Post on 01-May-2023
1 23
Molecular Biology ReportsAn International Journal on Molecularand Cellular Biology ISSN 0301-4851 Mol Biol RepDOI 10.1007/s11033-014-3197-3
Thymidylate synthase polymorphismsare associated to therapeutic outcomeof advanced non-small cell lung cancerpatients treated with platinum-basedchemotherapyAurea Lima, Vítor Seabra, SandraMartins, Ana Coelho, António Araújo &Rui Medeiros
1 23
Your article is protected by copyright and all
rights are held exclusively by Springer Science
+Business Media Dordrecht. This e-offprint
is for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.
Thymidylate synthase polymorphisms are associated to therapeuticoutcome of advanced non-small cell lung cancer patients treatedwith platinum-based chemotherapy
Aurea Lima • Vıtor Seabra • Sandra Martins •
Ana Coelho • Antonio Araujo • Rui Medeiros
Received: 9 January 2013 / Accepted: 24 January 2014
� Springer Science+Business Media Dordrecht 2014
Abstract Thymidylate synthase (TYMS) has three poly-
morphisms that may modulate thymidylate synthase (TS)
expression levels: (1) 28 base pairs (bp) variable number
tandem repeat (VNTR) (rs34743033); (2) single nucleotide
polymorphism (SNP) C[G at the twelfth nucleotide of the
second repeat of 3R allele (rs2853542); and (3) 6 bp
sequence deletion (1494del6, rs34489327). This study was
conducted to evaluate the influence of TYMS polymor-
phisms on the survival of Portuguese patients with
advanced non-small cell lung cancer (NSCLC) undergoing
platinum-based chemotherapy. Our results showed no sta-
tistically significant differences between VNTR genotypes;
although, considering the SNP C[G, homozygotes 3RG
presented a better prognostic at 36 months (p = 0.004) and
overall survival (p = 0.003) when compared to 2R3RG
patients. Patients with ‘‘median/high expression geno-
types’’ demonstrated a better survival at 12 months
(p = 0.041) when compared to ‘‘low expression
genotypes’’. Furthermore, 6 bp- carriers (p = 0.006)
showed a better survival at 12 months when compared to
6 bp? homozygotes patients. When analyzing TYMS hap-
lotypes, better survival at 12 months was observed for
patients carrying haplotypes with the 6 bp- allele
(2R6 bp-; p = 0.026 and 3RG6 bp-; p = 0.045). This is
the first report that evaluates the three major TYMS poly-
morphisms in the therapeutic outcome of NSCLC in Por-
tugal. According to our results, the TYMS polymorphisms
may be useful tools to predict which advanced NSCLC
patients could benefit more from platinum-based chemo-
therapy regimens.
Keywords NSCLC � Platinum-based chemotherapy �Polymorphisms � Therapeutic outcome � Thymidylate
synthase
A. Lima (&) � V. Seabra
IINFACTS/CESPU, Institute of Research and Advanced
Training in Health Sciences and Technologies, Department of
Pharmaceutical Sciences, Higher Institute of Health Sciences
(ISCS-N), Rua Central de Gandra 1317, 4585-116 Gandra PRD,
Portugal
e-mail: aurea.lima@iscsn.cespu.pt;
aurealima1010@hotmail.com
A. Lima � A. Coelho � A. Araujo � R. Medeiros
Molecular Oncology Group CI, Portuguese Institute of Oncology
of Porto (IPO-Porto), Rua Dr. Antonio Bernardino de Almeida,
4200-072 Porto, Portugal
A. Lima � R. Medeiros
Abel Salazar Institute for the Biomedical Sciences (ICBAS),
University of Porto, Rua de Jorge Viterbo Ferreira 228,
4050-313 Porto, Portugal
S. Martins
Institute of Molecular Pathology and Immunology of the
University of Porto (IPATIMUP), Rua Dr. Roberto Frias,
4200-465 Porto, Portugal
A. Coelho
Faculty of Medicine of University of Porto (FMUP), Al. Prof.
Hernani Monteiro, 4200-319 Porto, Portugal
A. Araujo
Medical Oncology Department, Portuguese Institute of
Oncology of Porto (IPO-Porto), Rua Dr. Antonio Bernardino de
Almeida, 4200-072 Porto, Portugal
R. Medeiros
Research Department, Portuguese League Against Cancer
(LPCC-NRNorte), Estrada Interior da Circunvalacao, 6657,
4200-177 Porto, Portugal
123
Mol Biol Rep
DOI 10.1007/s11033-014-3197-3
Author's personal copy
Introduction
Lung cancer is the most common type of cancer in Europe
[1] and non-small cell lung cancer (NSCLC) accounts for
75–85 % of all histological types. The high mortality rate
(80–85 % within 5 years) results from the lack of effective
screening tools allowing for early-stage diagnosis [2]; the
inability to identify subsets of patients that would benefit
from adjuvant chemotherapy (CT) or adjuvant targeted
therapies; and the slow development of new drug therapies.
