Cancer occurs due to mutations or damage in DNA that disrupt normal cell regulation, leading to uncontrolled cell division and growth(Reference Ledesma, Jung-Hynes and Schmit1). The acknowledged risk factors inadequately elucidate the disease’s manifestation patterns. Knowledge of the molecular causes of carcinogenesis has advanced significantly, predominantly originating from genetic mutations in most cases. Moreover, an increasing recognition exists regarding the influence of inflammation and the tissue microenvironment, especially on hormone-dependent cancers. However, the predominant genetic mutations implicated in cancer development are primarily non-hereditary. They are associated with the accumulation of somatic mutations and epigenetic changes prompted by incompletely understood environmental factors(Reference Davis, Donovan and Herberman2). In 2012, it was predicted that there were 14·1 million diagnosed cases and 8·2 million cancer-related deaths globally(Reference Torre, Bray and Siegel3). For example, in Canada, an estimated 28·2 % of deaths in 2021 were linked to cancer(Reference Brenner, Weir and Demers4). Glutathione S-transferases (GSTs), polymorphic biotransformation enzymes, augment detoxification processes and thwart DNA alterations(Reference Brenner, Weir and Demers4–Reference Irimie, Braicu and Pasca6). GST Mu 1 (GSTM1) and GST theta 1 (GSTT1) are notable members of the GST enzyme family. Polymorphic variants of GSTM1 or GSTT1 include the homozygous null genotypes for the null (deleted) allele(Reference Saitou and Ishida7). Due to a lack of enzyme activity, people with null genotypes may be more susceptible to DNA damage and mutations as well as malignancies of the bladder, lung, colon, head and neck, breast, kidney and prostate(Reference Liu, Zhang and Xie8–Reference Gaudet, Olshan and Poole11). There is increasing evidence that polymorphisms of low penetrant genes and environmental exposures such as dietary components and lifestyle may modulate the risk of cancers(Reference Klusek, Nasierowska-Guttmejer and Kowalik12,Reference Fortes, Mastroeni and Boffetta13) . Therefore, it was suggested that people with unfavourable genetic polymorphisms would be more susceptible to oxidative damage and could benefit more from the antioxidants included in food.(Reference Yuan, Ma and Liu14). This can facilitate the implementation of personalised dietary interventions in cancer prevention or treatment by identifying specific nutrients with protective effects on the human body.(Reference Irimie, Braicu and Pasca6). The role of GSTM1/GSTT1 polymorphisms was examined in dietary components and the risk of cancers. In most of these studies, GST genotypes modified relations between dietary factors and cancer susceptibility(Reference Skjelbred, Sæbø and Hjartåker15–Reference Gervasini, Jose and Carrillo18). However, only one review has investigated the association between GSTM1 and GSTT1 gene polymorphisms, dietary factors and cancer susceptibility, identifying an inverse correlation between the consumption of cruciferous vegetables and the risk of lung cancer among individuals with GSTM1 and GSTT1 null genotypes(Reference Lam, Ruczinski and Helzlsouer19). Understanding the interaction between genes and dietary patterns is crucial for tailoring individualised nutritional advice and enhancing public health initiatives(Reference Phillips20). Consequently, the objectives of this review are as follows: (1) to conduct a thorough exploration of published observational studies that have examined the relationship between nutrient intake, GSTM1 and GSTT1 gene polymorphisms and cancer risks in adult populations while summarising the key characteristics and findings of these studies; (2) to identify research gaps and underscore areas that warrant further investigation and (3) to propose recommendations regarding the consumption of various nutrients for cancer prevention.
Methods
We meticulously conducted and drafted this comprehensive systematic literature review in strict accordance with the structured framework specified in the esteemed Cochrane Handbook for Systematic Reviews of Interventions(Reference Chandler, Cumpston and Li21). Adhering to the rigorous standards of the internationally recognised Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines(Reference Page, Moher and Bossuyt22), we outlined the current review process. We took proactive steps by prospectively registering this significant literature review with PROSPERO (CRD42023452067), emphasising transparency and adherence to best practices in research methodology. Our commitment to following established protocols ensures the credibility and robustness of the findings presented in this systematic review.
Search strategy
We systematically searched three databases, MEDLINE/PubMed, Scopus and Web of Science – up to 30 April 2023, to identify relevant content for inclusion in the literature review. No restrictions were considered based on publication dates or languages. The predetermined search terms relating to diet, GSTM1 & GSTT1 gene polymorphisms and cancer risks are detailed in online Supplementary Table S1. Furthermore, an extensive review of reference lists from pertinent reviews and eligible papers was undertaken to mitigate the risk of overlooking relevant studies. To address the potential impact of publication bias, gray literature was integrated into the search results through searches conducted in institutional repositories, conference proceedings and preprint databases. Two researchers (SZM and RA) independently evaluated relevant papers’ titles and abstracts based on predetermined inclusion criteria. Both reviewers screened all identified articles independently, and their level of agreement was assessed through a pilot screening phase followed by regular meetings to resolve discrepancies and ensure consistency. Full-text articles were reviewed to identify studies that might be pertinent. Similarly, both reviewers independently conducted the full-text screening, with disagreements resolved through discussion and, if necessary, by a third reviewer (HM) reaching a consensus.
Eligibility criteria
Two reviewers, EK and SA, applied specific criteria to evaluate the titles and abstracts of each study and made selections based on the following criteria: (1) studies were required to be conducted on human subjects aged ≥ 18 years across all ethnicities and genders, (2) studies needed to be designed as observational studies encompassing case–control, nested case–control or case-cohort, cohort and cross-sectional studies, (3) studies must have investigated cancer incidence with outcomes reported as odds ratios, incidence rate ratios, relative risks or hazard ratios along with an appropriate assessment of variance, (4) the studies had to focus on various dietary factors as exposures, (5) the analysis should have examined the impact of food items and GSTM1 or GSTT1 genes polymorphisms (gene–diet interaction) on cancer incidence and (6) comprehensive statistical assessment of the interaction between genes and diet was required. Studies conducted on animal models, individuals below the age of 18, in vitro studies, review articles, case reports, letters, abstracts or conference papers, and intervention trials were excluded from the selection process.
