Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Oct 1.
Published in final edited form as: Fertil Steril. 2019 Jul 29;112(4):731–739.e1. doi: 10.1016/j.fertnstert.2019.05.018

Genetic predictors of chemotherapy-related amenorrhea in women with breast cancer

Kathryn J Ruddy a,*, Daniel J Schaid b, Ann H Partridge c, Nicholas B Larson b, Anthony Batzler b, Lothar Häberle d, Ralf Dittrich d, Peter Widschwendter e, Visnja Fink e, Emanuel Bauer e, Judith Schwitulla d, Matthias Rübner d, Arif B Ekici f, Viktoria Aivazova-Fuchs g, Elizabeth A Stewart h, Matthias W Beckmann d, Elizabeth Ginsburg i, Liewei Wang j, Richard M Weinshilboum j, Fergus J Couch k, Wolfgang Janni e, Brigitte Rack e, Celine Vachon l,**, Peter A Fasching d,**
PMCID: PMC6765404  NIHMSID: NIHMS1529950  PMID: 31371054

Abstract

Objective:

To study how genetics may play a role in determining risk of chemotherapy-related amenorrhea (CRA) in young women with breast cancer.

Design:

A genome-wide association study

Setting:

A pooled cohort including participants in three clinical trials (SUCCESS A, B or C) testing chemotherapy for early stage breast cancer

Patients:

Premenopausal women ≤45 years enrolled in one of these three trials were included if they had at least one menstrual case report form after chemotherapy ended, and if they were of European ancestry. Forms during and up to three months after receipt of gonadotropin-releasing hormone agonist were excluded.

Intervention(s):

None

Main Outcomes Measure(s):

The association of single nucleotide polymorphisms (SNPs) with post-chemotherapy menstruation adjusted for trial/arm, age, tamoxifen use, and nodal status.

Results:

The median age of the 1168 women was 41 years (range 19–45). Among these, 457 (39%) never resumed menses after chemotherapy. Older age, tamoxifen use, and node-negative disease were associated with increased risk of CRA. Adjusting for these, rs147451859, in an intron of PPCDC (phosphopantothenoylcysteine decarboxylase), and rs17587029, located 5’upstream of RPS20P11 (ribosomal protein S20 pseudogene 11), were associated with post-chemotherapy menstruation (relative risk, RR 1.74, 95% CI 1.43 − 2.12, p= 6.38 × 10−9, and RR 1.84, 95% CI 1.48 – 2.28, p=2.4×10−8, respectively).

Conclusion:

Genetic variation may contribute to risk of CRA. Better prediction of who will experience CRA may inform reproductive and treatment decision-making in young women with cancer.

Keywords: Breast neoplasms, amenorrhea, drug therapy, toxicity

Capsule:

This genome-wide association study identified a 1.7–1.8-fold increase in likelihood of post-chemotherapy menses in patients who carried the minor alleles of SNPs in PPCDC and near RPS20P11.

Introduction:

Standard chemotherapies for breast cancer damage the ovaries, causing temporary or permanent loss of menses (chemotherapy-related amenorrhea; CRA) in many young women. Long-term amenorrhea has been seen in more than half of premenopausal women treated with chemotherapy for breast cancer.(1–3) Premature loss of ovarian function can cause hot flashes, vaginal dryness,(4) and bone thinning, and may contribute to other health risks.(5)

Loss of ovarian function is a particular concern for women who wish to have biological children after their breast cancer treatment. As more women delay first child-bearing, and as potentially teratogenic adjuvant endocrine therapy durations increase for many patients, treatment-related infertility will be a growing problem. Most women diagnosed with breast cancer in their 30s desire future pregnancies,(6,7) but the birth rate of breast cancer survivors is only one third that of age-matched controls.(8) Because pregnancy after breast cancer does not increase cancer recurrence risk,(8–11) and the ability to complete desired child-bearing is an important quality of life issue, newly diagnosed patients who are interested in pursuing future pregnancies are often offered oocyte or embryo cryopreservation. Ovarian tissue cryopreservation is also used in this setting in some countries.(12) Some patients may instead decide to avoid chemotherapy if they strongly value future fertility.(13)

Unfortunately, it is very difficult to individualize counselling regarding risks of infertility, or about the value of expensive and invasive fertility preservation procedures, because currently available methods for predicting who will maintain ovarian function after cancer treatment are inadequate.(14) Older age and more gonadotoxic regimens are well-recognized risk factors for loss of ovarian function,(15) but there is significant individual heterogeneity in risk of CRA that is not accounted for by these factors.

Because age at natural menopause is highly heritable,(16) and because more than 40 SNPs associated with age at natural menopause have been found through GWAS,(17) we hypothesized that genetic variation may also be associated with CRA. Research on genetics of ovarian function after chemotherapy has been scant, with only candidate SNP assessments to date. One small study (n=176) suggested an inverse association of the CT genotype of rs1172822 in BRSK1 (BR serine threonine kinase 1) with anti-mullerian hormone (AMH), a biomarker of ovarian reserve, in childhood cancer survivors. Another small study (n=50) found that breast cancer survivors who were heterozygous (CT) at rs4149056 in SLC01B1(solute carrier organic anion transporter family member 1B) were less likely to experience CRA.(18) In the current study, our objective was to perform a comprehensive CRA GWAS in premenopausal breast cancer survivors, hypothesizing that we could identify genetic predictors of CRA.

