Research article

Follow-up of GSTM1, GSTT1, and NAT2 genotyped patients with knee or hip replacement

Selahattin Bozkurt¹, Silvia Selinski², Meinolf Blaszkewicz², Jörg Reinders², Jan G. Hengstler², Lukas Niggemann³, Klaus Golka²[*]

EXCLI J 2025;24:Doc677

Abstract

A total of 147 patients, genotyped for glutathione S-transferases M1 (GSTM1) and T1 (GSTT1) and for N-acetyltransferase 2 (NAT2), who had undergone total joint replacement of the knee or hip joint between August 2004 and June 2007, showed with 45% a remarkably lower portion of the GSTM1-negative genotype compared to both a local control (51%), an external control (52%) and the portion reported in the literature for the European population (50%). In contrast, the portions of GSTT1-positive (84%) and slow NAT2 (55.1%) patients of the initial collective were unremarkable, compared to both controls. To elucidate a possible impact of this interesting finding on the long-term outcome, the patients were contacted in December 2015. Afterwards, they were interviewed using a self-prepared questionnaire. The average follow-up time was 9 years. At the time of follow-up, 57 patients were deceased, 46 patients did not respond and 12 patients declined the interview. A total of 32 patients participated in the follow-up. The mean age of the followed-up patients was 75.9±8.3 years, whereas the mean age of all patients at the time of surgery was 70.9±9 years. The portions of the GSTM1-negative genotype (15 out of 32; 47%), the GSTT1-positive genotype (24 out of 32; 75%) and the slow NAT2 status (17 out of 32; 53%) in the followed-up patients were comparable to those of the initial collective. The follow-up results of the patients after 9 years were unable to clarify the significance of the observed lower portion of GSTM1-negative patients. In view of a recently published omics study reporting a reduced GSTM1 activity in tissue attached on hip implants explanted due to aseptic loosening, the striking portion of the GSTM1-negative genotype in this present study may encourage further investigation into the impact of this gene in patients with hip or knee replacement.

Keywords: osteoarthritis, hip arthroplasty, knee arthroplasty, glutathione S-transferase M1 (GSTM1), glutathione S-transferase T1 (GSTT1), N-acetyltransferase 2 (NAT2)

Introduction

Osteoarthritis of the knee or hip joints is a common condition in the general population and is becoming increasingly prevalent as the population ages. It is also of considerable socio-medical importance due to the high costs of both treatment and sick leave. In 2022, 177,826 initial hip joint implants and 137,030 initial knee joint implants were documented in Germany. In addition, there were 18,145 follow-up hip joint procedures and 14,379 follow-up knee joint procedures (Endoprothesenregister Deutschland (EPRD), 2023[11]).

Conservative treatment of osteoarthritis cannot currently reverse cartilage damage, as joint cartilage does not regenerate. This is impressively demonstrated by a study on the uptake of the radioactive carbon isotope ¹⁴C, which originated from surface nuclear weapons tests and, unlike most other tissues, was not increasingly incorporated into joint cartilage (Libby et al., 1964[40]). Therefore, in addition to taking pressure off the arthritic joint, conservative treatment primarily aims to relieve the pain.

When pain therapy has reached its limits, the only remaining effective therapeutic option is joint replacement. As the service life of artificial joints is limited (Evans et al., 2019[13][12]), it is of considerable interest to investigate factors that could be associated with a non-surgically induced reduced service life of the implants.

Aseptic loosening is the most common cause of total hip replacement and total knee replacement failure and revision surgery (Koks et al., 2020[34]). However, there is still no effective therapeutic target regarding prevention or conservative treatment. Although the impact of genetic factors on the development of osteoarthritis in the hip or knee joint is well known from twin studies (MacGregor et al., 2000[45]; Page et al., 2003[54]; Möller et al., 2015[47]; Skousgaard et al., 2016[62]), the underlying mechanism of aseptic loosening is largely unknown (Deng et al., 2017[7]). In contrast, Pan et al. (2014[55]) stated that genetic factors play an important role in early failure of total hip replacement due to aseptic loosening, e.g., SNPs (single-nucleotide polymorphisms) of GNAS1 (guanine nucleotide-binding protein, alpha-stimulating activity polypeptide 1), TNF-238 (tumor necrosis factor-alpha (TNF-α) 238 G/A promoter polymorphism), TNF-α (tumor necrosis factor-alpha), IL-6-174 (interleukin-6 174 G/C promoter polymorphism), MMP1 (matrix metalloproteinase-1), and MMP2 (matrix metalloproteinase-2).

However, several results of the first studies, e.g., the impact of GNAS1 or TNF genes on aseptic loosening could not be confirmed by later analyses (Koks et al., 2020[34]). A break-through in this area was a GWAS (genome-wide association study) based on 423 patients with total joint replacement, who were divided into 3 groups (without symptoms who received an implant at least one year before, patients undergoing implant surgery, patients receiving revision due to aseptic loosening) (Koks et al., 2020[34]). Totally 52 SNPs with a genome-wide suggestive p value less than 10^-5 were associated with implant loosening. The highest odds ratios (OR) were found for variants in the IFIT2/IFIT3 (interferon-induced protein with tetratricopeptide repeats 2 and 3, resp.) (OR 21.6, p= 4.35x10^-5), CERK (ceramide kinase) (OR 12.6, p=7.51x10^-7), and PAPPA (pappalysin) (OR 14.0, p=1.27x10^-5) genes. Variants of the 4 SNPs rs115871127, rs16823835, rs13275667 and rs2514486 predicted variability in the time to aseptic loosening.

In this study we focused on the genotype frequencies of the polymorphic xenobiotic-metabolising enzymes GSTM1 (glutathione S-transferase M1), GSTT1 (glutathione S-transferase T1), and NAT2 (N-acetyltransferase 2) in the follow-up group of patients with knee or hip replacement.

