Association Between Genes in the Nuclear Factor E2–Related Factor 2 Antioxidative Response Elements Pathway and Cancer-Related Fatigue in Women With Early-Stage Breast Cancer

Tara S. Davis

Theresa A. Koleck

Margaret Q. Rosenzweig

Christine Miaskowski

Kirk I. Erickson

Susan M. Sereika

Catherine M. Bender

Yvette P. Conley

antioxidants, breast neoplasms, DNA methylation, fatigue, polymorphism
ONF 2024, 51(4), 404-416. DOI: 10.1188/24.ONF.404-416

Objectives: To explore genes in the nuclear factor E2–related factor 2 antioxidative response elements (Nrf2-ARE) signaling pathway using a multiomics approach for associations with variability of cancer-related fatigue (CRF) in postmenopausal women with early-stage hormone receptor–positive breast cancer.

Sample & Setting: Postmenopausal women (N = 116) with early-stage hormone receptor–positive breast cancer were recruited from western Pennsylvania.

Methods & Variables: Candidate genes from the Nrf2-ARE pathway were investigated for associations with CRF occurrence and severity. Associations were evaluated using logistic regression for occurrence and linear regression for severity.

Results: The rs2706110 TT genotype in NFE2L2 was associated with a 3.5-fold increase in odds of CRF occurrence. The cytosine-phosphate-guanine (CpG) site cg22820568 in PRDX1 was associated with CRF occurrence and severity.

Implications for Nursing: Biomarkers based on Nrf2-ARE genes may help to identify women at increased risk for more severe CRF and to develop targeted interventions.

Jump to a section

    More than 3.8 million women in the United States are breast cancer survivors (Giaquinto et al., 2022). Cancer-related fatigue (CRF) is among the most common symptoms associated with breast cancer and its treatments (Berger et al., 2012; Mao et al., 2018; Sanft et al., 2023). One in four women with breast cancer will experience CRF at some point (Maass et al., 2021). The National Comprehensive Cancer Network defines CRF as “a distressing, persistent, subjective sense of physical, emotional, and/or cognitive tiredness or exhaustion related to cancer or cancer treatment that is not proportional to recent activity and interferes with usual functioning” (Sanft et al., 2023, p. 797). CRF is different from general fatigue because it is not likely to improve with adequate rest, may be more severe, and is more likely to cause interference in daily activities (Berger et al., 2015). Patients with breast tumors report among the highest prevalence of moderate to severe CRF (Kang et al., 2023). A growing body of evidence suggests that early identification and treatment of CRF across the trajectory of cancer treatment may decrease its occurrence and severity (Bower et al., 2019). The burden associated with CRF includes interference with activity and physical functioning, inability to work (Schmidt et al., 2019) and socialize, greater financial stress, and increased healthcare utilization (Behringer et al., 2016). These effects result in decrements in quality of life (Bower et al., 2000). In addition, CRF is associated with worse disease trajectories and reductions in survival (Groenvold et al., 2007).

    Despite the high prevalence of CRF and its debilitating effects, the mechanisms that underlie CRF are not fully understood. A multitude of factors, including genomic factors (e.g., DNA methylation variation [Xiao et al., 2021], single nucleotide polymorphisms [SNPs] [Kober et al., 2016]), are likely to be involved in the occurrence and severity of CRF. Inflammatory mechanisms are among the most studied mechanisms for CRF (Saligan et al., 2015; Wright et al., 2017). Of note, oxidative stress regulation, one of many inflammatory mechanisms, plays an important role in the development and progression of different cancers, including breast cancer (Hayes et al., 2020; Lee et al., 2017).

    One potential pathway involved in oxidative stress regulation that may help to explain the complexity of CRF in women with breast cancer is the nuclear factor E2–related factor 2 antioxidative response elements (Nrf2-ARE) signaling pathway. The Nrf2-ARE pathway maintains redox homeostasis through the regulation of intracellular antioxidant response and mediates anti-inflammatory processes for various conditions, such as neurodegenerative diseases (George et al., 2022) and aging (Liguori et al., 2018). In the context of cancer, Nrf2-ARE has tumor-suppressing and tumor-promoting properties (Rojo de la Vega et al., 2018). In general, when reactive oxygen species levels exceed antioxidant capacity in the cell, they cause destruction of a number of cell components (e.g., lipids, proteins, DNA) and subsequently lead to apoptosis or necrosis (Rojo de la Vega et al., 2018). However, a more aggressive cellular response can occur that initiates a significantly higher level of inflammation (Rojo de la Vega et al., 2018), which may result in CRF. On the other hand, in some situations, Nrf2-ARE protects cells from destruction (Schmidlin et al., 2021); this protective role may prevent the development of cancer and may decrease the occurrence and/or severity of CRF. However, pathways involved in antioxidative stress are not well studied and warrant additional investigation (García-González et al., 2023). To the current research team’s understanding, only three studies, including one done by the current team, have investigated associations between oxidative stress mechanisms and CRF (Davis et al., 2023; Dickinson et al., 2020; Repka & Hayward, 2018). It is hypothesized that antioxidative stress cellular responses to cellular death and tumor growth or suppression may contribute to CRF. Therefore, the aim of this study was to use a multiomics approach to explore the role of genes in the Nrf2-ARE pathway in the occurrence and severity of CRF in postmenopausal women with early-stage hormone receptor–positive breast cancer. Hormone receptor–positive breast cancer cells have receptor proteins for estrogen, progesterone, or both.

