Toxic Occupational Exposures and Membranous Nephropathy

Visual Abstract

Abstract

Background and objectives

Membranous nephropathy is a rare autoimmune kidney disease whose increasing prevalence in industrialized countries pleads for the involvement of an environmental factor in the development of the disease. In addition, the predominance of men in membranous nephropathy, classically attributed to biologic or genetic differences between men and women, could also be due to different occupational exposures. To support this hypothesis, we sought to describe the toxic occupational exposures of patients with membranous nephropathy.

Design, setting, participants, & measurements

In this observational epidemiologic study, we compared the occupations and toxic occupational exposures of 100 patients with membranous nephropathy with those of the general population, consisting of two cohorts of 26,734,000 and 26,500 French workers. We then compared the characteristics of patients exposed to an occupational toxic substance with those of unexposed patients.

Results

Patients with membranous nephropathy worked more frequently in the construction sector than the general population (33% versus 7%, P<0.001). This difference remained significant by age and sex. They were also more frequently exposed to toxic substances, such as asbestos (16% versus 5%, P<0.001), lead (9% versus 1%, P<0.001), or organic solvents (37% versus 15%, P<0.001), than the general population. The predominance of men in the subgroup of patients occupationally exposed to toxic substances was not observed in unexposed individuals (organic solvents: 80% men versus 41%, P<0.001; asbestos: 90% men versus 55%, P=0.004). In addition, patients with phospholipase A2 receptor 1 (PLA2R1) epitope spreading were more frequently exposed to asbestos and organic solvents than patients without epitope spreading (32% versus 7%, P=0.02 and 74% versus 43%, P=0.02, respectively), with a dose-dependent effect.

Conclusions

Patients with membranous nephropathy were more frequently exposed to certain occupational toxic substances, such as asbestos and organic solvents, than the general population. This occupational exposure was more frequent in men and in patients with PLA2R1 epitope spreading.

Clinical Trial registry name and registration number:

Immunopathological Analysis in a French National Cohort of Membranous Nephropathy (IHMN), NCT04326218.

Podcast

This article contains a podcast at https://www.asn-online.org/media/podcast/CJASN/2022_10_25_CJN02930322.mp3.

Introduction

Membranous nephropathy is one of the most common causes of nephrotic syndrome in nondiabetic White adults (1). This rare autoimmune kidney disease is characterized by subepithelial immune deposits containing IgG and complement fractions with alteration of the glomerular basement membrane structure (2,3). Patients with membranous nephropathy develop autoantibodies against podocyte proteins (4). The most common target antigen is M-type phospholipase A2 receptor 1 (PLA2R1), which is found in 70% of patients with membranous nephropathy (5–7). Of increasing prevalence (8,9), it preferentially affects men between 50 and 60 years old (3). In recent decades, Western countries have been confronted with an increasing incidence of autoimmune diseases. One of the main hypotheses is that this is the result of a complex interaction between genetic predisposition, immune dysregulation, and environmental factors (10). Numerous genome-wide association studies have identified that some single nucleotide polymorphisms within the PLA2R1, HLA-DQA1, and HLA-DRB1 genes are significantly associated with primary membranous nephropathy (11–17). More recently, another genome-wide association study has identified two new risk loci for primary membranous nephropathy encoding transcriptional master regulators of inflammation, NFKB1 and IRF4, suggesting that inflammation may play a part in membranous nephropathy physiopathology (18). Little is known about immune dysregulation in membranous nephropathy, but several studies suggested a disturbed balance between T helper 17 (Th17) and T regulatory cells (19–24). Environmental factors could play a triggering role in immune dysregulation in genetically predisposed individuals, leading to the development of autoimmune diseases (25). Exposure to certain environmental toxic substances, such as mercury (26,27), formaldehyde (28), and air pollution (22,29), has been identified as a potential cause of membranous nephropathy. Occupational exposure to organic solvents is a risk factor for autoimmune diseases (30–32). Although described nearly 50 years ago, exposure to organic solvents remains a controversial cause of GN and/or progression of CKD (33–37). At present, no study has attempted to establish a link between occupational exposure to toxic substances and membranous nephropathy. However, the increasing prevalence and predominance of men in membranous nephropathy could suggest an occupational environment trigger for this autoimmune disease (38). In this study, we sought to describe the occupational exposure of patients with membranous nephropathy to toxic substances in comparison with the general French population to identify potential environmental factors in the onset of the disease.