The gene coding for glutathione S-transferase M1 (GSTM1) is located on chromosome 1 and has a length of 21,226 bases (Gene Cards The human genome database, 2025[16]). The frequency of the GSTM1-negative genotype in Europeans is 50%. It ranges from 27% (Sub-Saharan Africans) to 54% (East Asians) (Nakanishi et al., 2022[48]). The enzyme GSTM1 metabolises highly reactive molecules (Bolt and Thier, 2006[5]; Ginsberg et al., 2009[20]). A classic example is epoxides resulting from polycyclic aromatic hydrocarbons metabolism by CYP450 (cytochrome P450) enzymes such as CYP1A1/2 or CYP1B1 (Figure 1(Fig. 1); Reference in Figure 1: Deutsche Forschungsgemeinschaft, 2008[8]).

A special feature of the enzyme GSTM1 is that it is obviously only relevant for the detoxification of substances whether the gene is present or not. However, single-nucleotide exchanges were also detected in this gene (Gene Cards The human genome database, 2025[16]). The first report on bladder cancer and GSTM1 showed an association with smoking habits (Bell et al., 1993[2]). Later, an association with environmental and / or occupational exposure was reported (Golka et al., 1997[22], 2009[21]; Hung et al., 2004[28]; Figueroa et al., 2015[14]). This association was no longer detectable after the end of exposure in the regions of Dortmund, Germany and Lutherstadt Wittenberg, Germany due to industrial structural changes with a dramatic reduction in emissions of combustion products due to reduction in the enzyme`s substrate (Ovsiannikov et al., 2012[53]; Krech et al., 2017[36]; Bergmann et al., 2025[3]). In a recent meta analysis based on 28,270 bladder cancer cases from 46 studies, OR was 1.37 (95% CI 1.19-1.59) for smokers and 1.26 (95% CI 1.08-1.48) for non-smokers (Yu et al., 2017[66]).

However, to date there have been hardly any studies on the role of this enzyme in osteoarthritis. Klein et al. (2012[33]) found a remarkably reduced portion of GSTM1-negative patients (45%) in 147 patients with hip or knee joint replacement compared to two control groups (local 51%, external 52%) and compared to the portion reported in the literature for the European population (50%) (Nakanishi et al., 2022[48]). In addition, the association of the GSTM1 genotype with cytogenetic abnormalities was analysed in knee joint synovial cells from patients with arthritis of different etiology (Ilyinskikh et al., 2017[29]) and in joint synovial fluid with the recovery of patients undergoing hip replacement (Lin et al., 2018[41]). GSTM1 also deactivates reactive oxygen species and lipid peroxidation products (Ellwanger et al., 2025[10]). For a comprehensive overview of the impact of the GSTM1 gene on different diseases, see Nakanishi et al. (2022[48]).

The gene coding for glutathione S-transferase T1 (GSTT1) is located on chromosome 22 and has a length of 8,179 bases (Gene Cards The human genome database, 2025[17]). The frequency of GSTT1-negative genotype in Europeans is 17%. It ranges from 14 % (Native Americans) to 48 % (East Asians) (Nakanishi et al., 2022[48]). An essential task of this enzyme is the conjugation of reduced glutathione to a wide number of exogenous and endogenous hydrophobic electrophiles (Gene Cards The human genome database, 2025[17]).

GSTT1 metabolises in particular small molecules with one or two carbon atoms, such as dichloromethane or ethylene oxide (Bolt and Thier, 2006[5]). A special feature of GSTT1 is that, as with the above described enzyme GSTM1, it is obviously only relevant for the detoxification of substances whether this gene is present or not. GSTT1 also deactivates reactive oxygen species and lipid peroxidation products (Ellwanger et al., 2025[10]). For a comprehensive overview of the impact of the GSTT1 gene on different diseases, see Nakanishi et al. (2022[48]).

The gene coding for N-acetyltransferase 2 (NAT2) is located on chromosome 8 and has a length of 14,918 bases (Gene Cards The human genome database, 2025[18]). Over 120 alleles (haplotypes) are known to date (NAT Committee, 2024[49]). With regard to the effect in the organism, a distinction is generally made between rapid and slow acetylation. Ruiz et al. (2012[57]) described an ultra-slow NAT2 status for the first time. Selinski et al. (2013[60]) reported that the ultra-slow NAT2 status increases the risk of urinary bladder cancer even if the classic slow NAT2 acetylation status, which includes all genotypes coding for a slow NAT2 acetylation status, no longer shows an increased risk of urinary bladder cancer. Compared to the slow NAT2 acetylation status, the ultra-slow NAT2 acetylation status has a metabolic capacity that is reduced by approx. 30% (Selinski et al., 2013[60]). This means that in ultra-slow NAT2 acetylators, the oxidative metabolic pathway is followed to a relevant extent, which ultimately leads to arylnitrenium ions, which, according to current knowledge, are the metabolites of carcinogenic aromatic amines that trigger bladder cancer (Hein et al., 1992[26]).

Classical substrates of NAT2 are aromatic amines and other substances that have an amino group on an aromatic ring or in which such a group is released during metabolism (Golka et al., 2004[24]). The substrates of NAT2 also include, for example, the tuberculostatic drug isoniazid (INH), the anticonvulsant clonazepam, the antihypertensive drug hydralazine and the anti-inflammatory drug sulfasalazine (Gene Cards The human genome database 2025[18]).

With regard to inflammatory diseases, there is a known correlation between NAT2 acetylation status and the risk of developing lupus erythematosus, an autoimmune inflammatory rheumatic disease from the group of collagenoses (von Schmiedeberg et al., 1999[65]; Santos et al., 2016[58]). In psoriasis, which is also an autoimmune disease and manifests as an inflammatory skin disease, an association between two SNPs of NAT2 and an earlier manifestation as well as a more severe course has been described (Dursun et al., 2018[9]). To the best of our knowledge, no studies have yet been published on a possible association between osteoarthritis of the hip or knee joint and NAT2 acetylation status.

