Severity of COVID-19 at elevated exposure to perfluorinated alkylates

Philippe Grandjean; Clara Amalie Gade Timmermann; Marie Kruse; Flemming Nielsen; Pernille Just Vinholt; Lasse Boding; Carsten Heilmann; Kåre Mølbak

doi:10.1371/journal.pone.0244815

. 2020 Dec 31;15(12):e0244815. doi: 10.1371/journal.pone.0244815

Severity of COVID-19 at elevated exposure to perfluorinated alkylates

Philippe Grandjean ^1,^2,^*, Clara Amalie Gade Timmermann ², Marie Kruse ², Flemming Nielsen ², Pernille Just Vinholt ³, Lasse Boding ⁴, Carsten Heilmann ⁵, Kåre Mølbak ^4,⁶

Editor: Jaymie Meliker⁷

PMCID: PMC7774856 PMID: 33382826

Abstract

Background

The course of coronavirus disease 2019 (COVID-19) seems to be aggravated by air pollution, and some industrial chemicals, such as the perfluorinated alkylate substances (PFASs), are immunotoxic and may contribute to an association with disease severity.

Methods

From Danish biobanks, we obtained plasma samples from 323 subjects aged 30–70 years with known SARS-CoV-2 infection. The PFAS concentrations measured at the background exposures included five PFASs known to be immunotoxic. Register data was obtained to classify disease status, other health information, and demographic variables. We used ordered logistic regression analyses to determine associations between PFAS concentrations and disease outcome.

Results

Plasma-PFAS concentrations were higher in males, in subjects with Western European background, and tended to increase with age, but were not associated with the presence of chronic disease. Of the study population, 108 (33%) had not been hospitalized, and of those hospitalized, 53 (16%) had been in intensive care or were deceased. Among the five PFASs considered, perfluorobutanoic acid (PFBA) showed an unadjusted odds ratio (OR) of 2.19 (95% confidence interval, CI, 1.39–3.46) for increasing severities of the disease. Among those hospitalized, the fully adjusted OR for getting into intensive care or expiring was 5.18 (1.29, 20.72) when based on plasma samples obtained at the time of diagnosis or up to one week before.

Conclusions

Measures of individual exposures to immunotoxic PFASs included short-chain PFBA known to accumulate in the lungs. Elevated plasma-PFBA concentrations were associated with an increased risk of a more severe course of COVID-19. Given the low background exposure levels in this study, the role of exposure to PFASs in COVID-19 needs to be ascertained in populations with elevated exposures.

Introduction

Elevated exposure to community pollution is associated with a worsened outcome of coronavirus disease 2019 (COVID-19) [1–4]. While replicated in different populations, this evidence relies solely on ecological study designs of air pollution without measures of individual exposures. Several environmental chemicals are known to suppress immune functions [5, 6] and worsen the course of infections [7]. Of particular relevance, the perfluorinated alkylate substances (PFASs) are persistent, globally disseminated chemicals known to be immunotoxic [8]. Thus, elevated blood-PFAS concentrations are associated with lower antibody responses to vaccinations in children [9] and in adults [10]. Also, infectious disease occurs more frequently in children with elevated exposure [11–13]. In support of the potential impact of these substances, a modeling study suggested that endocrine disruptors, including major PFASs, may interfere with proteins involved in critical pathways, such as IL-17, associated with severe clinical outcomes of the COVID-19 infection [14].

Substantial differences occur in the clinical course of the disease, and the reasons for this variability are only partially known [15, 16]. As a possible contributor, a deficient antibody response may be an important contributor to a more severe clinical course of the infection [17], as also suggested by the poorer prognosis in patients with bacterial co-infection [18]. The most serious clinical consequences are associated with male sex, older age, and the presence of co-morbidities, including obesity and diabetes [19–23]. In parallel, serum-PFAS concentrations are higher in men than in women and also tend to increase with age [8, 24]. Because elevated PFAS exposure has been linked to both obesity and diabetes [25, 26], these substances may potentially affect the progression of COVID-19 directly as well as indirectly.

Several PFASs can be reliably determined in human blood samples, where most of them show long biological half-lives of 2–3 years or more [27], thereby providing a measure of cumulated exposure. Still, blood concentrations may not accurately reflect the retention in specific organs, e.g., the short-chain perfluorobutanoic acid (PFBA), which accumulates in the lungs [28].

To assess if elevated background exposures to immunotoxic PFASs are associated with the clinical course of the infection, a study was undertaken in Denmark to determine individual plasma-PFAS concentrations in adults confirmed to be infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and examine the association with the severity of COVID-19 development.

Methods

Population

Plasma samples for PFAS analysis were obtained from medical biobanks that store excess material from diagnostic tests, viz., the Danish National Biobank at the Statens Serum Institut (SSI) and Odense University Hospital (OUH). Eligible subjects were identified from the Danish cohort of COVID-19 patients [29]. All cases were tested by quantitative polymerase-chain-reaction (PCR) and had a positive response for SARS-CoV-2 infection, as recorded in the Danish Microbiology Database (MiBa), a national database that contains both positive and negative results of the majority of microbiology testing done in Denmark [30].

The study included non-pregnant subjects aged 30–70 years at the time of the positive test by early March 2020 through early May 2020, provided that the biobanks could provide a plasma sample of 0.15 mL. Although most blood samples were obtained soon after SARS-CoV-2 infection was identified, we also included subjects, mainly those not hospitalized, whose plasma in the SSI biobank had been obtained up to 28 months earlier, i.e., less than a half-life for major PFASs [27]. We calculated the time interval from blood sampling to the time of diagnosis, of relevance mainly for non-hospitalized subjects. In those hospitalized, we computed the interval from admission to the time of sampling the plasma used for PFAS analysis.

All samples were coded, and the Personal Identification Number for each subject was separately transferred to the Danish Health Data Authority (FSEID-00005000) to allow linkage to demographic and medical information from the Danish Civil Registration System (CRS) [31], the Danish National Register of Patients (DNRP) [32], and the National Health Insurance Service Register [33]. We used the following classification of disease status: no hospital admission and completed infection within 14 days of testing positive, hospitalization with COVID-19 up to, or above, 14 days, admission to intensive care unit, or death. Presence of chronic disease was based on the following diagnoses in the register data: diabetes type I and II (ICD10 codes E10-E11), malignant cancers (C00-C99), cerebrovascular and coronary disease (I00-I99), pulmonary disease (J00-J99), and obesity (E66-E68). Renal disease (N0-N2) was treated as a separate covariate due to the possible impact of kidney function on plasma-PFAS concentrations [34]. The linked data set was analyzed via secure server without access to information on the Personal Identification Numbers of the subjects involved. For confidentiality reasons, all tabular information had to be based on at least five subjects.

The protocol was approved by the Regional Committee on Health Research Ethics (S-20200064), which also allowed the project to proceed without seeking informed consent from the subjects identified for study participation. Additional approvals were obtained from the Danish Data Protection Agency as well as institutional and regional authorities for the transfer blood samples and linkage of subject information to the PFAS analyses, while protecting confidentiality.

Chemical analysis

The plasma samples were analyzed in successive series for PFAS concentrations, including PFBA, perfluorooctane sulfonate (PFOS), perfluorooctanoate (PFOA), perfluorohexane sulfonate (PFHxS), and perfluoronanoate (PFNA), which are known from previous studies to be associated with immunotoxicity in humans [8, 35, 36]. We also determined plasma concentrations of PFASs so far not linked to immunotoxicity, i.e., short-chain perfluorobutanesulfonate (PFBS), perfluoroheptanesulfonate (PFHpS), perfluorodecanoate (PFDA), and perfluoroundecanoate (PFUdA) (results shown in the Supporting information). We used online solid-phase extraction followed by liquid chromatography and triple quadropole mass spectrometry (LC–MS/MS) at the University of Southern Denmark [37]. Accuracy of the analysis was ensured by inclusion of quality control (QC) samples comprising proficiency test specimens from the HBM4EU program organized by Interlaboratory Comparison Investigations (ICI) and External Quality Assurance Schemes (EQUAS). All results of the QC samples were within the acceptance range. The between-batch CVs for the actual series ranged between 3% and 14% for all compounds. Both PFOS and PFOA were quantified in all blood samples, and all PFASs were detectable in at least 30% of the samples. Results below the limit of detection (LOD, 0.03 ng/ml) were replaced by LOD/2 before uploading to the secure server at the Danish Health Data Authority, where linkage to other information took place.

Statistical analysis

Correlations between PFASs were examined using Spearman’s correlation coefficient. The PFAS concentrations were compared between demographic groups (age in years, sex, national origin, place of inclusion), presence of comorbidities, and number of days between blood sampling and diagnosis, and differences were tested using Kruskal-Wallis and Wilcoxon rank-sum test. Furthermore, associations of COVID-19 severity with age were tested using Kruskal-Wallis test, and relations with each of the variables sex, national origin, presence of comorbidities, and number of days between blood sampling and diagnosis were tested using χ² test. Associations between place of inclusion and COVID-19 severity could not be displayed and tested, as some cells contained less than five individuals.

Because COVID-19 severity was categorized, the association between the continuous plasma-PFAS concentrations and COVID-19 severity was tested in ordered logistic regression models. More than half the short-chain PFAS concentrations were below the LOD, and they were therefore treated as binary variables (below/above LOD). Potential confounding variables were identified based on a priori knowledge as summarized above and included age (continuous, years) sex, and national origin (Western European yes/no). Among those of Western European national origin, 94% were Danish, while most of the participants of non-Western European national origin were born in or of parents from Somalia (20% of the sample), Pakistan (13%), Iraq (12%), Morocco (11%), Eastern Europe (9%), and Turkey (9%). Kidney disease may affect PFAS elimination, and PFAS exposure could potentially increase the risk of certain other chronic diseases that may affect COVID-19 severity [8]. Kidney disease (yes/no) and other chronic disease (yes/no) were thus considered potential confounders to allow estimation of the direct, rather than the total effect of plasma-PFAS concentrations. Due to changes in PFAS exposures over time, the timing of blood sampling was included as covariate. Further, due to the short elimination half-life for short-chain PFASs [8], we carried out sensitivity analyses excluding plasma samples obtained more than one week before or after diagnosis. We also adjusted for the place of inclusion (OUH/SSI) but, under the circumstances of this study, detailed data on socioeconomic status (e.g., income, education or labor market affiliation) were unavailable for this study. Dichotomous analyses comparing severities of the disease were performed in logistic regression models.