    Methods and Variables

    Study Design

    This exploratory study employed a cross-sectional observational design that capitalized on existing data and biospecimens from postmenopausal women with early-stage breast cancer who participated in the Exercise Program in Cancer and Cognition (Gentry et al., 2018) and the Epigenomics of Neurocognitive Function in Breast Cancer studies (together referred to as “the parent study” in this article). The parent study’s aim was to investigate the effects of moderate intensity aerobic exercise on cognitive function in women with early-stage postmenopausal breast cancer within the first six months of an aromatase inhibitor treatment. For this study, data from the prerandomization time point were used. Extensive demographic, clinical, and CRF-related data, as well as stored biosamples and whole-genome methylation data, were available from the parent study. In the current study, SNP genotypes were generated and used with DNA methylation data to investigate targeted genes from the Nrf2-ARE pathway.

    Sample and Setting

    A total of 116 women were included in this analysis. Women were included if they had hormone receptor–positive breast cancer, were aged younger than 80 years, were diagnosed with stage 0, I, II, or IIIa breast cancer, and had available blood samples. Women were excluded if they had prior cancer diagnoses (except for some skin cancers), metastases, self-reported hospitalization for psychiatric illness within the past two years, history of neurologic illness, breast cancer surgery complications, reconstructive surgery within the study period, or history of chronic fatigue. This study was approved by the University of Pittsburgh Institutional Review Board.

    Patients were recruited during their initial postoperative appointment, which is typically two to three weeks following breast surgery and three to four weeks before beginning radiation therapy. Recruitment took place at the Comprehensive Breast Care Program of the University of Pittsburgh Cancer Institute and University of Pittsburgh Medical Center cancer treatment centers from October 2016 to September 2021. All patients provided written informed consent (Gentry et al., 2018).

    CRF Measures

    The Patient-Reported Outcomes Measurement Infor-mation System (PROMIS) Fatigue Short Form 8a is a valid and reliable self-report measure of fatigue in individuals with chronic disease including cancer (Cella et al., 2016). This measure is an eight-item, five-point Likert-type scale with responses ranging from 1 (not at all) to 5 (very much). T scores are calculated and standardized to a mean of 50 and an SD of 10 in reference to the U.S. general population. Higher T scores indicate greater fatigue. Two different CRF outcomes were evaluated, namely CRF occurrence (PROMIS Fatigue T score of 50 or greater) and CRF severity as a continuous score (Cella et al., 2014).

    Additional measures that were evaluated as potential covariates included age, employment status, marital or partner status, parental status, State and National Area Deprivation Indices (state index measured by decile, national index measured by percentile) (Kind & Buckingham, 2018), body mass index (BMI), estimated maximal oxygen consumption (VO2 max), radiation therapy, chemotherapy, pain severity (Brief Pain Inventory–Short Form, range = 0–10), pain interference (Brief Pain Inventory–Short Form, range = 0–10) (Cleeland & Ryan, 1994), anxiety (PROMIS Emotional Distress–Anxiety Short Form, T score range = 35–75) (Pilkonis et al., 2011), depression (Beck Depression Inventory–II, range = 0–63) (Beck et al., 1996), daytime sleepiness (Epworth Sleepiness Scale, range = 0–24) (Johns, 1991), and self-reported cognitive function (Patient Assessment of Own Functioning Inventory, range = 0–155) (Chelune et al., 1986). Because of the relatively small sample size, all covariates that were significantly correlated with CRF could not be included in the regression models. Therefore, a priori and data-driven covariate identification methods were used (Greenland & Pearce, 2015).

    Descriptive Statistical Analysis

    Analyses were completed using R statistical software, version 4.0.4 (R Core Team, 2021), and PLINK, version 1.9 (Purcell et al., 2007). Univariate statistics and visualization of dependent, independent, and potential covariate variables were performed based on level of measurement for each variable. Additional variables known to influence CRF were identified using a priori and data-driven bivariate analysis methods. Continuous covariates were analyzed for associations with CRF using Pearson correlation. Nominal variables were evaluated using the point biserial correlation test. Bivariate comparison analyses, consisting of two-sample t tests, chi-square tests of independence, and Fisher’s exact tests, were performed to evaluate equality of group means for continuous variables and group proportions for categorical variables. Using PLINK, allele frequencies were calculated and Hardy–Weinberg Equilibrium was assessed for each SNP. Additional descriptive and quality control analyses are described in the following sections.