More than 70 % of NSCLC patients are diagnosed with
advanced disease and, therefore, are good candidates for
neoadjuvant, adjuvant or palliative systemic treatment with
platinum-based CT [3]. Although the use of cisplatin or
carboplatin has proved to be effective in combination with
non-platinum CT agents, such as paclitaxel or gemcitabine,
considerable variation has been observed in response to
treatment [3]. Pharmacogenetics strives to identify genetic
variations that could be useful in treatment prediction and
has become an important field in cancer treatment.
Recently, genetic polymorphisms have been suggested to
alter drug metabolism and activity leading to differences in
toxicity and/or efficacy on patients’ treatment [4].
Human thymidylate synthase (TS) is a key enzyme in de
novo synthesis of 20-deoxythymidine-50-monophosphate
(dTMP), an essential precursor of deoxyribonucleic acid
(DNA) biosynthesis and important factor to DNA replica-
tion and repair [5]. Inhibition of TS leads to depletion of
dTMP, which contributes for the incorporation of uracil
into DNA leading to chromosome instability [5–7].
Moreover, as TS catalyzes the methylation of 20-deoxyur-
idine-50-monophosphate (dUMP) to dTMP, using 5,10-
methylene-tetrahydrofolate (MTHF), it is important for
several chemotherapeutics [5, 6, 8]. Literature has descri-
bed that three polymorphisms (rs34743033, rs2853542 and
rs34489327) in the untranslated regions (UTRs) of the
thymidylate synthase (TYMS, 18p11.32) seem to influence
TS expression levels: (1) in the TYMS enhancer region
(TSER) there is a variable number tandem repeat (VNTR)
(rs34743033) upstream the ATG initiation codon site [9].
Humans have, most frequently, two and three repeats (2R
and 3R alleles, respectively), with 3R alleles associated to
higher TS expression and translation efficiency when
compared to 2R alleles [7, 10–15]. The 28 base pairs (bp)
repeated element appears to function as a preferential
enhancer linkage site and, therefore, influences TYMS
transcription and TS expression; (2) a functional C[G
single nucleotide polymorphism (SNP) (rs2853542) was
found in the second repeat of 3R alleles (3RC vs. 3RG)
[15]. This SNP changes an upstream stimulatory factor
(USF) binding site, with 3RG alleles allowing its ability to
complex with the USF protein. In vitro, a similar tran-
scriptional activity and translation efficiency has been
found to 3RC and 2R alleles [15–17]; (3) TYMS 1494del6
(rs34489327) is a deletion/insertion polymorphism (DIP)
of 6 bp (CTTTAA) located at nucleotide 1494 in the 30-UTR. Previous in vitro studies suggested that this poly-
morphism is associated with decreased messenger ribonu-
cleic acid (mRNA) stability and lower intratumoral TS
expression [15, 17]. Literature has suggested that TYMS
1494del6 polymorphism is in linkage disequilibrium (LD)
with the TSER polymorphisms (VNTR and SNP) and,
therefore, it has been hypothesized that TYMS polymor-
phisms might have an impact on the efficacy of CT treat-
ment, influencing the overall survival of NSCLC patients
[5, 18, 19]. To evaluate the clinical usefulness of geno-
typing TYMS as a prognostic marker of response to plati-
num-based CT treatment, we analyzed a subset of
advanced NSCLC patients. Moreover, we performed a
haplotype analysis, including all three polymorphisms, in
order to determine if an association that more effectively
predicts clinical outcomes is present, for combined che-
motherapy with platinum plus non-platinum drugs.
Materials and methods
Subjects
In this study, a total of 130 consecutive patients from the
northern region of Portugal with advanced NSCLC (stages
IIIA, IIIB and IV) treated with platinum-based CT were
included without restrictions of age, sex, smoking status or
disease history. All patients included in the study were
Table 1 NSCLC patients variables included in the study
Variable Value
Gender
Male, n (%) 103 (79)
Female, n (%) 27 (21)
Age (years)
Mean ± SD 62.88 ± 9.48
Median 64.00
Smoking status
Current and former, n (%) 100 (77)
Never, n (%) 30 (23)
Tumor stage
III, n (%) 92 (71)
IV, n (%) 38 (29)
Histological type
Squamous cell carcinoma, n (%) 47 (36)
Adenocarcinoma, n (%) 60 (46)
Others, n (%) 23 (18)
NSCLC non-small cell lung cancer
Mol Biol Rep
123
Author's personal copy
referred for treatment with platinum derivatives according
to the guidelines of treatments at the date of enrollment.