The default assumption of dose-response linearity was tested by including PFAS squared along with PFAS in the regression models. No significant (p<0.05) deviation from linearity was found. The proportional odds assumption in the ordered logistic regression was tested by a likelihood-ratio test using the Stata omodel package. In a model adjusting for age, place of inclusion, and timing of blood sampling, the hypothesis of proportional odds was accepted (p>0.05) in all analyses. Odds ratios (ORs) between groups of COVID-19 severity were therefore calculated using logistic regression models.

Results

The predominant PFAS in plasma was PFOS, with an average concentration of 6.1 ng/mL (median, 4.7 ng/L), approximately equally distributed between the normal and branched isomers. Other PFASs quantified showed averages below 1 ng/mL. In a sensitivity analysis, one extreme PFHxS outlier at 12.9 ng/mL was omitted. The PFAS concentrations correlated well, with Spearman correlation coefficients generally above 0.5 (Table 1 and S1 Table), except for short-chained PFAS. PFOS on average contributed 69% of the total PFAS concentrations by weight and correlated particularly well with most other PFASs quantified.

Table 1. Spearman’s correlation coefficients for pairwise comparisons of detectable PFASs in plasma from 323 subjects included in the study.

	PFBA	PFHxS	PFOA	PFOS
PFHxS	0.0520
PFOA	0.0617	0.7072
PFOS	0.0591	0.8406	0.7248
PFNA	0.0127	0.7133	0.7759	0.8406

Open in a new tab

In general, serum-PFAS concentrations were higher at older ages, in men, and among those of Western European origin. Although the presence of chronic disease did not seem to be associated with PFAS, the plasma concentrations appeared to be higher in the presence of kidney disease (Table 2 and S2 Table).

Table 2. Median plasma-PFAS concentrations (25^th, 75^th percentiles) in ng/mL by population characteristics.

			PFAS (ng/mL) median (25th,75th percentile)
Population characteristics	n (%)	PFBA	PFHXS	PFOA	PFOS	PFNA
Total	323 (100)	<LOD (<LOD, 0.04)	0.48 (0.28, 0.71)	0.77 (0.43, 1.18)	4.86 (2.85, 8.29)	0.38 (0.23, 0.59)
Age (years)
30–39	37 (11)	<LOD (<LOD, 0.03)	0.32 (0.19, 0.46)	0.59 (0.43, 0.86)	3.30 (1.89, 5.27)	0.29 (0.21, 0.43)
40–49	64 (20)	<LOD (<LOD, 0.03)	0.35 (0.15, 0.57)	0.58 (0.35, 0.89)	3.11 (2.24, 5.06)	0.27 (0.19, 0.39)
50–59	106 (33)	<LOD (<LOD, <LOD)	0.50 (0.31, 0.75)	0.83 (0.43, 1.18)	5.41 (2.79, 8.84)	0.40 (0.24, 0.61)
60–70	116 (36)	<LOD (<LOD, 0.05)	0.56 (0.39, 0.89)	0.97 (0.56, 1.51)	6.11 (3.83, 9.60)	0.48 (0.30, 0.70)
p-value ^a		0.008	<0.001	<0.001	<0.001	<0.001
Sex
Male	174 (54)	<LOD (<LOD, 0.04)	0.59 (0.40, 0.87)	0.81 (0.51, 1.26)	5.96 (3.65, 10.17)	0.40 (0.25, 0.61)
Female	149 (46)	<LOD (<LOD, 0.04)	0.35 (0.17, 0.52)	0.70 (0.40, 1.04)	3.43 (2.06, 5.66)	0.36 (0.22, 0.56)
p-value ^b		0.713	<0.001	0.011	<0.001	0.131
Kidney disease
yes	34 (11)	<LOD (<LOD, 0.06)	0.55 (0.34, 0.77)	0.91 (0.54, 1.46)	5.60 (3.08, 8.38)	0.50 (0.24, 0.67)
no	289 (89)	<LOD (<LOD, 0.03)	0.47 (0.28, 0.71)	0.76 (0.43, 1.15)	4.76 (2.82, 8.10)	0.36 (0.23, 0.57)
p-value ^b		0.040	0.466	0.065	0.489	0.141
Other chronic disease
Yes	220(68)	<LOD (<LOD, 0.04)	0.47 (0.28, 0.68)	0.71 (0.42, 1.15)	4.70 (2.87, 7.99)	0.38 (0.23, 0.57)
No	103 (32)	<LOD (<LOD, 0.03)	0.51 (0.28, 0.76)	0.87 (0.47, 1.23)	5.35 (2.72, 8.41)	0.41 (0.23, 0.65)
p-value ^b		0.075	0.314	0.124	0.850	0.407
National origin
Western Europe	224 (69)	<LOD (<LOD, 0.04)	0.52 (0.35, 0.76)	0.91 (0.60, 1.29)	5.61 (3.40, 9.18)	0.43 (0.29, 0.64)
Other	99 (31)	<LOD (<LOD, 0.04)	0.34 (0.16, 0.57)	0.44 (0.31, 0.80)	2.86 (1.61, 5.13)	0.23 (0.16, 0.36)
p-value ^b		0.552	<0.001	<0.001	<0.001	<0.001
Place of inclusion
Odense	48 (15)	<LOD (<LOD, 0.06)	0.45 (0.32, 0.69)	0.67 (0.42, 0.95)	4.67 (3.29, 8.09)	0.36 (0.24, 0.45)
Copenhagen	275 (85)	<LOD (<LOD, 0.03)	0.48 (0.28, 0.72)	0.79 (0.44, 1.20)	4.89 (2.72, 8.31)	0.39 (0.23, 0.62)
p-value ^b		0.003	0.967	0.203	0.697	0.299
Timing of blood sampling
After diagnosis—1 week before	193 (60)	<LOD (<LOD, 0.04)	0.48 (0.30, 0.71)	0.70 (0.40, 1.11)	4.63 (2.83, 7.65)	0.34 (0.23, 0.56)
>1 week—1 year before	46 (14)	<LOD (<LOD, 0.03)	0.45 (0.21, 0.66)	0.82 (0.38, 1.35)	4.81 (2.36, 8.62)	0.38 (0.20, 0.65)
> 1year before diagnosis	84 (26)	<LOD (<LOD, 0.03)	0.50 (0.30, 0.72)	0.87 (0.57, 1.22)	5.48 (3.10, 10.28)	0.45 (0.28, 0.65)
p-value ^a		0.185	0.756	0.085	0.209	0.053

Open in a new tab

^a Variables with more than two categories tested using Kruskal-Wallis rank test.

^b Binary variables tested using Wilcoxon rank-sum test.

In the study population, males, older subjects, and those with chronic disease, were more frequently represented among subjects with severe COVID-19, while there was no difference in regard to national origin for disease severity (Table 3). The PFAS-associations with disease severity were similar in Western Europeans and subjects with other backgrounds (P > 0.2 for population differences).

Table 3. COVID-19 severity by population characteristics.

	COVID-19 severity
Population characteristics	No. of subjects	No hospitalization	Hospitalization	Intensive care unit and/or deceased
Total No. of subjects (%)	323 (100)	108 (33)	162 (50)	53 (16)
Age (years) median (25th,75th percentile)	55 (46, 62)	49 (41, 57)	57 (51, 63)	62 (53, 67)
P value ^a	<0.001
Sex
Male, n (%)	174 (100)	44 (25)	94 (54)	36 (21)
Female, n (%)	149 (100)	64 (43)	68 (46)	17 (11)
P value ^b	0.002
Kidney disease
Yes, n (%)	34 (11)	7 (21)	13 (38)	14 (41)
No, n (%)	289 (89)	101 (35)	149 (52)	39 (13)
P value ^b	<0.001
Other chronic disease
Yes, n (%)	220 (100)	54 (25)	119 (54)	47 (21)
No, n (%)	103 (100)	54 (52)	43 (42)	6 (6)
P value ^b	<0.001
National origin
Western Europe, n (%)	224 (100)	76 (34)	113 (50)	35 (16)
Other, n (%)	99 (100)	32 (32)	49 (49)	18 (18)
P value ^b	0.844
Days between blood sampling and diagnosis
median (25th,75th percentile)	0 (-1, 393)	335 (22.5, 639.5)	0 (-1, 0)	0 (-2, 1)
P value ^a	<0.001

Open in a new tab

^a Associations tested using Kruskal-Wallis rank test.

^b Associations tested using Pearson’s chi-squared test.

A more severe disease outcome was associated with higher plasma-PFBA concentrations, also after adjustment for all covariates (Table 4 and S4 Table). None of the other PFASs showed a similar tendency. If leaving out presence of chronic disease as a non-significant predictor, the adjusted OR for PFBA was 1.77 (95% CI, 1.09, 2.87). More importantly, when excluding samples collected earlier than one week before the time of diagnosis (148 samples), or more than one week later (5 samples), stronger ORs emerged for PFBA (Table 4). Counter to the a priori hypothesis, some PFASs, including PFHxS, seemed associated with a lower risk, but this tendency was weakened when relying on plasma samples collected in close connection to the diagnosis of corona infection (Table 4 and S3 Table).

Table 4. Ordered logistic regression OR of increased Covid-19 severity for an increase by 1 ng/mL in plasma-PFAS concentrations.

PFAS	No. of subjects	OR (95% CI)		No. of subjects	OR (95% CI)
PFAS	No. of subjects	Crude	Adjusted for main covariates^a	No. of subjects	Exposure at time of diagnosis^a^,^b
PFBA (>LOD/<LOD)	104/219	2.19 (1.39, 3.46)	1.57 (0.96, 2.58)	61/109	2.10 (1.02, 4.33)
PFHxS (ng/mL)	323	0.85 (0.63, 1.15)	0.52 (0.29, 0.91)	170	0.52 (0.24, 1.14)
PFHxS ^c (ng/mL)	322	1.00 (0.62, 1.61)	0.52 (0.29, 0.93)	169	0.53 (0.22, 1.27)
PFOA (ng/mL)	323	0.99 (0.72, 1.36)	0.83 (0.57, 1.20)	170	0.62 (0.36, 1.08)
PFOS (ng/mL)	323	1.00 (0.96, 1.04)	0.97 (0.92, 1.02)	170	0.98 (0.89, 1.07)
PFNA (ng/mL)	323	1.18 (0.67, 2.09)	1.04 (0.54, 2.02)	170	0.73 (0.25, 2.11)

Open in a new tab

^a Adjusted for age, sex, kidney disease, other chronic disease, national origin, place of testing, and days between blood sampling and diagnosis.

