Abstract
Background/Aim: Enzymatic variants involved in fluoropyrimidine metabolism have been associated with adverse events (AEs). We assessed the association between C677T (rs1801133) and A1298 (rs1801131) methylenetetrahydrofolate reductase (MTHFR) polymorphisms and AEs in patients with first-line fluoropyrimidine-based chemotherapy. Patients and Methods: Fifty patients with metastatic colorectal cancer were prospectively followed-up during the first 4 cycles of fluoropyrimidine-based treatment to assess AEs. Germline DNA was analyzed to determine the C677T and A1298C MTHFR polymorphisms. The associations between MTHFR polymorphisms and toxicity were examined. Results: Individuals carrying at least one mutant allele of the MTHFR C677T polymorphism had increased risk to experience anemia (OR=1.69, 95% CI=1.13-2.53, p=0.005), neutropenia (OR=2.27, 95% CI=1.47-3.42, p<0.001) thrombocytopenia (OR=1.91, 95% CI=1.30-2.70, p<0.001), neuropathy (OR=1.77, 95% CI=1.16-2.70, p=0.02), diarrhea (OR=1.69, 95% CI=1.13-2.53, p=0.005), and hand-foot syndrome (OR=1.56, 95% CI=1.08-2.27, p=0.013), compared to patients carrying the wild type alleles. The presence of the mutant allele C of the MTHFR A1298C polymorphism was associated with increased risk of anemia (OR=2.75, 95% CI=1.01-7.48, p=0.02) and thrombocytopenia (OR=3.14, 95% CI=1.01-9.78, p=0.03); however, the prevalence of this allele in the sample was quite low (20%). Conclusion: MTHFR C677T and A1298C polymorphisms predicted toxicity in a subset of Mestizo patients with colorectal adenocarcinoma.
- Metastatic colorectal cancer
- methylenetetrahydrofolate reductase
- SNPs
- fluoropyrimidine-based chemotherapy
- side effects
- Mestizo population
Colorectal cancer (CRC) is the second most prevalent neoplastic disorder worldwide, and the second leading cause of cancer-related death in both sexes (1). Although during the last decades new therapies have become available for the treatment of this disease, fluoropyrimidine-based chemotherapy still remains the backbone of treatment for these patients (2). However, adverse drug reactions to fluoropyrimidine-based chemotherapy (i.e., diarrhea, mucositis, vomiting, myelosuppression, neuropathy, and hand-foot syndrome) represent a clinical challenge due to the relatively high frequency of dose reductions or treatment discontinuations among affected patients (3).
Fluoropyrimidines, mainly capecitabine and its active metabolite 5-fluorouracil (5-FU), exert their cytotoxic effects in two different ways (Figure 1). First, 5-FU metabolites are extensively incorporated into RNA and DNA molecules, disrupting their normal functions. In addition, 5-FU inhibits the thymidylate synthase, in a reaction where the 5,10-methylenetetrahydrofolate (MTHF) acts as a methyl donor by forming a stable ternary complex with both the active 5-FU metabolite (5-fluoro-2-deoxyuridine-5-monophosphate), and the target enzyme (4). This inhibition precludes the de novo synthesis of thymidylate, which is required for DNA replication and repair. Intracellular concentrations of 5,10-MTHF are highly regulated by the MTHF reductase (MTHFR). This enzyme catalyzes the irreversible conversion of 5,10-MTHF to 5-methyltetrahydrofolate, and reduces the concentration of 5,10-MTHF available for binding to the ternary complex (5).
It has been previously shown that two common single-nucleotide polymorphisms (SNPs), C677T (rs1801133) and A1298C (rs1801131) reduce enzyme activity and lead to an altered distribution of intracellular folates (6). Individuals homozygous for the MTFHR C677T variant (TT) produce a thermolabile form of the protein with reduced catalytic activity as a result of an Ala-to-Val substitution at codon 222 (6, 7). The C677T variant is associated with a reduction in MTHFR activity by 30% in heterozygotes (CT) and 70% in homozygotes (TT) (7). Similarly, the MTHFR A1298C polymorphism induces a Glu-to-Ala substitution (Glu429Ala) in a regulatory domain resulting in a mild reduction (30%-40%) in the catalytic activity of the enzyme (8). Heterozygote patients for both SNPs (677CT/1298AC) also share a synergic decrease in the MTHFR activity (9).