    DNA Methylation Data Collection and Statistical Analysis

    Genes from the Nrf2-ARE pathway involved in the reduction of oxidative damage and/or breast cancer were chosen using Ingenuity Pathway Analysis, version 21.0.1. Transporter genes for exogenous chemicals that were not known to be associated with aromatase inhibitor drugs were excluded. Whole-methylome data were generated using the Infinium MethylationEPIC, version 1.0, BeadChip, and were put through a substantial quality control pathway. Quality control procedures included calculating beta values from the ratio of the intensity of methylated probes to the intensity from total probes and correcting for background and dye bias. Beta values were normalized using the preprocessFunnorm function from the minfi package, version 1.40.0, in Bioconductor. Normalization was verified with cluster analysis on technical replicates. Probes were removed if they were cross-reactive, located on the Y chromosome, or located near a known SNP. Cell type heterogeneity was not corrected for in this study. Cytosine-phosphate-guanine (CpG) site data were extracted for each candidate gene within a +/- 2,000 base pair (cis) window to allow for evaluation of proximal regulatory regions. M values were used because of their statistically valid properties for differential analysis of methylation levels (Xie et al., 2019) in the context of the outcome of interest (i.e., CRF).

    Regression analyses were used to identify methylated regions associated with CRF occurrence (logistic regression) and severity (linear regression). Regression models were adjusted for covariates identified in this study. Unadjusted and adjusted p values were calculated using the Benjamini–Hochberg procedure (Benjamini & Hochberg, 1995). A false discovery rate of 0.05 was applied to control for multiple testing.

    SNP Data Collection and Statistical Analysis

    From the genes identified in the Nrf2-ARE pathway, functional polymorphisms were selected using the University of California, Santa Cruz, Genomics Institute (n.d.), Genome Browser and through a literature search using the terms functional polymorphism AND the full name of each gene. SNPs known to influence fatigue and/or breast cancer were prioritized.

    DNA was extracted from banked blood samples taken from the patients at enrollment. A simple salting out procedure was used to extract DNA. Allelic discrimination was performed using the Applied Biosystems QuantStudio™ 3 Real-Time PCR System and TaqMan™ allelic discrimination assays. To ensure rigor of the data, genotypes were blindly double-called, compared, reconciled, or rerun.

    The quality control pathway for the genotype data was completed using the PLINK toolset (Purcell et al., 2007) to identify per-marker missingness (greater than 5%), per-individual missingness (greater than 5%), and per-marker minor allele frequency (less than 5%), as well as to calculate per-marker deviation from Hardy–Weinberg Equilibrium. Genetic associations were tested under an additive model for which SNPs were treated as ordinal variables based on genotype (SNP genotype was coded as 0 for homozygous wild type, 1 for heterozygous, and 2 for homozygous minor allele).

    To investigate the effect of the prioritized SNPs on CRF outcomes, regression models were employed. The CRF occurrence and CRF severity outcomes were evaluated using logistic regression and linear regression modeling, respectively. Models were adjusted for covariates and a Benjamini–Hochberg procedure, (Benjamini & Hochberg, 1995), and a false discovery rate correction of 0.05 was applied to control for multiple testing.

    Genetically Driven Methylation Variation Measurement and Statistical Analysis

    To understand whether genetic variation influenced DNA methylation variation, the associations between genetic and local methylation variations (i.e., methylation quantitative trait loci [meQTLs]) were analyzed. The meQTL analyses were limited to significant SNPs that were associated with CRF occurrence and severity and located within a +/- 2,000 base pair (cis) window of each CpG site. Unadjusted linear regression modeling was used to examine the associations between the identified SNPs and DNA methylation M value levels. A significant meQTL was identified if a CpG site was regulated by at least one SNP (Gaunt et al., 2016). Using the Benjamini–Hochberg procedure (Benjamini & Hochberg, 1995), a false discovery rate correction of 0.05 was applied to control for multiple testing. The mQTLdb database was used to interpret significant meQTLs (Gaunt et al., 2016).


    Patient Characteristics

    As shown in Table 1, the 116 participants had a mean age of 62.4 years (SD = 7.1); 90% self-reported their race as White; 98% self-reported their ethnicity as non-Hispanic/Latina; 61% were currently employed; 68% were married and/or partnered; and 87% were parents. Participants had a mean BMI of 29.8 kg/m2 (SD = 6.7) and a mean VO2 max of 21.9 ml/kg/minute (SD = 4.6). Most patients had stage I hormone receptor–positive breast cancer (60%), and 84% had received radiation therapy prior to surgery. Because the parent study was conducted during the COVID-19 pandemic, relaxation of the initial study inclusion and exclusion criteria was allowed, resulting in about 16% of the patients having received chemotherapy and less than 2% having received aromatase inhibitor therapy prior to enrollment. Additional analyses were performed to compare the entire cohort to those who did not receive chemotherapy or aromatase inhibitor therapy prior to study entry (n = 96). These results are summarized in Supplemental Tables 1–3 online, with minor differences noted.