Methods:

Study population

The SUCCESS A (, clinicaltrials.gov) and SUCCESS B studies (), which enrolled between September 2005 and March 2007, randomized patients with early stage breast cancer (N1-3 and M0, or N0 with >T2, grade 3, hormone receptor-negative, or age <35) to receive gemcitabine or not (1000 mg/m2 on day 1 and day 8 of each three week cycle), concurrent with three cycles of docetaxel 100 mg/m2 every three weeks following three cycles of epirubicin 100 mg/m2, cyclophosphamide 500 mg/m2, and 5-fluorouracil 500 mg/m2 every three weeks (FEC).(19) Patients with human epidermal growth factor receptor 2 (HER2)-negative tumors were enrolled in SUCCESS A and secondarily randomized to receive zoledronic acid for two or five years.(20) Patients with HER2-positive tumors were enrolled on SUCCESS B and randomized to the same chemotherapy regimens without zoledronic acid, and all patients received trastuzumab. 3754 patients were recruited to SUCCESS A, and 793 to SUCCESS B.(21)

The SUCCESS C trial (), which enrolled between February 2009 and August 2011, randomized 3547 patients with early stage breast cancer (N1-3 and M0, or high risk N0 defined as >T2, grade 3, hormone receptor-negative, or age <35) to either: 1) 6 cycles of docetaxel 75 mg/m2 and cyclophosphamide 600 mg/m2 every three weeks; or 2) 3 cycles of FEC followed by 3 cycles of docetaxel 100 mg/m2 every three weeks; patients with body mass index (BMI) 24–40 kg/m2 were secondarily randomized to a lifestyle intervention vs. observation.(22) (Supplemental Table 1)

Patients who were premenopausal and age ≤45 at diagnosis, who had available GWAS data, were eligible for analysis. Because some were prescribed gonadotropin-releasing hormone agonist (GNRHa) as part of their cancer treatment, we required at least one post-chemotherapy menstrual data collection outside of any period of receipt of GNRHa (starting the day after the first GNRHa injection and finishing three months after the final GNRHa injection to allow washout of both monthly and every-three-month doses). This study was approved by the institutional review board (IRB) at Mayo Clinic.

Clinical data assessment

SUCCESS trial follow-up visits after completion of chemotherapy were every 3 months until year 2, then every 6 months until year 5, then yearly until year 8, resulting in a maximum of 17 visits. Menstrual data were collected on case report forms at each visit (classified as “menstruating” or “not menstruating”). We excluded menstrual data collected during and up to three months after receipt of GNRHa. Our analysis focused on the presence or absence of menses at the repeated follow-up visits.

DNA extraction and genotyping

Whole blood samples were collected in citrate-phosphate-dextrose-adenine tubes (Sarstedt AG, Nubrecht, Germany). Germline DNA was extracted via automated magnetic bead-based chemagic MSM I technique (PerkinElmer Chemagen, Baesweiler, Germany) per manufacturer’s instructions. Genotyping was performed on different arrays for the three trials. SUCCESS A subjects (N=643) were genotyped on the Illumina HumanOmniExpress at the Center for Inherited Disease Research, Johns Hopkins University, and quality control was performed at the University of Washington.(23) SUCCESS B (N=83) and SUCCESS C (N=442) participants were genotyped on the Illumina Infinium OncoArray, conducted either by the Breast Cancer Association Consortium (BCAC)(24) or the Mayo Clinic Medical Genome Facility.

Each of the genotyped batches was processed by a Mayo Clinic quality control pipeline that implemented the following filters: 1) removed unmapped SNPs, duplicate SNPs, SNPs on pseudoautosomal XY region, and mitochondrial SNPs; 2) removed SNPs with a call rate < 98%; 3) removed samples with a call rate < 95%; 4) removed SNPs with Hardy-Weinberg p values < 10−5; 5) removed duplicate samples detected by estimating identity-by-descent using King robust;(25) 6) removed SNPs with minor allele frequency < 0.005; 7) removed samples where sex was inconsistent with X chromosome SNPs; 8) removed tri-allelic and small insertion/deletions before imputation; 9) used STRUCTURE software(26) and 1000 Genome data as a reference to remove subjects that were not likely to have European ancestry (estimated probability of European ancestry < 0.7).

Unmeasured genotypes were imputed by the University of Michigan imputation server.(27) Because the subjects had European ancestry, we used the Haplotype Reference Consortium (HRC) reference panel to impute genotypes. SNPs that did not match the HRC, ambiguous palindromic SNPs with allele frequencies between 0.4 and 0.6, SNPs with non-matching alleles, and SNPs with mismatched allele frequencies were removed. Imputed SNPs with dosage r2 < 0.3 were considered low-quality and removed from analyses. Imputed allele frequencies were required to have allele frequencies within 0.15 units from the HRC panel, and GWAS analyses included 5,430,816 measured and imputed SNPs that had minor allele frequencies at least 0.05.