Simplified overview of Capecitabine and 5-Fluorouracil metabolic pathways and targets. Enzymes: DHFR: dihydrofolate reductasa; MS: methionine synthase; MTHFR: 5,10-methylenetetrahydrofolate reductase; SHMT: serine-hydroxymethyltransferase; TP: thymidine phosphorylase; TYMS: thymidylate synthase. Metabolites: 5-CH3-FH4: 5’-methyl-tetrahydrofolate; 5-FU: 5-fluourouracil; 5,10-MTHF: 5,10-methylenetetrahydrofolate; 5’DFUR: 5’-deoxyfluorouracil;; dTMP: deoxythymidine 5’-monophosphate; dUMP: deoxyuridine 5’-monophosphate; FdUMP: 5-fluoro-2-deoxyuridine-5’-monophosphate; FH2: dihydrofolate; FH4: tetrahydrofolate.
Decreased enzyme activity results in an increased intracellular concentration of 5,10-MTHF that could enhance the stable formation of the ternary complex, resulting in increased toxicity of fluoropyrimidines (5). Indeed, clinical data support this hypothesis (10-12), although with conflicting results (13-16). Previous studies have reported that frequencies of mutant alleles of the aforementioned SNPs vary according to ethnic origin (17), with described frequencies for the MTFHR C677T mutant allele ranging from 10% (in Asians) to 57% (in Mexicans Mestizos), and for the 1298 CC allele in a range from 14% (in Asian) to 37% (in Caucasians) (18-20). This high variability is considered an important cause of inconsistencies among available data regarding the toxicity of fluoropyrimidine-based chemotherapy and its relationship with MTHFR polymorphisms (21). Therefore, in this prospective study we aimed to estimate the effect of the germline MTHFR polymorphisms C677T and A1298C on the toxicity induced by fluoropyrimidine-based chemotherapy in patients with metastatic colorectal adenocarcinoma from a Mestizo cohort in Costa Rica.
Patients and Methods
Patients and clinical data. A prospective study was carried out at Hospital San Juan de Dios, San José, Costa Rica (Caja Costarricense de Seguro Social), from January to July 2019. We included patients with histologically confirmed diagnosis of stage IV adenocarcinoma of the colon or rectum, who were treatment naïve in the metastatic setting. Eligible patients were required to have adequate organ function according to their attending oncologist, life expectancy more than 3 months, and good performance status (ECOG performance status 0-2). Patients received one of the following chemotherapy schemes: i) FOLFOX (5-FU 400 mg/m2 as a bolus, plus leucovorin 400 mg/m2 and oxaliplatin 85 mg/m2 on day one; followed by 5-FU 2400 mg/m2 as a 46-h continuous infusion every 15 days); ii) CAPEOX (Capecitabine 1000 mg/m2 twice daily for 14 days and oxaliplatin 130 mg/m2 every 21 days); iii) FOLFIRI (5-FU 400 mg/m2 as a bolus, plus leucovorin 400 mg/m2 and irinotecan 180 mg/m2 on day 1, followed by 5-FU 2400 mg/m2 as a 46-h continuous infusion every 15 days).
The assessment of treatment-induced toxicity was monthly performed by the oncologist in charged by determining patient symptoms, physical examination, and haemogram and biochemical tests during the first four cycles of treatment (84 days). Patients were followed up from the start of their treatment until the end of their fourth cycle of cytotoxic chemotherapy. Toxic effects were graded according to the National Cancer Institute – Common Terminology Criteria for Adverse Events version 4 criteria (CTCAE v.4) (19). Toxicity grades were dichotomized as either mild to moderate (grade 0-2), or severe (grade 3 or 4). Hand-foot syndrome was categorized into mild to moderate (grade 0 or 1), or severe (grade 2 or 3).
Ethical approval was obtained from the Institution's Ethics Committee (Comité Ético Científico Central, Caja Costarricense de Seguro Social, protocol No. R017-SABI-00126). All patients signed informed consents before inclusion. (NCT registration number: 03852290).
MTHFR polymorphisms. Venous blood samples were collected into tubes containing 1.7 mM EDTA and stored at -20o C until testing. Germline DNA was extracted automatically using the QIAamp DNA Blood Mini Kit with QIAcube (Qiagen, Valencia, CA, USA). DNA quality was evaluated by Nanodrop 260/280 and 260/230 ratios (Thermo Fisher Scientific®) and concentration was quantified with a Qubit Fluorometer (Life Technologies®) according to the manufacturer's recommendations.