The aim of this study was to investigate the frequencies of GSTM1, GSTT1, and NAT2 genotypes in the follow-up group.

Materials and Methods

This follow-up study is based on a study group of 147 patients who had undergone total joint replacement of the knee or hip joint between August 2004 and June 2007 at the department of surgery, St. Elisabeth Hospital, Iserlohn, Germany (Klein et al., 2012[33]). First, the patients were contacted in December 2015 and then interviewed using a self-prepared questionnaire, given in the supplement8565_supplementary data.pdf. The control group of 129 patients without indication for joint replacement of the initial study group was from the surgical department of the St. Elisabeth Hospital, Iserlohn, Germany (Klein et al., 2012[33]). NAT2 genotypes of the Iserlohn patients were separately published (Selinski et al., 2011[61]). The external control group of urological patients without known cancer was from the department of urology of the Josefs Hospital, Dortmund, Germany (Ovsiannikov et al., 2012[53]), which is about 25 km away.

The study was approved by the Ethics Committee of the Westphalia-Lippe Medical Association and of the Westphalian Wilhelms University, Münster (2015-492-f-S) and by the Institutional Review Board.

Analytical Methods

Genotyping for GSTM1, GSTT1 and NAT2 was performed during the initial study period between August 2004 and June 2007 according to standard methods. In short, DNA of cases and controls was isolated from venous blood samples according to standard methods in the Dortmund Institute (QIAamp DNA Maxi Kit Qiagen, Hilden, Germany). Genotyping for the presence or absence of the GSTM1 and GSTT1 genes was performed by duplex polymerase chain reaction (PCR), as described by Kempkes et al. (1996[31]). NAT2 genotyping was performed by PCR and RFLP-based methods (Selinski et al., 2013[60]).

Statistical Methods

Percentages of the different genotypes of the three investigated polymorphic enzymes GSTM1, GSTT1 and NAT2 were determined in the groups of followed-up, deceased and not contactable patients. Age at death of the deceased patients with artificial knee or hip replacement was plotted using Kaplan-Meier curves. The free software R version 4.2.1 was used for all calculations and plotting Kaplan-Meier curves (The R Foundation, 2022[63]).

Results

At the time of follow-up, 57 patients were deceased, 46 patients did not respond and 12 patients declined the interview. The age of the deceased patients is presented in Figure 2(Fig. 2). A total of 32 patients participated in the follow-up. The average follow-up time was 9 years.

The mean age of the followed-up patients was 75.9±8.3 years, whereas the mean age of all patients at the time of surgery was 70.9±9 years.

The portions of the GSTM1-negative genotype (15 out of 32; 47%), the GSTT1-positive genotype (24 out of 32; 75%) and the slow NAT2 status (17 out of 32; 53%) in the followed-up patients (Figure 3(Fig. 3)) were comparable to that of the initial collective that showed with 45% a remarkably lower portion of the GSTM1-negative genotype compared to both a local control (51%) and an external control (52%) (Figure 4(Fig. 4); References in Figure 4: Klein et al., 2012[33]; Ovsiannikov et al., 2012[53]; Selinski et al., 2011[61]) and the portion reported in the literature for the European population (50%) (Nakanishi et al., 2022[48]). The portions of GSTT1-positive (84%) and slow NAT2 (55.1%) patients of the initial collective were unremarkable, compared to both controls.

A review of the routine postoperative X-ray images of the joints revealed that although they provided the information required by the surgeon to assess the correct position of the implant, they did not provide any relevant additional information for the study.

Discussion

Twin studies showed the impact of genetic factors on the development of osteoarthritis in the hip or knee joint (MacGregor et al., 2000[45]; Page et al., 2003[54]; Möller et al., 2015[47]; Skousgaard et al., 2016[62]). However, the genetic effect was reported to be greater in the case of osteoarthritis of the hip joint (Magnusson et al., 2017[46]). Besides a considerable number of articles, several GWAS have been published on the impact of polymorphisms (Kerkhof et al., 2010[32]; arcOGEN Consortium, 2012[1]; Boer et al., 2021[4]; Kulm et al., 2022[38], 2023[37]; Henkel et al., 2023[27]; Hatzikotoulas et al., 2025[25]).

Henkel et al. (2023[27]) described 28 variants for knee joint osteoarthritis and 34 variants for hip joint osteoarthritis in a GWAS based on 61,151 cases of knee joint osteoarthritis, including 22,522 cases with joint replacement, and 38,068 cases of hip joint osteoarthritis, including 20,221 cases with joint replacement. Among them, no genes coding for xenobiotic-metabolising enzymes were reported.

The observed odds ratios were low. For surgically treated knee joints, the odds ratios for the variants in the additive model were a maximum of 1.11; for the only variant analysed according to the recessive model, the odds ratio was 3.57. For surgically treated hip joint arthrosis, the odds ratios were significantly higher. Here, in the analysis according to the recessive model, 6 variants were above 1.11, with a maximum of 2.56. In the three variants analysed according to the recessive model, the odds ratio was between 1.14 and 5.60.

Most recently, Hatzikotoulas et al. (2025[25]) published a GWAS based on 489,975 osteoarthritis cases, including all locations of osteoarthritis, and 1,472,094 controls. This study included 97,328 hip osteoarthritis cases, 49,874 total hip replacement cases, 172,256 knee osteoarthritis cases, and 48,161 total knee replacement cases. The authors reported an association of 146 variants for knee joint osteoarthritis, 92 variants for total knee replacement, 151 variants for hip joint osteoarthritis and 136 variants for total hip replacement. Furthermore, they identified eight biological processes enriched for effector genes. Although the study of Hatzikotoulas et al. (2025[25]) included drug targets, only CYP26B1 (cytochrome P450 26B1) involved in the degradation and ALDH1A2 (aldehyde dehydrogenase 1A2) involved in the synthesis of all-trans-retinoic acid were reported as genes coding for metabolizing enzymes associated with hip or knee osteoarthritis. Furthermore, a decrease in COX1 (cytochrome C oxidase 1, synonym: cyclooxygenase-1) (synonym: PTGS1, prostaglandin-endoperoxide synthase 1) expression in degraded compared with intact osteoarthritis-affected chondrocytes was reported.