    Additional Measures Evaluated as Potential Covariates

    A summary of the correlated and a priori variables that were included in regression models is presented in Supplemental Table 4 online. In the bivariate analyses, although estimated VO2 max (p = 0.02), BMI (p = 0.04), pain interference (p ≤ 0.001), pain severity (p = 0.02), anxiety (p ≤ 0.001), depression (p ≤ 0.001), daytime sleepiness (p ≤ 0.001), and self-reported cognitive function (p < 0.001) were positively correlated with CRF severity, being married or partnered (p = 0.01) was negatively correlated with CRF severity. No correlations were found between age, employment status, parental status, radiation therapy, chemotherapy, or State or National Area Deprivation Indices and CRF severity. However, because it is well established that age influences the regulation of oxidative stress, age was included in all the regression models. Because pain severity was correlated with pain interference, pain severity was not included in the models. Although self-reported cognitive function was correlated with CRF severity, because of the small sample size, this characteristic was not included in the analyses. The characteristics used as covariates in the adjusted analyses were age, BMI, VO2 max, pain interference, anxiety, depression, and daytime sleepiness.

    Genetic Extraction Information and Quality Control Procedures

    A total of 142 CpG sites and 18 SNPs were evaluated across seven candidate genes and are summarized in Table 2. After applying a variety of quality control procedures, all 142 CpG sites and 16 SNPs from 116 patients were analyzed.


    Following identification of per-marker missingness of greater than 5%, two SNPs were excluded. Although one SNP deviated from Hardy–Weinberg Equilibrium (p < 0.001), this SNP was not eliminated from analyses on the basis that this study’s cohort does not represent the general population because it is enriched for breast cancer. Therefore, a total of 16 SNPs from 116 women who provided blood samples passed all the genotype quality control procedures and were analyzed (see Supplemental Figure 1 online).


    DNA Methylation Analysis Results

    Three CpG sites were significantly associated with CRF occurrence (see Table 3). Eight CpG sites were significantly associated with CRF severity (see Table 4). All of the DNA methylation results for CRF occurrence and severity are presented in Supplemental Tables 5 and 6 online, respectively.


    SNP Analysis Results

    Although no associations were found with CRF severity, significant associations were found for two SNPs (rs2706110 in nuclear factor erythroid 2–like 2 [NFE2L2] and rs10432782 in superoxide dismutase 1 [SOD1]) with CRF occurrence (see Table 5). All of the SNP results for CRF occurrence and severity are presented in Supplemental Tables 7 and 8 online, respectively.


    meQTL Analysis Results

    The two SNPs that were significantly associated with CRF occurrence (rs2706110 in NFE2L2 and rs10432782 in SOD1) were tested for local CpG sites. Local CpG sites were any CpG sites within 2,000 base pairs up- and downstream from the gene. NFE2L2 rs2706110 was identified to be an meQTL for the following five local NFE2L2 CpG sites: cg01417537, cg09248506, cg08334199, cg05303734, and cg03988329 (see Table 6). No meQTLs were found for SOD1 rs10432782.



    To the knowledge of this research team, this study is the first to evaluate for associations between Nrf2-ARE genes and CRF in women with early-stage postmenopausal hormone receptor–positive breast cancer. The most notable finding is the association between CRF and NFE2L2. In particular, rs2706110 in NFE2L2 was associated with CRF occurrence. In addition, this SNP was identified as an meQTL for the following five CpG sites in this gene: cg01417537, cg09248506, cg08334199, cg5303734, and cg03988329. In addition, the CpG site PRDX1 (cg09131901) was found to be significantly associated with both CRF occurrence and severity (see Figure 1).



    The SNP rs2706110 in NFE2L2 was significantly associated with CRF occurrence and was found to be an meQTL for five CpG sites. For CRF occurrence, three of these CpG sites were hypermethylated (cg01417537, cg09248506, cg05303734) and two were hypomethylated (cg08334199 and cg03988329). These CpG sites are visualized in a map of NFE2L2 that demonstrates the distribution of these sites within the gene (see Figure 2).