The MTHFR C677T and A1298C SNPs were determined by polymerase chain reaction (PCR) – restriction fragment length polymorphism (RFLP) analysis. For the MTHFR C677T polymorphism the sequences of primers were 5’-TGAAGGAGAAGGTGTCTGCGGGA-3’ and 5’-AGGACGGTGCGGTGAGAGTG-3’. The PCR products were amplicons of 198 bp and were digested with 1U of Hinf I for 8 h. For the MTHFR A1298C polymorphism, the primer sequences were 5’ CTTTGGGGAGCTGAAGGACTACTAC 3’ and 5’ CACTTTGTGACCATTCCGGTTTG 3’. The PCR products were amplicons of 163 bp and were digested by 1U Mbo II. The PCR was performed in thermal cycler (Verity, Applied Biosystem) using 50-100 ng/sample and PCR conditions were an initial preheating step of 95°C for 3 min, followed by 30 cycles of 95°C for 1 min, an annealing step at 55°C for 1 min and an extension step at 72°C for 1 min. A last step of extension was performed at 72°C for 7 min. Regarding the C677T SNP, since mutation creates a Hinf I restriction site, the amplicon of the wild-type allele (198 bp) is not cut, while the mutant allele is cut into 2 fragments of 175 and 23 bp, respectively. The A1298C mutation abolishes an Mbo II restriction site, therefore, the amplicon of the wild-type allele showed fragments of 56, 30, 28 and 19 bp, while the mutant allele only showed three segments of 84, 30 and 19 bp. The digested PCR products were separated on ethidium bromide-stained 3% gels and visualized under ultra-violet light. A positive control was included in all the electrophoresis samples for comparative purposes.
Statistical analysis. Genotype distributions were tested for agreement with those expected under the Hardy-Weinberg equilibrium using chi-squared test. The association between genotypes under a dominant model (TT+CT vs. CC for the MTHFR C677T and AC+CC vs. AA for the MTHFR A1298C) and the presence of severe (grade 3 or 4) side effects were performed using the chi-squared or Fisher's exact test when applicable. t-Test analysis was used to determine whether age was significantly different between individuals who experienced at least one serious adverse event and those who did not. Univariate logistic regression analysis was performed to calculate the odds ratio (OR) for toxicity and its corresponding 95% confidence interval (95% CI). A p-value less than 0.05 was considered statistically significant. Analyses were performed with the SPSS software version 21.0 for Mac (Chicago, IL, USA).
Results
Clinical characteristics. A total of 50 patients were included in this study. Their demographic and clinical characteristics are resumed in Table I. The majority of recruited individuals were female (n=27, 54%) with good performance status at treatment initiation. Most patients had left-sided or rectal primary tumors (n=38, 76%) that were usually removed before therapy (n=32, 64%). Synchronous metastatic disease with liver and/or lung involvement was the most frequent clinical presentation (n=38, 76%). The majority of patients were treated with a chemotherapy combination of 5-FU or capecitabine and oxaliplatin (n=43, 86%).
Genotyping. C677T MTHFR genotypic frequencies were distributed as follows: CC (30%), CT (50%), and TT (20%). The C allele frequency was 55%, following the Hardy-Weinberg equilibrium (HWE) (p=0.91). A1298C MTHFR genotypic frequencies were: AA (70%), AC (26%), and CC (4%). The A allele frequency was 80%, also following the HWE (p=0.58). No patients were found to be homozygous for both loci. Figure 2 shows representative agarose gel electrophoresis profiles of PCR-RFLP for the MTHFR A1298C and C677T polymorphisms, respectively.
Toxicity. Overall, the frequency of severe toxicity was 52% (n=26). Mild or moderate adverse events were recorded in 80% of the patients (n=40). Mean age did not differ between patients who experienced at least one severe adverse event and those who did not (60.8±10.3 vs. 57.3±13.2 years, p=0.32). Neuropathy, blood marrow suppression, and diarrhea were the most incident side effects (Table II).
The correlations between MTHFR polymorphisms and grades of toxicity are presented in Table II. Significant correlations were found between the presence of at least one mutant allele of the MTHFR C677T polymorphism and the occurrence of severe neutropenia (p<0.001), anemia (p=0.003), thrombocytopenia (p<0.001), neuropathy (p=0.007), fatigue (p=0.007), diarrhea (p=0.009), hand-foot syndrome (p=0.012), nausea (p=0.016), and vomiting (p=0.011). Similarly, we found a statistical significant difference in the occurrence of severe anemia (p=0.031) and thrombocytopenia (p=0.025) among patients carrying at least one C mutant allele of the MTHFR A1298C polymorphism in comparison with wild-type individuals.