To overcome the generally low power of single-nucleotide polymorphisms (Golka et al., 2011[23]), polygenic risk scores (PRS) have been applied (Sedaghati-Khayat et al., 2022[59]). On the other hand, Gill et al. (2024[19]) reported that reduced genomic heterozygosity is associated with a higher risk of osteoarthritis. Hatzikotoulas et al. (2025[25]) also evaluated the predictive potential of risk scores. The best performing genetic risk score (GRS) was obtained for hip osteoarthritis (area under the curve, AUC, 58.6%). Therefore, such scores will not achieve a noticeable practical significance beyond research.

Studies on the impact of polymorphisms on the service life of the implant and/or long-term satisfaction with the implant have not been published, with the exception of studies of Omar et al. (2017[52]), who applied transcriptome-wide high-density microarray analysis on periprosthetic tissue from 12 hips with chronic infection, using 12 hips with aseptic loosening as controls, and Liu et al. (2025[42]). Liu et al. (2025[42]) applied integrated transcriptomics, proteomics and untargeted metabolomics analyses in 8 cases of aseptic loosening of implants after hip replacement. They observed characteristic metabolite changes (biosynthesis of guanine, L‑glycine and adenosine) and decreased CRLF1 (cytokine receptor-like factor 1) and GSTM1 activities in tissues attached to explanted artificial hip specimens, compared to 8 controls who received primary total hip replacement due to avascular necrosis of femoral head or femoral neck fracture.

Service life of the implant and satisfaction with an artificial knee or hip joint can vary greatly. Early loosening of implants is essentially caused by five interfering factors: patient, material, implant design, fixation, and surgical technique (Katzer and Löhr, 2003[30]). However, the underlying mechanism of aseptic loosening is largely unknown (Deng et al., 2017[7]). Polymorphic enzymes do not only influence a number of diseases with regard to the risk of disease (NHGRI-EBI GWAS Catalog, 2025[50]), but in some cases also its course.

In the initial study, patients were genotyped for glutathione S-transferases M1 (GSTM1) and T1 (GSTT1) as well as for N-acetyltransferase 2 (NAT2) to investigate the impact of the polymorphic enzymes on the indication for joint replacement (Klein et al., 2012[33]) - pain that can no longer be controlled with medication. These patients showed a remarkably lower portion of the GSTM1-negative genotype (45%) compared to two control groups (local 51%, external 52%) and the portion reported in the literature for the European population (50%) (Nakanishi et al., 2022[48]).

In the scientific discussion with colleagues, the question raised was what the lower frequency of the GSTM1-negative genotype and thus an increased metabolic activity with regard to the detoxification of reactive metabolic products could mean in the long term with regard to the service life of the joint replacement and patient satisfaction.

Apart from the study by Klein et al. (2012[33]), no studies listed in PubMed or Google Scholar on the frequency of the GSTM1-negative genotype in this patient group have been published to date.

One reason for this could be that this genotype was not included as standard on SNP chips used in GWAS, at least not in the past. This is because a decisive criterion for the inclusion of a reference SNP or relevant genetic information on a SNP chip is the good differentiation of the signal. As an example, this meant that previously the ultra-slow NAT2 genotype defined by SNPs rs1041983 and rs1799930 was not genotyped in GWAS for bladder cancer (Figueroa et al., 2016[15]) - even though the slow NAT2 phenotype is the oldest known genetic risk factor for the development of urinary bladder cancer when exposed to carcinogenic aromatic amines (Lower et al., 1979[43]). Due to its fundamental meaning in molecular epidemiology, the article of Lower et al. (1979[43]) was reprinted and commented (Lower et al., 2007[44]; Vineis, 2007[64]; Olden, 2007[51]; Rothman et al., 2007[56]).

The GSTM1 genotype, which is a risk factor for bladder cancer, was therefore determined indirectly via a proxy marker in some GWAS for bladder cancer, for example. A proxy marker (chr1:110229772) effectively tagged the GSTM1 deletion (Koutros et al., 2023[35]).

In a patient population with an average age of 70.9±9 years, the question arises as to when a follow-up should be carried out. A higher probability of participation speaks in favour of a short follow-up period, where lower number of deaths in the patient population as well as the robustness of the contacted patients is to be expected. An argument against a relatively short follow-up period is that no findings can be obtained on the long-term stability of the implants and the long-term satisfaction of the patients with regard to quality of life.

In the present study, the follow-up period was 9 years. During this period, 57 patients had died and 46 could not be contacted.

The selected follow-up period covers at least a considerable part of the period in which early implant loosening occurs. The frequency of failure of hip replacements varies among countries. During the first 5 years, New Zealand had the highest survival rate, with almost 97%, while Denmark had the lowest with 95.5%. After 10 years, Australia revealed the highest survival rate of implants (95%), whereas Denmark had the lowest, namely 92.8%. Fifteen years after implantation, 94% of hip replacements in Australia survived, whereas the survival rate in Denmark was much lower with 89% (Clar et al., 2024[6]). These findings are also of clinical interest in patient collectives at an older age. However, early loosening is not clearly defined (Kvarda et al., 2024[39]).

The main result of this study is that the portion of GSTM1-negative patients (47%) is comparable to the portion in the initial study group, albeit with a small number of cases. The conspicuous distribution in the follow-up group does not provide any concrete evidence of a possible cause.