^b Excluding individuals who had blood sampled more than one week before or after diagnosis.

^c PFHxS >10 ng/mL excluded.

In dichotomous analyses comparing severities of the disease (S4 Table), detectable PFBA in plasma also showed a clear association with a more severe clinical course of the disease, most pronounced for odds between hospitalization and admission to intensive care unit/death, especially when based on plasma samples obtained at the time of diagnosis or up to one week before where the adjusted OR was 5.18 (1.29, 20.72). No such tendency was seen for the other PFASs detected (S4 Table). The association between PFBA and disease severity was similar for men and women (Fig 1).

Fig 1 — Results are shown for 44 men and 64 women with up to two weeks of hospitalization, 94 men and 68 women with longer hospitalization, and 36 men and 17 women admitted to the intensive care unit (ICU) or deceased (P = 0.003).

Discussion

The present study aimed at determining the potential aggravation of COVID-19 associated with elevated exposures to PFASs. Several of these substances are known immunotoxicants in laboratory animals [35] and in humans [8, 9]. In addition to immunotoxicity, major PFASs can potentially interfere with major pathways that are predictive of a serious clinical outcome of the infection [14]. An association of PFAS exposure with disease severity therefore appears biologically plausible.

Among the PFASs, presence of detectable PFBA in plasma showed the strongest positive association with the severity of the disease. This finding may at first seem surprising, as this PFAS has a short elimination half-life in the blood and is often considered of less importance to health [27]. However, in tissue samples from autopsies, PFBA is the only PFAS that is substantially accumulated in the lungs [28]. Given the persistence of the PFASs in general, the unique retention of PFBA in lung tissue may offer a clue to interpreting the findings in this study.

Some odds ratios for PFBA were weakened after adjustment for covariates. However, adjustment for all covariates may result in over-adjustment bias. Thus, older age and male sex are known to be strong predictors of higher blood-PFAS concentrations, and simple adjustment for these factors could potentially result in a bias toward the null. As PFAS exposure has been linked to important comorbidities, such as diabetes and obesity [25, 26], both of which may exacerbate the virus infection, adjustment for chronic disease may also not be justified. Leaving it out slightly strengthened the PFBA association with the disease severity. The strongest associations for PFBA, but not for other PFASs, appeared when focusing on the most representative blood samples obtained close to the time of diagnosis.

An additional consideration is that the present study relates to low background exposure levels, in comparison with PFAS concentrations to findings in, e.g., U.S. adults [38]. Given the wide occurrence of highly contaminated drinking water in other countries [39], the present study results should not be interpreted as evidence that most PFASs do not contribute to a worsened clinical course of COVID-19.

The results for PFBA in this study appear to parallel the findings in regard to other environmental toxicants, viz., air pollutants [1–4] and suggest a need to ascertain the impact of relevant occupational or environmental exposures on COVID-19 severity. Of note, the evidence on air pollution relies solely on ecological study designs without measures of individual levels of exposure, while the present study benefitted from measurements of plasma-PFAS concentrations of all study subjects.

In regard to limitations, the study population may not be representative of corona-positive subjects, as inclusion in the study depended solely on the existence of plasma from diagnostic blood samples at the participating hospitals. Thus, subjects with chronic disease or more severe COVID-19 likely had more frequent hospital visits or longer admissions and thereby a greater chance of having plasma available for inclusion in this study. With a corona-related fatality rate of Danish blood donors below 70 years of age at 89 per 100,000 infections [40], the presence of 17 deaths in the present material (i.e., against 0.3 deaths expected) confirms that the blood samples represent a highly selected population. Still, a total of 108 subjects were known to have been infected, though not hospitalized. In many cases, their plasma had been stored on previous occasions, and the PFAS concentrations may reflect slightly higher exposures in the recent past [8], which could possibly explain the apparent protective effects of some PFASs, although adjustment for the time interval since sample collection was included in the analyses. However, the strongest associations for PFBA, but not for other PFASs, were seen when excluding samples not obtained in close temporal connection with the infection.

The study population included mostly older subjects who were more frequently male, and a large proportion of foreign-born subjects and second-generation immigrants (Table 3), thereby possibly deviating from the background population of corona-infected patients in Denmark. Still, the results do not suggest major biases affecting PFAS exposure and its association with COVID-19 outcomes.

Among immigrants, adverse associations appeared slightly stronger, also after adjustments, in accordance with national origin, perhaps as related to demographic or social factors, resulting in a greater likelihood also to PFAS-associated aggravation of the infection. Difference in age, sex, or comorbidities did not explain this tendency, but is in agreement with previous findings of ethnic differences in vulnerability [41]. However, national origin may be a surrogate marker for other factors, such as exposure at work or exposure within crowded households, as immigrant origin tends to be associated with certain occupations including front-line workers and living in areas with higher population density [42]. Still, in agreement with higher PFAS exposure being associated with higher socioeconomic position [43], we found that the association between PFBA exposure and disease severity was independent of national origin.

Conclusions

Increased plasma-PFBA concentrations were associated with a greater severity of COVID-19 prognosis, and this tendency remained after adjustment for sex, age, comorbidities, national origin, sampling location and time. Although occurring in fairly low concentrations in plasma, PFBA is known to accumulate in the lungs. Thus, as immunotoxic substances, the PFASs may well contribute to the severity of COVID-19. The present findings on a short-chain PFAS at background exposures suggest a need to ascertain if elevated exposures to environmental immunotoxicants may worsen the outcome of the SARS-CoV-2 infection.

Supporting information

S1 Table. Spearman’s correlation coefficients for pairwise comparisons of detectable PFASs in plasma from 323 subjects included in the study.

(DOCX)

Click here for additional data file.^{(24.6KB, docx)}

S2 Table. Median plasma concentrations of additional PFASs (25^th-75^th percentiles) in ng/mL by population characteristics.

(DOCX)

Click here for additional data file.^{(23.1KB, docx)}

S3 Table. Ordered logistic regression odds ratios (ORs) of increased COVID-19 severity at an increase by 1 ng/mL in plasma concentrations of additional PFASs.

(DOCX)

Click here for additional data file.^{(20.1KB, docx)}

S4 Table. Logistic regression odds ratios (ORs) of increased COVID-19 severity at an increase by 1 ng/mL in plasma concentrations of all PFASs detected.

(DOCX)

Click here for additional data file.^{(27KB, docx)}

Acknowledgments

The Danish Departments of Clinical Microbiology, the Danish Microbiology Database and the section for Data Integration and Analysis at Statens Serum Institut collated the national COVID-19 data. Plasma samples were identified and provided by Statens Serum Institut and the Department of Clinical Biochemistry and Pharmacology at Odense University Hospital.

Data Availability

Data cannot be shared publicly because of the need to protect personal data. Anonymous data are available from the secure server at the Danish Health Data Authority, pending necessary approvals from the Authority (instructions at www.sundhedsdata.dk) and the Regional Committee on Health Research Ethics for researchers who meet the criteria for access to confidential data. Aggregated data underlying the results presented in the study are available from the corresponding author, provided that no data are based on less than 5 individuals.