Demographic and clinical characteristics of the studied population (n=50).
Table III shows the univariate regression analysis for toxicity in a dominant model for each SNP (TT+CT vs. CC and CC+AC vs. AA, for the 677 and 1298 MTHFR genotypes, respectively). Patients with at least one mutant allele for the MTHFR C677T genotype were more likely to experience hematological toxicity [neutropenia (OR=2.27, 95% CI=1.47-3.42, p<0.001), anemia (p=0.005), and thrombocytopenia (OR=1.91, 95% CI=1.30-2.70, p<0.001)], neuropathy (OR=1.77, 95% CI=1.16-2.70, p=0.02), diarrhea (OR=1.69, 95% CI=1.13-2.53, p=0.005), hand-foot syndrome (OR=1.56, 95% CI=1.08-2.27, p=0.013), nausea (OR=1.55, 95% CI=1.15-2.10, p=0.021), and vomiting (OR=1.48, 95% CI=1.11-1.99, p=0.041). Similarly, the presence of the mutant allele C of the MTHFR A1298C polymorphism was significantly associated with anemia (OR=2.75, 95% CI=1.01-7.48, p=0.02) and thrombocytopenia (OR=3.14, 95% CI=1.01-9.78, p=0.03). However, the prevalence of the mutant allele C of the MTHFR A1298C polymorphism was quite low in the studied population (20%).
Discussion
Our findings support the association between MTHFR C677T (rs1801133) and A1298C (rs1801131) SNPs and fluoro - pyrimidine-related toxicity in a cohort of Costa Rican Mestizo patients with metastatic colorectal cancer. Specifically, patients with at least one mutated allele were more likely to experience severe toxicity than those with wild-type alleles. Although these findings have been described previously with contradictory results, few reports have evaluated this association in Latin-American populations. Indeed, our findings are novel to show the prevalence and influence of these SNPs on 5-FU-related toxicity in a Mestizo population.
The association between toxicity and MTHFR polymorphisms is controversial. In line with our results, Chua and colleagues reported higher gastrointestinal toxicity of 5-FU based chemotherapy in TT homozygous patients with metastatic colorectal cancer in comparison with heterozygous and wild-type subjects of the MTHFR C677T polymorphism (23). Similarly, Derwinger et al., reported that colorectal cancer patients carrying one mutated allele of the aforementioned SNP had more dose reductions or treatment modifications than those carrying the MTHFR 677 CC genotype, as a consequence of side effects and poor tolerability to 5-FU based chemotherapy (24).
Although other authors have also described analogous results to our findings (11, 25), some researchers have published no association between MTHFR SNPs and toxicity (13-16). It is important to highlight that the majority of studies that did not report any association between toxicity and SNPs had a lower percentage of patients with grade 3 to 4 toxicity than those studies in which this correlation was found (15 vs. 50%). From a pharmacological point of view, patients carrying the mutated forms of the MTHFR C677T and A1298C SNPs have a decreased enzyme activity resulting in higher inhibition of the thymidylate synthase by fluoropyrimidines, which could explain the increase of side effects (4-6).
Representative electrophoretic profiles of PCR-restriction fragment length polymorphism (RFLP) for 5,10-methylenetetrahydrofolate reductase (MTHFR) polymorphisms. Each column represents an individual patient; the last three columns represent controls (wild type homozygous, heterozygous, and mutant homozygous, respectively). (a) MTHFR A1298C and (b) MTHFR C677T. M: DNA ladder marker.
Frequency of mild to moderate (n=40) and severe (n=26) adverse events according to 5,10-methylenetetrahydrofolate reductase (MTHFR) polymorphisms.
Univariate analysis of mild/moderate or severe adverse events according to the presence of 5,10-methylenetetrahydrofolate reductase (MTHFR) polymorphisms.
The inconsistency among our data and other studies can also rely in several variables such as age, sample size, the chemotherapy scheme, individual folate status, and ethnic origin (21). Among these factors, ethnic background could be considered of paramount importance, since the prevalence of mutant alleles has been shown to vary greatly between different ethnic populations. Specifically, previous studies have reported that the frequency of the MTFHR C677T mutant allele ranged from 10% in Asian populations to 57% in Mexicans Mestizos, and that of the 1298 CC allele ranged from 14% (in Asian) to 37% (in Caucasians) (17-20).