Clinical stage and pathologic evaluation were classified
according to the International System for Staging Lung
Cancer [20, 21]. Gender, age, smoking status, clinical stage
and histological classification of patients are described in
Table 1.
Prospective clinical follow-up time was available for
patients and defined as the time elapsed between diagnosis
and the last clinical evaluation or patient death. Procedures
were in accordance with the ethical standards of the Hel-
sinki Declaration, the study was approved by the Institu-
tional Ethical Committee and the subjects were included
after providing signed informed consent.
Blood samples were obtained with standard venipunc-
ture technique using ethylenediaminetetraacetic acid
(EDTA) containing tubes and genomic DNA extracted
using a commercial kit (Quiagen� QIAamp� DNA Blood
Mini Kit), according to the manufacturer instructions.
TYMS genotyping
The three polymorphisms (rs34743033, rs2853542 and
rs34489327) in the TYMS untranslated regions were gen-
otyped as following: (1) 28 bp VNTR was amplified by
polymerase chain reaction (PCR) as previously described
by Kawakami and Omura [13]; (2) SNP C[G at the twelfth
nucleotide of the second repeat of 3R allele was genotyped
by PCR-restriction fragment length polymorphism (RFLP)
as previously described by Kawakami and Watanabe [16];
(3) TYMS 1494del6 was genotyped by PCR–RFLP as
previously described by Ulrich and Bigler [17]. For quality
control, 10 % of all TYMS genotypes were confirmed by
automated sequencing in a 3130xl Genetic Analyzer using
the Kit BigDye Terminator v3.1 (Applied� Biosystems)
and results were 100 % concordant.
Statistical analysis
Statistical analyses were performed with the SPSS� soft-
ware (Version 15.0 of 2006, LEAD Technologies�, Inc.;
Chicago). Differences on TYMS genotypes were calculated
by the v2 test with a 5 % statistical significance (p \ 0.05).
To estimates the LD between pairs of alleles at TSER and
TYMS 1494del6 loci, D0 coefficients were calculated in Ar-
lequin 3.11 [22] with 100,000 number of steps in Markov
chain. The measure is interpretable as the proportion of the
maximum possible level of association between two loci,
given the allele frequencies, ranging from 0 (linkage equi-
librium) to 1 (complete LD) [23].
Hazard ratios (HRs) with 95 % confidence intervals
(CIs) were calculated. Genotypes were analyzed as a three-
group categorical variable (codominant model), and they
were also grouped according to the recessive model. TSER
genotypes were classify according to their theoretical TS
functional status as described previously (‘‘low’’: 2R2R,
2R3RC and 3RC3RC, ‘‘median’’: 2R3RG and 3RC3RG
and ‘‘high’’: 3RG3RG expression genotypes) [5, 19, 24].
For the haplotype analysis, allelic phase of genotyped
polymorphisms was inferred by PHASE 2.1 [25].
The influence of the different TYMS genotypes on
patients’ outcome was compared for 12, 36 months and
overall survival (OS) of the patients.
Survival curves were estimated by using the Kaplan–
Meier method. Differences between individual curves were
evaluated by multivariate analyses using Cox proportional
hazards regression models adjusted for disease stage and
NSCLC histological type. For this analysis, survival time
was defined as the time between diagnosis and an event
(either the last clinical evaluation or patient death).
Results
TSER genotypes
Observed genotype frequencies for rs34743033 and
rs2853542 polymorphisms are shown in Table 2. Survival
analysis showed no differences among patients carrying
different VNTR genotypes at 12, 36 months and for OS in
both codominant and recessive models. SNP genotyping of
3R alleles did not provide additional prognostic informa-
tion. Cox regression was not computed for functional 3R at
12 months because all patients in the reference group
(2R3R) were alive at that time. For functional 3R analysis,
3RG homozygotes presented a better prognosis when
compared to 2R3RG patients at both 36 months survival
(HR = 0.03; 95 % CI 0.00–0.34, p = 0.004) and for OS
(HR = 0.03; 95 % CI 0.00–0.31, p = 0.003). Finally,
classification of alleles according to their theoretical TS
functional status has suggested that patients with ‘‘median/
high expression genotypes’’ have a better prognosis than
those with ‘‘low expression genotypes’’, with survival at
12 months (p = 0.041) as shown in Fig. 1.
TYMS 1494del6 genotypes
TYMS 1494del6 observed genotype frequencies are pre-
sented in Table 2. By using a codominant model, a poor
survival rate at 12 months was observed for 6 bp?
homozygotes when compared to 6 bp?6 bp- patients
(p = 0.020) (Fig. 2A). The presence of the 6 bp- allele
was associated with a reduced risk of death. In fact, at
Mol Biol Rep
123
Author's personal copy
12 months no deaths were observed in all the 6 bp-
homozygotes. The Cox regression between 6 bp? homo-
zygotes and 6 bp- homozygotes was not computed
because all 6 bp- homozygotes patients were alive at
12 months. In the recessive model, survival curves at
12 months suggested that 6 bp- carriers were associated
with a better prognosis (p = 0.006), as shown in Fig. 2B.