The reduced GSTM1 activity reported by Liu et al. (2025[42]) in cells attached to explanted artificial joints due to early loosening may also result from non-genetic causes such as implant-related changes in the microenvironment. Therefore, factors like substances released from the artificial joint or the cement used or biological processes triggered by the implants may be considered.

Overall, with regard to the possible systemic influence of GSTM1 on osteoarthritis of the knee or hip joint, it must also be taken into account that noticeable associations with SNPs may be caused by indirect effects, such as an impact on obesity risk factors.

In view of the first description of a decreased activity of the GSTM1 enzyme in tissues attached to explanted artificial hip specimens from patients with aseptic loosening of the artificial hip joint compared to controls by Liu et al. (2025[42]), the findings of this study and the preceding study by Klein et al. (2012[33]) may encourage further research on the impact of GSTM1 in patients with hip or knee replacement.

Notes

Dedicated to our librarian Dipl.-Bibl. Susanne Lindemann on the occasion of her retirement

Declaration

Author Contributions

Conceptualization S.B., S.S., J.G.H., and K.G., methodology S.S., M.B., and J.R., software S.S., laboratory analysis M.B., data analysis S.B., S.S., investigation S.B., S.S., and K.G., resources L.N., J.G.H., data curation J.R., S.S., writing-original draft preparation S.B., S.S., M.B., and K.G., writing-review and editing S.B., L.N., and K.G., visualization S.B., K.G., supervision L.N., J.G.H, project administration J.G.H., K.G. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board and by the Ethics Committee of the Westphalia-Lippe Medical Association and of the Westphalian Wilhelms University, Münster (2015-492-f-S).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data are available on reasonable request from the corresponding author.

Conflicts of Interest

The authors declare no conflict of interest.

Using Artificial Intelligence (AI)

The authors have not used artificial intelligence-assisted technologies.

References

1. arcOGEN Consortium; arcOGEN Collaborators; Zeggini E, Panoutsopoulou K, Southam L, Rayner NW, Day-Williams AG, Lopes MC, et al. Identification of new susceptibility loci for osteoarthritis (arcOGEN): a genome-wide association study. Lancet. 2012;380(9844):815-23. doi: 10.1016/S0140-6736(12)60681-3

2. Bell DA, Taylor JA, Paulson DF, Robertson CN, Mohler JL, Lucier GW. Genetic risk and carcinogen exposure: a common inherited defect of the carcinogen-metabolism gene glutathione S-transferase M1 (GSTM1) that increases susceptibility to bladder cancer. J Natl Cancer Inst. 1993;85(14):1159-64. doi: 10.1093/jnci/85.14.1159

3. Bergmann S, Blaszkewicz M, Reinders J, Selinski S, Kowald B, Volkert F, et al. NAT2, GSTM1, and GSTT1 genotypes and urinary bladder cancer risk after structural changes in the industrial region of Lutherstadt Wittenberg. Abstract P036. 10th German Pharm-Tox Summit 2025. Naunyn-Schmiedeberg's Arch Pharmacol. 2025. https://doi.org/10.1007/s00210-025-03881-x. Accessed 27 April 2025

4. Boer CG, Hatzikotoulas K, Southam L, Stefánsdóttir L, Zhang Y, Coutinho de Almeida R, Deciphering osteoarthritis genetics across 826,690 individuals from 9 populations. Cell. 2021;184(18):4784-818.e17. doi: 10.1016/j.cell.2021.07.038. Erratum in: Cell. 2021;184(24):6003-6005. doi: 10.1016/j.cell.2021.11.003

5. Bolt HM, Thier R. Relevance of the deletion polymorphisms of the glutathione S-transferases GSTT1 and GSTM1 in pharmacology and toxicology. Curr Drug Metab. 2006;7(6):613-28. doi: 10.2174/138920006778017786

6. Clar C, Leitner L, Koutp A, Hauer G, Rasic L, Leithner A, et al. The worldwide survival rate of total hip arthroplasties is improving: a systematic comparative analysis using worldwide hip arthroplasty registers. EFORT Open Rev. 2024;9(8):745-50. doi: 10.1530/EOR-23-0080

7. Deng Z, Wang Z, Jin J, Wang Y, Bao N, Gao Q, et al. SIRT1 protects osteoblasts against particle-induced inflammatory responses and apoptosis in aseptic prosthesis loosening. Acta Biomater. 2017;49:541-54. doi: 10.1016/j.actbio.2016.11.051

8. Deutsche Forschungsgemeinschaft. Ständige Senatskommission zur Prüfung gesundheitsschädlicher Arbeitsstoffe. Polyzyklische aromatische Kohlenwasserstoffe (PAH). In: The MAK collection for occupational health and safety | Gesundheitsschädliche Arbeitsstoffe, Toxikologisch-arbeitsmedizinische Begründung von MAK-Werten, 45. Lieferung, Wiley, 2008. https://doi.org/10.1002/3527600418.mb0223orgd0045 Accessed 27 April 2025

9. Dursun R, Dursun HG, Zamani AG, Yıldırım MS, Çınar İ. NAT2 Gene polymorphisms in Turkish patients with psoriasis vulgaris. Biomed Res Int. 2018;2018:3258708. doi: 10.1155/2018/3258708

10. Ellwanger JH, Ziliotto M, Chies JAB. Toxicogenomics of glutathione S-transferase (GST) gene family members: Chemical-gene interactions and potential implications of gene deletions. Comput Biol Med. 2025;189:110025. doi: 10.1016/j.compbiomed.2025.110025

11. Endoprothesenregister Deutschland (EPRD) Bericht 2023 Mit Sicherheit mehr Qualität. doi: 10.36186/reporteprd082023 https://www.eprd.de/fileadmin/user_upload/Dateien/Publikationen/Berichte/Jahresbericht2023-Status5_2023-10-24_F.pdf. Accessed 27 April 2025