Funding Statement

Novo Nordisk Foundation (NNF20SA0062871) (LB, KM). National Institute for Environmental Health Sciences (ES027706) (PG). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Fattorini D, Regoli F. Role of the chronic air pollution levels in the Covid-19 outbreak risk in Italy. Environ Pollut 2020;264:114732 10.1016/j.envpol.2020.114732 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Zhu Y, Xie J, Huang F, Cao L. Association between short-term exposure to air pollution and COVID-19 infection: Evidence from China. Sci Total Environ 2020;727:138704 10.1016/j.scitotenv.2020.138704 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Bashir MF, Ma BJ, Bilal, et al. Correlation between environmental pollution indicators and COVID-19 pandemic: A brief study in Californian context. Environ Res 2020;187:109652 10.1016/j.envres.2020.109652 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Wu X, Nethery RC, Sabath MB, Braun D, Dominici F. Air pollution and COVID-19 mortality in the United States: Strengths and limitations of an ecological regression analysis. Sci Adv 2020;6 10.1126/sciadv.abd4049 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Rooney AA, Luebke RW, Selgrade MK, Germolec DR. Immunotoxicology and its application in risk assessment. Exp Suppl 2012;101:251–87. 10.1007/978-3-7643-8340-4_9 [DOI] [PubMed] [Google Scholar]
6.World Health Organization W. Guidance for Immunotoxicity Risk Assessment for Chemicals: IPCS Harmonization Project. Geneva, Switzerland: 2012. [Google Scholar]
7.DeWitt JC, Germolec DR, Luebke RW, Johnson VJ. Associating Changes in the Immune System with Clinical Diseases for Interpretation in Risk Assessment. Curr Protoc Toxicol 2016;67:18.1–1-.1.22. 10.1002/0471140856.tx1801s67 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.European Food Safety Authority. Risk to human health related to the presence of perfluoroalkyl substances in food Panel on Contaminants in the Food Chain. Parma: EFSA: 2020:e06223. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Grandjean P, Andersen EW, Budtz-Jorgensen E, et al. Serum vaccine antibody concentrations in children exposed to perfluorinated compounds. JAMA 2012;307:391–7. 10.1001/jama.2011.2034 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Kielsen K, Shamim Z, Ryder LP, et al. Antibody response to booster vaccination with tetanus and diphtheria in adults exposed to perfluorinated alkylates. J Immunotoxicol 2016;13:270–3. 10.3109/1547691X.2015.1067259 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Dalsager L, Christensen N, Husby S, et al. Association between prenatal exposure to perfluorinated compounds and symptoms of infections at age 1–4 years among 359 children in the Odense Child Cohort. Environ Int 2016;96:58–64. 10.1016/j.envint.2016.08.026 [DOI] [PubMed] [Google Scholar]
12.Granum B, Haug LS, Namork E, et al. Pre-natal exposure to perfluoroalkyl substances may be associated with altered vaccine antibody levels and immune-related health outcomes in early childhood. J Immunotoxicol 2013;10:373–9. 10.3109/1547691X.2012.755580 [DOI] [PubMed] [Google Scholar]
13.Goudarzi H, Miyashita C, Okada E, et al. Prenatal exposure to perfluoroalkyl acids and prevalence of infectious diseases up to 4years of age. Environ Int 2017;104:132–8. 10.1016/j.envint.2017.01.024 [DOI] [PubMed] [Google Scholar]
14.Wu Q, Coumoul X, Grandjean P, Barouki R, Audouze K. Endocrine disrupting chemicals and COVID-19 relationships: a computational systems biology approach. Environ Int 2020;(in press):106232 10.1016/j.envint.2020.106232 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Grasselli G, Zangrillo A, Zanella A, et al. Baseline Characteristics and Outcomes of 1591 Patients Infected With SARS-CoV-2 Admitted to ICUs of the Lombardy Region, Italy. JAMA 2020;323:1574–81. 10.1001/jama.2020.5394 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Liang W, Liang H, Ou L, et al. Development and Validation of a Clinical Risk Score to Predict the Occurrence of Critical Illness in Hospitalized Patients With COVID-19. JAMA Intern Med 2020;180:1081–9. 10.1001/jamainternmed.2020.2033 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Atyeo C, Fischinger S, Zohar T, et al. Distinct Early Serological Signatures Track with SARS-CoV-2 Survival. Immunity 2020;53:524–32 e4. 10.1016/j.immuni.2020.07.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Goncalves Mendes Neto A, Lo KB, Wattoo A, et al. Bacterial infections and patterns of antibiotic use in patients with COVID-19. J Med Virol 2020;(in press). 10.1002/jmv.26441 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Petrilli CM, Jones SA, Yang J, et al. Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease 2019 in New York City: prospective cohort study. BMJ 2020;369:m1966 10.1136/bmj.m1966 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Zhou Y, He Y, Yang H, et al. Development and validation a nomogram for predicting the risk of severe COVID-19: A multi-center study in Sichuan, China. PLoS One 2020;15:e0233328 10.1371/journal.pone.0233328 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Zhao X, Zhang B, Li P, et al. Incidence, clinical characteristics and prognostic factor of patients with COVID-19: a systematic review and meta-analysis. medRxiv 2020:2020.03.17.20037572. [Google Scholar]
22.Cummings MJ, Baldwin MR, Abrams D, et al. Epidemiology, clinical course, and outcomes of critically ill adults with COVID-19 in New York City: a prospective cohort study. Lancet 2020;395:1763–70. 10.1016/S0140-6736(20)31189-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Tartof SY, Qian L, Hong V, et al. Obesity and Mortality Among Patients Diagnosed With COVID-19: Results From an Integrated Health Care Organization. Ann Intern Med 2020;173:773–81. 10.7326/M20-3742 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Yu CH, Riker CD, Lu SE, Fan ZT. Biomonitoring of emerging contaminants, perfluoroalkyl and polyfluoroalkyl substances (PFAS), in New Jersey adults in 2016–2018. Int J Hyg Environ Health 2020;223:34–44. 10.1016/j.ijheh.2019.10.008 [DOI] [PubMed] [Google Scholar]
25.Sun Q, Zong G, Valvi D, Nielsen F, Coull B, Grandjean P. Plasma Concentrations of Perfluoroalkyl Substances and Risk of Type 2 Diabetes: A Prospective Investigation among U. S. Women. Environ Health Perspect 2018;126:037001 10.1289/EHP2619 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Liu G, Dhana K, Furtado JD, et al. Perfluoroalkyl substances and changes in body weight and resting metabolic rate in response to weight-loss diets: A prospective study. PLoS Med 2018;15:e1002502 10.1371/journal.pmed.1002502 [DOI] [PMC free article] [PubMed] [Google Scholar]
27.ATSDR. Draft toxicological profile for perfluoroalkyls. Atlanta, GA: Agency for Toxic Substances and Disease Registry; 2018. [PubMed] [Google Scholar]
28.Perez F, Nadal M, Navarro-Ortega A, et al. Accumulation of perfluoroalkyl substances in human tissues. Environ Int 2013;59:354–62. 10.1016/j.envint.2013.06.004 [DOI] [PubMed] [Google Scholar]
29.Reilev M, Kristensen KB, Pottegard A, et al. Characteristics and predictors of hospitalization and death in the first 11 122 cases with a positive RT-PCR test for SARS-CoV-2 in Denmark: a nationwide cohort. Int J Epidemiol 2020. 10.1093/ije/dyaa140 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Voldstedlund M, Haarh M, Molbak K, MiBa Board of R. The Danish Microbiology Database (MiBa) 2010 to 2013. Euro Surveill 2014;19 10.2807/1560-7917.es2014.19.1.20667 [DOI] [PubMed] [Google Scholar]
31.Schmidt M, Pedersen L, Sorensen HT. The Danish Civil Registration System as a tool in epidemiology. Eur J Epidemiol 2014;29:541–9. 10.1007/s10654-014-9930-3 [DOI] [PubMed] [Google Scholar]
32.Schmidt M, Schmidt SA, Sandegaard JL, Ehrenstein V, Pedersen L, Sorensen HT. The Danish National Patient Registry: a review of content, data quality, and research potential. Clin Epidemiol 2015;7:449–90. 10.2147/CLEP.S91125 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Andersen JS, de Fine Olivarius N, Krasnik A. The Danish National Health Service Register. Scand J Public Health 2011;39:34–7. 10.1177/1403494810394718 [DOI] [PubMed] [Google Scholar]
34.Jain RB, Ducatman A. Perfluoroalkyl substances follow inverted U-shaped distributions across various stages of glomerular function: Implications for future research. Environ Res 2019;169:476–82. 10.1016/j.envres.2018.11.033 [DOI] [PubMed] [Google Scholar]
35.DeWitt JC, Blossom SJ, Schaider LA. Exposure to perfluoroalkyl and polyfluoroalkyl substances leads to immunotoxicity: epidemiological and toxicological evidence. J Expo Sci Environ Epidemiol 2019;29:148–56. 10.1038/s41370-018-0097-y [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Zeng XW, Li QQ, Chu C, et al. Alternatives of perfluoroalkyl acids and hepatitis B virus surface antibody in adults: Isomers of C8 Health Project in China. Environ Pollut 2020;259:113857 10.1016/j.envpol.2019.113857 [DOI] [PubMed] [Google Scholar]
37.Haug LS, Thomsen C, Becher G. A sensitive method for determination of a broad range of perfluorinated compounds in serum suitable for large-scale human biomonitoring. J Chromatogr A 2009;1216:385–93. 10.1016/j.chroma.2008.10.113 [DOI] [PubMed] [Google Scholar]
38.CDC. National Report on Human Exposure to Environmental Chemicals, Updated Tables, January 2019. Atlanta, GA: Centers for Disease Control and Prevention; 2019. [Google Scholar]
39.Hu XC, Andrews DQ, Lindstrom AB, et al. Detection of Poly- and Perfluoroalkyl Substances (PFASs) in U.S. Drinking Water Linked to Industrial Sites, Military Fire Training Areas, and Wastewater Treatment Plants. Environ Sci Technol Lett 2016;3:344–50. 10.1021/acs.estlett.6b00260 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Erikstrup C, Hother CE, Pedersen OBV, et al. Estimation of SARS-CoV-2 infection fatality rate by real-time antibody screening of blood donors. Clin Infect Dis 2020:in press. 10.1093/cid/ciaa849 [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Williamson EJ, Walker AJ, Bhaskaran K, et al. Factors associated with COVID-19-related death using OpenSAFELY. Nature 2020;584:430–6. 10.1038/s41586-020-2521-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Lassale C, Gaye B, Hamer M, Gale CR, Batty GD. Ethnic disparities in hospitalisation for COVID-19 in England: The role of socioeconomic factors, mental health, and inflammatory and pro-inflammatory factors in a community-based cohort study. Brain Behav Immun 2020;88:44–9. 10.1016/j.bbi.2020.05.074 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Montazeri P, Thomsen C, Casas M, et al. Socioeconomic position and exposure to multiple environmental chemical contaminants in six European mother-child cohorts. Int J Hyg Environ Health 2019;222:864–72. 10.1016/j.ijheh.2019.04.002 [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0244815.r001

Decision Letter 0

Jaymie Meliker

18 Nov 2020

PONE-D-20-32774

Severity of COVID-19 at elevated exposure to perfluorinated alkylates

PLOS ONE

Dear Dr. Grandjean,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jan 02 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Jaymie Meliker, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In ethics statement in the manuscript and in the online submission form, please provide additional information about the patient records/samples used in your retrospective study. Specifically, please ensure that you have discussed whether all samples were fully anonymized before you accessed them and/or whether the IRB or ethics committee specifically waived the requirement for informed consent.

3. Please note that PLOS does not permit references to “data not shown.” Authors should provide the relevant data within the manuscript, the Supporting Information files, or in a public repository. If the data are not a core part of the research study being presented, we ask that authors remove any references to these data.

4.We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

5.Thank you for stating the following in the Competing Interests section:

[Apart from PG having served as health expert in lawsuits on environmental contamination, which does not affect the adherence to all PLOS ONE policies, the authors have no competing interests to declare, financial or otherwise. ].

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors report associations between PFAS and clinical progression of COVID19 among 323 Danish patients with SARS-COV-2 infection and archived blood specimens. The report focuses on higher odds for more severe clinical COVID19 disease associated with greater concentrations of blood PFBA, a short-chain and short-lived PFAS, in covariate adjusted models. Yet, consistent results are also presented for a “protective” PFHxS association, stronger it appears than that for PFBA. Despite the highly selected nature of the study population (as acknowledged by the authors), and the limited sample size, this paper has the potential to make an important contribution to the developing PFAS-human immunotoxicity literature, and especially its impact on the COVID19 pandemic. Still, there are several points that will benefit from additional development and other that merit clarification. Overall, a more balanced interpretation of the literature and the current results will be helpful.

