Longitudinal Outcomes of Pi*MZ versus Pi*MM Subjects Stratified by Former and Current Smoking Status

To the Editor:

Barjaktarevic and colleagues (1) used the SPIROMICS (Subpopulations and Intermediate Outcome Measures in COPD) database to compare lung function and emphysema outcomes at enrollment (baseline) and longitudinally between ever-smokers (⩾20 pack-years) with a Pi*MM (normal) or Pi*MZ α-1 antitrypsin (AAT) genotype. At baseline, they found a lower forced expiratory volume in 1 second (FEV₁) and more emphysema (i.e., lower lung density) in the Pi*MZ group, which is not surprising because all were ever-smokers and these findings are consistent with a prior meta-analysis (2). However, on longitudinal follow-up (median of 4.4 visits over 4.8 yr), there was no significant difference in the rate of decline in the FEV₁ or lung density between the genotypes. The lack of difference in the follow-up period between Pi*MM and Pi*MZ ever-smokers may not be unexpected because nearly two thirds of the individuals in both groups were former smokers at baseline and therefore no longer exposed to the inactivating effects of cigarette smoke–derived oxidants on AAT function. Indeed, several studies have shown that the steeper decline in lung function in smokers may be attenuated when smoking cessation is achieved (3, 4).

The authors additionally stratified the longitudinal analyses by smoking status (current vs. former smokers) at study enrollment, which will be the focus of this commentary. Longitudinal analysis in the combined group (Pi*MZ plus Pi*MM) revealed, intuitively, that current smokers had a significantly greater rate of decline in FEV₁ than former smokers (–43.4 ml/yr vs. –29.4 ml/yr, respectively), i.e., a 14 ml/yr greater loss in FEV₁ in current smokers than former smokers during the follow-up period. In contrast, when the current smokers were differentially stratified by genotype, the rate of FEV₁ decline in the Pi*MM group (–44.0 ml/yr) was 1.8 times greater than that of the Pi*MZ group (–24.7 ml/yr), an absolute difference of 19.3 ml/yr, which is counterintuitive. Although this difference was not statistically significant (perhaps being underpowered), it could be clinically significant. We wonder whether this seemingly paradoxical finding may be due to changing smoking habits of Pi*MZ versus Pi*MM individuals in the follow-up period. Although the authors noted that “very few participants changed their smoking status over time,” the finding that Pi*MM current smokers had a greater rate of decline in FEV₁ than Pi*MZ current smokers may lead one to posit that Pi*MM current smokers tend to remain active smokers (or smoked more cigarettes and/or for a longer time than Pi*MZ subjects) whereas Pi*MZ current smokers tend to quit (or quit sooner than Pi*MM subjects) during the follow-up period. This notion is supported by studies showing that a greater severity of AAT genotype is associated with a higher quit rate of smoking, perhaps due to more convincing smoking cessation counseling (5, 6).

Based on our reasoning, we wonder whether the SPIROMICS database has adequate granularity to quantify the change in smoking status longitudinally and to analyze how the change may impact the rate of decline of lung function and lung density. If so, the information could refute or support our hypotheses that: (1) current smokers with Pi*MM have quantitatively greater tobacco exposure during the follow-up period than current smokers with Pi*MZ and, (2) adjusted for smoking habits in the follow-up period, Pi*MZ individuals (vs. Pi*MM) would show, contrary to these findings, greater FEV₁ decline due to the compounded effects of lower AAT levels and negative effects of continued smoking on AAT function. Regardless of whether our hypotheses are validated, the investigators’ important findings support the maxim that smoking cessation should be stressed equally to all individuals, regardless of their AAT genotype.