Data extraction
Two reviewers, EK and SA, conducted separate data extraction processes autonomously. Subsequently, a third reviewer (HM) validated the extracted data. The extracted data included the following information: author’s surname, study location and publication years, gender distribution, total number of subjects including participants and cases of cancer, average age, duration of the study, dietary intake assessment method and type, cancer site, gene polymorphism type, comparative analysis, controlled variables and concise overview of effect sizes. To ensure a high level of agreement between the two reviewers, a standardised data extraction form was used, and the reviewers held regular calibration meetings to discuss and resolve any ambiguities. Discrepancies were handled through collaborative discussions between the two reviewers, with unresolved disagreements adjudicated by the third reviewer (HM).
Quality assessment
Two independent reviewers (SZM and RA) evaluated the methodological quality of the included studies using the Risk of Bias in Non-randomised Studies of Interventions framework(Reference Sterne, Hernán, McAleenan, Higgins, Thomas and Chandler23). The Cochrane guidelines were employed to assess potential biases in the included papers(Reference Sterne, Hernán and Reeves24). A third reviewer (HM) resolved any discrepancies and verified them.
Data synthesis
The meta-analysis was unfeasible due to the variability of cancer outcomes, limited studies providing data on individual dietary exposures and the diversity among studies (encompassing different dietary exposures and types of cancer). Consequently, a narrative synthesis was performed to delineate the association between GSTM1 or T1 gene polymorphisms, dietary factors and cancer susceptibility. The researchers conducted a qualitative analysis within the narrative synthesis to show the relationships between genetic variations and dietary components, focusing on dietary elements such as acrylamide, fruits, vegetables, plant-based foods, total meat consumption, red meat consumption, coffee consumption and green tea consumption. We have compiled and summarized the data from the articles in Table 1.
Table 1. Overview of the reviewed source
DCH, Diet, Cancer and Health; EPIC, European Prospective Investigation into Cancer and Nutrition; IRR, incidence rate ratio; GSTT1: Glutathione S-transferase T1; GSTM1: Glutathione S-transferase M1; HR, hazard ratio; mo, month; Q, quartiles; RR, risk ratio; ref, references; SMHS, Shanghai Men’s Health Study; SNP, single nucleotide polymorphisms; T, tertiles; WNYDS, (Reference Joseph, Moysich and Freudenheim28)Western New York Diet Study.
a Age (1), parity (n children) (2), ever use of oral contraceptives (3), ever use of postmenopausal hormone use (4), body mass index (5), current smoking (6), quantity of smoking (7), smoking intensity (8), family history of endometrial cancer (9), energy intake (10), intakes of fruits & vegetables (11), physical activity (12), red meat intake (13), total meat (14), education (15), income (16), occupation (17), alcohol consumption (18), family history of cancer (19), town (20), menopausal status (21), age at menarche (22), family history of breast cancer (23), gender (24), hair color (25), skin phototype (26), common nevi (27), sunburns episodes in childhood (28),race (29), family history of prostate cancer (30), total vegetable consumption (31), study site (32), saturated fat (33) and folate intake (34).
Results
Study identification and selection
Our study resulted in 5306 records, comprising 731 from PubMed, 3094 from Scopus and 1481 from Web of Sciences. After eliminating 1776 duplicate entries and excluding 1174 studies involving animals (see Figure 1), we reviewed the titles and the abstracts of the remaining 2356 articles. The screening stage resulted in the exclusion of 2273 papers deemed irrelevant. A thorough assessment of the full texts of the remaining 83 publications resulted in the exclusion of 62 studies for various reasons, as detailed in online Supplementary Table S3. Ultimately, 21 articles involving 25 576 participants met the eligibility criteria for the systematic review(Reference Liu, Zhang and Xie8,Reference Gaudet, Olshan and Poole11–Reference Fortes, Mastroeni and Boffetta13,Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17,Reference Ambrosone, McCann and Marshall25–Reference Zhao, Lin and Grossman39) . Overall, our investigation included fifteen case–control studies(Reference Liu, Zhang and Xie8,Reference Gaudet, Olshan and Poole11–Reference Fortes, Mastroeni and Boffetta13,Reference Ambrosone, McCann and Marshall25,Reference Carpenter, Yu and London26,Reference Joseph, Moysich and Freudenheim28–Reference Moore, Brennan and Karami31,Reference Slattery, Kampman and Samowitz33,Reference Tijhuis, Wark and Aarts35,Reference Turner, Smith and Sachse36,Reference Zhao, Lin and Grossman39,Reference Wang, Giovannucci and Hunter40) , four nested case–control studies(Reference Seow, Yuan and Sun32,Reference Steinbrecher, Rohrmann and Timofeeva34,Reference van der Hel, Peeters and Hein37,Reference Vogtmann, Xiang and Li38) and two case-cohort studies(Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17,Reference Hogervorst, van den Brandt and Godschalk27) , published between 1998(Reference Lin, Probst-Hensch and Louie29) and 2019(Reference Klusek, Nasierowska-Guttmejer and Kowalik12). The median sample size across the studies analysed was 1105, with a participant range extending from 204 to 25 576. The duration of follow-up exhibited variability across the studies, ranging from a minimum of 1 year(Reference Zhao, Lin and Grossman39) to a maximum of 15 years(Reference Liu, Zhang and Xie8). Although the majority of the studies included both male and female participants, there were three studies.