Major points:

1. There are no line numbers or page numbers, this makes review/providing feedback challenging

2. Although the terms “ordered” and “ordinal” logistic regression are used throughout, I believe these to be synonymous. If so, the interpretation of the odds ratio from the ordinal logistic is generally for the cumulative odds; i.e., falling into a higher ordered category or less relative to the reference group. This is a subtle but important distinction of interpretation that should be clarified throughout.

3. I’m not sure that the title is appropriate; the outcome does not appear to be “severity of COVID-19” per se (e.g., viral load, etc.), given that comorbidities may also play an important role in the extent of treatment. The author might consider revising the title to more closely reflect the study outcome, such as “clinical course of COVID-19 infection” or something along those lines.

4. The links to air pollution studies, seem tenuous at best. The paper starts out with a focus on air pollution and limitations to the exposure assessment strategies that have been implemented for studies of COVID19. This seems like an “apples to oranges” comparison, and leaves the reader anticipating a study of air pollutants, not PFAS. I do appreciate the use of air pollution studies as a “proof of concept” but at present the focus is too great in my opinion.

5. Methods, “…we also included subjects, mainly those not hospitalized, whose plasma…” How many? While agreed that this may be less of an issue for the long-chain PFAS with ½-lived measured in years, even this difference might introduce or obfuscate a signal if systematically different between those with more/less severe disease (essentially a selection bias). Furthermore, the serum half-life for PFBA is measured in days. Is this a problem in terms of potential iatrogenic exposures to PFAS compounds during hospital stay/treatment among those with more severe disease? Specifically, where is PFBA used/found?

6. Methods, Chemical Analysis, “…were replaced by LOD/2…” Imputation of values below the LOD can lead to bias, in particular with a large proportion of values imputed (this shifts the mean and truncates the variance). The authors should repeat the analysis using the original, machine-read values for the analysis.

7. Methods, Statistical Analysis, “Associations between comorbidities…” Why can’t the results be presented without the corresponding cross-tabs, as differences with confidence intervals for example?

8. Methods, Statistical Analysis, A brief description of ordered (ordinal?) logistic regression could be provided as this approach is uncommon. While two terms are used here and in the abstract I am under the impression that they are synonymous, though unclear (see comment #2 above). Please clarify.

9. Methods, Statistical Analysis, “The assumption of PFAS linearity…” With only n=323, a more liberal p<0.10 seems more appropriate to assess PFAS linearity if a potential non-linear association is suspected. While I appreciate the latter when it comes to endocrine disrupting effects I am not familiar with non-linear associations reported for immune effects; is this exploratory? Still, this would seem to have been infeasible for dichotomized PFBA and PFBS, a limitation of the approach.

10. Methods, Statistical Analysis, “Potential confounding variable were identified based on a priori…” What a priori knowledge, specifically? What criteria to define a confounder and what supporting literature/citations?

11. Methods, Statistical Analysis, Did the authors adjust for plasma/serum albumin or renal clearance? What is the relationship between serum albumin and kidney function and COVI19 or and COVID19. Reports of compromised renal function among COVID19 patients might impact PFBA clearance for example (in particular given the short half-life), leading to reverse causal association. While the latter is less likely for inflated albumin due to infection (as the other PFAS were not associated with more severe disease progression), the issue merits attention.

12. Method, Statistical Analysis, How were diagnoses confirmed? Is there a difference in the diagnosis among those hospitalized vs. those not hospitalized?

13. Methods, Statistical Analysis. Was adjustment made for socioeconomic status? This is an important predictor of COVID19 outcomes in some countries and often a predictor of environmental exposures, including PFAS. If not adjusted, the authors should at least touch upon the relevance/importance.

14. Discussion, “Among…the PFASs, presence of detectable…” PFHxS showed the strongest association according to Table 4.

15. Discussion “The associations of PFASs…” The results do not support this conclusion, only PFBA above the LOD was associated with a more severe clinical course of COVID19, the other PFAS, including the persistent long-chain PFAS, were null or even protective (e.g., for PFHxS).

16. Discussion, this appears to be a (mostly) cross-sectional study; is reverse causation a concern here with respect in particular to renal clearance/serum albumin PFAS binding (see comment #11 above)

17. Discussion, “Low background exposure results…” How did the PFAS levels compare to other reports of associations with immune function?

18. Discussion, “Among immigrants, adverse associations appeared…” Is there any education/income data that might help to tease apart a sociodemographic disparity here? If not, is there national data that the authors can use to place this issue into a socioeconomic and/or residential context?

19. Conclusions, The authors should also report on the “protective” effect of PFHxS, a long-lived PFAS.

Minor points

1. Abstract: “Methods”: “…at the background exposures…” Unclear.

2. Abstract, “Results”: increasing severities of disease; again, does the outcome variable truly capture “disease severity.”?

20. Introduction, I am somewhat concerned about reliance on citing preprints (four of these that I found) given that these publications have not yet been “vetted” by peers and very well may change, in some circumstances dramatically, before eventual publication. This is analogous to citing four “submitted” papers, appropriate on occasion and perhaps this is one of those occasions. Given the novelty of the COVID19 pandemic and the nascent literature I’m not quite sure how to get around this but perhaps the papers can be cited in text as “submitted” or “preprint” to clarify for the reader.

3. Introduction, “…which accumulates in the lung..” This was shown by a single study in a single Spanish population. While I agree that it is important a very important point, it has not been conclusively established to my knowledge and should be reworded to reflect this, such as “..shown to accumulate in the lung.”

4. Introduction, a more comprehensive treatment of the PFAS immune function literature should be undertaken. Specifically, is the specific immune response to COVID19 threatened by PFAS according to previous work or is this a pending data gap too?

5. Methods, is exogenous contamination of specimens, for example Teflon caps, a concern for blood specimens?

6. Methods, “…no hospital admission and completed infection within 14 days of testing positive…” How was this determined?

7. Methods, is there a sufficient number of participants to stratify according to presence absence of obesity/chronic disease? Even in unadjusted analysis? Some prior work indicates difference in male/female PFAS-immunotoxicity; any difference between PFAS associations in men/women in this study?

8. Methods, Chemical Analysis, “Succeeding series…” Unclear.

9. Methods, Chemical Analysis, “…for PFAS concentrations, including…” Why these five PFAS specifically?

10. Methods, Chemical Analysis, “Because more than half the PFBA and PFBS…” PFBS was not mentioned in the “Chemical Analysis” section above or reported in the results as far as I can tell.

11. Methods, Statistical Analysis, “…presence of comorbidities…” What comorbidities did participants have, specifically?

12. Methods, Statistical Analysis, “…demographic groups…” What demographic groups?

13. Methods, Statistical Analysis, “The test could not be fitted…” Why not?

14. Methods, Statistical Analysis, “Among those of Western European national origin…” This seems better suited to the Results section.

15. Statistical Analysis, “…risk of certain chronic disease…” Which chronic diseases, specifically?

16. Results, “Correlated well..” Unclear.

17. Discussion, The first paragraph should summarize the major findings/and implications of the research. This interpretation is better suited to later components.

18. Discussion, “… the only PFAS that is substantially accumulated in the …” This has been shown to substantially accumulate in one study, please clarify this point (see comment #3 above)

19. Discussion, “…major pathways that are predictive…” What pathways, specifically?

20. Discussion, “Thus older age and male sex…” These are also strong predictors of COVId19 severity, making the likely to be confounding variables and so adjustment was appropriate and the diminished associations simply reflecting unbiased effect estimates. This seems like circular reasoning and should be clarified or removed from the manuscript.

21. Discussion, “As PFAS exposure has been linked to important comorbidities…not be justified.” Agreed, suggest stratification/interaction analysis if plausible, but comorbidities are probably not confounders here. A DAG would help to clarify this issue.

Reviewer #2: Title: Severity of COVID-19 at elevated exposure to perfluorinated alkylates

The aim of the present study was to assess if elevated background exposures to immunotoxic PFASs are associated with the severity of COVID-19 development. PFAS levels were measured in plasma samples from 323 subjects aged 30-70 years with known SARS-CoV-2 infections. Logistic regression analyses were performed to study the association between PFAS levels and disease severity. Elevated plasma-PFBA concentrations were associated with an increased risk of more severe course of CIVID-19. Since there are studies showing that PFASs can cause immunosuppression in humans, the present study may contribute with valuable information on the interplay between environmental chemicals – immune responses – disease severity of Covid-19.

Abstract

A1. “The course of coronavirus disease 2019 ( COVID-19) seems to be aggravated by air pollution, and some industrial chemicals, such as the perfluorinated alkylate substances (PFASs), are immunotoxic and may contribute as well.” What is the last part of the sentence referring to (PFASs or some industrial chemicals)? The sentence should be rewritten to make it more clear.

A2. “We used ordinal and ordered logistic regression analyses to determine associations between PFAS concentrations and disease outcome.“ Maybe I am wrong here, but isn’t ordered logistic regression the same as ordinal logistic regression?

Materials and methods

M1. Statistical analysis, 2nd section: PFBS – write the full name. Alternatively delete since no findings on PFBS are shown in the manuscript.

Results

R1. “PFBA was lower, but the origin of the samples was only weakly associated with plasma-PFAS concentrations (Table 2).” Since PFBA is used as a binary variable (below/above LOD), the authors should consider showing PFBA as n(%) where n is the number of samples above LOD. Showing the levels as mean and 95% CI does not give any information that are helpful

when interpreting the findings in Table 2.

R2. “A more severe disease outcome was associated with higher plasma-PFBA concentrations, also after adjustment for all covariates (Table 4).” This statement should be moderated in that this association becomes non-significant after adjustment for covariates (1.62(0.99, 2.64)). Consider moving this sentence from the Discussion to the Results section “The associations of PFASs with a more serious course of COVID-19 are weakened after adjustment for covariates, and some regression coefficients and ORs are below 1.”

R3. “In additional analyses, marginal changes in the ORs occurred when plasma samples obtained more than 60 days before diagnosis were excluded.” This may not be enough to adjust for the short half-life of 72 hours for PFBA. See also comment D3 below.