No statistically significant differences were observed at
36 months and for OS in the analyzed models.
Linkage disequilibrium and haplotype analysis
TSER and TYMS 1494del6 polymorphisms are in LD in our
population (p = 0.00058). Alleles 2R and 6 bp?
(D0 = 0.43) as well as 3RG and 6 bp- (D0 = 0.45) are the
most linked ones.
When analyzing TSER-TYMS 1494del6 haplotypes,
different survivals were observed among patients carrying
(1) the 2R6 bp? and 2R6 bp- haplotypes (p = 0.026) and
Table 2 Multivariate analysis of TSER and TYMS 1494del6 polymorphisms in NSCLC patients
Genotype N (%) Survival at 12 months Survival at 36 months Overall survival
HR 95 % CI p* HR 95 % CI p* HR 95 % CI p*
VNTR genotypes (n = 130)
Codominant model
2R2R 26 (20) 1.00 1.00 1.00
2R3R 63 (48) 1.67 (0.45–6.14) 0.445 1.67 (0.66–4.24) 0.275 1.10 (0.50–2.41) 0.812
3R3R 41 (32) 1.05 (0.29–3.76) 0.938 0.98 (0.47–2.34) 0.958 0.78 (0.36–1.67) 0.517
Alleles grouped as a recessive model
2R carriers* 89 (69) 1.00 1.00 1.00
3R3R 41 (31) 0.74 (0.26–2.05) 0.560 0.68 (0.35–1.35) 0.277 0.66 (0.37–1.14) 0.137
TSER genotypes (n = 130)
Functional 2R (n = 81)
2R2R 26 (32) 1.00 1.00 1.00
2R3RC 40 (49) 1.05 (0.29–3.82) 0.942 1.24 (0.48–3.15) 0.657 0.93 (0.41–2.09) 0.856
3RC3RC 15 (19) 0.79 (0.17–3.61) 0.763 0.86 (0.30–2.51) 0.786 0.91 (0.34–2.41) 0.847
Funcional 3R (n = 49)
2R3RG 23 (47) 1.00a 1.00 1.00
3RC3RG 21 (43) 0.50 (0.08–3.15) 0.463 0.33 (0.10–1.09) 0.069
3RG3RG 5 (10) 0.03 (0.00–0.34) 0.004 0.03 (0.00–0.31) 0.003
Alleles grouped according to the functional status** (n = 130)
Low expression 81 (62) 1.00 1.00 1.00
Median/High expression 49 (38) 0.32 (0.09–1.15) 0.041 0.59 (0.28–1.24) 0.067 0.86 (0.49–1.51) 0.605
TYMS 1494del6 genotypes (n = 130)
Codominant model
6 bp?6 bp? 60 (46) 1.00 1.00 1.00
6 bp?6 bp- 54 (42) 0.17 (0.04–0.75) 0.020 0.61 (0.30–1.24) 0.170 0.71 (0.39–1.28) 0.259
6 bp-6 bp- 16 (12) b 0.49 (0.14–1.65) 0.251 0.78 (0.32–1.91) 0.583
Recessive model
6 bp?6 bp? 60 (46) 1.00 1.00 1.00
6 bp- carriers*** 70 (54) 0.12 (0.03–0.55) 0.006 0.59 (0.30–1.15) 0.121 0.74 (0.43–1.28) 0.278
CI confidence interval, HR hazard ratio, NSCLC non-small cell lung cancer, SNP single nucleotide polymorphism, TSER thymidylate synthase
enhancer region, TYMS thymidylate synthase (gene), VNTR variable number tandem repeat
* 2R carriers include homozygotes 2R and heterozygotes 2R3R. ** Low expression genotypes are 2R2R, 2R3RC and 3RC3RC. Median
expression genotypes are 2R3RG and 3RC3RG. High expression genotype is 3RG3RG. *** 6 bp- carriers include homozygotes 6 bp- and
heterozygous 6 bp?6 bp-
p* values correspond to multivariate Cox models adjusted for disease stage and to histological type of NSCLCa Cox regression was not computed because all patients in the reference group (2R3RG) were alive at 12 monthsb Cox regression was not computed because all 6 bp-6 bp- patients were alive at 12 months
Mol Biol Rep
123
Author's personal copy
(2) the 2R6 bp? and 3RG6 bp- haplotypes (p = 0.045)
(Table 3; Fig. 3). In addition, a better prognosis was
observed at 12 months for patients carrying haplotypes
with the 6 bp- allele.