(Reference Ambrosone, McCann and Marshall25,Reference van der Hel, Peeters and Hein37,Reference Hogervorst, van den Brandt and Godschalk41) that exclusively focused on women and two studies(Reference Joseph, Moysich and Freudenheim28,Reference Vogtmann, Xiang and Li38) That solely included male participants. Among the twenty-one articles reviewed, eight were conducted in the USA(Reference Gaudet, Olshan and Poole11,Reference Ambrosone, McCann and Marshall25,Reference Carpenter, Yu and London26,Reference Joseph, Moysich and Freudenheim28,Reference Lin, Probst-Hensch and Louie29,Reference Slattery, Kampman and Samowitz33,Reference Zhao, Lin and Grossman39,Reference Wang, Giovannucci and Hunter40) , eight in various European countries(Reference Klusek, Nasierowska-Guttmejer and Kowalik12,Reference Fortes, Mastroeni and Boffetta13,Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17,Reference Hogervorst, van den Brandt and Godschalk27,Reference Steinbrecher, Rohrmann and Timofeeva34–Reference van der Hel, Peeters and Hein37) , two in China(Reference Liu, Zhang and Xie8,Reference Vogtmann, Xiang and Li38) , one in Singapore(Reference Seow, Yuan and Sun32), one in Brazil(Reference Marchioni, Gattás and Curioni30) and one was a multicenter study(Reference Moore, Brennan and Karami31). The studies covered a range of cancers, with four focused on colorectal cancer(Reference Klusek, Nasierowska-Guttmejer and Kowalik12,Reference Seow, Yuan and Sun32,Reference Turner, Smith and Sachse36,Reference Vogtmann, Xiang and Li38) , three on lung cancer(Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17,Reference Carpenter, Yu and London26,Reference Wang, Giovannucci and Hunter40) , two each on the breast(Reference Ambrosone, McCann and Marshall25,Reference van der Hel, Peeters and Hein37) , head and neck(Reference Joseph, Moysich and Freudenheim28,Reference Marchioni, Gattás and Curioni30) and prostate cancers(Reference Joseph, Moysich and Freudenheim28,Reference Steinbrecher, Rohrmann and Timofeeva34) , as well as colorectal adenoma(Reference Lin, Probst-Hensch and Louie29,Reference Tijhuis, Wark and Aarts35) . A single study was identified for each of the following types of cancers: endometrial.(Reference Hogervorst, van den Brandt and Godschalk41), kidney(Reference Moore, Brennan and Karami31), colon(Reference Slattery, Kampman and Samowitz33), bladder(Reference Zhao, Lin and Grossman39), cutaneous melanoma(Reference Fortes, Mastroeni and Boffetta13) and adult leukaemia(Reference Liu, Zhang and Xie8). Vegetable consumption was the most frequently evaluated dietary exposure, with fifteen studies addressing this(Reference Gaudet, Olshan and Poole11,Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17,Reference Ambrosone, McCann and Marshall25,Reference Carpenter, Yu and London26,Reference Joseph, Moysich and Freudenheim28,Reference Lin, Probst-Hensch and Louie29,Reference Moore, Brennan and Karami31–Reference Turner, Smith and Sachse36,Reference Vogtmann, Xiang and Li38–Reference Wang, Giovannucci and Hunter40) . The remaining exposures were red meat (n 4)(Reference Klusek, Nasierowska-Guttmejer and Kowalik12,Reference Marchioni, Gattás and Curioni30,Reference Turner, Smith and Sachse36,Reference van der Hel, Peeters and Hein37) , fruit (n 3)(Reference Gaudet, Olshan and Poole11,Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17,Reference Turner, Smith and Sachse36) , total meat (n 2)(Reference Marchioni, Gattás and Curioni30,Reference van der Hel, Peeters and Hein37) , coffee (n 2)(Reference Fortes, Mastroeni and Boffetta13,Reference Slattery, Kampman and Samowitz33) , green tea (n 1)(Reference Liu, Zhang and Xie8), plant-based food (n 1)(Reference Liu, Zhang and Xie8) and dietary acrylamide (n 1)(Reference Hogervorst, van den Brandt and Godschalk41). In examining habitual dietary intake, nine studies employed validated FFQ(Reference Liu, Zhang and Xie8,Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17,Reference Ambrosone, McCann and Marshall25,Reference Hogervorst, van den Brandt and Godschalk27,Reference Joseph, Moysich and Freudenheim28,Reference Marchioni, Gattás and Curioni30,Reference Turner, Smith and Sachse36–Reference Vogtmann, Xiang and Li38) . Furthermore, other studies utilised semi-quantitative FFQ (n 5)(Reference Fortes, Mastroeni and Boffetta13,Reference Carpenter, Yu and London26,Reference Lin, Probst-Hensch and Louie29,Reference Seow, Yuan and Sun32,Reference Zhao, Lin and Grossman39) , validated semi-quantitative FFQ (n 2)(Reference Steinbrecher, Rohrmann and Timofeeva34,Reference Wang, Giovannucci and Hunter40) , FFQ (n 3)(Reference Klusek, Nasierowska-Guttmejer and Kowalik12,Reference Moore, Brennan and Karami31,Reference Tijhuis, Wark and Aarts35) , validated CARDIA diet history questionnaire (n 1)(Reference Slattery, Kampman and Samowitz33) and block FFQ (n 1)(Reference Gaudet, Olshan and Poole11). No eligible studies were identified through the review of reference lists or searches for grey literature.
Figure 1. Literature search and review flow diagram for selection of studies.
Risk of bias
As indicated in online Supplementary Table S2, it was observed that fifteen of the selected studies exhibited a notable risk of bias, while the remaining studies were determined to have a moderate risk of bias based on the assessment using the Risk of Bias in Non-randomised Studies of Interventions tool. The main factors contributing to the increased bias level were the presence of uncontrolled confounders, participant selection and exposure assessment.