Materials and Methods

Study Design

First, we conducted an observational cross-sectional epidemiologic survey of occupational exposures of patients with membranous nephropathy followed at Nice University Hospital. We included patients with histologically proven (kidney biopsy–proven) membranous nephropathy. Noninclusion criteria were (1) refusal to participate, (2) language barrier or any other reason that did not allow a standardized collection of data by the questionnaire, and (3) minors or adults under protective supervision for whom informed consent could not be obtained. The data from this epidemiologic survey were then compared with the general French population.

In a second phase, we carried out a comparative observational study of the exposed/unexposed type. The clinical and biologic characteristics of patients with membranous nephropathy exposed to certain occupational toxic substances were compared with unexposed patients with membranous nephropathy.

Data Collection

Exposures.

An occupational health physician collected the epidemiologic data of patients with membranous nephropathy using a heteroquestionnaire (Supplemental Material) (available in French and English) completed during semiguided individual interviews. This data collection was conducted from February to June 2021. The heteroquestionnaire collected details of each patient’s current and past occupational activities: occupation at diagnosis and for how many years; history of occupations before diagnosis and their duration; possible occupational toxic exposures; and nephrotoxic prevention measures implemented by their employer’s occupational health services. Each patient received the same information during the interview, and medical and/or work-related terms were explained. The occupational physician collected exposure data by classifying the level of exposure into three categories: low, moderate, or high exposure. This classification was on the basis of a quantitative estimate following the patient’s detailed description of his or her professional activity. It summarizes a combination of (1) the frequency of exposure during the professional activity, (2) the cumulative length of exposure (in years), (3) the estimated concentration of the toxic substance(s) during the exposure situations, and (4) the adherence or not with the wearing of personal protective equipment (respiratory, skin, and digestive protection). The patients’ jobs and details of their occupational exposure are detailed in Table 1. For example, high exposure to organic solvents could be defined for a mechanic using motor oil >2 h/d indoors without personal protective equipment and for >10 years. The same exposure with personal protective measures in a ventilated area could be considered moderate. In contrast, if the same exposure occurs only occasionally (e.g., once a month), the exposure will be considered low. A second questionnaire focused on nonoccupational toxic exposures to search for possible overexposure outside the workplace (i.e., at home or during leisure activities).

Table 1.

Details of patients' jobs and occupational exposures

Business Sectors	Patient Jobs	Type of Product Containing Solvents	Type of Product Containing Lead	Type of Product Containing Asbestos
Agricultural workers	Farmers, gardeners			Renovation of an asbestos farm
Service workers	Office workers, teachers, postmen, cleaners, hairdressers, hospital workers	Bleach, cleaners, disinfectants, household sprays, dyes, ammonia perms, thioglycolic acid, printing ink
Industry workers	Seamstresses and tailors, technicians in the chemical industry, appliance technicians, gas station attendants	Pharmaceutical products, printing ink, methyl ethyl ketone, glue, resins, degreasers, cleaners, bleach, kerosene, alcohol vapor, nitrogenous solvents, solvent vapor, hydrochloric acid, silver photo developers, fuel, gasoline	Lead nitrate
Construction workers	Electricians, painters, car mechanics, masons, plumbers, carpenters, locksmiths, welders	Form oil, motor oil, oil-based paint, paint thinners, glue, bleach, white spirit, tile cleaning acids, paint stripper, hydrochloric acid, oxidizers, degreasers, acetone, ammonia, corrosive cleaning acids, fuel, petroleum products, brake fluid, varnish, benzine cans, cement, asphalt, tar, glue, shoe polish, paint strippers, dyes	Lead paint, pipes and cables	Renovation of old roofs, mortars on the basis of asbestos plaster, asbestos insulation materials (old buildings), asbestos brake pads and clutches, welding/brazing of asbestos-bearing materials

The occupations of each patient and the details of their occupational exposure were collected by the occupational medicine department of the Nice University Hospital.