Discussion
TS is an essential enzyme for DNA biosynthesis, replica-
tion and repair, hence, it is an important target for CT drugs
and its over-expression has been described as associated
with CT resistance in cancer [6–8]. Currently, despite the
large number of pharmacogenomics studies, there is no
available protocol for selecting cancer patients at risk for
drug resistance prior to CT.
Three polymorphisms (rs34743033, rs2853542 and
rs34489327) on TYMS have been shown to influence TS
expression levels and CT response [26, 27] although results
have not been consistent [28]. Considering the influence of
Fig. 1 Kaplan–Meier plots according to the TS functional status at
12 months. p values correspond to multivariate Cox models adjusted
for disease stage and to histological type of NSCLC
Fig. 2 Kaplan–Meier plots for
NSCLC patients according to
their TYMS 1494del6 genotypes
with survival at 12 months, as a
codominant (A) and recessive
model (B). p values correspond
to multivariate Cox models
adjusted for disease stage and to
histological type of NSCLC
Table 3 Multivariate analysis for TSER-TYMS 1494del6 haplotypes of NSCLC patients
Frequency Survival at 12 months Survival at 36 months Overall survival
N (%) HR 95 % CI p* HR 95 % CI p* HR 95 % CI p*
TSER-TYMS 1494del6 haplotype (n = 130)
2R6 bp? 37 (28) 1.00 1.00 1.00
3RC6 bp? 32 (25) 0.82 (0.37–1.83) 0.629 0.87 (0.47–1.62) 0.668 0.92 (0.55–1.56) 0.769
3RG6 bp? 19 (15) 0.31 (0.65–1.45) 0.135 0.56 (0.20–1.59) 0.278 0.73 (0.32–1.68) 0.462
2R6 bp- 22 (17) 0.23 (0.06–0.84) 0.026 0.55 (0.25–1.20) 0.133 0.55 (0.28–1.08) 0.082
3RC6 bp- 12 (9) 0.24 (0.03–1.85) 0.171 0.64 (0.25–1.62) 0.343 0.79 (0.39–1.61) 0.519
3RG6 bp- 8 (6) 0.12 (0.02–0.95) 0.045 0.54 (0.23–1.25) 0.151 0.69 (0.36–1.34) 0.276
bp base pairs, CI confidence interval, HR hazard ratio, NSCLC non-small cell lung cancer, TSER thymidylate synthase enhancer region, TYMS
thymidylate synthase (gene)
* p values correspond to multivariate Cox models adjusted for disease stage and to histological type of NSCLC
Mol Biol Rep
123
Author's personal copy
these polymorphisms on TS levels, it is extremely impor-
tant to study their role in NSCLC, since the degree and
duration of TS inhibition with CT drugs may depend on its
expression levels and, therefore, may influence patients’
response.
TSER genotypes
TSER polymorphisms have been shown to influence the total
amount of active protein, although significant individual
variability is observed [29]. Preliminary data suggested
that 3R homozygotes patients have higher TS mRNA
expression than those homozygotes for 2R allele [12, 30].
Moreover, a poor response to CT is observed in patients with
3R allele [26, 27]. Although published results have not been
consistent, studies have shown that patients with low TS
levels seem to have a worse outcome when treated with
adjuvant CT, probably due to the high incidence of adverse
drug reactions associated with 2R allele, that, consequently,
can lead to poor survival [28, 31]. These inconsistent results
might be explained by the variation of TYMS gene copy
number due to loss of heterozygosis or gene amplification as
showed by others [32–36]. In our study, results showed no
significant differences among patients carrying different
VNTR genotypes at 12, 36 months survival and OS, in both
codominant and recessive models, similarly to the results
shown by Lecomte and Ferraz [18].
To the best of our knowledge, the putative relationship
between the VNTR and the survival time needs further
clarification. Some authors suggested that this could be
improved with the study of the SNP C[G at the second
repeat of the 3R allele, since it is reported that the presence
of 3RG allele is associated with higher transcriptional
activity and translation efficiency due to an increased
ability of this allele to complex with the USF protein [15,
16]. In our study, 3RG homozygotes showed better sur-
vival, corroborating the results of Edler et al. [28]. These
results underpin the importance of 3RG allele SNP’s on TS
functional role. On the other hand, since there is no dif-
ference on survival analysis between patients with 2R3RG
and 3RC3RG genotypes, these results are in accordance
with in vitro studies were a similar transcriptional activity
and translation efficiency was found for those 2R and 3RC
alleles [16]. Moreover, and considering the theoretical
functional status of TS, accordingly with the possible
genotypes for the 50-UTR, our results indicated that
patients with ‘‘median/high expression genotypes’’ have a
better prognosis that those with ‘‘low expression geno-
types’’ (Table 2; Fig. 1). In agreement with others [29, 37]
these results may be explained by the influence of 3RG
allele, since homozygotes 3RG have two 3RG alleles in
comparison to 2R3RG and 3RC3RG patients, both with
only one 3RG allele. However, different results regarding
NSCLC patients treated with pemetrexed were reported,
showing longer progression free survival times for patients
with ‘‘low expression genotypes’’ [38, 39]. According with
these results, it is important to elucidate de influence of
TSER genotypes on the TS target CT outcome.