Vegetable type (raw/cooked), GSTM1/GSTT1 polymorphisms
The study explored the connection between vegetable consumption (both raw and cooked), GSTM1/GSTT1 polymorphisms and the risk of cancer development. This analysis comprised a total of eleven case–control studies(Reference Gaudet, Olshan and Poole11,Reference Ambrosone, McCann and Marshall25,Reference Carpenter, Yu and London26,Reference Joseph, Moysich and Freudenheim28,Reference Lin, Probst-Hensch and Louie29,Reference Moore, Brennan and Karami31,Reference Slattery, Kampman and Samowitz33,Reference Tijhuis, Wark and Aarts35,Reference Turner, Smith and Sachse36,Reference Zhao, Lin and Grossman39,Reference Wang, Giovannucci and Hunter40) , one case cohort(Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17) and three nested case–control studies(Reference Seow, Yuan and Sun32,Reference Steinbrecher, Rohrmann and Timofeeva34,Reference Vogtmann, Xiang and Li38) . On the whole, the studies investigated various types of cancers, namely, lung (n 3)(Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17,Reference Carpenter, Yu and London26,Reference Wang, Giovannucci and Hunter40) , prostate (n 2)(Reference Joseph, Moysich and Freudenheim28,Reference Steinbrecher, Rohrmann and Timofeeva34) , breast (n 1)(Reference Ambrosone, McCann and Marshall25), head and neck (n 1)(Reference Gaudet, Olshan and Poole11), bladder (n 1)(Reference Zhao, Lin and Grossman39), colon (n 1)(Reference Slattery, Kampman and Samowitz33) and kidney (n 1)(Reference Moore, Brennan and Karami31), as well as colorectal adenoma (n 2)(Reference Lin, Probst-Hensch and Louie29,Reference Tijhuis, Wark and Aarts35) . The sample size varied from 329(Reference Chandler, Cumpston and Li21) to 3477(Reference Gaudet, Olshan and Poole11) within the studies. Cruciferous vegetables were the most frequently studied category (n 11)(Reference Ambrosone, McCann and Marshall25,Reference Carpenter, Yu and London26,Reference Joseph, Moysich and Freudenheim28,Reference Lin, Probst-Hensch and Louie29,Reference Moore, Brennan and Karami31,Reference Seow, Yuan and Sun32,Reference Turner, Smith and Sachse36,Reference Vogtmann, Xiang and Li38,Reference Wang, Giovannucci and Hunter40) , followed by various vegetables as a whole (n 3)(Reference Gaudet, Olshan and Poole11,Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17,Reference Turner, Smith and Sachse36) , glucosinolates (n 1)(Reference Steinbrecher, Rohrmann and Timofeeva34) and isothiocyanates (n 1)(Reference Zhao, Lin and Grossman39). In their hospital-based case–control study, Ambrosone et al. did not identify any interaction between cruciferous vegetable consumption and the GSTM1 and GSTT1 null/present genotypes concerning breast cancer risk(Reference Ambrosone, McCann and Marshall25). Similarly, there was no observed interaction between head and neck cancer(Reference Gaudet, Olshan and Poole11), colon cancer(Reference Slattery, Kampman and Samowitz33) and bladder cancer(Reference Zhao, Lin and Grossman39) and cruciferous vegetable intake and GSTM1 and GSTT1 null/positive polymorphisms. The case-cohort research conducted by Sørensen et al. found no notable link between the risk of lung cancer and the GSTM1 or GSTT1 genotypes, whether they were null or non-null, in relation to vegetable intake. This discovery indicates that changes in these glutathione S-transferase genes do not notably impact lung cancer risk regarding dietary practices.(Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17). Carpenter et al. conducted a case–control study within a hospital setting, indicating that high consumption of isothiocyanates among individuals with a GSTM1 homozygous deletion was associated with a significant reduction in the risk of developing lung cancer (OR = 0·52; 95 % CI = 0·31, 0·86) compared with those possessing at least one copy of GSTM1 (OR = 0·77; 95 % CI = 0·49, 1·21)(Reference Carpenter, Yu and London26). Wang et al., through a case–control study, demonstrated that amplified intake of cruciferous vegetables was linked to a considerable decrease in lung cancer risk among non-smokers with the GSTM1 present genotype (CR = 0·25; 95 % CI: 0·07, 0·9)(Reference Wang, Giovannucci and Hunter40). Lastly, Lin et al.’s research conducted in southern California revealed that the consumption of cruciferous vegetables reduced the occurrence of colorectal adenoma in individuals with the GSTM1 null genotype (OR = 0·52; CI = 0·29, 0·93)(Reference Lin, Probst-Hensch and Louie29). Moore et al. conducted a study spanning seven centers in Central and Eastern Europe, which revealed that decreased consumption of cruciferous vegetables was associated with elevated kidney cancer risk in individuals with GSTT1 null genotype (OR = 0·54; 95 % CI: 0·31, 0·93)(Reference Moore, Brennan and Karami31). Conversely, Joseph et al. found in a population-based case–control study no significant interaction between cruciferous vegetable consumption and GSTM1 and GSTT1-deletion or present genotypes in prostate cancer risk(Reference Joseph, Moysich and Freudenheim28). In a study by Steinbrecher et al. within the European Prospective Investigation into Cancer and Nutrition-Heidelberg cohort, individuals with a null GSTM1 genotype exhibited a notably decreased risk of prostate cancer with higher glucosinolates intake (OR = 0·56; 95 % CI: 0·35, 0·87)(Reference Steinbrecher, Rohrmann and Timofeeva34). Vogtmann et al. and Seow et al. conducted nested case–control studies that yielded no evidence of an interaction between cruciferous vegetable intake and GSTM1 and GSTT1 null or not null polymorphisms concerning the risk of colorectal cancer(Reference Seow, Yuan and Sun32,Reference Vogtmann, Xiang and Li38) . Conversely, Turner et al.’s case–control study in the UK demonstrated a significant reduction in the risk of colorectal cancer among individuals with the GSTT1 null genotype who consumed cruciferous vegetables or vegetables in general (OR = 0·4; 95 % CI: 0·2, 0·8)(Reference Turner, Smith and Sachse36).