Statistical Methods

The post-chemotherapy follow-up visits measuring presence or absence of menstruation represent recurrent events, and a common way to analyze recurrent events is by well-known extensions of the Cox regression model.(28,29) We used the coxph R software package that implements the Cox model for recurrent events per subject. The start and stop times for each recurrent event surrounded the time of follow-up visit, so that the risk sets for the Cox model included women with the same follow-up visit. This model measured the association of covariates and SNPs with occurrence of menses in terms of relative risk. Furthermore, this accounted for within-subject correlations by estimating a robust variance estimate of the relative risk. We examined patient characteristics for associations with menses occurrence using multivariable selection and p< 0.05. Backward selection was used to remove covariates that were not significant in a joint model.

Genome-wide association tests were performed for each SNP using the Cox regression model adjusting for significant covariates. SNP genotypes were modeled according to the dose of the minor allele. Statistical significance was defined as p < 5×10−8.(30,31) Quantile-quantile (Q-Q) plots graphically showed the distribution of observed test-statistics versus expected under a null distribution. Manhattan plots were used to plot p values for all SNP associations across chromosomes. Regional association plots (locus zoom)(32) detailed the genetic regions around the genome-wide significant SNPs, providing gene annotations, estimated recombination rates, and pairwise correlations between the surrounding SNPs and the SNP of interest.

Functional annotation for significant findings, including predicted chromatin states from the Roadmap Epigenomics Project and reported expression quantitative trait loci (eQTL) associations, was queried using HaploReg v4.1 and RegulomeDB v1.1.(33–35) We evaluated cis- and trans-eQTL associations (false discovery rate < 0.05) for candidate SNPs using data from an eQTL investigation of microarray-based gene and exon expression levels in whole blood in 5257 Framingham Heart Study participants.(36) To explore whether the SNPs that were found to be associated with menses were likely to be involved in chemotherapy processing or ovarian reserve pathways, we examined the SNP associations with nausea, grade 3–4 leukopenia, and grade 3–4 neutropenia (adjusting for covariates used in the menses model).

Results:

Patient Characteristics

Of the 8,094 patients who enrolled in SUCCESS A/B/C, 1,675 were premenopausal and under age 45, and 1,168 had informative menstrual and genetic data. Of these 1,168, 457 (39%) never reported a menstrual period after chemotherapy, whereas 711 (61%) had at least one menstrual period recorded. Only 18 (1.5%) had received GNRHa prior to chemotherapy, but more than one third received GNRHa later. Univariate results for associations between patient/tumor characteristics and whether or not at least one menstrual period occurred after chemotherapy are presented in Table 1. More than 95% received six cycles of chemotherapy, and there were no differences between those who menstruated at least once after chemotherapy and those who did not with regard to study or treatment arm.

Table 1.

Characteristics of patients who were amenorrheic throughout post-chemotherapy follow-up vs. those who menstruated at least once

Amenorrheic
(N=457)		 Women who resumed menses
(N=711)
N (%) N (%)
Age
 ≤35 59 (12.9%) 133 (18.7%)
 36–40 109 (23.9%) 189 (26.6%)
 41–45 289 (63.2%) 389 (54.7%)
Tumor nodal status
 Node-positive 263 (57.5%) 331 (46.6%)
 Node-negative 194 (42.5%) 380 (53.4%)
Tumor ER status
 Negative 105 (23%) 289 (40.6%)
 Positive 352 (77%) 422 (59.4%)
Tumor PR status
 Negative 117 (25.6%) 304 (42.8%)
 Positive 340 (74.4%) 407 (57.2%)
Tumor HER2 status
 Negative 378 (82.7%) 536 (75.4%)
 Positive 77 (16.8%) 167 (23.5%)
 Indeterminate 2 (0.4%) 8 (1.1%)
Number of chemotherapy cycles
 1 0 (0.0%) 1 (0.1%)
 2 2 (0.4%) 2 (0.3%)
 3 3 (0.7%) 1 (0.1%)
 4 1 (0.2%) 11 (1.6%)
 5 5 (1.1%) 8 (1.1%)
 6 446 (97.6%) 688 (96.8%)
Tamoxifen use
 No 92 (20.1%) 269 (37.8%)
 Yes 365 (79.9%) 442 (62.2%)
Study
 SUCCESS A Arm A 102 (22.3%) 239 (33.6%)
 SUCCESS A Arm B 121 (26.5%) 181 (25.5%)
 SUCCESS B Arm A 15 (3.3%) 27 (3.8%)
 SUCCESS B Arm B 9 (2.0%) 32 (4.5%)
 SUCCESS C Arm A 99 (21.7%) 119 (16.7%)
 SUCCESS C Arm B 111 (24.3%) 113 (15.9%)
GNRHa use
No 257 (56.2%) 468 (65.8%)
Yes 200 (43.8%) 243 (34.2%)
Open in a new tab

PR= progesterone receptor; ER= estrogen receptor; HER2= human epidermal growth factor receptor 2; GNRHa= gonadotropin releasing hormone agonist

Potential covariates were screened for associations with post-chemotherapy menses (as reported on case report forms). Univariate associations are presented in Supplemental Table 2, and the final Cox regression model is presented in Supplemental Table 3. To test the effect of treatment, we contrasted Arm A with Arm B within each of the trials. We treated the combination of study and treatment arm as a stratification factor when evaluating the remaining covariates.