We compared the epidemiologic data of patients with membranous nephropathy with the general population in two national surveys. We analyzed the main sectors of occupational activity and personal toxic exposures. We used two cohorts of French workers as controls. The French National Institute of Statistics and Economic Studies (INSEE) 2018 provides the occupational activity of the French general population by sex and age groups. After excluding unemployed individuals, this information is available for 26,734,000 participants. The 2017 French survey on Medical Surveillance of Employees’ Exposure to Occupational Risks (Surveillance Médicale des Expositions des salariés aux Risques professionnels [SUMER]) provides data on occupational toxic exposure. The statistical department of the French Ministry of Labor conducts this national survey every 7 years among the national working population of employees. In 2017, it included 26,500 participants. To note, in developing the questionnaire for patients with membranous nephropathy, we used similar evaluation criteria to those of the 2017 SUMER survey (i.e., exposition duration, personal and collective protective equipment, and intensity [measured or estimated] of exposure). To be consistent with the 2017 SUMER survey data, we decided to analyze both lifetime exposure of patients with membranous nephropathy data and also over the year 2017.

Membranous Nephropathy.

We collected data concerning the medical history of membranous nephropathy using patients’ records and, whenever necessary, direct patient questioning. The data collected included demographic characteristics, medical history and comorbidities, and clinical presentation of membranous nephropathy, as well as biologic parameters at diagnosis and during follow-up.

Among the 100 patients included, 45 were enrolled during an active phase of the disease (i.e., urine protein-creatinine ratio >3.5 g/g). In these participants, also included in a local cohort (Immunopathological Analysis in a French National Cohort of Membranous Nephropathy [IHMN]), 1 ml of whole blood was collected and stimulated with immune ligands (anti-CD3 as T cells stimulant and R848 as TLR 7/8 agonist) on single lyophilized spheres (LyoSphere; Qiagen) within 8 hours of blood collection. Stimulated blood samples were incubated for 16–24 hours at 37°C and then centrifuged at 2000–3000×g for 15 minutes to harvest the stimulated plasma. Plasma levels of IL-17A were measured using the custom-designed cartridges Ella (ProteinSimple) following the manufacturer’s instructions.

Outcomes

First, we aim to describe the main occupational toxic substances to which patients with membranous nephropathy were exposed and then compare them with those of the general population. Second, using subgroup analyses of patients with membranous nephropathy, we compared the characteristics of patients exposed to an occupational toxic substance with those who were not, including disease progression (epitope spreading and kidney function).

Statistical Analyses

Qualitative variables are presented as numbers and percentages and compared using a Pearson chi-squared test or a Fisher exact test in the case of a small sample. Quantitative variables with a Gaussian distribution (age) are presented as means and SDs and compared by the unpaired t test. Quantitative variables with a non-Gaussian distribution (creatinine, cytokines, and duration of exposure) are presented as medians and interquartile ranges (25th to 75th percentile) and compared with a nonparametric two-tailed test (Mann–Whitney U). The Shapiro–Wilk normality test was used to determine if a variable had Gaussian distribution. We performed a comparison in frequency by sex and age groups to compare the main job categories between patients with membranous nephropathy and the general population of French workers. Frequencies observed in the membranous nephropathy cohort for sex and age groups were used to calculated frequencies of the main job categories for the two cohorts. A multivariable logistic regression model with backward selection (threshold =0.15) was used to investigate independent factors that influence PLA2R1 spreading. Of the 100 patients, ten creatinine values at diagnosis were missing. We ensured that the numbers of missing data were similar between patient groups using a Fisher exact test. The ten missing data were thus simply removed from the analysis. Statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, Inc., San Diego, CA) for unadjusted analysis and SAS Enterprise Guide 7.1 for multivariable analysis. All comparisons were two tailed, and the statistical differences were considered significant when P<0.05.

Results

Cohort Description

We included 100 patients with biopsy-proven membranous nephropathy in this retrospective Nice-based cohort study. The men-women sex ratio was 1.7, and the mean age at diagnosis of membranous nephropathy was 52±15 years. The median follow-up at the time of the study and thus at the time of the questionnaire was 7 (3–13) years. Sixty-two were active or past smokers. Eleven had membranous nephropathy secondary to another disease, such as infection, lupus, or cancer. For 89 patients, the etiologic workup remained negative; 63 had anti-PLA2R1 antibodies, two had antibodies to thrombospondin type 1 domain containing 7A, and 25 had neither anti-PLA2R1 nor antithrombospondin type 1 domain containing 7A antibodies. Among patients with membranous nephropathy mediated by anti-PLA2R1 antibodies, 31 had PLA2R1 epitope spreading beyond the CysR domain. Among the 100 participants, six patients achieved spontaneous complete remission, 70 had persistent nephrotic syndrome and/or required immunosuppressive therapy to achieve remission, and 24 progressed to kidney failure requiring replacement therapy with dialysis and/or kidney transplantation. Patients’ characteristics are detailed in Table 2.