Fruit consumption, GSTM1/GSTT1 polymorphisms
Two case–control and one cohort studies conducted by Gaudet et al. (2004), Turner et al. (2004) and Sørensen et al. (2007) investigated the association between fruit consumption, GST genes polymorphisms and the risk of developing colorectal cancer(Reference Gaudet, Olshan and Poole11), lung cancer(Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17) and head and neck cancer(Reference Gaudet, Olshan and Poole11). The participant numbers varied from 149(Reference Gaudet, Olshan and Poole11) to 500(Reference Turner, Smith and Sachse36). Sørensen et al. observed a significant reduction in lung cancer risk associated with fruit consumption among individuals with one (OR = 0·82; 95 % CI: 0·73, 0·93) or two (OR = 0·82; 95 % CI: 0·69, 0·97) functional alleles of GSTM1, as well as carriers of two copies of GSTT1 (OR = 0·86; 95 % CI: 0·76, 0·97)(Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17). In contrast, Gaudet et al. (2004) and Turner et al. (2004) reported that neither present nor null genotypes of both GSTM1 and GSTT1 significantly modified fruit intake and head and neck cancer and colorectal cancer risk, respectively.
Red meat consumption, GSTM1/GSTT1 polymorphisms
Several research studies have examined the correlation between the consumption of red meat, genetic variations in GST genes and the susceptibility to colorectal, breast and head and neck cancers. Specifically, Klusek et al., Marchioni et al. and Turner et al. (2004) performed case-cohort studies(Reference Klusek, Nasierowska-Guttmejer and Kowalik12,Reference Marchioni, Gattás and Curioni30,Reference Turner, Smith and Sachse36) and van der Hel et al. conducted a nested case–control study(Reference van der Hel, Peeters and Hein37). The participant sample sizes varied, ranging from 103 individuals in the study by Marchioni et al. (2011) to 500 individuals in the study by Turner et al. (2004). The findings by Klusek et al. (2019) indicated a significant increase in the risk of colorectal cancer among individuals with the GSTM1 null genotype who consumed high amounts of red meat (OR = 3·8; 95 % CI: 1·6, 9·1), while no such association was observed for individuals with the GSTT1 null genotype in the same study. The other three studies by van der Hel et al. (2004), Marchioni et al. (2011) and Turner et al. (2004) did not identify any significant interactions between red meat intake, GSTM1 null and not null polymorphisms and the risk of breast, head and neck and colorectal cancers, respectively.
Total meat consumption, GSTM1/GSTT1 polymorphisms
One nested case–control study(Reference van der Hel, Peeters and Hein37) and another case–control study(Reference Marchioni, Gattás and Curioni30) examined the correlation among total meat consumption, GST genotypes and the susceptibility to breast (n 1)(Reference van der Hel, Peeters and Hein37) and head and neck (n 1) cancers. The studies included 229 and 103 cases, respectively. However, neither of these investigations observed a significant interaction between overall meat intake and GSTM1 and GSTT1 null genotype in cancer risk.
Coffee consumption, GSTM1/GSTT1 polymorphisms
Two case–control studies by Fortes et al. (2013) with 304 cases(Reference Fortes, Mastroeni and Boffetta13) and Slattery et al. (2000) with 1579 cases(Reference Slattery, Kampman and Samowitz33) examined the association between coffee consumption, GST gene polymorphisms and the risk of cutaneous melanoma (n 1) as well as colon cancer (n 1). Fortes et al. (2013) reported a noteworthy decrease in the risk of cutaneous melanoma among individuals with the null genotype of GSTM1 and GSTT1 (homozygous deletion for GSTM1 and GSTT1) who had a high coffee consumption. Slattery et al. (2000) did not observe a significant interaction between coffee intake and GSTM1 genotypes concerning the risk of colon cancer.
Dietary Acrylamide, GSTM1/GSTT1 polymorphisms
A single case-cohort investigation involving 315 cases examined the relationship between dietary acrylamide, GST gene variations and susceptibility to endometrial cancer.(Reference Hogervorst, van den Brandt and Godschalk41). The findings from this study indicated the absence of a significant relationship between dietary acrylamide and GST null genotype effects on endometrial cancer.
Green tea consumption, GSTM1/GSTT1 polymorphisms
A case–control study by Liu et al. explored the relationship between green tea consumption, GST polymorphisms and leukaemia risk in a cohort of 442 adult cases and controls. The research revealed a notable decrease in leukaemia risk among individuals with the GSTT1 null genotype who followed particular green tea consumption patterns.(Reference Liu, Zhang and Xie8).
Plant-Based food consumption, GSTM1/GSTT1 polymorphisms
An investigation conducted in a hospital setting reported a case–control study that examined the association between the intake of plant-based foods, genetic variations in GST genes and the probability of developing head and neck cancer.(Reference Marchioni, Gattás and Curioni30). The findings of this research did not indicate a substantial relationship between the consumption of plant-based foods and specific GSTT1 and GSTM1 null polymorphisms concerning the risk of head and neck cancer.