R4. “If not adjusted for the presence of chronic disease, the adjusted OR for PFBA was 1.77 (95% CI, 1.09, 2.87).” What is the rationale for excluding chronic disease as a covariate in this analysis? Please give an explanation for not adjusting for chronic disease. See also comment D2 below.

R5. “In dichotomous analyses comparing severities of the disease, detectable PFBA in plasma also showed a clear association with a more severe clinical course of the disease, most pronounced for odds between hospitalization and admission to intensive care unit/death (data not shown).” Isn’t this the finding shown in Table 4 and already mentioned above? Please clarify.

R6. To make it easier for the readers to interpret all the findings reported in the Results section, the author should consider showing results of all additional analyses as supplementary material.

Discussion

D1. “Given the persistence of the PFASs in general, the unique retention of PFBA in lung tissue may offer a clue to interpreting the findings in this study.” This is a very interesting theory. Are there more information on this issue to support such a theory, e.g.:

- Are there any studies showing the correlation between blood and lung levels of PFBA? This would be interesting to know since you are looking at plasma levels and not lung levels of PFBA.

- According to Pérez et al. 2013, lung was the tissue showing the highest accumulation of PFASs. Of the PFASs included in the present paper, Pérez et al reported mean concentrations of PFBA and PFOS of 807 and 28.4 ng/g lung tissue. In addition, even though the percentage of samples with detected values of PFOA fell down to 45%, the contribution of PFOA to the total PFASs in lung was quite important, in comparison other tissues and analytes. Since PFBA was not the only PFAS to be retained in the lungs, the authors could include this information in their discussion.

D2. “The associations of PFASs with a more serious course of COVID-19 are weakened after adjustment for covariates, and some regression coefficients and ORs are below 1. However, adjustment for all covariates may result in over-adjustment bias. Thus, older age and male sex are known to be strong predictors of higher blood-PFAS concentrations, and simple adjustment for these factors could potentially result in a bias toward the null. As PFAS exposure has been linked to important comorbidities, such as diabetes and obesity, both of which may exacerbate the virus infection, adjustment for chronic disease may also not be justified. Leaving it out slightly strengthened the PFBA association with the disease severity.” The authors should consider using a more structured approach of selecting covariates to be included in the adjusted analyses and not just include/exclude covariates since such an approach may lead to both over- and under-adjustments. Using Directed Acyclic Graphs (DAGs) to select covariates to include in the adjusted analyses, may decrease the chance of adjustment bias. Entering the covariates in the present study into Dagitty.net, this simple DAG returns a minimal sufficient adjustment set for estimating the total effect of PFASs on disease severity that includes age and sex only. The authors should also include possible other important covariates they may have information on. The Dagitty.net code for the DAG is:

dag {

"Disease severity Covid-19" [outcome,pos="1.400,1.621"]

Age [pos="-2.025,-1.064"]

Diabetes [pos="0.150,-0.740"]

Obesity [pos="-0.917,-0.547"]

PFASs [exposure,pos="-2.200,1.597"]

Sex [pos="-1.758,0.282"]

Age -> "Disease severity Covid-19"

Age -> Diabetes

Age -> Obesity

Age -> PFASs

Diabetes -> "Disease severity Covid-19"

Obesity -> "Disease severity Covid-19"

Obesity -> Diabetes

PFASs -> "Disease severity Covid-19"

PFASs -> Diabetes

PFASs -> Obesity

Sex -> "Disease severity Covid-19"

Sex -> PFASs

}

D3. “In many cases, their plasma had been stored on previous occasions, and the PFAS concentrations may reflect slightly higher exposures in the recent past. Although adjustment for the time interval since sample collection was included in the analyses, its impact on the results was negligible.” Since PFBA has a half-life of only 72 hours, this may affect your findings in that in the “No hospitalization group”, the number of days between sampling and diagnosis was 355 (22.5, 639.5) whereas it was 0 for the two other categories of severity. PFHxS, PFOS, PFOA and PFNA have half-life ranging from 2.5 to 8.5 years and may be representative of the PFAS levels at time of disease, whereas PFBA levels resemble more a “snap-shot”. Thus, the PFBA levels up to 28 months before disease onset may not be a reliable measure compared to the other PFASs included in the manuscript. Excluding samples that were taken 60 days before diagnosis may not be sufficient to adjust for the short half-life of 72 hours for PFBA. The authors should expand their discussion accordingly.

D4. “Among immigrants, adverse associations appeared slightly stronger, also after adjustments, thus suggesting that national origin, perhaps as related to demographic or social factors, may result in a greater vulnerability to PFAS-associated aggravation of the infection. Difference in age, sex, or comorbidities did not explain this tendency, but is in agreement with previous findings of ethnic differences in vulnerability.” These findings are not included in the Results section. If possible, this data should be included in the manuscript or as supplementary data.

D5. In adjusted analysis, there is a statistically significant inverse relationship between PFHxS and disease severity (0.52 (0.29, 0.91)), as well as for persons of Western European origin (0.46 (0.23, 0.89)). This is not mentioned specifically either in Results or Discussion. The author should include text on these findings in both these sections.

Conclusion

C1. “Thus, given the immunotoxicity of the PFASs, exposure to these persistent industrial chemicals may contribute to the severity of COVID-19.” This statement should be moderated in that PFBA is statistically significant related to disease severity in the crude analysis only. Moreover, PFHxS is inversely related to disease severity. If the authors decide to keep this statement, a clarification should be added.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2020 Dec 31;15(12):e0244815. doi: 10.1371/journal.pone.0244815.r002

Author response to Decision Letter 0

12 Dec 2020

Responses to Reviewer Comments

Response: Thanks for the detailed and helpful comments.

Major points:

1. There are no line numbers or page numbers, this makes review/providing feedback challenging

Response: We have added the requested format change.

Response: We have corrected ordinal to ordered logistic regression throughout the manuscript.

Response: We appreciate the advice, but we have not studied in detail the clinical course over time. We believe that severity is a better choice for the title.

Response: We believe that the air pollution studies support the notion that community exposures to environmental contaminants can generally worsen the outcome of the COVID-19. Our own study adds to this association by documenting individual exposure levels, which could not be done in the air pollution studies. We have adjusted the wording so that it is clear that our work does not focus on air pollution.

Response: As indicated in Table 1, especially the non-hospitalized subjects had blood samples drawn much before the infection. We have now stratified the data and focused an analysis on samples collected close to the time of infection. The exclusion mainly of older blood samples results in stronger findings and also moves the OR for PFHxS closer to 1. Additional PFASs are shown in the Supplementary information.

Response: We appreciate the advice, but signals below the LOD are too uncertain, and there are therefore no machine-read values that can be used. This applies to PFBS as well in the Supplementary information.

Response: The comorbidities are now described in detail in the main text. Along with place of inclusion and COVID-19 severity, they are categorical variables, and thus cannot be shown as differences.

Response: We have included a brief description of ordered logistic regression in the methods section.

Response: We agree with the reviewer that a linear association is most likely, which is why we did not per default include loglinear PFAS or PFAS squared. However, as is standard when performing linear regression models, we tested our assumptions. With 323 individuals we believe that we would be able to detect a fairly strong non-linear associations using a significance level of 0.05. For dichotomized PFASs our interpretation was not based on a linear assumption, and there was thus no need to test the assumption.

Response: We have now emphasized in the text that the potential confounders were identified from the information and references already provided.

Response: Given the conditions of the study of anonymized data, we did not have access to patient charts or individual data on kidney function markers. We have now treated kidney disease as a covariate separate from other chronic disease and commented on its association with plasma-PFAS concentrations. .

12. Method, Statistical Analysis, How were diagnoses confirmed? Is there a difference in the diagnosis among those hospitalized vs. those not hospitalized?

Response: All diagnoses were obtained from the national register, which has a high validity, and there is no reason to suspect any difference in validity between groups of subjects included in the study.

Response: We appreciate the comment, as it relates to an increased risk of infection associated with socioeconomic status. However, it is not clear from the literature whether it predicts a more serious disease as well. Due to the limitations of the register data available to us, we focused on national origin as the most important socioeconomic predictor of corona infection. In agreement with previous studies, we find lower PFAS exposures in subjects of other national origin, but also that this factor did not affect the disease severity.

14. Discussion, “Among…the PFASs, presence of detectable…” PFHxS showed the strongest association according to Table 4.

Response: We have now clarified that we refer to positive associations, as reflected in the a priori hypothesis. As we are examining the possible association with immunotoxic exposures, the PFHxS association is not plausible. In the restricted analysis, we find PFHxS not to be a significant factor in regard to disease severity. The same applies to other PFASs in the Supplementary information.

Response: We agree that PFBA is in accordance with the hypothesis, but PFHxS is not (and is not accumulated in the lungs). The text has been adjusted for clarity, while taking into regard the further analyses that show no significant effect of PFHxS, nor other PFASs.

Response: We refer to our response above, that the samples obtained early after diagnosis could not have been affected by pathologies occurring in late disease development.

17. Discussion, “Low background exposure results…” How did the PFAS levels compare to other reports of associations with immune function?

Response: We hesitate to provide detailed information on exposures in other studies, as the data are already available to the reader. However, we have added a sentence to support the notion that the population is not highly exposed, while referring to national data from the U.S.

Response: We refer to a previous study on ethnicity as a risk factor for the infection. Our own opportunities to explore these factors are limited due to the size of the material and to the circumstances of studying sensitive information anonymously. We note that ethnicity status was not an important predictor of the PFAS associations with disease severity. Further, as noted above, we did not have access to other socioeconomic data

19. Conclusions, The authors should also report on the “protective” effect of PFHxS, a long-lived PFAS.

Response: We already report this association, which was substantially weakened in the further analyses, and we note that a protective effect is counter to expectation.

Minor points

1. Abstract: “Methods”: “…at the background exposures…” Unclear.

Response: The term is commonly used to indicate non-occupational exposure not caused by local sources. We have commented on this in the Discussion and would prefer not to add additional wording here.

2. Abstract, “Results”: increasing severities of disease; again, does the outcome variable truly capture “disease severity.”?

Response: We can’t think of a better word, as clinical course would seem misleading.

Response: We shall comply with the journal policy in this regard. We note that the COVID-19 pandemic has increased the need for access to relevant reports in advance of peer review that may substantially delay public access to the findings. Incidentally, two of the in-press references cited now have more definite journal information, and two of the initial three medRxiv articles are now available as journal articles, as indicated in the reference list, so there is now only one left.