12. Evans JT, Evans JP, Walker RW, Blom AW, Whitehouse MR, Sayers A. How long does a hip replacement last? A systematic review and meta-analysis of case series and national registry reports with more than 15 years of follow-up. Lancet. 2019;393(10172):647-54. doi: 10.1016/S0140-6736(18)31665-9

13. Evans JT, Walker RW, Evans JP, Blom AW, Sayers A, Whitehouse MR. How long does a knee replacement last? A systematic review and meta-analysis of case series and national registry reports with more than 15 years of follow-up. Lancet. 2019;393(10172):655-63. doi: 10.1016/S0140-6736(18)32531-5

14. Figueroa JD, Koutros S, Colt JS, Kogevinas M, Garcia-Closas M, Real FX, et al. Modification of occupational exposures on bladder cancer risk by common genetic polymorphisms. J Natl Cancer Inst. 2015;107(11):djv223. doi: 10.1093/jnci/djv223

15. Figueroa JD, Prokunina-Olsson L, Koutros S, Garcia-Closas M, Chanock S, Silverman DT, et al. Response. J Natl Cancer Inst. 2016;108(3):djv441. doi: 10.1093/jnci/djv441

16. Gene Cards The human genome database GSTM1 Gene - Glutathione S-Transferase Mu 1 https://www.genecards.org/cgi-bin/carddisp.pl?gene=GSTM1. 2025a. Accessed 27 April 2025

17. Gene Cards The human genome database GSTT1 Gene Glutathione S-Transferase Theta 1 https://www.genecards.org/cgi-bin/carddisp.pl?gene=GSTT1. 2025b. Accessed 27 April 2025

18. Gene Cards The human genome database NAT2 Gene - N-Acetyltransferase 2 https://www.genecards.org/cgi-bin/carddisp.pl?gene=NAT2. 2025c. Accessed 27 April 2025

19. Gill R, Liu M, Sun G, Furey A, Spector T, Rahman P, et al. Genomic heterozygosity is associated with a lower risk of osteoarthritis. BMC Genomics. 2024;25(1):85. doi: 10.1186/s12864-024-10015-9

20. Ginsberg G, Smolenski S, Hattis D, Guyton KZ, Johns DO, Sonawane B. Genetic polymorphism in glutathione transferases (GST): Population distribution of GSTM1, T1, and P1 conjugating activity. J Toxicol Environ Health B Crit Rev. 2009;12(5-6):389-439. doi: 10.1080/10937400903158375

21. Golka K, Hermes M, Selinski S, Blaszkewicz M, Bolt HM, Roth G, et al. Susceptibility to urinary bladder cancer: relevance of rs9642880[T], GSTM1 0/0 and occupational exposure. Pharmacogenet Genomics. 2009;19(11):903-6. doi: 10.1097/FPC.0b013e328331b554

22. Golka K, Reckwitz T, Kempkes M, Cascorbi I, Blaskewicz M, Reich SE, et al. N-acetyltransferase 2 (NAT2) and glutathione S-transferase µ (GSTM1) in bladder-cancer patients in a highly industrialized area. Int J Occup Environ Health. 1997;3(2):105-10. doi: 10.1179/107735297800407686

23. Golka K, Selinski S, Lehmann ML, Blaszkewicz M, Marchan R, Ickstadt K, et al. Genetic variants in urinary bladder cancer: collective power of the "wimp SNPs". Arch Toxicol. 2011;85(6):539-54. doi: 10.1007/s00204-011-0676-3

24. Golka K, Wiese A, Assennato G, Bolt HM. Occupational exposure and urological cancer. World J Urol. 2004;21(6):382-91. doi: 10.1007/s00345-003-0377-5

25. Hatzikotoulas K, Southam L, Stefansdottir L, Boer CG, McDonald ML, Pett JP, et al. Translational genomics of osteoarthritis in 1,962,069 individuals. Nature. 2025;641(8065):1217-24. doi: 10.1038/s41586-025-08771-z

26. Hein DW, Rustan TD, Doll MA, Bucher KD, Ferguson RJ, Feng Y, et al. Acetyltransferases and susceptibility to chemicals. Toxicol Lett. 1992;64-65 Spec No:123-30. doi: 10.1016/0378-4274(92)90181-i

27. Henkel C, Styrkársdóttir U, Thorleifsson G, Stefánsdóttir L, Björnsdóttir G, Banasik K, et al. Genome-wide association meta-analysis of knee and hip osteoarthritis uncovers genetic differences between patients treated with joint replacement and patients without joint replacement. Ann Rheum Dis. 2023;82(3):384-92. doi: 10.1136/ard-2022-223199

28. Hung RJ, Boffetta P, Brennan P, Malaveille C, Hautefeuille A, Donato F, et al. GST, NAT, SULT1A1, CYP1B1 genetic polymorphisms, interactions with environmental exposures and bladder cancer risk in a high-risk population. Int J Cancer. 2004;110(4):598-604. doi: 10.1002/ijc.20157

29. Ilyinskikh NN, Ilyinskikh EN, Udartsev EY. [Age-related traits of in synovial cells of knee-joint in residents of the North of Siberia with arthritis of different etiology depending on the GSTM1 gene polymorphism of glutathion-S-transferase]. Adv Gerontol. 2017;30(6):868-872.