TYMS 1494del6 genotypes
Another polymorphism described on TYMS consists in a
6 bp DIP in the 30-UTR [17]. Although the function of this
polymorphism is not entirely known, there are evidences
suggesting that the deletion is associated with a decrease in
TS mRNA stability and expression [17, 40, 41]. Similarly
to previous studies, our results suggested that the 6 bp-
allele is associated with a better prognosis [19]. At
12 months, for both models, a better survival time was
observed for 6 bp?6 bp- patients when compared to
6 bp? homozygotes; no deaths were observed in the all
6 bp- homozygotes at that time; and survival curves at
12 months suggested that 6 bp- allele was associated with
a better prognosis. Therefore, it seems that 6 bp- allele is
important to predict the survival time of NSCLC patients.
Fig. 3 Kaplan–Meier plots to
TSER-TYMS 1494del6
haplotypes at 12 months.
A 2R6 bp? and 2R6 bp-
haplotypes. B 2R6 bp? and
3RG6 bp- haplotypes. p values
correspond to multivariate Cox
models adjusted for disease
stage and to histological type of
NSCLC
Mol Biol Rep
123
Author's personal copy
Moreover, recent studies reported similar results in NSCLC
patients treated with pemetrexed [39].
Linkage disequilibrium and haplotype analysis
Our study has confirmed that TYMS polymorphisms were in
LD as suggested by others [5, 18, 19, 40, 42, 43]. The associ-
ation was higher with haplotypes harboring the 6 bp- allele,
suggesting a prominent role of the 30-UTR polymorphism in
predicting the prognosis of advanced NSCLC patients. At
12 months, our results suggested that the haplotypes with the
6 bp- allele were associated with a better response to CT,
better prognostic and, consequently, reduced risk of death, as
previously reported in other populations [19, 44, 45].
From the haplotype analysis we can infer that 2R6 bp-
and 3RC6 bp- haplotypes were different when survival at
12 months was considered (Table 3; Fig. 3). This can be
explained by the presence of the 6 bp- allele which can
interact differently with 2R and 3RC alleles, in agreement
with Lurje and Zhang [24]. Thus, our results demonstrated
the importance of locus 6 bp- in a better prognosis and
suggested that TYMS haplotypes analysis needs to be
considered in the evaluation of TYMS polymorphisms on
NSCLC therapeutic response.
Final conclusions
In this study we have addressed the possible role of three
TYMS polymorphisms in the prognosis of Portuguese patients
with advanced NSCLC undergoing platinum-based CT regi-
mens. As a result, we have attempted to establish if TYMS
polymorphisms, using genotype and haplotype-based
approaches, lead to differences in clinical outcome of patients.
This is the first report that evaluates the three major TYMS
polymorphisms related with TS expression in therapeutic
outcome of a NSCLC Portuguese population. According to
our results, the genotyping of TYMS polymorphisms may be a
useful tool to predict which advanced NSCLC patients could
benefit more from platinum-based CT regimens and also
emphasize the importance of analyzing patients’ TYMS
haplotypes. Nevertheless, the clinical utility of analysing
3RG6 bp- and 2R6 bp- haplotypes deserves further eval-
uation, as more work is necessary before coming to a positive
conclusion. Although, this is a matter of controversy and some
caution is required when translating this approach into the
clinic, in order to optimise and individualize therapeutic
options as an approach to predict prognosis and therapy
outcomes.
Acknowledgments The authors wish to acknowledge the Ministry
of Health of Portugal (CFICS-Project 31/2007) and Astrazeneca
Foundation for the financial support; Liga Portuguesa Contra o
Cancro—Centro Regional do Norte (Portuguese League Against
Cancer) for the support to the lab; and to Fundacao para a Ciencia e
Tecnologia (FCT) for the Doctoral Grant (SFRH/BD/64441/2009) for
Aurea Lima. Authors would also like to acknowledge Hugo Sousa
(Ph.D.) for his critics in the final version of the manuscript.