Discussion
Diets play a crucial role as environmental determinants in the onset of cancer. While dietary requirements are unique to each individual, there exists a link between genetic diversity and the individual’s dietary needs and nutritional status(Reference Key, Bradbury and Perez-Cornago42). The components of one’s diet can impact metabolic pathways, which are modulated by genetic variations, thereby affecting nutritional status and health outcomes. An essential aim of personalised medicine for cancer patients is to establish tailored nutritional recommendations based on individual genetic, metabolomics and microbiome profiling. GST genetic variants are implicated in aberrant signalling pathways that can contribute to the development of malignancies. However, as critical environmental elements, dietary factors can influence metabolic pathways(Reference Allocati, Masulli and Di Ilio43,Reference Mena, Ortega and Estrela44) . The primary objective of this investigation was to examine the interactions between nutrients and genes to propose tailored nutritional approaches for the prevention and management of cancer, focusing on GSTM1/GSTT1 polymorphisms. The study examined the interplay between genetic determinants and dietary factors influencing cancer susceptibility. Through a systematic review, we explored the association between GST genetic polymorphisms and diet across various cancer types, marking the first comprehensive analysis of its kind. Our findings indicate a correlation between GST genetic polymorphisms, nutrients and their distinct impacts on cancer development and vulnerability. Previous research on the GSTM1/GSTT1 polymorphisms in different ethics and cancers with various diets demonstrated controversial results.
In this study, a total of twenty-one papers employing an observational study design were included. The relationship between GST polymorphisms and nutrition has been investigated in various forms of cancer, given their significant role in cancer prevention and treatment.(Reference McIlwain, Townsend and Tew45). The interaction between genes and diet highlighted in these studies is essential for enhancing patients’ life expectancy. Therefore, it is recommended that further research in personalised medicine and tailored dietary interventions for cancer patients be conducted to leverage these findings for better patient outcomes in terms of survival and life expectancy(Reference Krzyszczyk, Acevedo and Davidoff46). The superfamily of GSTs encompasses five primary classes: Alpha, mu, pi, theta, and zeta. The enzyme GSTT1 is crucial in conjugating reduced glutathione with a wide array of electrophilic and hydrophobic compounds.(Reference Townsend and Tew47). The genes GSTT1 and GSTT2, classified under the theta class, have been linked to human cancer development. Similarly, the enzymes GSTM1, which are also part of the GST superfamily, exhibit high levels of polymorphism and are located within a gene cluster on chromosome 1p13·3(Reference Nebert and Vasiliou48). These genetically diverse forms can influence an individual’s susceptibility to toxins and carcinogens and the efficacy and toxicity of certain medications.(Reference Autrup49). The presence of GSTM1 null mutations has been associated with an increased occurrence of malignancies, possibly due to heightened vulnerability to environmental toxins and carcinogens(Reference Landi50). Perinously a meta-analysis was carried out to examine the correlation between cancer susceptibility and GSTs in individuals who smoke and consume alcohol. It was determined that the GSTM1-null genotype exhibits a noteworthy association with escalated cancer susceptibility, both independently and when combined with smoking. The GSTT1-null genotype is likewise significantly linked to heightened cancer risks, especially when combined with alcohol consumption(Reference Hu, Li and Huang51).Our qualitative analysis provides evidence suggesting that the interplay between GSTM1/GSTT1 polymorphisms, along with the consumption of fruits, red or processed meat, coffee, dietary acrylamide, green tea, and plant-based foods, can have a debatable impact on the predisposition to different types of cancer.
Various studies have investigated the relationship between vegetable consumption and the likelihood of developing different types of cancer. The results have been conflicting. Ambrosone et al., Sørensen et al., Vogtmann et al., and Seow et al. have collectively suggested that there is no significant link between vegetable consumption and the presence of GSTM1/GSTT1 polymorphisms in cancer risk.(Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17,Reference Ambrosone, McCann and Marshall25,Reference Vogtmann, Xiang and Li38) . However, Seow et al. found that a high intake of isothiocyanates from cruciferous vegetables is negatively correlated with colorectal cancer risk. This connection is particularly notable in individuals with GSTM1 and T1 null genotypes due to their slower processing and elimination of these compounds compared to others.(Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17). Carpenter et al. (2009) found that individuals with GSTM1 homozygous deletions who consumed isothiocyanates experienced a reduction in lung cancer risk. Additionally, Lin et al. (1998) suggested that cruciferous vegetables may play a role in reducing colorectal adenoma incidence in individuals with the GSTM1 null genotype. This discrepancy in findings is likely attributed to factors such as population diversity and biases related to recall and participant selection(Reference Shu, Yu and Jin52). Moreover, research by Moore et al. (2007) indicated that individuals carrying the GSTT1 null genotype may benefit from increasing their consumption of cruciferous vegetables to reduce the risk of kidney cancer. Turner et al. (2004) demonstrated that incorporating cruciferous vegetables into the diet could lower the risk of colorectal cancer. Wang et al. (2004) reported that a high intake of cruciferous vegetables was associated with decreased lung cancer risk among non-smokers. Sørensen et al. (2007) found that fruits and vegetables can mitigate lung cancer risk, specifically in individuals carrying at least one functional GSTM1 allele. While the relationship between vegetable consumption and cancer risk is complex and still not fully understood, research suggests that consuming cruciferous vegetables may offer some protection against certain types of cancer, particularly in individuals with specific genetic polymorphisms. A study by Sørensen et al. revealed subtle indications of a difference in the probability of lung cancer among those with one functional allele of either GSTT1 or GSTM1 compared to individuals with two alleles. This research suggests a potential correlation between higher consumption of fruits and vegetables and GSTM1 polymorphisms, which refer to variations in the DNA sequence that can affect the function of the GSTM1 gene. According to the study, individuals with a genetic predisposition to