    Nuclear factor erythroid 2–related factor 2 (Nrf2), encoded by NFE2L2, is an important transcription factor that binds to antioxidant response elements to regulate various detoxifying and antioxidant defense genes (Moon & Giaccia, 2015). Nrf2 is a master regulator for several genes that contain antioxidant response element enzymes involved in the redox cycle (e.g., peroxiredoxin, thioredoxin, catalase, superoxide dismutase). These enzymes are responsible for mediating the reduction of reactive oxygen species to prevent oxidative stress–induced cell damage (Moon & Giaccia, 2015; Yamamoto et al., 2018). When reactive oxygen species are not regulated by the redox cycle, oxidative stress may occur and lead to DNA damage and, in some cases, tumorigenesis. In contrast, when Nrf2 is upregulated, it has protective effects and may prevent the initiation of tumor development caused by oxidative stress–induced injury in healthy cells (Hayes et al., 2020; Wu et al., 2019). On the other hand, upregulation of Nrf2 in malignant cells may be protective for these cells and promote tumor progression. In previous studies (Hartikainen et al., 2012; Karihtala & Soini, 2007), higher levels of antioxidative enzymes were found to be present in tumor tissue compared to nontumor tissue. These findings suggest that depending on the tissue of origin, Nrf2 may be pro- or antitumorigenic.

    Although no studies have evaluated the relationship between rs2706110 and CRF, Hartikainen et al. (2012) investigated its effect in the context of breast cancer. In this study, survivors with the rs2706110 AA genotype had an increased risk of breast cancer (Hartikainen et al., 2012). In the current study, the rs2706110 TT genotype was associated with a 3.5-fold increase in odds of CRF occurrence (odds ratio = 3.54, 95% confidence interval [1.28, 11.07], p = 0.02). This finding suggests that this SNP may play a role in the occurrence of CRF in patients with breast cancer.

    PRDX1 cg09131901

    cg09131901 in peroxiredoxin 1 (PRDX1) is a CpG site that was associated with CRF occurrence and severity. Peroxiredoxin enzymes, including peroxiredoxin 1, which is encoded by PRDX1 (Neumann et al., 2003), are responsible for more than 90% of the detoxification of the cellular reactive oxygen species, particularly hydrogen peroxide (H2O2) (Hayes et al., 2020). Peroxiredoxin 1 is primarily involved in scavenging for H2O2 and converting it into water and oxygen, blocking the circulation of reactive oxygen species (Hayes et al., 2020). When H2O2 is at extreme levels, the system becomes overwhelmed and inactivates PRDX1. This action results in the inability of cells to regulate oxidative stress efficiently (Wood et al., 2003). This process creates chronic levels of H2O2, and results in the progression and metastasis of breast cancer (Ding et al., 2017). This finding in the current study suggests the importance of the protective role of the PRDX1 gene and may also be important for understanding the occurrence and severity of CRF in patients with breast cancer.

    Oxidative Stress and Fatigue

    A study by Richards et al. (2000) suggests that oxidative stress overabundance may contribute to myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and its symptom presentation. Like CRF, ME/CFS is a complex condition that involves multiple symptoms. In addition to fatigue, some patients with CRF and ME/CFS report sleep disturbance, pain, cognitive dysfunction, and mood disturbance (Aoun Sebaiti et al., 2022). In a study that compared patients with and without chronic fatigue (Lee et al., 2018), serum levels of antioxidative proteins regulated by NFE2L2 (i.e., catalase and superoxide dismutase) were significantly lower in patients with idiopathic chronic fatigue. This study suggests that NFE2L2 (Nrf2) and downstream targeted proteins (e.g., peroxiredoxin, thioredoxin, superoxide dismutase) have a role in fatigue.

    Strengths and Limitations

    Some strengths of this study include capitalizing on data from a single-blinded randomized clinical trial, as well as having access to existing whole-genome methylation data and banked serial blood samples to generate additional omics data (i.e., SNPs). Also, the data and methods used in this study were novel in the context of CRF. Although the current study has many strengths, it does have limitations. The sample size was relatively small, primarily comprised postmenopausal women who reported their race as White, and consisted of women recruited from western Pennsylvania. Therefore, the generalizability of these findings is limited. Replication of these findings is warranted in a larger and more diverse sample.


    Conclusion and Implications for Nursing

    The findings from this study add to the understanding of the biologic mechanisms underlying variability in CRF occurrence and severity among patients with breast cancer. This study sheds light on the importance of the genes involved in the Nrf2-ARE pathway and patient-reported CRF, which is a novel finding. If replicated, these results have the potential to identify patients who are most at risk for the development of CRF early in the disease trajectory. Knowing which patients with cancer are at risk for development of CRF, particularly severe CRF, and knowing what intervention could target its underlying mechanism(s) could assist with clinical management and improved quality of life for patients with cancer.

    The authors gratefully acknowledge all the women who participated in this study as well as the University of Pittsburgh School of Nursing laboratory manager, Sandra Deslouches, for her laboratory assistance.