References
1. Alberg AJ, Brock MV, Samet JM (2005) Epidemiology of lung
cancer: looking to the future. J Clin Oncol 23(14):3175–3185.
doi:10.1200/JCO.2005.10.462
2. Beane J, Spira A, Lenburg ME (2009) Clinical impact of high-
throughput gene expression studies in lung cancer. J Thorac
Oncol 4(1):109–118. doi:10.1097/JTO.0b013e31819151f8
3. National Comprehensive Cancer Network (2012) NCCN clinical
practice guidelines in oncology (NCCN guidelines)—non-small
cell lung cancer. Version 2. http://www.tri-kobe.org/nccn/guide
line/lung/english/small.pdf. Accessed 29 April 2013
4. Watters JW, McLeod HL (2003) Cancer pharmacogenomics:
current and future applications. Biochim Biophys Acta
1603(2):99–111. doi:10.1016/S0304-419X(03)00003-9
5. Lima A, Azevedo R, Sousa H, Seabra V, Medeiros R (2013)
Current approaches for TYMS polymorphisms and their impor-
tance in molecular epidemiology and pharmacogenetics. Phar-
macogenomics 14(11):1337–1351. doi:10.2217/pgs.13.118
6. Carreras CW, Santi DV (1995) The catalytic mechanism and
structure of thymidylate synthase. Annu Rev Biochem 64:721–762.
doi:10.1146/annurev.bi.64.070195.003445
7. Horie N, Aiba H, Oguro K, Hojo H, Takeishi K (1995) Functional
analysis and DNA polymorphism of the tandemly repeated
sequences in the 50-terminal regulatory region of the human gene
for thymidylate synthase. Cell Struct Funct 20(3):191–197
8. Danenberg PV (1977) Thymidylate synthetase—a target enzyme
in cancer chemotherapy. Biochim Biophys Acta 473(2):73–92.
doi:10.1016/0304-419X(77)90001-4
9. Marsh S, McKay JA, Cassidy J, McLeod HL (2001) Polymor-
phism in the thymidylate synthase promoter enhancer region in
colorectal cancer. Int J Oncol 19(2):383–386
10. Kaneda S, Takeishi K, Ayusawa D, Shimizu K, Seno T, Altman S
(1987) Role in translation of a triple tandemly repeated sequence in
the 50-untranslated region of human thymidylate synthase mRNA.
Nucleic Acids Res 15(3):1259–1270. doi:10.1093/nar/15.3.1259
11. Kawakami K, Salonga D, Park JM, Danenberg KD, Uetake H,
Brabender J, Omura K, Watanabe G, Danenberg PV (2001)
Different lengths of a polymorphic repeat sequence in the thy-
midylate synthase gene affect translational efficiency but not its
gene expression. Clin Cancer Res 7(12):4096–4101
12. Pullarkat ST, Stoehlmacher J, Ghaderi V, Xiong YP, Ingles SA,
Sherrod A, Warren R, Tsao-Wei D, Groshen S, Lenz HJ (2001)
Thymidylate synthase gene polymorphism determines response
and toxicity of 5-FU chemotherapy. Pharmacogenomics J
1(1):65–70
13. Kawakami K, Omura K, Kanehira E, Watanabe Y (1999) Poly-
morphic tandem repeats in the thymidylate synthase gene is
associated with its protein expression in human gastrointestinal
cancers. Anticancer Res 19(4B):3249–3252
14. Kaneda S, Nalbantoglu J, Takeishi K, Shimizu K, Gotoh O, Seno T,
Ayusawa D (1990) Structural and functional analysis of the human
thymidylate synthase gene. J Biol Chem 265(33):20277–20284
15. Mandola MV, Stoehlmacher J, Muller-Weeks S, Cesarone G, Yu
MC, Lenz HJ, Ladner RD (2003) A novel single nucleotide
polymorphism within the 50 tandem repeat polymorphism of the
thymidylate synthase gene abolishes USF-1 binding and alters
transcriptional activity. Cancer Res 63(11):2898–2904
16. Kawakami K, Watanabe G (2003) Identification and functional
analysis of single nucleotide polymorphism in the tandem repeat
Mol Biol Rep
123
Author's personal copy
sequence of thymidylate synthase gene. Cancer Res 63(18):
6004–6007
17. Ulrich CM, Bigler J, Velicer CM, Greene EA, Farin FM, Potter
JD (2000) Searching expressed sequence tag databases: discovery
and confirmation of a common polymorphism in the thymidylate
synthase gene. Cancer Epidemiol Biomark Prev 9(12):1381–1385
18. Lecomte T, Ferraz JM, Zinzindohoue F, Loriot MA, Tregouet
DA, Landi B, Berger A, Cugnenc PH, Jian R, Beaune P, Laurent-
Puig P (2004) Thymidylate synthase gene polymorphism predicts
toxicity in colorectal cancer patients receiving 5-fluorouracil-
based chemotherapy. Clin Cancer Res 10(17):5880–5888. doi:10.