lung cancer may reduce their risk by incorporating more fruits and vegetables into their diet, as these foods may help counteract the adverse effects of the GSTM1 polymorphisms. This underscores the importance of maintaining a healthy diet in mitigating the risk of lung cancer, particularly for those who are genetically susceptible to the disease(Reference Sørensen, Raaschou-Nielsen and Brasch-Andersen17). Moreover, Ying Gao et al. worked on the meta-analysis and indicated a positive association between the combined effects of GSTM1/GSTT1 polymorphisms and lung cancer risk(Reference Gaudet, Olshan and Poole11). In certain epidemiological studies, consuming well-cooked meat has been associated with an increased risk of breast cancer(Reference Lo, Park and Sinha53). Research has shown that women with the GSTM1 null genotype are at a higher risk of developing breast cancer. The consumption of red meat does not significantly increase the risk of breast cancer in women, regardless of whether they have the GSTM1 genetic makeup or not. Moreover, there is no clear correlation between the amount of red meat consumed and the risk of developing breast cancer. The presence of the GSTM1 genetic makeup does not affect the relationship between red meat consumption and breast cancer risk(Reference van der Hel, Peeters and Hein37). The meta-analysis in 2016 by Zhiwang Song investigated the relationship between genetic polymorphisms of glutathione S-transferase (GST) M1, GSTT1, and GSTP1 and the risk of breast cancer. It was suggested that GSTT1 null genotype and GSTP1 Ile105Val polymorphism may be potential genetic risk factors for breast cancer(Reference Song, Shao and Feng54). A meta-analysis showed a correlation between the glutathione S-transferase M1 (GSTM1) polymorphism suggesting that individuals with the GSTM1 null genotype may be more susceptible to developing CRC(Reference Huang, Zeng and Zhao55). According to the studies conducted by van der Hel et al., Marchioni et al., and Turner et al., there is no substantial evidence to suggest a significant correlation between the consumption of red meat and the GST genotypes concerning breast, head, neck, and colorectal cancers. These studies have analysed various factors such as dietary habits, genetic makeup, and cancer incidence rates, and the findings suggest that red meat intake may not be a significant risk factor for the development of these types of cancer. However, it is essential to note that other factors such as lifestyle choices, environmental exposures, and genetic predispositions can also play a role in the development of cancer, and further research is needed to fully understand the relationship between red meat intake and cancer risk(Reference Marchioni, Gattás and Curioni30,Reference Turner, Smith and Sachse36,Reference van der Hel, Peeters and Hein37) . Van der Hel et al. observed robust and statistically significant effects of red meat consumption only in postmenopausal women with the GSTM1 genotype present. Klusek et al. also noted that high consumption of red meat significantly interacts with carriers of the GSTM1 null genotype, advising them to avoid high red meat intake to reduce susceptibility to colorectal cancer, which is not significant for the GSTT1 null genotype(Reference Klusek, Nasierowska-Guttmejer and Kowalik12).
Two research studies were undertaken to examine whether there is a link between total meat consumption and GSTM1/GSTT1 polymorphisms in cancer risk. The results of both studies demonstrated no significant correlation between cancer risk, total meat intake, and GST polymorphisms. Van der Hel, et al.’s study, found that individuals with a GSTM1 null genotype faced a higher risk of breast cancer due to insufficient detoxification of carcinogenic substances. While there was no statistical significance, a minor increase in the risk of breast cancer development was observed in those who consumed red meat(Reference van der Hel, Peeters and Hein37).
An investigation was conducted to find the connection between coffee consumption and the presence of GSTM1/GSTT1 polymorphisms. A significant reduction in the risk of developing cutaneous melanoma was observed in individuals with the GSTM1 and GSTT1 genotypes who drank large amounts of coffee. Cutaneous melanoma is a skin cancer that can be caused by prolonged exposure to UV radiation. Nonetheless, no meaningful connection was discovered between the GSTM1 genotype, coffee intake, and colon cancer. The study by Fortes et al. demonstrated that elevated coffee intake can lower the likelihood of cutaneous melanoma among individuals with the GSTT1 null genotype. The GSTT1 null genotype refers to the absence of the GSTT1 gene in an individual’s DNA. This gene encodes for an enzyme that helps the body to detoxify harmful substances, including carcinogens. Thus, people with the GSTT1 null genotype are more vulnerable to the damaging effects of UV radiation, which may result in the development of cutaneous melanoma. This investigation represents the first documented case of elevated coffee consumption reducing the risk of cutaneous melanoma in individuals with the GSTM1 and GSTT1 null genotypes. The study sheds light on the potential protective effects of coffee against skin cancer, especially among individuals with specific genetic profiles. More research is needed to confirm and explore the findings(Reference Fortes, Mastroeni and Boffetta13).
The study conducted by Hogervorst et al. in 2016 investigated the relationship between acrylamide exposure, genetic susceptibility, and the risk of developing endometrial cancer. individuals with the GSTM1 null genotype might convert acrylamide into glycinamide at a higher rate. Glycinamide is considered less toxic than acrylamide but still poses risks. This conversion process indicates that individuals with functional alleles may have a more effective detoxification pathway, potentially reducing their risk of developing endometrial cancer upon exposure to acrylamide. As to the findings, people with the GSTM1 null genotype tend to convert acrylamide into glycinamide at a higher rate. This suggests that exposure to acrylamide increases the risk of endometrial cancer in carrier women with at least one copy of the GSTM1 gene. However, the relationship between GSTM1 and GSTT1 followed a consistent interaction pattern, regardless of the underlying biological mechanisms that governed their interactions with GSTs. These results emphasise the significance of comprehending the genetic components that lead to cancer development and could guide future approaches to prevention and therapy.(Reference Hogervorst, van den Brandt and Godschalk41). Furthermore, Xiuxiu Yin et al. 2017 conducted a meta-analysis to investigate the relationship between genetic polymorphisms in glutathione S-transferases M1 (GSTM1) and T1 (GSTT1) genes and the risk of endometrial cancer (EC) and found no significant association between GSTM1 null genotype and an increased risk of endometrial cancer (EC)(Reference Yin and Chen56).