30. Katzer A, Löhr JF. Early loosening of total hip replacement. Dtsch Arztebl. 2003;100:A 784–90 [Heft 12]. German

31. Kempkes M, Golka K, Reich S, Reckwitz T, Bolt HM. Glutathione S-transferase GSTM1 and GSTT1 null genotypes as potential risk factors for urothelial cancer of the bladder. Arch Toxicol. 1996;71(1-2):123-6. doi: 10.1007/s002040050366

32. Kerkhof HJ, Lories RJ, Meulenbelt I, Jonsdottir I, Valdes AM, Arp P, et al. A genome-wide association study identifies an osteoarthritis susceptibility locus on chromosome 7q22. Arthritis Rheum. 2010;62(2):499-510. doi: 10.1002/art.27184

33. Klein T, Selinski S, Blaszkewicz M, Hengstler JG, Golka K. Indication for joint replacement and glutathione S-transferases M1 and T1 genotypes. J Toxicol Environ Health A. 2012;75(8-10):597-601. doi: 10.1080/15287394.2012.675313

34. Koks S, Wood DJ, Reimann E, Awiszus F, Lohmann CH, Bertrand J, et al. The genetic variations associated with time to aseptic loosening after total joint arthroplasty. J Arthroplasty. 2020;35(4):981-88. doi: 10.1016/j.arth.2019.11.004

35. Koutros S, Kiemeney LA, Pal Choudhury P, Milne RL, Lopez de Maturana E, Ye Y, et al. Genome-wide association study of bladder cancer reveals new biological and translational insights. Eur Urol. 2023;84(1):127-37. doi: 10.1016/j.eururo.2023.04.020

36. Krech E, Selinski S, Blaszkewicz M, Bürger H, Kadhum T, Hengstler JG, et al. Urinary bladder cancer risk factors in an area of former coal, iron, and steel industries in Germany. J Toxicol Environ Health A. 2017;80(7-8):430-38. doi: 10.1080/10937404.2017.1304719

37. Kulm S, Kaidi AC, Kolin D, Langhans MT, Bostrom MP, Elemento O, et al. Genetic risk factors for end-stage hip osteoarthritis treated with total hip arthroplasty: a genome-wide association study. J Arthroplasty. 2023;38(10):2149-53.e1. doi: 10.1016/j.arth.2023.05.006

38. Kulm S, Kolin DA, Langhans MT, Kaidi AC, Elemento O, Bostrom MP, et al. Characterization of genetic risk of end-stage knee osteoarthritis treated with total knee arthroplasty: a genome-wide association study. J Bone Joint Surg Am. 2022;104(20):1814-20. doi: 10.2106/JBJS.22.00364

39. Kvarda P, Mills A, Shepherd D, Schneider T. Lack of consensus on the definition of aseptic loosening in total ankle replacement: a narrative systematic review. J Clin Med. 2024;13(3):786. doi: 10.3390/jcm13030786

40. Libby WF, Berger R, Mead JF, Alexander GV, Ross JF. Replacement rates for human tissue from atmospheric radiocarbon. Science. 1964;146(3648):1170-2. doi: 10.1126/science.146.3648.1170

41. Lin X, Xu T, Wu B, Hu B, Qin M. Correlation of GSTM1 gene deletion in joint synovial fluid with the recovery of patients undergoing artificial hip replacement. Exp Ther Med. 2018;16(5):3821-26. doi: 10.3892/etm.2018.6661

42. Liu YK, Dong YH, Liang XM, Qiang S, Li ME, Sun Z, et al. Application of integrated omics in aseptic loosening of prostheses after hip replacement. Mol Med Rep. 2025;31(3):65. doi: 10.3892/mmr.2025.13430

43. Lower GM Jr, Nilsson T, Nelson CE, Wolf H, Gamsky TE, Bryan GT. N-acetyltransferase phenotype and risk in urinary bladder cancer: approaches in molecular epidemiology. Preliminary results in Sweden and Denmark. Environ Health Perspect. 1979;29:71-9. doi: 10.1289/ehp.792971

44. Lower GM Jr, Nilsson T, Nelson CE, Wolf H, Gamsky TE, Bryan GT. N-acetyltransferase phenotype and risk in urinary bladder cancer: approaches in molecular epidemiology. Preliminary results in Sweden and Denmark. Environmental Health Perspectives;1979:71-79. Int J Epidemiol. 2007;36(1):11-8. doi: 10.1093/ije/dyl290

45. MacGregor AJ, Antoniades L, Matson M, Andrew T, Spector TD. The genetic contribution to radiographic hip osteoarthritis in women: results of a classic twin study. Arthritis Rheum. 2000;43(11):2410-6. doi: 10.1002/1529-0131(200011)43:11<2410::AID-ANR6>3.0.CO;2-E

46. Magnusson K, Scurrah K, Ystrom E, Ørstavik RE, Nilsen T, Steingrímsdóttir ÓA, et al. Genetic factors contribute more to hip than knee surgery due to osteoarthritis - a population-based twin registry study of joint arthroplasty. Osteoarthritis Cartilage. 2017;25(6):878-84. doi: 10.1016/j.joca.2016.12.015

47. Möller S, Overgaard S. Probability and heritability estimates on primary osteoarthritis of the hip leading to total hip arthroplasty: a nationwide population based follow-up study in Danish twins. Arthritis Res Ther. 2015;17:336. doi: 10.1186/s13075-015-0854-4

48. Nakanishi G, Pita-Oliveira M, Bertagnolli LS, Torres-Loureiro S, Scudeler MM, Cirino HS, et al. Worldwide systematic review of GSTM1 and GSTT1 null genotypes by continent, ethnicity, and therapeutic area. OMICS. 2022;26(10):528-541. doi: 10.1089/omi.2022.0090

49. NAT Committee. Human NAT2 alleles (haplotypes) Updated May 2024. https://nat.mbg.duth.gr/Huma120n_NAT2_alleles.htm. Accessed 27 April 2025

50. NHGRI-EBI GWAS Catalog https://ftp.ebi.ac.uk/pub/databases/gwas/timeseries/current/GWAS_Catalog_diagram.pdf. 2025. Accessed 27 April 2025