    About the Authors

    Tara S. Davis, PhD, RN, is a postdoctoral iCURE Fellow on the Patient Outcomes Team in the Neuro-Oncology Branch of the Center for Cancer Research at the National Cancer Institute in Bethesda, MD; Theresa A. Koleck, PhD, RN, is an assistant professor in the Department of Health Promotion and Development and Margaret Q. Rosenzweig, PhD, FNP-BCTM, FAAN, is a professor in the Department of Acute and Tertiary Care, both in the School of Nursing at the University of Pittsburgh in Pennsylvania; Christine Miaskowski, RN, PhD, FAAN, is a professor in the Department of Physiological Nursing in the School of Nursing at the University of California, San Francisco; Kirk I. Erickson, PhD, is the director of translational neuroscience at AdventHealth Research Institute in Orlando, FL; Susan M. Sereika, PhD, is a professor and associate dean for research and education support services; Catherine M. Bender, PhD, RN, FAAN, is a professor in the Department of Health and Community Services; and Yvette P. Conley, FAAN, PhD, is a professor and the associate dean for research and scholarship, all in the School of Nursing at the University of Pittsburgh. This research was funded, in part, by the National Institute of Nursing Research and the National Cancer Institute of the National Institutes of Health (T32NR009759, R01CA196762, and R01CA221882), The University of Pittsburgh School of Nursing Ruth and Bill Fine PhD Student Award, Janice Scully Dorman Endowed Omics Research Award, Jayne F. Wiggins Memorial Scholarship Award, and the International Society of Nurses in Genetics. Davis, Koleck, Rosenzweig, Erickson, Bender, and Conley contributed to the conceptualization and design. Bender and Conley completed the data collection. Koleck and Sereika provided statistical support. Davis, Rosenzweig, Sereika, and Conley provided the analysis. Davis, Koleck, Rosenzweig, Miaskowski, Erickson, Bender, and Conley contributed to the manuscript preparation. Davis can be reached at, with copy to (Submitted January 2024. Accepted March 26, 2024.)


    Aoun Sebaiti, M., Hainselin, M., Gounden, Y., Sirbu, C.A., Sekulic, S., Lorusso, L., . . . Authier, F.J. (2022). Systematic review and meta-analysis of cognitive impairment in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). Scientific Reports, 12(1), 2157.

    Beck, A.T., Steer, R.A., Ball, R., & Ranieri, W.F. (1996). Comparison of Beck Depression Inventories -IA and -II in psychiatric outpatients. Journal of Personality Assessment, 67(3), 588–597.

    Behringer, K., Goergen, H., Müller, H., Thielen, I., Brillant, C., Kreissl, S., . . . Borchmann, P. (2016). Cancer-related fatigue in patients with and survivors of Hodgkin lymphoma: The impact on treatment outcome and social reintegration. Journal of Clinical Oncology, 34(36), 4329–4337.

    Benjamini, Y., & Hochberg, Y. (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B (Methodological), 57(1), 289–300.

    Berger, A.M., Gerber, L.H., & Mayer, D.K. (2012). Cancer-related fatigue: Implications for breast cancer survivors. Cancer, 118(Suppl. 8), 2261–2269.

    Berger, A.M., Mooney, K., Alvarez-Perez, A., Breitbart, W.S., Carpenter, K.M., Cella, D., . . . Smith, C. (2015). Cancer-related fatigue, version 2.2015. Journal of the National Comprehensive Cancer Network, 13(8), 1012–1039.

    Bower, J.E., Asher, A., Garet, D., Petersen, L., Ganz, P.A., Irwin, M.R., . . . Crespi, C.M. (2019). Testing a biobehavioral model of fatigue before adjuvant therapy in women with breast cancer. Cancer, 125(4), 633–641.

    Bower, J.E., Ganz, P.A., Desmond, K.A., Rowland, J.H., Meyerowitz, B.E., & Belin, T.R. (2000). Fatigue in breast cancer survivors: Occurrence, correlates, and impact on quality of life. Journal of Clinical Oncology, 18(4), 743–753.

    Cella, D., Choi, S., Garcia, S., Cook, K.F., Rosenbloom, S., Lai, J.-S., . . . Gershon, R. (2014). Setting standards for severity of common symptoms in oncology using the PROMIS item banks and expert judgment. Quality of Life Research, 23(10), 2651–2661.

    Cella, D., Lai, J.-S., Jensen, S.E., Christodoulou, C., Junghaenel, D.U., Reeve, B.B., & Stone, A.A. (2016). PROMIS Fatigue item bank had clinical validity across diverse chronic conditions. Journal of Clinical Epidemiology, 73, 128–134.

    Chelune, G.J., Heaton, R.K., & Lehman, R.A.W. (1986). Neurological and personality correlates of patients’ complaints of disability. In G. Goldstein & R.E. Tarter (Eds.), Advances in clinical neuropsychology (vol. 3, pp. 95–126). Springer.

    Cleeland, C.S., & Ryan, K.M. (1994). Pain assessment: Global use of the Brief Pain Inventory. Annals of the Academy of Medicine, Singapore, 23(2), 129–138.

    Davis, T., Koleck, T., Conway, A., Bender, C., & Conley, Y. (2023). Genetic variability of oxidative stress and DNA repair genes associated with pre-treatment cancer-related fatigue in women with breast cancer. Supportive Care in Cancer, 31(6), 345.