In a multivariate model, node-negative disease was associated with increased chance of menses (RR=1.18, 95% CI 1.00–1.39, p= 0.05), while receipt of tamoxifen (RR=0.53, 95% CI 0.44−0.61, p = 5×10−15) and older age (age 41–45 vs. age ≤ 35 (RR=0.62, 95% CI 0.52−0.75, p = 5×10−7); age 36–40 years vs. age ≤ 35 (RR=0.77, 95% CI 0.62−0.95, p= 0.01) were associated with reduced chance of menses.

GWAS Results

The Q-Q plot for the associations is shown in Supplemental Figure 1. Although the QQ plot shows a slight departure of p values from the null expected values, we confirmed that this is not likely due to population stratification because the top four eigenvectors that capture population stratification were not associated with the occurrence of menses. SNP associations, adjusted for age, tamoxifen use, nodal status, and study, are shown in Figure 1. The Manhattan plot illustrates that two SNPs (rs147451859 and rs17587029) achieved genome-wide significance. SNP rs147451859 on chromosome 15 (p =6.6×10−9) is located in an intron of PPCDC (Figure 2). The minor allele of rs147451859 had an allele frequency (MAF) of 0.07, and each allele was associated with 1.74-fold increased chance of menses. SNP rs17587029 on chromosome 2 (p=2.4×10−8) is 5’upstream of RPS20P11, has a MAF of 0.05, and is associated with 1.84-fold greater chance of menses. At any given assessable time point after chemotherapy, menses were identified in 53.6% of carriers of the minor allele (heterozygous or homozygous) for rs147451859 and in 35.6% of noncarriers (homozygous wildtype), as well as in 62.3% of carriers of rs17587029 and 35.9% of noncarriers. The likelihood of at least one menstrual period after chemotherapy was 71.8% for carriers of rs147451859 and 59.7% for noncarriers. The likelihood of at least one menstrual period after chemotherapy was 78.9% for carriers of rs17587029 and 59.9% for noncarriers. There were no other SNPs with suggestive associations (p<1×10−7) (Table 2; Figure 1).

Figure 1.

Figure 1.

Open in a new tab

Manhattan plot of SNP associations with post-chemotherapy menses, adjusted for SUCCESS trial/arm, age, tamoxifen use, and nodal status; the solid line represents the genome-wide significance threshold (p<5×10−8), and the dashed line represents a suggestive significance threshold (p<1×10−7).

Figure 2.

Figure 2.

Open in a new tab

Regional associations of SNPs with post-chemotherapy menses

Table 2.

Top SNP associations with post-chemotherapy menses, adjusted for SUCCESS trial/arm, age, tamoxifen use, and nodal status

rsID Chr Position Common/minorallele MAF RR SE of log RR p Hugo Gene Name Distance to Gene (bp) SNP Location
rs147451859 15 75331443 C/G 0.07 1.74 0.10 6.38×10−9 PPCDC 0 Intron
rs17587029 2 113382054 A/G 0.05 1.84 0.11 2.87×10−8 RPS20P11 2196 5′upstr
rs6488809 12 16154238 G/T 0.06 1.74 0.11 3.24×10−7 DERA 0 Intron
rs1410669 6 72450435 A/G 0.07 1.71 0.11 4.77×10−7 RIMS1 146215 5′upstr
rs11096688 2 21085410 C/T 0.10 1.45 0.08 6.80×10−7 C2orf43 62583 5′upstr
rs73419340 7 93018017 G/A 0.06 0.49 0.15 7.98×10−7 CCDC132 29679 3′downstr
rs1189020 14 56876959 G/T 0.06 1.61 0.10 9.27×10−7 PELI2 108928 3′downstr
rs1152563 14 56879264 G/A 0.06 1.61 0.10 9.52×10−7 PELI2 111233 3′downstr
rs1152562 14 56879149 C/T 0.06 1.61 0.10 9.52×10−7 PELI2 111118 3′downstr
rs1189021 14 56875202 A/G 0.06 1.60 0.10 9.79×10−7 PELI2 107171 3′downstr
rs78264913 12 29886853 T/C 0.05 1.53 0.09 1.14×10−6 TMTC1 0 Intron
rs78264913 12 29886853 T/C 0.05 1.53 0.09 1.14×10−6 TMTC1 0 Intron
rs1189135 14 56883950 G/A 0.06 1.64 0.10 1.25×10−6 PELI2 115919 3′downstr
rs77569618 5 6784195 T/A 0.09 1.48 0.08 1.28×10−6 PAPD7 27034 3′downstr
rs74994781 7 93018140 C/A 0.06 0.51 0.14 1.33×10−6 CCDC132 29802 3′downstr
rs1119997 8 132121917 A/G 0.46 1.30 0.05 1.40×10−6 ADCY8 69082 5′upstr
rs117140727 7 93018485 T/C 0.06 0.51 0.14 1.46×10−6 CCDC132 30147 3′downstr
rs73419344 7 93020108 T/C 0.06 0.52 0.14 1.55×10−6 CCDC132 31770 3′downstr
rs73419345 7 93020498 T/A 0.06 0.52 0.14 1.58×10−6 CCDC132 32160 3′downstr
rs55842191 12 16224098 C/T 0.05 1.65 0.10 1.64×10−6 DERA 33783 3′downstr
rs73059274 12 16219164 T/C 0.05 1.66 0.11 1.70×10−6 DERA 28849 3′downstr
Open in a new tab

