’t seem to vary by cumulative tobacco exposure. Emphysema on CT and bronchodilator reversibility at baseline also were related to faster decline.
The investigators also looked at serum markers that could be useful both for prognosis or to provide potential therapeutic targets. But after testing a battery of biomarkers, only levels of CC-16 (Clara cell protein, which might reflect the presence of intact small airways) were associated with FEV1 rate of change. Even this association was “weak,” however, authors acknowledged and thus didn’t comment further on the meaning of this finding.
So, back to our patient. Does this 3-year large cohort study bring us any closer to predicting her course?
Not entirely. But as Dr. Jeffrey Drazen, the NEJM editor handling the article, said, “Though not perfect, these data provide a contemporary snapshot of lung function loss in patients with COPD.” Perhaps the outlook for lung function in the disease is not as bleak as we had thought, nevertheless the ECLIPSE study makes one message for our patient even clearer: Stop smoking!
Does this study change how you counsel or treat a patient with COPD? Are you surprised at all by the results?