    Dickinson, K., Case, A.J., Kupzyk, K., & Saligan, L. (2020). Exploring biologic correlates of cancer-related fatigue in men with prostate cancer: Cell damage pathways and oxidative stress. Biological Research for Nursing, 22(4), 514–519.

    Ding, C., Fan, X., & Wu, G. (2017). Peroxiredoxin 1—An antioxidant enzyme in cancer. Journal of Cellular and Molecular Medicine, 21(1), 193–202.

    García-González, D., Medino-Muñoz, J., Romero-Elías, M., García-Foncillas, J., & Ruiz-Casado, A. (2023). Biological mechanisms of cancer-related fatigue in breast cancer survivors after treatment: A scoping review. Journal of Cancer Survivorship. Advance online publication.

    Gaunt, T.R., Shihab, H.A., Hemani, G., Min, J.L., Woodward, G., Lyttleton, O., . . . Relton, C.L. (2016). Systematic identification of genetic influences on methylation across the human life course. Genome Biology, 17, 61.

    Gentry, A.L., Erickson, K.I., Sereika, S.M., Casillo, F.E., Crisafio, M.E., Donahue, P.T., . . . Bender, C.M. (2018). Protocol for Exercise Program in Cancer and Cognition (EPICC): A randomized controlled trial of the effects of aerobic exercise on cognitive function in postmenopausal women with breast cancer receiving aromatase inhibitor therapy. Contemporary Clinical Trials, 67, 109–115.

    George, M., Tharakan, M., Culberson, J., Reddy, A.P., & Reddy, P.H. (2022). Role of Nrf2 in aging, Alzheimer’s and other neurodegenerative diseases. Ageing Research Reviews, 82, 101756.

    Giaquinto, A.N., Sung, H., Miller, K.D., Kramer, J.L., Newman, L.A., Minihan, A., . . . Siegel, R.L. (2022). Breast cancer statistics, 2022. CA: A Cancer Journal for Clinicians, 72(6), 524–541.

    Greenland, S., & Pearce, N. (2015). Statistical foundations for model-based adjustments. Annual Review of Public Health, 36, 89–108.

    Groenvold, M., Petersen, M.A., Idler, E., Bjorner, J.B., Fayers, P.M., & Mouridsen, H.T. (2007). Psychological distress and fatigue predicted recurrence and survival in primary breast cancer patients. Breast Cancer Research and Treatment, 105(2), 209–219.

    Hartikainen, J.M., Tengström, M., Kosma, V.-M., Kinnula, V.L., Mannermaa, A., & Soini, Y. (2012). Genetic polymorphisms and protein expression of NRF2 and sulfiredoxin predict survival outcomes in breast cancer. Cancer Research, 72(21), 5537–5546.

    Hayes, J.D., Dinkova-Kostova, A.T., & Tew, K.D. (2020). Oxidative stress in cancer. Cancer Cell, 38(2), 167–197.

    Johns, M.W. (1991). A new method for measuring daytime sleepiness: The Epworth Sleepiness Scale. Sleep, 14(6), 540–545.

    Kang, Y.-E., Yoon, J.-H., Park, N.-H., Ahn, Y.-C., Lee, E.-J., & Son, C.-G. (2023). Prevalence of cancer-related fatigue based on severity: A systematic review and meta-analysis. Scientific Reports, 13(1), 12815.

    Karihtala, P., & Soini, Y. (2007). Reactive oxygen species and antioxidant mechanisms in human tissues and their relation to malignancies. APMIS, 115(2), 81–103.

    Kind, A.J.H., & Buckingham, W.R. (2018). Making neighborhood-disadvantage metrics accessible—The neighborhood atlas. New England Journal of Medicine, 378(26), 2456–2458.

    Kober, K.M., Smoot, B., Paul, S.M., Cooper, B.A., Levine, J.D., & Miaskowski, C. (2016). Polymorphisms in cytokine genes are associated with higher levels of fatigue and lower levels of energy in women after breast cancer surgery. Journal of Pain and Symptom Management, 52(5), 695–708.E4.

    Lee, J.D., Cai, Q., Shu, X.O., & Nechuta, S.J. (2017). The role of biomarkers of oxidative stress in breast cancer risk and prognosis: A systematic review of the epidemiologic literature. Journal of Women’s Health, 26(5), 467–482.

    Lee, J.-S., Kim, H.-G., Lee, D.-S., & Son, C.-G. (2018). Oxidative stress is a convincing contributor to idiopathic chronic fatigue. Scientific Reports, 8(1), 12890.

    Liguori, I., Russo, G., Curcio, F., Bulli, G., Aran, L., Della-Morte, D., . . . Abete, P. (2018). Oxidative stress, aging, and diseases. Clinical Interventions in Aging, 13, 757–772.