Chr= chromosome; MAF= minor allele frequency; RR= relative risk of menses occurrence; bp= base pairs; downstr= downstream; upstr= upstream

Review of rs147451859 functional annotation from HaploReg revealed no cis-eQTL associations for PPCDC, although cis- and trans-eQTL associations with multiple genes such as CA1 (carbonic anhydrase 1), SCAMP2 (secretory carrier-associated membrane protein 2), and MAN2C1 (mannosidase alpha class 2C member 1)/NEIL1(endonuclease VIII-Like 1) were identified in eQTL results from the Framingham Heart Study. Although this variant overlaps predicted active enhancer regions in multiple tissue types, current evidence of regulatory impact is minimal (RegulomeDB score = 5).

Multiple eQTL association signals have previously been reported for rs17587029, including SLC20A1 (sodium-dependent phosphate transporter 1), ZC3H6 (zinc finger CCCH-type containing 6), and POLR1B (RNA polymerase I subunit B) in whole blood along with SLC20A1, CHCHD5 (coiled-coil-helix-coiled-coil-helix domain containing 5), POLR1B, and TTL (tubulin tyrosine ligase) in brain tissues.(36–38) This variant also overlaps an ENCODE DNAse hypersensitivity peak cluster and is predicted to disrupt a putative PPAR (peroxisome proliferator-activated receptors) transcription factor binding motif (RegulomeDB score = 2b). Further evaluation of cis-eQTL associations for both SNPs with the above-listed genes in the GTEx Project normal ovary samples (n=122) did not reveal any significant associations (all p > 0.05).(39)

Associations with other adverse events

We found that the minor alleles of both SNPs (rs147451859 and rs17587029) associated with an increased chance of post-chemotherapy menses were also associated with a decreased chance of having nausea, grade 3–4 leukopenia, and grade 3–4 neutropenia while on chemotherapy, adjusted for all covariates used in the menses model (please see Supplemental Table 4).

Discussion:

Chemotherapy-related ovarian insufficiency impacts the fertility, menopausal symptoms, and potentially long-term health of many survivors of premenopausal breast cancer. We confirmed previous findings that older age and tamoxifen use are associated with CRA, as well as the fact that age alone is not an adequate single predictor of post-chemotherapy ovarian function (given that approximately one third of those aged ≤35 never resumed menses and more than half of those aged 41–45 did menstruate at least once after chemotherapy). Importantly, we found that two SNPs, rs147451859 on chromosome 15 and rs17587029 on chromosome 2, were associated with resumption of menses after chemotherapy.

If confirmed, these SNPs may be included in models to better predict who will recover menses after chemotherapy, and to inform fertility preservation and cancer treatment decisions at the time of a premenopausal breast cancer diagnosis. The absolute increase in the likelihood of at least one menstrual period after chemotherapy was 12.1% for carriers of rs147451859 and 19.0% for carriers of rs17587029, and the absolute increase in likelihood of menses at any given time point was 18% for carriers of rs147451859 and 26.4% for carriers of rs17587029.

These loci may provide critical insights into the mechanisms underlying the inter-individual variation in loss of ovarian function and infertility in breast cancer survivors, though no functional studies were performed in this study. Our finding that both rs147451859 and rs17587029 were also associated with less likelihood of nausea, grade 3–4 leukopenia, and grade 3–4 neutropenia suggests that at least part of the mechanism of their association with post-chemotherapy menses is via a role in general chemotherapy processing pathways. In addition, this finding suggests that further investigation of the genetic predictors of hematologic and non-hematologic toxicities of chemotherapy could be fruitful, and this might eventually inform prophylactic management strategies (such as use of more aggressive anti-emetic medications starting with cycle 1 for those with an increased genetic risk of nausea, and/or prophylactic granulocyte colony-stimulating factor injections for certain patients receiving regimens that do not usually require that to maintain adequate white blood cell counts).

SNP rs147451859 is in an intron of PPCDC, which encodes a protein that is involved in the metabolism of water-soluble vitamins and cofactors. This protein is necessary for the biosynthesis of coenzyme A. A molecule of coenzyme A carrying an acetyl group, acetyl-CoA, is a cofactor required for the drug metabolizing enzyme N-acetyltransferase.(40) Therefore, variation in drug metabolism may be as important to CRA as ovarian reserve. Alternatively or in addition, the impact of coenzyme A activity on ovarian cell metabolism may influence the susceptibility of this tissue to damage from chemotherapy.

rs17587029 is located 5’upstream of ribosomal protein S20 pseudogene 11 (RPS20P11). Pseudogenes are presumed to be nonfunctional due to mutations and truncation by premature stop codons,(41) but mutations in a pseudogene may still regulate the expression of other protein-coding genes. Pseudogenes are known to regulate gene expression in mouse oocytes by impacting the RNA interference pathway.(42, 43) rs17587029 may regulate multiple nearby protein-coding genes, including SLC20A1. SLC20A1 is highly expressed in breast tissue, and tumor over-expression of SLC20A1 has been associated with poor clinical outcome in estrogen receptor (ER)-positive breast cancer cases.(44) Further research will be needed to assess how RPS20P11 may affect ovarian reserve in humans.