51. Olden K. Commentary: From phenotype, to genotype, to gene-environment interaction and risk for complex diseases. Int J Epidemiol. 2007;36(1):18-20. doi: 10.1093/ije/dyl292

52. Omar M, Klawonn F, Brand S, Stiesch M, Krettek C, Eberhard J. Transcriptome-wide high-density microarray analysis reveals differential gene transcription in periprosthetic tissue from hips with chronic periprosthetic joint infection vs aseptic loosening. J Arthroplasty. 2017;32(1):234-40. doi: 10.1016/j.arth.2016.06.036

53. Ovsiannikov D, Selinski S, Lehmann ML, Blaszkewicz M, Moormann O, Haenel MW, et al. Polymorphic enzymes, urinary bladder cancer risk, and structural change in the local industry. J Toxicol Environ Health A. 2012;75(8-10):557-65. doi: 10.1080/15287394.2012.675308

54. Page WF, Hoaglund FT, Steinbach LS, Heath AC. Primary osteoarthritis of the hip in monozygotic and dizygotic male twins. Twin Res. 2003;6(2):147-51. doi: 10.1375/136905203321536272

55. Pan F, Hua S, Luo Y, Yin D, Ma Z. Genetic susceptibility of early aseptic loosening after total hip arthroplasty: the influence of TIMP-1 gene polymorphism on Chinese Han population. J Orthop Surg Res. 2014;9:108. doi: 10.1186/s13018-014-0108-1

56. Rothman N, Garcia-Closas M, Hein DW. Commentary: Reflections on G. M. Lower and colleagues' 1979 study associating slow acetylator phenotype with urinary bladder cancer: meta-analysis, historical refinements of the hypothesis, and lessons learned. Int J Epidemiol. 2007;36(1):23-8. doi: 10.1093/ije/dym026

57. Ruiz JD, Martínez C, Anderson K, Gross M, Lang NP, García-Martín E, et al. The differential effect of NAT2 variant alleles permits refinement in phenotype inference and identifies a very slow acetylation genotype. PLoS One. 2012;7(9):e44629. doi: 10.1371/journal.pone.0044629

58. Santos EC, Pinto AC, Klumb EM, Macedo JM. Polymorphisms in NAT2 (N-acetyltransferase 2) gene in patients with systemic lupus erythematosus. Rev Bras Reumatol Engl Ed. 2016;56(6):521-29. doi: 10.1016/j.rbre.2016.09.015

59. Sedaghati-Khayat B, Boer CG, Runhaar J, Bierma-Zeinstra SMA, Broer L, Ikram MA, et al. Risk assessment for hip and knee osteoarthritis using polygenic risk scores. Arthritis Rheumatol. 2022;74(9):1488-96. doi: 10.1002/art.42246

60. Selinski S, Blaszkewicz M, Ickstadt K, Hengstler JG, Golka K. Refinement of the prediction of N-acetyltransferase 2 (NAT2) phenotypes with respect to enzyme activity and urinary bladder cancer risk. Arch Toxicol. 2013;87(12):2129-39. doi: 10.1007/s00204-013-1157-7

61. Selinski S, Blaszkewicz M, Lehmann ML, Ovsiannikov D, Moormann O, Guballa C, et al. Genotyping NAT2 with only two SNPs (rs1041983 and rs1801280) outperforms the tagging SNP rs1495741 and is equivalent to the conventional 7-SNP NAT2 genotype. Pharmacogenet Genomics. 2011;21(10):673-8. doi: 10.1097/FPC.0b013e3283493a23

62. Skousgaard SG, Skytthe A, Möller S, Overgaard S, Brandt LP. Sex differences in risk and heritability estimates on primary knee osteoarthritis leading to total knee arthroplasty: a nationwide population based follow up study in Danish twins. Arthritis Res Ther. 2016;18:46. doi: 10.1186/s13075-016-0939-8

63. The R Foundation for Statistical Computing. Platform: x86_64-pc-linux-gnu (64-bit) Copyright (C) 2022

64. Vineis P. Commentary: First steps in molecular epidemiology: Lower et al. 1979. Int J Epidemiol. 2007;36(1):20-2. doi: 10.1093/ije/dyl291

65. von Schmiedeberg S, Fritsche E, Rönnau AC, Specker C, Golka K, Richter-Hintz D, et al. Polymorphisms of the xenobiotic-metabolizing enzymes CYP1A1 and NAT-2 in systemic sclerosis and lupus erythematosus. Adv Exp Med Biol. 1999;455:147-52. doi: 10.1007/978-1-4615-4857-7_21

66. Yu C, Hequn C, Longfei L, Long W, Zhi C, Feng Z, et al. GSTM1 and GSTT1 polymorphisms are associated with increased bladder cancer risk: Evidence from updated meta-analysis. Oncotarget. 2017;8(2):3246-58. doi: 10.18632/oncotarget.13702

 

 

File-Attachments

1. 8565_supplementary data.pdf (316,81 KB)
   Supplementary data

 

 

Figure 1: Detoxification of an epoxide by glutathione S-transferase (GST) catalysed GSH conjugation (based on Deutsche Forschungsgemeinschaft, 2008)

Figure 2: Age at death of the 57 deceased patients with artificial knee or hip replacement

Figure 3: Portions of GSTM1-negative, GSTT1-positive, and slow NAT2 genotypes in followed-up, deceased, and not contactable patients with artificial knee or hip replacement

Figure 4: Portions of GSTM1-negative, GSTT1-positive, and slow NAT2 genotypes in the initial collective with artificial hip or knee replacement, in local hospital and in external controls (Klein et al., 2012; Ovsiannikov et al., 2012; Selinski et al., 2011)

 

[*] Corresponding Author:

Klaus Golka, Leibniz Research Centre for Working Environment and Human Factors at TU Dortmund (IfADo), Dortmund, Germany; Tel: ++49 (0)231 413 411, eMail: golka@ifado.de