Response: The study referred to is the only one published so far and encompassed samples from 20 autopsies. PFBA had by far the highest concentrations in lung tissue, higher than any of the other 20 PFASs analyzed, and with a median much higher than other PFASs. We highlight that this is so far a single study.

Response: We hesitate to expand on this information, as we are already referring to the NTP review. Also, the specific mode of action of PFAS immunotoxicity is unknown.

5. Methods, is exogenous contamination of specimens, for example Teflon caps, a concern for blood specimens?

Response: All sample processing was done by robot, and Teflon caps are not used.

6. Methods, “…no hospital admission and completed infection within 14 days of testing positive…” How was this determined?

Response: This information was retrieved from the patient register as described.

Response: We believe that we already provide this information in regard to sex dependence, but we are restricted about tabular data on obesity.

8. Methods, Chemical Analysis, “Succeeding series…” Unclear.

Response: We ran the analyses in sequence to avoid any risk of drift.

9. Methods, Chemical Analysis, “…for PFAS concentrations, including…” Why these five PFAS specifically?

Response: These are the five that have been found to associate with immunotoxicity in humans, as we now clarify.

10. Methods, Chemical Analysis, “Because more than half the PFBA and PFBS…” PFBS was not mentioned in the “Chemical Analysis” section above or reported in the results as far as I can tell.

Response: PFBS is left out in the main text, as it is not known to be immunotoxic in humans, see instead Supplementary information.

11. Methods, Statistical Analysis, “…presence of comorbidities…” What comorbidities did participants have, specifically?

Response: Given the conditions of the study, we were unable to extract information on specific comorbidities of individuals, and we therefore grouped the conditions, as described.

12. Methods, Statistical Analysis, “…demographic groups…” What demographic groups?

Response: Thank you for pointing out this imprecision. It has now been clarified in the text.

13. Methods, Statistical Analysis, “The test could not be fitted…” Why not?

Response: This was most likely due to too many strata, but the omodel package in Stata does not provide detailed information why a fit was not achieved.

14. Methods, Statistical Analysis, “Among those of Western European national origin…” This seems better suited to the Results section.

Response: We appreciate the comment, but we believe that the descriptive information on the study population fits better in the Methods section.

15. Statistical Analysis, “…risk of certain chronic disease…” Which chronic diseases, specifically?

Response: The text has now been expanded with the list of diseases.

16. Results, “Correlated well..” Unclear.

Response: The explanation is given in the table that the sentence refers to.

17. Discussion, The first paragraph should summarize the major findings/and implications of the research. This interpretation is better suited to later components.

Response: We prefer to highlight the study purpose before summarizing the results. It is not clear why the reviewer prefers a different sequence of presentation.

18. Discussion, “… the only PFAS that is substantially accumulated in the …” This has been shown to substantially accumulate in one study, please clarify this point (see comment #3 above)

Response: We believe that the present wording is appropriate and that the reference provided reflects the evidence. Please see our response to #3 above.

19. Discussion, “…major pathways that are predictive…” What pathways, specifically?

Response: We hesitate to include a discussion of the pathways, as this matter is not central to our study, and relevant detail is provided in the reference cited. We now mention IL-17 as one of the pathways.

Response: This point is exactly what we want to make, i.e., that PFASs (except for PFBA) are generally higher in men and the elderly, and that fits the tendency observed for COVID-19 severity.

Response: We already tried out a DAG but found that it in this case is not providing clarity due to several factors of possible importance, and we therefore decided to leave it out.

Reviewer #2: Title: Severity of COVID-19 at elevated exposure to perfluorinated alkylates

Abstract

Response: We have edited the text to highlight industrial chemicals.

Response: Yes, as we have now clarified.

Materials and methods

M1. Statistical analysis, 2nd section: PFBS – write the full name. Alternatively delete since no findings on PFBS are shown in the manuscript.

Response: PFBS is included in the supplementary tables, as it is not known to be a human immunotoxicant, and it does not accumulate in the lungs like PFBA.

Results

Response: We believe that the revised presentation satisfies this comment.

Response: We agree that the two-sided p-value is marginally above 0.05 and have edited the text, also in accordance with the new and restricted analyses.

Response: We agree and have conducted further stratifications and now present further restricted analyses.

Response: Chronic disease was not a significant predictor (Table 2) and did not change the estimate, but the restricted analysis seems more appropriate.

Response: We have edited the text accordingly and now provide Supplementary Information.

R6. To make it easier for the readers to interpret all the findings reported in the Results section, the author should consider showing results of all additional analyses as supplementary material.

Response: We now include Supporting information.

Discussion

- Are there any studies showing the correlation between blood and lung levels of PFBA? This would be interesting to know since you are looking at plasma levels and not lung levels of PFBA.

Response: The autopsy study did not include blood samples, unfortunately, so we cannot comment.

Response: We appreciate the detail, but we are concerned about providing much additional information from a published study. PFBA by far reached the highest concentrations in lung tissue, and other PFASs were lower by more than an order of magnitude.

dag {"Disease severity Covid-19" [outcome,pos="1.400,1.621"]

Age [pos="-2.025,-1.064"]

Diabetes [pos="0.150,-0.740"]

Obesity [pos="-0.917,-0.547"]

PFASs [exposure,pos="-2.200,1.597"]

Sex [pos="-1.758,0.282"]

Age -> "Disease severity Covid-19"

Age -> Diabetes

Age -> Obesity

Age -> PFASs

Diabetes -> "Disease severity Covid-19"

Obesity -> "Disease severity Covid-19"

Obesity -> Diabetes

PFASs -> "Disease severity Covid-19"

PFASs -> Diabetes

PFASs -> Obesity

Sex -> "Disease severity Covid-19"

Sex -> PFASs}

Response: We are grateful for the advice and we did indeed use a DAG as part of the planning but decided against inserting it into the manuscript.

Response: We agree on this point, and we have therefore included further restricted analyses that take into account the short elimination half-life of PFBA in blood.

Response: Thank you, we now include Supporting Information to justify this conclusion.

Response: As indicated above, we now emphasize that this association is counter to the a priori hypothesis, and it disappears in the restricted analysis, possibly because of higher concentrations in “old” samples.

Conclusion

Response: As responded above, we have modified the text.

Attachment

Submitted filename: PLoSoneReviewCommentsDec7.docx

Click here for additional data file.^{(37.8KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0244815.r003

Decision Letter 1

Jaymie Meliker

17 Dec 2020

Severity of COVID-19 at elevated exposure to perfluorinated alkylates

PONE-D-20-32774R1

Dear Dr. Grandjean,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jaymie Meliker, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

PLoS One. doi: 10.1371/journal.pone.0244815.r004

Acceptance letter

Jaymie Meliker

21 Dec 2020

PONE-D-20-32774R1

Severity of COVID-19 at elevated exposure to perfluorinated alkylates

Dear Dr. Grandjean:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Jaymie Meliker

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Spearman’s correlation coefficients for pairwise comparisons of detectable PFASs in plasma from 323 subjects included in the study.

(DOCX)

Click here for additional data file.^{(24.6KB, docx)}

S2 Table. Median plasma concentrations of additional PFASs (25^th-75^th percentiles) in ng/mL by population characteristics.

(DOCX)

Click here for additional data file.^{(23.1KB, docx)}

S3 Table. Ordered logistic regression odds ratios (ORs) of increased COVID-19 severity at an increase by 1 ng/mL in plasma concentrations of additional PFASs.

(DOCX)

Click here for additional data file.^{(20.1KB, docx)}

S4 Table. Logistic regression odds ratios (ORs) of increased COVID-19 severity at an increase by 1 ng/mL in plasma concentrations of all PFASs detected.

(DOCX)

Click here for additional data file.^{(27KB, docx)}

Attachment

Submitted filename: PLoSoneReviewCommentsDec7.docx

Click here for additional data file.^{(37.8KB, docx)}

Data Availability Statement

[pone.0244815.ref001] 1.Fattorini D, Regoli F. Role of the chronic air pollution levels in the Covid-19 outbreak risk in Italy. Environ Pollut 2020;264:114732 10.1016/j.envpol.2020.114732 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref002] 2.Zhu Y, Xie J, Huang F, Cao L. Association between short-term exposure to air pollution and COVID-19 infection: Evidence from China. Sci Total Environ 2020;727:138704 10.1016/j.scitotenv.2020.138704 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref003] 3.Bashir MF, Ma BJ, Bilal, et al. Correlation between environmental pollution indicators and COVID-19 pandemic: A brief study in Californian context. Environ Res 2020;187:109652 10.1016/j.envres.2020.109652 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref004] 4.Wu X, Nethery RC, Sabath MB, Braun D, Dominici F. Air pollution and COVID-19 mortality in the United States: Strengths and limitations of an ecological regression analysis. Sci Adv 2020;6 10.1126/sciadv.abd4049 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref005] 5.Rooney AA, Luebke RW, Selgrade MK, Germolec DR. Immunotoxicology and its application in risk assessment. Exp Suppl 2012;101:251–87. 10.1007/978-3-7643-8340-4_9 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref006] 6.World Health Organization W. Guidance for Immunotoxicity Risk Assessment for Chemicals: IPCS Harmonization Project. Geneva, Switzerland: 2012. [Google Scholar]