Table 2.

Characteristics of the study population

Patients' Characteristics	Participants, n=100
Baseline characteristics
Men, n (%)	63 (63)
Sex ratio	1.7
Age at inclusion, yr	61 (15)
Age at membranous nephropathy diagnosis, yr	52 (15)
20–39, n (%)	20 (20)
40–64, n (%)	60 (60)
65–99, n (%)	20 (20)
Age at diagnosis according to the sex
Women	52 (18)
Men	51 (13)
Job categories, n (%)
Agricultural workers	6 (6)
Industry workers	14 (14)
Service workers	43 (43)
Construction workers	31 (31)
No professional activity	6 (6)
Active or former smoker, n (%)	62 (62)
Membranous nephropathy presentation, n (%)
Secondary membranous nephropathy	11 (11)
Autoimmune disease (e.g., lupus)	8 (8)
Infection	2 (2)
Malignancies	1 (1)
Primary membranous nephropathy	89 (89)
PLA2R1-mediated membranous nephropathy	63 (63)
THSD7A-mediated membranous nephropathy	2 (2)
Double-negative membranous nephropathy	25 (25)
PLA2R1 epitope spreading, n=63
Yes	31 (49)
No	30 (48)
Missing data	2 (3)
Course of membranous nephropathy
Spontaneous complete remission	6 (6)
Persistent nephrotic syndrome and/or need for immunosuppressive therapy	70 (70)
Kidney failure	24 (24)

The numbers (and percentages) of patients are indicated for categorical variables, and mean (SD) are shown for continuous variables. Excluding the two patients for whom we did not have data on their PLA2R1 epitope spreading, there were no missing data. PLA2R1, phospholipase A2 receptor 1; THSD7A, thrombospondin type 1 domain containing 7A.

Occupational Exposures of Patients with Membranous Nephropathy

Main Job Categories.

The main job categories of the 94 patients with membranous nephropathy who had a professional activity during their lifetime were compared with those of the 26,734,000 French workers in 2018 (INSEE data) and are represented in Figure 1. Patients with membranous nephropathy worked significantly more in the construction (33% versus 7%, P<0.001) and agricultural sectors (6% versus 3%, P=0.04) and less in the service sector (46% versus 77%, P<0.001) than the general population of French workers. A detailed comparison of demographic and socioprofessional characteristics between patients with membranous nephropathy and French workers is provided in Supplemental Table 1. We also performed a comparison of the main job categories between French workers and the subgroup of the 83 patients with primary membranous nephropathy who had a professional activity (i.e., excluding patients with secondary membranous nephropathy), and we found the same results (Supplemental Figure 1). Because patients with membranous nephropathy and the general population of French workers had noncomparable demographic characteristics (sex and age), we performed a comparison in frequency (sex and age groups) of the main job categories; patients with membranous nephropathy over 30 years old worked significantly more in the construction sector than the general population of French workers (Table 3). In this cohort, considering service workers under 50 as the reference population, we estimated that the risk of developing membranous nephropathy is eight times higher in construction workers, regardless of age and sex (Supplemental Table 2).

Figure 1.

Comparison of the main job categories between patients with membranous nephropathy and the general population of French workers. The data source for French workers is the French National Institute of Statistics and Economic Studies 2018 and covers 26,734,000 individuals. Individuals who have never worked were not included in the analysis. Patients with membranous nephropathy worked significantly more in the construction (P<0.001) and agricultural sectors (P=0.04) and less in the service sector (P<0.001) than the general population of French workers. Statistical significance was determined by a chi-squared test or a Fisher exact test according to patient sample size. There were no missing data.

Table 3.

Comparison in frequency (sex and age) of the main job categories between patients with membranous nephropathy and the general population of French workers