These findings are independent of GWAS signals associated with age at menarche or age at natural menopause. Additional studies will be needed to confirm these findings. It will be important to assess whether any of the other SNPs that did not reach genome-wide significance in this analysis might be identified as significant in larger data sets. Although it is difficult to obtain transvaginal ultrasound data, it would also be interesting to assess if these SNPs correlate with a low number of primordial follicles (a more direct measure of ovarian reserve) in healthy women, as well as before and after chemotherapy.

Limitations of this study include the relatively small sample size, missing menstrual data due to incomplete case report forms, and administration of GNRHa (which suppresses menses and limits our ability to understand menstrual patterns over time). Future trials could facilitate more definitive studies of the genetics of chemotherapy-related amenorrhea by prospectively and rigorously collecting patient-reported menstrual data in premenopausal patients whose cancers will not be treated with GNRHa (e.g., those with hormonally insensitive cancers). In addition, because chemotherapy-related amenorrhea is an imperfect surrogate for infertility, future research should assess the genetics of other biomarkers for infertility such as post-chemotherapy AMH levels.

In conclusion, we found a per allele 1.7–1.8-fold increase in likelihood of post-chemotherapy menses for SNPs in PPCDC and near RPS20P11. These SNPs help to identify who has a less than 40% versus greater than 60% chance of menses at any given time during the eight years following chemotherapy for breast cancer. Confirmatory studies are ongoing, as are functional assessments of whether or not the PPCDC finding may be related to differences in coenzyme A activity that lead to more rapid metabolism of chemotherapies (and therefore less ovarian toxicity). Further investigation of the genetics of CRA may inform individualized decision-making regarding breast cancer treatments (e.g., whether or not to receive chemotherapy for a small estimated reduction in risk of recurrence) as well as fertility preservation techniques.

Supplementary Material

1

Supplemental Figure 1. Q-Q plot

Funding:

This work was supported by CTSA Grant Numbers KL2TR000136–09 and KL2TR002379 from the National Center for Advancing Translational Sciences (NCATS), a component of the National Institutes of Health (NIH). Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NIH. In addition, grant funding for SUCCESS-A genotyping and quality control came from GARNET (Genomics and Randomized Trials Network): U01HG004438, U01HG005137, U01HG005157. Genotyping of SUCCESS-C was funded in part by the Breast Cancer Research Foundation. These funding sources had no involvement in study design, collection, analysis, interpretation of data, writing, or decision to submit this article for publication.

Disclosures: Dr. Stewart reports a consulting or advisory role with AbbVie, Bayer, Myovant, and Allergan (unrelated to this work) and royalties from UpToDate and Med Learning Group; Dr. Ginsburg reports patents, royalties, or other intellectual property from UptoDate and Springer, Inc., providing expert testimony for Adler and Cohen, and receiving research funding from Serono Inc. (all unrelated to this work); Dr. Fasching reports NIH-NHGRI funding relevant to this work, honoraria from and a consulting or advisory role with Novartis, Roche, Amgen, Celgene, Pfizer (unrelated to this work); all other authors have no conflicts of interest to disclose.

Abbreviations:

CRA

Chemotherapy-related amenorrhea

GWAS

genome-wide association study

SNPs

single nucleotide polymorphisms

GNRHa

gonadotropin-releasing hormone agonist

BCAC

Breast Cancer Association Consortium

HRC

Haplotype Reference Consortium

Q-Q

Quantile-quantile (Q-Q)

eQTL

expression quantitative trait loci

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

CONSORT diagram: This report describes a genome-wide association study that pools menstrual data from several clinical trials. It does not require a CONSORT diagram because we did not perform a clinical trial.

References:

