Erratum: COVID-19–related Genes in Sputum Cells in Asthma: Relationship to Demographic Features and Corticosteroids

Our article, published in the July 1, 2020, issue of the Journal (1), contained an error in the number of healthy control subjects. The paper reported on 330 asthma participants in the SARP-3 (NHLBI Severe Asthma Research Program-3) cohort and 79 healthy control subjects (57 recruited by the University of California San Francisco [UCSF] Airway Clinical Research Center and 22 recruited by SARP-3). We recently discovered that a coding error resulted in sputum cell RNA from 47 mild asthma patients being included in the UCSF healthy subject group. To correct the error, we removed the 47 mild asthma patients and reanalyzed the data. After performing the reanalysis including the 22 healthy subjects from SARP and 10 healthy subjects from UCSF (total of 32 healthy controls) (revised Table 1), we found that our study conclusions remain the same. As illustrated in revised Figures 1 and 2, sputum cell gene expression for COVID-19–related genes (ACE2 [angiotensin-converting enzyme 2] and TMPRSS2 [transmembrane protease serine 2]) are not significantly different in asthma and health (revised Figure 1A and 1B), and sputum cell gene expression for ACE2 and TMPRSS2 are significantly correlated with one another (revised Figure 2A). The reanalysis shows that the P value for the increase in asthma for sputum cell ICAM1 expression (a comparator/control gene) compared with health increased from 0.005 to 0.09 (revised Figure 1C). The main data for the paper, as originally presented in Figures 3 and 4 and which relied on data analyses that were restricted to the asthma cohort, do not need correction.

Table 1.

Demographic Features of Subjects with Asthma and Healthy Control Subjects

Characteristic	Healthy (n = 32)	Asthma (n = 330)	P Value
Age, yr	38.8 (13.3)	48.5 (13.8)	<0.001
Sex, F, n (%)	24 (75)	230 (69)	0.53
Race, n (%)*			0.002
AIAN	0 (0)	2 (1)
Asian	6 (19)	13 (4)
African American	4 (13)	77 (23)
White	18 (58)	217 (66)
Native Hawaiian or other Pacific Islander	0 (0)	0 (0)
Mixed race	3 (10)	23 (7)
BMI, kg/m2	25.9 (5.6)	32.4 (8.7)	<0.001
Spirometry†
FEV1, % predicted	96.2 (10.1)	72.8 (19.3)	<0.001
FVC, % predicted	97.2 (12.8)	85.2 (17.0)	<0.001
FEV1%/FVC%	99.0 (5.4)	84.5 (12.1)	<0.001

Definition of abbreviations: AIAN = American Indian and Alaska Native; BMI = body mass index.

Data are shown as mean (SD) unless otherwise noted.

^*AIAN patients are not included in the mixed effects models because only 2 patients identified as AIAN. One healthy control subject did not answer the race questionnaire.

^†Spirometry values are before administration of bronchodilator medications.

Figure 1.

Sputum gene expression at the initial study visit in participants with asthma (n = 330) and healthy participants (n = 32). (A and B) No difference in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)–related genes, ACE2 (angiotensin-converting enzyme 2) and TMPRSS2 (transmembrane protease serine 2), between participants with asthma and healthy participants. (C) Gene expression for the rhinovirus binding protein ICAM1 (intercellular adhesion molecule 1) trended higher in asthma participants compared with healthy participants.

Figure 2A.

Sputum gene expression of ACE2 (angiotensin-converting enzyme 2) and TMPRSS2 (transmembrane protease serine 2) is strongly correlated in healthy participants. Best fit line was fitted using a cubic smoothing spline.

We have made several additional changes throughout the article to reflect the corrected data. For the convenience of the readers, the Journal has replaced the online version of the article with a corrected version (a redlined version showing the changes may be accessed from the Supplements tag at the top of the HTML view of the article).

We apologize to the readership for these errors.