Job Categories	French Workers, 100% (n=26,734,000)	Patients with Membranous Nephropathy, 100% (n=94)	P Value
Men (63% a ), %
Age 15–29 (10%a)
Service workers	5 (1,296,599)	3 (3)	ns
Industry workers	0.9 (235,259)	1 (1)	ns
Construction workers	0.4 (117,630)	2 (2)	ns
Agricultural workers	0.2 (42,774)	(0)	ns
Age 30–49 (34%a)
Service workers	16 (4,408,437)	10 (9)	ns
Industry workers	3 (802,020)	3 (3)	ns
Construction workers	2 (401,010)	7 (7)	0.01b
Agricultural workers	0.5 (141,690)	1 (1)	ns
Age ≥50 (56%a)
Service workers	27 (7,260,954)	16 (15)	0.01b
Industry workers	5 (1,320,660)	5 (5)	ns
Construction workers	2 (660,330)	12 (11)	0.02b
Agricultural workers	0.9 (235,259)	2 (2)	ns
Women (37% a ), %
Age 15–29 (10%a)
Service workers	3 (761,919)	2 (2)	ns
Industry workers	0.5 (139,017)	1 (1)	ns
Construction workers	0.3 (69,508)	1 (1)	ns
Agricultural workers	0.1 (24,061)	(0)	ns
Age 30–49 (34%a)
Service workers	10 (2,590,525)	5 (5)	ns
Industry workers	3 (470,518)	2 (2)	ns
Construction workers	0.9 (235,259)	4 (4)	0.01b
Agricultural workers	0.3 (80,202)	1 (1)	ns
Age ≥50 (56%a)
Service workers	16 (4,264,073)	10 (9)	ns
Industry workers	3 (775,286)	3 (3)	ns
Construction workers	1 (387,643)	6 (6)	0.01b
Agricultural workers	0.5 (136,343)	1 (1)	ns

Main job categories between patients with membranous nephropathy and the general population of French workers are compared using a Fisher exact test. Differences between groups were considered to be statistically significant when P values were <0.05. ns, not significant.

^aThe frequency observed in the membranous nephropathy cohort was used to calculate the frequency (by sex and age groups) of the main job categories for both cohorts.

^bSignificant associations.

Occupational Toxic Agents.

Compared with the general population of French workers (2017 SUMER survey), patients with membranous nephropathy were more frequently exposed to three toxic substances during their professional activity: lead (14% lifetime prevalence and 9% 2017 prevalence for patients with membranous nephropathy versus 1% prevalence for the general population; P<0.001 and P<0.001, respectively), asbestos (21% lifetime prevalence and 16% 2017 prevalence for patients with membranous nephropathy versus 5% prevalence for the general population; P<0.001 and P<0.001, respectively), and organic solvents (57% lifetime prevalence and 37% 2017 prevalence for patients with membranous nephropathy versus 15% prevalence for the general population; P<0.001 and P<0.001, respectively) (Figure 2). It should be noted that environmental and domestic nonoccupational toxic exposures identified through a second questionnaire were negligible (data not shown).

Figure 2.

Comparison of (A) lifetime and (B) 2017 prevalence of occupational exposure between patients with membranous nephropathy and the general population of French workers. The data source for French workers is the 2017 French Medical Surveillance of Employees’ Exposure to Occupational Risks survey, which is a national survey of occupational exposures developed by the French Ministry of Labor that reported the proportion of workers exposed to toxic agents among the employed population between April 2016 and September 2017. Patients with membranous nephropathy were more frequently exposed to lead, asbestos, and organic solvents during their professional activity than the general population of French workers. Statistical significance was determined by a chi-squared test or a Fisher exact test according to patient sample size. There were no missing data.

Relationship between Membranous Nephropathy Characteristics and Occupational Exposure to Toxic Substances

Demographic Characteristics.

As shown in Figure 3, patients exposed to occupational toxic substances showed a predominance of men for two toxic agents: organic solvents (80% men in exposed patients versus 41% men in unexposed patients, P<0.001) and asbestos (90% men in exposed patients versus 55% men in unexposed patients, P=0.004). Interestingly, the sex ratio of unexposed patients with membranous nephropathy was balanced. For the three toxic substances studied, the age at diagnosis was not different between exposed and unexposed patients. Moreover, no difference in smoking habits was found between patients exposed to asbestos or lead and nonexposed patients (P=0.21 and P=0.36, respectively). In contrast, patients exposed to organic solvents were more likely to be active or past smokers than unexposed patients (P=0.03).

Figure 3.

Sex ratio by toxic occupational exposure to (A) organic solvents, (B) asbestos, and (C) lead. Patients with membranous nephropathy exposed to occupational toxic substances seemed to be more often men than women: organic solvents (80% men in exposed patients versus 41% men in unexposed patients, P<0.001), asbestos (90% men in exposed patients versus 55% men in unexposed patients, P=0.004), and lead (85% men in exposed patients versus 59% men in unexposed patients, P=0.12). Statistical significance was determined by a chi-squared test or a Fisher exact test according to patient sample size. There were no missing data.