  • 1.Abusief ME, Missmer SA, Ginsburg ES, Weeks JC, Partridge AH: The effects of paclitaxel, dose density, and trastuzumab on treatment-related amenorrhea in premenopausal women with breast cancer. Cancer 116:791–8, 2010 [DOI] [PubMed] [Google Scholar]
  • 2.Swain SM, Land SR, Ritter MW, Costantino JP, Cecchini RS, Mamounas EP, et al. : Amenorrhea in premenopausal women on the doxorubicin-and-cyclophosphamide-followed-by-docetaxel arm of NSABP B-30 trial. Breast cancer research and treatment 113:315–20, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Stachs A, Hartmann S, Gerber B: Preservation of Fertility or Ovarian Function in Patients with Breast Cancer or Gynecologic and Internal Malignancies. Geburtshilfe Frauenheilkd 77:861–869, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Rocca WA, Gazzuola-Rocca L, Smith CY, Grossardt BR, Faubion SS, Shuster LT, et al. : Accelerated Accumulation of Multimorbidity After Bilateral Oophorectomy: A Population-Based Cohort Study. Mayo Clin Proc 91:1577–1589, 2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Rosenberg SM, Tamimi RM, Gelber S, Ruddy KJ, Bober SL, Kereakoglow S, et al. : Treatment-related amenorrhea and sexual functioning in young breast cancer survivors. Cancer 120:2264–71, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Senkus E, Gomez H, Dirix L, Jerusalem G, Van Tienhoven G, Westenberg AH, et al. : Attitudes of young patients with breast cancer toward fertility loss related to adjuvant systemic therapies. EORTC study 10002 BIG 3–98. Psycho-oncology 23:173–82, 2014 [DOI] [PubMed] [Google Scholar]
  • 7.Paluch-Shimon S, Pagani O, Partridge AH, Abulkhair O, Cardoso MJ, Dent RA, et al. : ESO-ESMO 3rd international consensus guidelines for breast cancer in young women (BCY3). Breast 35: 203–217, 2017 [DOI] [PubMed] [Google Scholar]
  • 8.Kroman N, Jensen MB, Wohlfahrt J, Ejlertsen B: Pregnancy after treatment of breast cancer--a population-based study on behalf of Danish Breast Cancer Cooperative Group. Acta oncologica 47:545–9, 2008 [DOI] [PubMed] [Google Scholar]
  • 9.Lambertini M, Martel S, Campbell C, Guillaume S, Hilbers FS, Schuehly U, et al. : Pregnancies during and after trastuzumab and/or lapatinib in patients with human epidermal growth factor receptor 2-positive early breast cancer: Analysis from the NeoALTTO (BIG 1–06) and ALTTO (BIG 2–06) trials. Cancer 125(2):307–316, 2018. [DOI] [PubMed] [Google Scholar]
  • 10.Cordoba O, Bellet M, Vidal X, Cortes J, Llurba E, Rubio IT, et al. : Pregnancy after treatment of breast cancer in young women does not adversely affect the prognosis. Breast 21:272–5, 2012 [DOI] [PubMed] [Google Scholar]
  • 11.Azim HA Jr, Kroman N, Paesmans M, Gelber S, Rotmensz N, Ameye L, et al. : Prognostic impact of pregnancy after breast cancer according to estrogen receptor status: a multicenter retrospective study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 31:73–9, 2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Beckmann MW, Dittrich R, Lotz L, Oppelt PG, Findeklee S, Hildebrandt T, et al. : Operative techniques and complications of extraction and transplantation of ovarian tissue: the Erlangen experience. Arch Gynecol Obstet 295:1033–1039, 2017 [DOI] [PubMed] [Google Scholar]
  • 13.Ruddy KJ, Gelber SI, Tamimi RM, Ginsburg ES, Schapira L, Come SE, et al. : Prospective study of fertility concerns and preservation strategies in young women with breast cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 32:1151–6, 2014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Findeklee S, Lotz L, Heusinger K, Hoffmann I, Dittrich R, Beckmann MW: Fertility Protection in Female Oncology Patients: How Should Patients Be Counseled? Geburtshilfe Frauenheilkd 75:1243–1249, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Abusief ME, Missmer SA, Ginsburg ES, Weeks JC, Partridge AH: Relationship between reproductive history, anthropometrics, lifestyle factors, and the likelihood of persistent chemotherapy-related amenorrhea in women with premenopausal breast cancer. Fertility and sterility 97:154–9, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Murabito JM, Yang Q, Fox C, Wilson MUSSY33#, Cupples LA: Heritability of age at natural menopause in the Framingham Heart Study. The Journal of clinical endocrinology and metabolism 90:3427–30, 2005 [DOI] [PubMed] [Google Scholar]
  • 17.Day FR, Ruth KS, Thompson DJ, Lunetta KL, Perfjakova N, Chasman DI, et al. : Large-scale genomic analyses link reproductive aging to hypothalamic signaling, breast cancer susceptibility and BRCA1-mediated DNA repair. Nature genetics 47:1294–303, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Reimer T, Kempert S, Gerber B, Thiesen HJ, Hartmann S, Koczan D: SLCO1B1*5 polymorphism (rs4149056) is associated with chemotherapy-induced amenorrhea in premenopausal women with breast cancer: a prospective cohort study. BMC Cancer 16:337, 2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Widschwendter P, Friedl TW, Schwentner L, DeGregorio N, Jaeger B, Schramm A, et al. : The influence of obesity on survival in early, high-risk breast cancer: results from the randomized SUCCESS A trial. Breast cancer research : BCR 17:129, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Janni W, Schneeweiss A, Haeberle L, Fasching PA, Sommer HL, Rezai M, et al. : Adjuvant gemcitabine for high-risk breast cancer (BC) patients: Final survival results of the randomized phase III SUCCESS-A study, American Society of Clinical Oncology Chicago, IL, 2014 [Google Scholar]