    Maass, S.W.M.C., Brandenbarg, D., Boerman, L.M., Verhaak, P.F.M., de Bock, G.H., & Berendsen, A.J. (2021). Fatigue among long-term breast cancer survivors: A controlled cross-sectional study. Cancers, 13(6), 1301.

    Mao, H., Bao, T., Shen, X., Li, Q., Seluzicki, C., Im, E.-O., & Mao, J.J. (2018). Prevalence and risk factors for fatigue among breast cancer survivors on aromatase inhibitors. European Journal of Cancer, 101, 47–54.

    Moon, E.J., & Giaccia, A. (2015). Dual roles of NRF2 in tumor prevention and progression: Possible implications in cancer treatment. Free Radical Biology and Medicine, 79, 292–299.

    Neumann, C.A., Krause, D.S., Carman, C.V., Das, S., Dubey, D.P., Abraham, J.L., . . . Van Etten, R.A. (2003). Essential role for the peroxiredoxin Prdx1 in erythrocyte antioxidant defence and tumour suppression. Nature, 424(6948), 561–565.

    Pilkonis, P.A., Choi, S.W., Reise, S.P., Stover, A.M., Riley, W.T., & Cella, D. (2011). Item banks for measuring emotional distress from the Patient-Reported Outcomes Measurement Information System (PROMIS®): Depression, anxiety, and anger. Assessment, 18(3), 263–283.

    Purcell, S., Neale, B., Todd-Brown, K., Thomas, L., Ferreira, M.A.R., Bender, D., . . . Sham, P.C. (2007). PLINK: A tool set for whole-genome association and population-based linkage analyses. American Journal of Human Genetics, 81(3), 559–575.

    R Core Team (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing.

    Repka, C.P., & Hayward, R. (2018). Effects of an exercise intervention on cancer-related fatigue and its relationship to markers of oxidative stress. Integrative Cancer Therapies, 17(2), 503–510.

    Richards, R.S., Roberts, T.K., McGregor, N.R., Dunstan, R.H., & Butt, H.L. (2000). Blood parameters indicative of oxidative stress are associated with symptom expression in chronic fatigue syndrome. Redox Report, 5(1), 35–41.

    Rojo de la Vega, M., Chapman, E., & Zhang, D.D. (2018). NRF2 and the hallmarks of cancer. Cancer Cell, 34(1), 21–43.

    Saligan, L.N., Olson, K., Filler, K., Larkin, D., Cramp, F., Yennurajalingam, S., . . . Mustian, K. (2015). Erratum to: The biology of cancer-related fatigue: A review of the literature. Supportive Care in Cancer, 23(9), 2853.

    Sanft, T., Day, A., Ansbaugh, S., Armenian, S., Baker, K.S., Ballinger, T., . . . Freedman-Cass, D.A. (2023). NCCN Guidelines® Insights: Survivorship, version 1.2023. Journal of the National Comprehensive Cancer Network, 21(8), 792–803.

    Schmidlin, C.J., Shakya, A., Dodson, M., Chapman, E., & Zhang, D.D. (2021). The intricacies of NRF2 regulation in cancer. Seminars in Cancer Biology, 76, 110–119.

    Schmidt, M.E., Scherer, S., Wiskemann, J., & Steindorf, K. (2019). Return to work after breast cancer: The role of treatment-related side effects and potential impact on quality of life. European Journal of Cancer Care, 28(4), e13051.

    University of California, Santa Cruz, Genomics Institute. (n.d.). Genome browser.

    Wood, Z.A., Poole, L.B., & Karplus, P.A. (2003). Peroxiredoxin evolution and the regulation of hydrogen peroxide signaling. Science, 300(5619), 650–653.

    Wright, F., Hammer, M., Paul, S.M., Aouizerat, B.E., Kober, K.M., Conley, Y.P., . . . Miaskowski, C. (2017). Inflammatory pathway genes associated with inter-individual variability in the trajectories of morning and evening fatigue in patients receiving chemotherapy. Cytokine, 91, 187–210.

    Wu, S., Lu, H., & Bai, Y. (2019). Nrf2 in cancers: A double-edged sword. Cancer Medicine, 8(5), 2252–2267.

    Xiao, C., Beitler, J.J., Peng, G., Levine, M.E., Conneely, K.N., Zhao, H., . . . Miller, A.H. (2021). Epigenetic age acceleration, fatigue, and inflammation in patients undergoing radiation therapy for head and neck cancer: A longitudinal study. Cancer, 127(18), 3361–3371.

    Xie, C., Leung, Y.-K., Chen, A., Long, D.-X., Hoyo, C., & Ho, S.-M. (2019). Differential methylation values in differential methylation analysis. Bioinformatics, 35(7), 1094–1097.

    Yamamoto, M., Kensler, T.W., & Motohashi, H. (2018). The KEAP1-NRF2 system: A thiol-based sensor-effector apparatus for maintaining redox homeostasis. Physiological Reviews, 98(3), 1169–1203.