Epitope Spreading.

Because epitope spreading is a poor prognostic factor in PLA2R1-associated membranous nephropathy (39,40), we investigated whether certain factors influence PLA2R1 epitope spreading, such as occupational exposure to toxic substances. After unadjusted analysis, two variables appeared to be significantly associated with epitope spreading: asbestos exposure (ten of 31 spreaders [32%] versus two of 30 nonspreaders [7%], P=0.02) and organic solvent exposure (23 of 31 spreaders [74%] versus 13 of 30 nonspreaders [43%], P=0.02). The association between PLA2R1 epitope spreading and asbestos exposure was confirmed by multivariable analysis (odds ratio, 6.67; 95% confidence interval, 1.32 to 33.69; P=0.02). In addition, patients exposed to organic solvents and asbestos with moderate or high intensity were more likely to have PLA2R1 epitope spreading than patients with no or low exposure (P=0.04 and P=0.04, respectively). The data are summarized in Table 4.

Table 4.

Comparison of baseline characteristics and occupational exposure between patients with membranous nephropathy and phospholipase A2 receptor 1 epitope spreading and those with membranous nephropathy without phospholipase A2 receptor 1 epitope spreading

Patients' Characteristics and Exposure	Spreading, n=31	No Spreading, n=30	Unadjusted P Value
Baseline characteristics
Men	25 (81%)	17 (57%)	0.06
Sex ratio men-women	4.2	1.3
Age at diagnosis, yr	52 (16)	55 (15)	0.52
Job categories
Construction workers	14 (45%)	7 (23%)	0.11
Other sector of activity	17 (55%)	23 (77%)
Smoking	19 (61%)	20 (67%)	0.79
Occupational exposure
Asbestos	10 (32%)	2 (7%)	0.02a
Lead	4 (13%)	2 (7%)	0.67
Organic solvents	23 (74%)	13 (43%)	0.02a
Level of exposure
Asbestos			0.04a
High/moderate exposure	9 (29%)	2 (7%)
Low/no exposure	22 (71%)	28 (93%)
Lead			0.19
High/moderate exposure	4 (13%)	1 (3%)
Low/no exposure	27 (87%)	29 (97%)
Organic solvents			0.04a
High/moderate exposure	20 (65%)	11 (37%)
Low/no exposure	11 (36%)	19 (66%)

Categorial (number and percentage) and continuous (mean and SD) variables are shown for patients with membranous nephropathy with or without epitope spreading. Baseline characteristics and occupational exposures were compared using a Fisher exact test and unpaired t test for categorial and continuous variables, respectively. All variables with P<0.15 in unadjusted analysis were used for multivariable logistic regression (i.e., men, construction worker, exposure to asbestos, and exposure to organic solvents). After backward selection, only asbestos was retained in the multivariable model (P=0.02). The proportion of patients with or without PLA2R1 epitope spreading is then shown for each exposure level and for each toxic agent in number and percentage and compared using a Fisher exact test. Differences between groups were considered to be statistically significant when P values were <0.05. There were no missing data.

^aSignificant associations.

Kidney Prognosis.

We found no significant difference in the kidney function at diagnosis according to the presence or absence of exposure during occupational activity. Creatinine at diagnosis in patients exposed to organic solvents compared with nonexposed patients was 1.09 (0.97–1.68) mg/dL and 1.01 (0.77–1.68) mg/dL, respectively (P=0.30). Creatinine at diagnosis in patients exposed to lead compared with nonexposed patients was 1.11 (1.02–1.60) mg/dL and 1.03 (0.78–1.69) mg/dL, respectively (P=0.38). Creatinine at diagnosis in patients exposed to asbestos compared with nonexposed patients was 1.39 (1.01–1.82) μmol/L and 1.02 (0.78–1.67) mg/dL, respectively (P=0.09). To note, ten creatinine values at diagnosis were missing, but the proportions of missing values between groups were not different. We did not find any effect of toxic occupational exposures on the progression of membranous nephropathy to kidney failure (Supplemental Table 3).

T Helper 17 Cytokine Profile.