Indeed, previous studies have revealed an association between the MTHFR SNPs and clinical outcomes. Martin and colleagues have demonstrated that these variant alleles have different effects on breast cancer prognosis depending on the studied population (26). In addition, data from Schwab and colleagues, which included patients with distinct types and stages of cancer, as well as different dosing regimens, showed that MTHFR polymorphisms played a limited role for 5-FU related toxicity (13). Similarly, Ruzzo and colleagues reported a lack of association between the presence of mutant alleles of the aforementioned SNPs and chemotherapy-induced toxicity with either FOLFOX or FOLFIRI, in advanced colorectal cancer patients (14, 15). However, it is worth noting that the incidence of severe toxicity was considerably lower in these studies in comparison with our results (15-23% vs. 50%). This could be due to the different populations that were analyzed (ethnicity, sample size). In line with our results, those studies showing a relationship between the MTHFR SNPs and fluoropyrimidine-related toxicity share a similar percentage of blood marrow suppression, diarrhea, and neuropathy (11, 23-25).
Combined analysis of both SNPs in our cohort did not show any patient with both mutant variants, suggesting linkage disequilibrium between C677T and A1298C variants, as previously described by other authors (27). However, the small sample size of our study could disregard the association between both SNPs.
Although some authors have acknowledged that in colorectal cancer the influence of MTHFR polymorphisms on toxicity can be blurred by the effect of oxaliplatin or irinotecan (28), our findings, as well as previous data, consistently described the association between the MTHFR 677 CT/TT genotype and increased toxicity in patients treated with either FOLFOX or FOLFIRI (29).
Regarding the association between the A1298C SNP and toxicity, the results of previous studies are also controversial. Although some authors have shown that individuals with advanced colorectal cancer and presence of the MTHFR 1298 CC genotype have significantly higher fluoropyrimidine-related toxicity compared to those with MTHFR 1298 AA genotype (14, 30), other authors have not shown any significant relationship between the occurrence of adverse events and this particular SNP (10, 13, 15-16). In our study, only a significant association between this SNP and anemia and thrombocytopenia was detected; however, we cannot claim this result as confirmatory due to the low prevalence of the mutant allele C in our cohort.
The mutant allelic frequency of the MTHFR C677T polymorphism in this study was slightly higher than that reported for Caucasian populations, and very similar to that described for Mestizos from Mexico (18). Similarly, the mutant allele frequency of the MTHFR A1298C polymorphism was very similar to that reported for Latin American populations, and lower than that described for patients of Caucasian origin (17, 18). These differences support the hypothesis of different toxicities and outcomes in patients treated with fluoropyrimidines, as a consequence of genetic variants of each studied population.
Our study has some caveats such as the relatively small sample size and unicenter design. Nevertheless, our population was in Hardy-Weinberg equilibrium, and we were able de detect a high number of individuals with the variant allele of the MTHFR C677T polymorphism. Although we did not test for other SNPs in fluoropyrimidines-metabolizing enzymes such as dihydropyrimidine dehydrogenase, and thymidylate synthase, we consider that our results provide a first approach to identify potential genetic markers of fluoropyrimidine toxicity in our population.
In summary, our results suggest that MTHFR C677T SNP may predict toxicity in Costa Rican patients with metastatic colorectal cancer and treated with fluoropyrimidine-based chemotherapy. These findings could help clinicians to foresee severe adverse events related to the use of this cytotoxic therapy among patients with metastatic colorectal cancer. Besides, our results provide a valuable tool in clinical practice to identify patients at risk, who require a closed follow-up in order to anticipate potential life-threatening toxicities. Further larger studies are warranted to validate these clinical findings.
Acknowledgements
The Authors thank Asociación Costarricense de Oncología Médica for the financial support for this project. Grant number: ACOMED-03-2019.
Footnotes
Authors' Contributions
Allan Ramos-Esquivel: Conceptualization, methodology, data acquisition, statistical analyses, interpretation of results, writing, and editing. Ricardo Chinchilla: data acquisition, reviewing, and editing. Marta Valle: supervision, conceptualization, methodology, reviewing, and editing.
The abstract of this paper was presented at the American Association for Cancer Research (AACR) Annual Meeting 2020.
Conflicts of Interest
None to declare.
Ethical Disclosure
The Authors have obtained appropriate institutional review board approval for all human experimental investigations (IRB Approval Number: R017-SABI-00126). An informed consent was obtained from the participants involved.
- Received May 15, 2020.
- Revision received June 7, 2020.
- Accepted June 24, 2020.
- Copyright© 2020, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.