[pone.0244815.ref007] 7.DeWitt JC, Germolec DR, Luebke RW, Johnson VJ. Associating Changes in the Immune System with Clinical Diseases for Interpretation in Risk Assessment. Curr Protoc Toxicol 2016;67:18.1–1-.1.22. 10.1002/0471140856.tx1801s67 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref008] 8.European Food Safety Authority. Risk to human health related to the presence of perfluoroalkyl substances in food Panel on Contaminants in the Food Chain. Parma: EFSA: 2020:e06223. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref009] 9.Grandjean P, Andersen EW, Budtz-Jorgensen E, et al. Serum vaccine antibody concentrations in children exposed to perfluorinated compounds. JAMA 2012;307:391–7. 10.1001/jama.2011.2034 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref010] 10.Kielsen K, Shamim Z, Ryder LP, et al. Antibody response to booster vaccination with tetanus and diphtheria in adults exposed to perfluorinated alkylates. J Immunotoxicol 2016;13:270–3. 10.3109/1547691X.2015.1067259 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref011] 11.Dalsager L, Christensen N, Husby S, et al. Association between prenatal exposure to perfluorinated compounds and symptoms of infections at age 1–4 years among 359 children in the Odense Child Cohort. Environ Int 2016;96:58–64. 10.1016/j.envint.2016.08.026 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref012] 12.Granum B, Haug LS, Namork E, et al. Pre-natal exposure to perfluoroalkyl substances may be associated with altered vaccine antibody levels and immune-related health outcomes in early childhood. J Immunotoxicol 2013;10:373–9. 10.3109/1547691X.2012.755580 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref013] 13.Goudarzi H, Miyashita C, Okada E, et al. Prenatal exposure to perfluoroalkyl acids and prevalence of infectious diseases up to 4years of age. Environ Int 2017;104:132–8. 10.1016/j.envint.2017.01.024 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref014] 14.Wu Q, Coumoul X, Grandjean P, Barouki R, Audouze K. Endocrine disrupting chemicals and COVID-19 relationships: a computational systems biology approach. Environ Int 2020;(in press):106232 10.1016/j.envint.2020.106232 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref015] 15.Grasselli G, Zangrillo A, Zanella A, et al. Baseline Characteristics and Outcomes of 1591 Patients Infected With SARS-CoV-2 Admitted to ICUs of the Lombardy Region, Italy. JAMA 2020;323:1574–81. 10.1001/jama.2020.5394 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref016] 16.Liang W, Liang H, Ou L, et al. Development and Validation of a Clinical Risk Score to Predict the Occurrence of Critical Illness in Hospitalized Patients With COVID-19. JAMA Intern Med 2020;180:1081–9. 10.1001/jamainternmed.2020.2033 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref017] 17.Atyeo C, Fischinger S, Zohar T, et al. Distinct Early Serological Signatures Track with SARS-CoV-2 Survival. Immunity 2020;53:524–32 e4. 10.1016/j.immuni.2020.07.020 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref018] 18.Goncalves Mendes Neto A, Lo KB, Wattoo A, et al. Bacterial infections and patterns of antibiotic use in patients with COVID-19. J Med Virol 2020;(in press). 10.1002/jmv.26441 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref019] 19.Petrilli CM, Jones SA, Yang J, et al. Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease 2019 in New York City: prospective cohort study. BMJ 2020;369:m1966 10.1136/bmj.m1966 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref020] 20.Zhou Y, He Y, Yang H, et al. Development and validation a nomogram for predicting the risk of severe COVID-19: A multi-center study in Sichuan, China. PLoS One 2020;15:e0233328 10.1371/journal.pone.0233328 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref021] 21.Zhao X, Zhang B, Li P, et al. Incidence, clinical characteristics and prognostic factor of patients with COVID-19: a systematic review and meta-analysis. medRxiv 2020:2020.03.17.20037572. [Google Scholar]

[pone.0244815.ref022] 22.Cummings MJ, Baldwin MR, Abrams D, et al. Epidemiology, clinical course, and outcomes of critically ill adults with COVID-19 in New York City: a prospective cohort study. Lancet 2020;395:1763–70. 10.1016/S0140-6736(20)31189-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref023] 23.Tartof SY, Qian L, Hong V, et al. Obesity and Mortality Among Patients Diagnosed With COVID-19: Results From an Integrated Health Care Organization. Ann Intern Med 2020;173:773–81. 10.7326/M20-3742 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref024] 24.Yu CH, Riker CD, Lu SE, Fan ZT. Biomonitoring of emerging contaminants, perfluoroalkyl and polyfluoroalkyl substances (PFAS), in New Jersey adults in 2016–2018. Int J Hyg Environ Health 2020;223:34–44. 10.1016/j.ijheh.2019.10.008 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref025] 25.Sun Q, Zong G, Valvi D, Nielsen F, Coull B, Grandjean P. Plasma Concentrations of Perfluoroalkyl Substances and Risk of Type 2 Diabetes: A Prospective Investigation among U. S. Women. Environ Health Perspect 2018;126:037001 10.1289/EHP2619 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref026] 26.Liu G, Dhana K, Furtado JD, et al. Perfluoroalkyl substances and changes in body weight and resting metabolic rate in response to weight-loss diets: A prospective study. PLoS Med 2018;15:e1002502 10.1371/journal.pmed.1002502 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref027] 27.ATSDR. Draft toxicological profile for perfluoroalkyls. Atlanta, GA: Agency for Toxic Substances and Disease Registry; 2018. [PubMed] [Google Scholar]

[pone.0244815.ref028] 28.Perez F, Nadal M, Navarro-Ortega A, et al. Accumulation of perfluoroalkyl substances in human tissues. Environ Int 2013;59:354–62. 10.1016/j.envint.2013.06.004 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref029] 29.Reilev M, Kristensen KB, Pottegard A, et al. Characteristics and predictors of hospitalization and death in the first 11 122 cases with a positive RT-PCR test for SARS-CoV-2 in Denmark: a nationwide cohort. Int J Epidemiol 2020. 10.1093/ije/dyaa140 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref030] 30.Voldstedlund M, Haarh M, Molbak K, MiBa Board of R. The Danish Microbiology Database (MiBa) 2010 to 2013. Euro Surveill 2014;19 10.2807/1560-7917.es2014.19.1.20667 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref031] 31.Schmidt M, Pedersen L, Sorensen HT. The Danish Civil Registration System as a tool in epidemiology. Eur J Epidemiol 2014;29:541–9. 10.1007/s10654-014-9930-3 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref032] 32.Schmidt M, Schmidt SA, Sandegaard JL, Ehrenstein V, Pedersen L, Sorensen HT. The Danish National Patient Registry: a review of content, data quality, and research potential. Clin Epidemiol 2015;7:449–90. 10.2147/CLEP.S91125 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref033] 33.Andersen JS, de Fine Olivarius N, Krasnik A. The Danish National Health Service Register. Scand J Public Health 2011;39:34–7. 10.1177/1403494810394718 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref034] 34.Jain RB, Ducatman A. Perfluoroalkyl substances follow inverted U-shaped distributions across various stages of glomerular function: Implications for future research. Environ Res 2019;169:476–82. 10.1016/j.envres.2018.11.033 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref035] 35.DeWitt JC, Blossom SJ, Schaider LA. Exposure to perfluoroalkyl and polyfluoroalkyl substances leads to immunotoxicity: epidemiological and toxicological evidence. J Expo Sci Environ Epidemiol 2019;29:148–56. 10.1038/s41370-018-0097-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref036] 36.Zeng XW, Li QQ, Chu C, et al. Alternatives of perfluoroalkyl acids and hepatitis B virus surface antibody in adults: Isomers of C8 Health Project in China. Environ Pollut 2020;259:113857 10.1016/j.envpol.2019.113857 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref037] 37.Haug LS, Thomsen C, Becher G. A sensitive method for determination of a broad range of perfluorinated compounds in serum suitable for large-scale human biomonitoring. J Chromatogr A 2009;1216:385–93. 10.1016/j.chroma.2008.10.113 [DOI] [PubMed] [Google Scholar]

[pone.0244815.ref038] 38.CDC. National Report on Human Exposure to Environmental Chemicals, Updated Tables, January 2019. Atlanta, GA: Centers for Disease Control and Prevention; 2019. [Google Scholar]

[pone.0244815.ref039] 39.Hu XC, Andrews DQ, Lindstrom AB, et al. Detection of Poly- and Perfluoroalkyl Substances (PFASs) in U.S. Drinking Water Linked to Industrial Sites, Military Fire Training Areas, and Wastewater Treatment Plants. Environ Sci Technol Lett 2016;3:344–50. 10.1021/acs.estlett.6b00260 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref040] 40.Erikstrup C, Hother CE, Pedersen OBV, et al. Estimation of SARS-CoV-2 infection fatality rate by real-time antibody screening of blood donors. Clin Infect Dis 2020:in press. 10.1093/cid/ciaa849 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref041] 41.Williamson EJ, Walker AJ, Bhaskaran K, et al. Factors associated with COVID-19-related death using OpenSAFELY. Nature 2020;584:430–6. 10.1038/s41586-020-2521-4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref042] 42.Lassale C, Gaye B, Hamer M, Gale CR, Batty GD. Ethnic disparities in hospitalisation for COVID-19 in England: The role of socioeconomic factors, mental health, and inflammatory and pro-inflammatory factors in a community-based cohort study. Brain Behav Immun 2020;88:44–9. 10.1016/j.bbi.2020.05.074 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0244815.ref043] 43.Montazeri P, Thomsen C, Casas M, et al. Socioeconomic position and exposure to multiple environmental chemical contaminants in six European mother-child cohorts. Int J Hyg Environ Health 2019;222:864–72. 10.1016/j.ijheh.2019.04.002 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Severity of COVID-19 at elevated exposure to perfluorinated alkylates

Philippe Grandjean

Clara Amalie Gade Timmermann

Marie Kruse

Flemming Nielsen

Pernille Just Vinholt

Lasse Boding

Carsten Heilmann

Kåre Mølbak

Roles

Abstract

Background

Methods

Results

Conclusions

Introduction

Methods

Population

Chemical analysis

Statistical analysis

Results

Table 1. Spearman’s correlation coefficients for pairwise comparisons of detectable PFASs in plasma from 323 subjects included in the study.

Table 2. Median plasma-PFAS concentrations (25th, 75th percentiles) in ng/mL by population characteristics.

Table 3. COVID-19 severity by population characteristics.

Table 4. Ordered logistic regression OR of increased Covid-19 severity for an increase by 1 ng/mL in plasma-PFAS concentrations.

Fig 1. Proportion of plasma samples with detectable PFBA concentrations at different disease severities.

Discussion

Conclusions

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Jaymie Meliker

Roles

Author response to Decision Letter 0

Decision Letter 1

Jaymie Meliker

Roles

Acceptance letter

Jaymie Meliker

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

Table 2. Median plasma-PFAS concentrations (25^th, 75^th percentiles) in ng/mL by population characteristics.