We previously described an association between exposure to fine particles PM_2.5 and cytokines produced after in vitro nonspecific stimulation of the immune cells of patients with membranous nephropathy (22). Patients exposed to high levels of PM_2.5 exhibited a higher level of cytokine from the Th17 pathway (22). Cytokine assays were performed for 45 patients in the cohort and analyzed according to job categories. Agricultural workers had significantly lower IL-17A levels than construction and service workers (14.5 [7.8–18.5] pg/ml versus 65.0 [21.5–163.0] pg/ml, P=0.01 and 98.5 [27.8–172.8] pg/ml, P=0.01, respectively). Results are shown in Figure 4.

Figure 4.

IL-17A levels secreted by immune cells after in vitro stimulation. Immune cells were stimulated at 37°C for 24 hours in whole blood with anti-CD3 agonist mAbs and a Toll-like receptor 7/8 agonist. IL-17A levels were measured in cellular supernatants. Statistical significance of the differences between job categories was assessed using the Mann–Whitney U nonparametric test. **P=0.01.

Discussion

Membranous nephropathy is a rare autoimmune kidney disease. Its increasing prevalence in industrialized countries (25,29) pleads for the involvement of environmental factors acting as an adjuvant at the origin of an immune dysregulation that could favor the onset of the disease. In addition, the predominance of men (38) in membranous nephropathy, classically attributed to biologic or genetic differences between men and women, could also be due to different occupational exposures. Here, we highlighted that patients with membranous nephropathy work more frequently in the construction sector with exposure to occupational toxic substances, such as asbestos, lead, and organic solvents. The association between organic solvents and GN had already been suggested (36,37) but remains controversial (35). We here show for the first time an association between membranous nephropathy and exposure to lead and asbestos. We suggest that the well-known predominance of men may be due to a subgroup of patients exposed to occupational toxic substances. Toxic exposures outside the workplace were negligible. These observations need to be confirmed in a large prospective cohort of patients with membranous nephropathy. This is one of the objectives of an ongoing prospective national French research program called the IHMN study. We will assess whether an exposure to environmental toxic substances alters membranous nephropathy prognosis.

We found an association between exposure to organic solvents or asbestos and PLA2R1 epitope spreading, with a dose-dependent effect. PLA2R1 epitope spreading is a biomarker of poor prognosis and of nonresponse to rituximab therapy (39,40). However, in contrast to the Jacob et al. (36) study, exposure to occupational toxic agents had no effect on the progression of membranous nephropathy to kidney failure in our study, possibly due to the improved management of this disease since the discovery of anti-PLA2R1 antibodies (4,5) and the use of effective immunosuppressive therapies, such as rituximab (41–43). We investigated whether other confounding factors might be involved in the pathogenesis of the membranous nephropathy, and we found that patients exposed to organic solvents were also more often active or past smokers than nonexposed patients. Patients with membranous nephropathy exposed to other toxic substances and with PLA2R1 epitope spreading were not more likely to be smokers. Therefore, smoking is unlikely to be directly involved in the pathogenesis of the disease.

In this study, patients working in the agricultural sector secreted less IL-17A after in vitro stimulation of immune cells than patients working in other sectors. In a previous study, our team showed that increased IL-17A levels were associated with a higher risk of thrombosis and relapse (22), which could not be confirmed here, likely due to a small sample size. These increased IL-17A levels were more likely in patients living in urban areas exposed to high levels of fine particles PM_2.5 (22). An environmental factor may induce an immune dysregulation with an imbalance of the Th17-T regulatory ratio, causing a more inflammatory membranous nephropathy that would increase the risk of thrombosis and relapse. Several environmental factors, such as exposure to organic solvents or exposure to PM_2.5, can be suspected. This immune dysregulation toward the Th17 pathway induced by such environmental factors must be confirmed in vitro on cell culture, as already shown in other pathologies, such as hypersensitivity dermatitis with organic solvents (44).

Nevertheless, our study has several limitations. First, it is a retrospective study with a long follow-up period, which may induce a bias in the patients' memory of their previous occupations and the toxic substances to which they may have been exposed. However, only prolonged exposures over several years seem relevant, and it is unlikely that patients will forget an occupation that they have performed for years. Moreover, as membranous nephropathy is a rare disease, it seems difficult to design a prospective exposed/unexposed study. Second, although the sample size is reasonable for a rare disease, it is a single-center study, and the limited size of the patient cohort induces an imbalance with general population cohorts. It would be interesting to conduct the same study in other geographic areas and/or a retrospective analytic study of matched case-control type to confirm these results. Third, general population data were extracted from publicly available but incomplete databases. It would have been interesting to have continuous age rather than age ranges, as well as health data. Fourth, although the occupational health physicians who conducted the questionnaires were experienced, it cannot be excluded that their questions may have guided the patients' responses. Fifth, the duration of patient follow-up was variable, leading to differences in management (pre- and postrituximab eras) and delays that do not allow conclusions to be drawn regarding the subsequent occurrence of kidney failure in newly diagnosed patients. Finally, we used backward selection in the multivariable model, although this may increase the type 1 error. Most of these limitations will be controlled in the national French prospective IHMN study previously described.