  • 21.Andergassen U, Kasprowicz NS, Hepp P, Schindlbeck C, Harbeck N, Kiechle M, et al. : Participation in the SUCCESS-A Trial Improves Intensity and Quality of Care for Patients with Primary Breast Cancer. Geburtshilfe und Frauenheilkunde 73:63–69, 2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Rack B, Andergassen U, Neugebauer J, Salmen J, Hepp P, Sommer H, et al. : The German SUCCESS C Study - The First European Lifestyle Study on Breast Cancer. Breast care 5:395–400, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Mailman MD, Feolo M, Jin Y, Kimura M, Tryka K, Bagoutdinov R, et al. : The NCBI dbGaP database of genotypes and phenotypes. Nature genetics 39:1181–6, 2007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Milne RL, Kuchenbaecker KB, Michailidou K, Beesley J, Kar S, Lindstrom S, et al. : Identification of ten variants associated with risk of estrogen receptor negative breast cancer. Nat Genet, 2017. December; 49(12):1767–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Manichaikul A, Mychaleckyj JC, Rich SS, Daly K, Sale M, Chen WM: Robust relationship inference in genome-wide association studies. Bioinformatics 26:2867–73, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Pritchard JK, Stephens M, Donnelly P: Inference of population structure using multilocus genotype data. Genetics 155:945–59, 2000 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Das S, Forer L, Schonherr S, Sidore C, Locke AE, Kwong A, et al. : Next-generation genotype imputation service and methods. Nat Genet 48:1284–1287, 2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Amorim LD, Cai J: Modelling recurrent events: a tutorial for analysis in epidemiology. . International journal of epidemiology 44:324–33, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Ozga AK, Kieser M, Rauch G: A systematic comparison of recurrent event models for application to composite endpoints. BMC medical research methodology 18:2, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Manolio TA: Genomewide association studies and assessment of the risk of disease. N Engl J Med 363:166–76, 2010 [DOI] [PubMed] [Google Scholar]
  • 31.Pe’er I, Yelensky R, Altshuler D, Daly MJ: Estimation of the multiple testing burden for genomewide association studies of nearly all common variants. Genetic epidemiology 32:381–5, 2008 [DOI] [PubMed] [Google Scholar]
  • 32.Pruim RJ, Welch RP, Sanna S, Teslovich TM, Chines PS, Gliedt TP, et al. : LocusZoom: regional visualization of genome-wide association scan results. Bioinformatics 26:2336–7, 2010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Ward LD, Kellis M: HaploReg: a resource for exploring chromatin states, conservation, and regulatory motif alterations within sets of genetically linked variants. Nucleic Acids Res 40:D930–4, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Boyle AP, Hong EL, Hariharan M, Cheng Y, Schaub MA, Kasowski M, et al. : Annotation of functional variation in personal genomes using RegulomeDB. Genome Res 22:1790–7, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Roadmap Epigenomics Consortium, Kundaje A, Meuleman W, Ernst J, Bilenky M, Yen A, et al. : Integrative analysis of 111 reference human epigenomes. Nature 518:317–30, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Joehanes R, Zhang X, Huan T, Yao C, Ying SX, Nguyen QT, et al. : Integrated genome-wide analysis of expression quantitative trait loci aids interpretation of genomic association studies. Genome Biol 18:16, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Westra HJ, Peters MJ, Esko T, Yaghootkar H, Schumann C, Keltunen J, et al. : Systematic identification of trans eQTLs as putative drivers of known disease associations. Nat Genet 45:1238–1243, 2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Zou F, Chai HS, Younkin CS, Allen M, Crook J, Pankratz VS, et al. : Brain expression genome-wide association study (eGWAS) identifies human disease-associated variants. PLoS Genet 8:e1002707, 2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Consortium GT: The Genotype-Tissue Expression (GTEx) project. Nat Genet 45:580–5, 2013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Daugherty M, Polanuyer B, Farrell M, Scholle M, Lykidis A, de Crecy-Lagard V, et al. : Complete reconstitution of the human coenzyme A biosynthetic pathway via comparative genomics. J Biol Chem 277:21431–9, 2002 [DOI] [PubMed] [Google Scholar]
  • 41.Balasubramanian S, Zheng D, Liu YJ, Fang G, Frankish A, Carriero N, et al. : Comparative analysis of processed ribosomal protein pseudogenes in four mammalian genomes. Genome Biol 10:R2, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Tam OH, Aravin AA, Stein P, Girard A, Murchison EP, Cheloufi S, et al. : Pseudogene-derived small interfering RNAs regulate gene expression in mouse oocytes. Nature 453:534–8, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Watanabe T, Totoki Y, Toyoda A, Kaneda M, Kuramochi-Miyagawa S, Obata Y, et al. : Endogenous siRNAs from naturally formed dsRNAs regulate transcripts in mouse oocytes. Nature 453:539–43, 2008 [DOI] [PubMed] [Google Scholar]
  • 44.Sato K, Akimoto K: Expression Levels of KMT2C and SLC20A1 Identified by Information-theoretical Analysis Are Powerful Prognostic Biomarkers in Estrogen Receptor-positive Breast Cancer. Clin Breast Cancer 17:e135–e142, 2017 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

1

Supplemental Figure 1. Q-Q plot

NIHMS1529950-supplement-1.pdf (336KB, pdf)

RESOURCES