The relationship between GST genotypes, plant-based diets and head and neck cancer was investigated. Nevertheless, according to the study’s findings, there was no discernible correlation between these parameters. Nonetheless, Liu and colleagues conducted another study highlighting a connection between green tea consumption, the GSTT1 null genotype and the risk of developing adult leukaemia. However, in the case of the GSTM1 null genotype, no such correlation was seen. The findings are supported by the functional significance of GSTT1 and GSTM1 genes in the human body. These genes play a critical role in metabolising various environmental carcinogens, which may reduce DNA damage caused by reactive metabolites to hematopoietic stem and progenitor cells. The lack of GSTT1 dramatically raises vulnerability to diepoxybutane-induced sister chromatid exchanges, according to in vitro studies, while the GSTM1 gene does not impact the observed DNA damage effects. These results imply that consuming green tea may be a preventative measure against adult leukaemia, particularly for individuals with the GSTT1 null genotype. More research is needed to understand further the relationship between these factors and the mechanisms that underlie their effects.(Reference Liu, Zhang and Xie8).
The study conducted by Marchioni and colleagues aimed to investigate the relationship between the intake of plant-based foods, GST gene polymorphisms and the risk of head and neck cancer development. According to their study’s results, no statistically significant association exists between the prevalence of head and neck cancer, the consumption of dietary plant-based foods and GST genotypes. These findings suggest that dietary habits and GST genotypes do not play a pivotal role in the development of head and neck cancer. The results of this research could potentially inform future studies on the topic and help to develop more effective prevention and treatment strategies. In other words, the study did not find any evidence to suggest that individuals with specific GST gene polymorphisms who consume more plant-based foods are at a higher or lower risk of developing head and neck cancer. Despite the lack of significant findings, the study provides valuable insights into the potential impact of plant-based diets on developing head and neck cancer. It highlights the importance of further research in this area(Reference Marchioni, Gattás and Curioni30).
Strengths and limitations
This study represents a significant milestone in the field of cancer research, as it undertakes a comprehensive review of the relationship between various dietary patterns and the presence of GSTM1/GSTT1 polymorphisms in cancer risk. Through this study, we aim to gain a deeper understanding of the complex interplay between genetic factors and dietary habits and how they can impact cancer risk. The findings of this study will not only shed light on the underlying mechanisms of cancer development but also provide valuable insights into the prevention and management of this devastating disease. As a result, it provides a comprehensive analysis of the existing primary research literature in this area. Observational studies have indicated that reducing red meat intake and increasing consumption of vegetables, coffee, fruits and green tea may lower cancer risk, particularly among individuals with the GSTM1/GSTT1 polymorphisms. There is currently a lack of substantial evidence to establish a significant association between the consumption of total meat, plant-based foods, dietary acrylamide and the risk of developing cancer. While some studies have suggested a potential link, more extensive research is required to confirm reliable conclusions. Therefore, it is always advisable to maintain a balanced and varied diet that includes a healthy mix of both plant-based and animal-based foods. The present study exhibits similar shortcomings to prior systematic reviews, such as a constrained selection of eligible papers for specific associations being studied, a notable degree of heterogeneity in diverse study characteristics and a scarcity of studies carrying out repeated assessments of dietary intake and polymorphism.
Future research direction
Further investigation is crucial to determine how the GSTM1 and GSTT1 variants affect cancer risk across various populations and ethnic groups. To obtain a thorough comprehension of GST gene variations that are involved in cancer, it is essential to explore how they interact with environmental factors like dietary habits and exposure to pollutants. Evaluating the long-term implications of GST gene variants on the formation and progression of cancer requires conducting longitudinal studies. Future research will benefit from larger sample sizes and consistent methodology to enhance statistical power and the reliability of findings. To advance personalised medicine for cancer prevention and treatment, functional research must be conducted to elucidate how variations in GST genes and dietary factors influence cancer risks. We can enhance our understanding of cancer development by identifying genetic variations in GST and other related factors associated with cancer onset.
Conclusion
Our systematic review found insufficient evidence to support the modifying influences of GSTM1/GSTT1 polymorphisms on the relationship between dietary factors and cancer susceptibility. Limited research exists regarding the relationship between cancer risk, dietary patterns and genetic variations. Given that metabolic genotypes can impact risk-modifying factors, controlled trials involving various populations are warranted to confirm the gene–diet interaction and enhance comprehension of the disease’s aetiology. Studies suggest that high consumption of red meat may increase the risk of colorectal cancer in people with the GSTM1 null genotype. Consuming fruits, green tea, coffee and cruciferous vegetables may reduce the incidence of adult leukaemia, cutaneous melanoma and lung cancer in people with GSTM1/GSTT1 polymorphisms. Nevertheless, the existing studies on the interplay between dietary factors, GSTM1 and T1 gene polymorphisms and cancer risk are currently limited. Comprehensive studies exploring the correlation of various dietary patterns and genetic variations with different types of cancer are scarce. The presence of polymorphisms in genes responsible for nutrient metabolism can significantly impact how nutrition affects cancer development. Modifying dietary recommendations for cancer prevention and control based on individual genotypes is imperative. Advancements in nutritional genomics can pave the way for personalised preventive and therapeutic strategies for cancer patients.
Supplementary material
For supplementary material/s referred to in this article, please visit https://doi.org/10.1017/S0007114525000327
Acknowledgments
This research was supported by the Tehran University of Medical Sciences and Health Services (Grant number: 67506).
The entire team of authors collaborated in conceptualising and designing the review. R. A. K. and S. Z-M. conducted the literature search, while E. K., S. A. and S. Z-M. were responsible for data collection. E. K., S. A. and S. Z-M. drafted the initial version of the manuscript. M. Y., H. M. and S. R. B-A. contributed significantly to the critical revision and interpretation of the manuscript and the data. All authors carefully reviewed and confirmed the final version of the manuscript.
The authors collectively confirmed that they have no conflicts of interest to declare.
This article does not involve any studies conducted on human participants or animals by any of the authors.