Although various studies report the existence of a dose-response relationship between exposure to organic solvents and the occurrence of GN (33,34,36,45–47) as well as the beneficial effect obtained by avoiding this exposure (46), professional practices do not seem to have been modified, and the nephrotoxicity of this exposure is still not recognized by the French health authorities. This lack of recognition as an occupational disease could be partly explained by the fact that no direct correlation has yet been established in the absence of in vitro studies on cell cultures or in vivo studies in animals.

In conclusion, patients with membranous nephropathy are more frequently exposed to certain occupational toxic substances (organic solvents, lead, and asbestos) than the general population. This predominance of men in exposures, particularly high levels of asbestos exposure, may favor PLA2R1 epitope spreading. These results justify adherence to the use of personal protective equipment and increased surveillance of proteinuria in exposed workers. The occurrence of membranous nephropathy in the context of high occupational exposure to organic solvents or asbestos, without any other obvious etiology, seems to justify recognition of an occupational disease.

Disclosures

S. Benzaken reports an advisory or leadership role on the BMS Advisory Committee and other interests or relationships with the French Health Authority and the Regional Health Agency: Public Health National Committee. S. Benzaken’s spouse reports employment with the Imaging Center: La Grande Bleue, Nice, France. V. Esnault reports consultancy agreements with AstraZeneca, Bayer, BMS-Pfizer, Boehringer-Ingelheim, and Novartis; research funding from AstraZeneca, BMS-Pfizer, Boehringer-Ingelheim, Fresenius, Hemotech, and Novartis; honoraria from Amgen, Bayer, BMS-Pfizer, Boehringer-Ingelheim, Fresenius, Lilly, and Novartis; and advisory or leadership roles for AstraZeneca, Bayer, and Boehringer-Ingelheim. B. Seitz-Polski reports consultancy agreements with Novartis and patents with Euroimmun and is the coinventor on the patents “Methods and kits for monitoring membranous nephropathy” and “Prognosis and monitoring of membranous nephropathy based on the analysis of PLA2R1 epitope profile and spreading.” All remaining authors have nothing to disclose.

Funding

This work was supported by GIRCI Méditerranée Grant 18-GIRCI-03 and grants from ORKID (grant AAP-ORKID-2019).

Author Contributions

B. Seitz-Polski conceptualized the study; S. Agbekodo and M. Cremoni were responsible for data curation; B. Seitz-Polski was responsible for investigation; S. Agbekodo and M. Cremoni were responsible for formal analysis; M. Cremoni, B. Seitz-Polski, and K. Zorzi were responsible for methodology; M. Teisseyre was responsible for validation; S. Benzaken, J.-H. Planchard, and M. Teisseyre were responsible for visualization; S. Benzaken and B. Seitz-Polski were responsible for funding acquisition; B. Seitz-Polski provided supervision; M. Cremoni wrote the original draft; and S. Agbekodo, V. Brglez, V. Esnault, and B. Seitz-Polski reviewed and edited the manuscript.

Data Sharing Statement

Data supporting the conclusions of this study are available upon request to M. Cremoni and B. Seitz-Polski.

Supplemental Material

This article contains supplemental material online at http://cjasn.asnjournals.org/lookup/suppl/doi:10.2215/CJN.02930322/-/DCSupplemental.

Supplemental Material. Questionnaire.

Supplemental Table 1. Comparison of demographic and socioprofessional characteristics between patients with membranous nephropathy and the general population of French workers.

Supplemental Table 2. Estimated risk of membranous nephropathy for each occupation compared with the referent occupation (service sector) adjusted for age and sex.

Supplemental Table 3. Comparison of baseline characteristics and occupational exposure between patients with membranous nephropathy and kidney failure and those with membranous nephropathy without kidney failure.

Supplemental Figure 1. Comparison of the main job categories between patients with primary membranous nephropathy and the general population of French workers.