1158/1078-0432
19. Dotor E, Cuatrecases M, Martinez-Iniesta M, Navarro M, Vilar-
dell F, Guino E, Pareja L, Figueras A, Mollevi DG, Serrano T, de
Oca J, Peinado MA, Moreno V, Germa JR, Capella G, Villanueva
A (2006) Tumor thymidylate synthase 1494del6 genotype as a
prognostic factor in colorectal cancer patients receiving fluoro-
uracil-based adjuvant treatment. J Clin Oncol 24(10):1603–1611.
doi:10.1200/JCO.2005.03.5253
20. Mountain CF (1986) A new international staging system for lung
cancer. Chest 89(4 Suppl):225S–233S. doi:10.1378/chest.89.4_
Supplement.225S
21. Mountain CF (1997) Revisions in the international system for
staging lung cancer. Chest 111(6):1710–1717
22. Excoffier L, Laval G, Schneider S (2005) Arlequin (version 3.0):
an integrated software package for population genetics data
analysis. Evol Bioinform Online 1:47–50. doi:10.4137/EBO.S0
23. Schaid DJ, Rowland CM, Tines DE, Jacobson RM, Poland GA
(2002) Score tests for association between traits and haplotypes
when linkage phase is ambiguous. Am J Hum Genet 70(2):
425–434. doi:10.1086/338688
24. Lurje G, Zhang W, Yang D, Groshen S, Hendifar AE, Husain H,
Nagashima F, Chang HM, Fazzone W, Ladner RD, Pohl A, Ning
Y, Iqbal S, El-Khoueiry A, Lenz HJ (2008) Thymidylate synthase
haplotype is associated with tumor recurrence in stage II and
stage III colon cancer. Pharmacogenet Genomics 18(2):161–168.
doi:10.1097/FPC.0b013e3282f4aea6
25. Stephens M, Donnelly P (2003) A comparison of bayesian
methods for haplotype reconstruction from population genotype
data. Am J Hum Genet 73(5):1162–1169. doi:10.1086/379378
26. Johnston PG, Lenz HJ, Leichman CG, Danenberg KD, Allegra
CJ, Danenberg PV, Leichman L (1995) Thymidylate synthase
gene and protein expression correlate and are associated with
response to 5-fluorouracil in human colorectal and gastric tumors.
Cancer Res 55(7):1407–1412
27. Lenz HJ, Hayashi K, Salonga D, Danenberg KD, Danenberg PV,
Metzger R, Banerjee D, Bertino JR, Groshen S, Leichman LP,
Leichman CG (1998) p53 Point mutations and thymidylate syn-
thase messenger RNA levels in disseminated colorectal cancer:
an analysis of response and survival. Clin Cancer Res 4(5):
1243–1250
28. Edler D, Glimelius B, Hallstrom M, Jakobsen A, Johnston PG,
Magnusson I, Ragnhammar P, Blomgren H (2002) Thymidylate
synthase expression in colorectal cancer: a prognostic and pre-
dictive marker of benefit from adjuvant fluorouracil-based che-
motherapy. J Clin Oncol 20(7):1721–1728
29. Kawakami K, Ishida Y, Danenberg KD, Omura K, Watanabe G,
Danenberg PV (2002) Functional polymorphism of the thymi-
dylate synthase gene in colorectal cancer accompanied by fre-
quent loss of heterozygosity. Jpn J Cancer Res 93(11):1221–1229
30. DiPaolo A, Chu E (2004) The role of thymidylate synthase as a
molecular biomarker. Clin Cancer Res 10(2):411–412. doi:10.
1158/1078-0432
31. Kristensen MH, Pedersen PL, Melsen GV, Ellehauge J, Mejer J
(2010) Variants in the dihydropyrimidine dehydrogenase,
methylenetetrahydrofolate reductase and thymidylate synthase
genes predict early toxicity of 5-fluorouracil in colorectal cancer
patients. J Int Med Res 38(3):870–883. doi:10.1177/1473230010
03800313
32. Brody JR, Hucl T, Gallmeier E, Winter JM, Kern SE, Murphy
KM (2006) Genomic copy number changes affecting the thymi-
dylate synthase (TYMS) gene in cancer: a model for patient
classification to aid fluoropyrimidine therapy. Cancer Res
66(19):9369–9373. doi:10.1158/0008-5472.CAN-06-2165
33. Etienne MC, Chazal M, Laurent-Puig P, Magne N, Rosty C,
Formento JL, Francoual M, Formento P, Renee N, Chamorey E,
Bourgeon A, Seitz JF, Delpero JR, Letoublon C, Pezet D, Milano
G (2002) Prognostic value of tumoral thymidylate synthase and
p53 in metastatic colorectal cancer patients receiving fluoroura-
cil-based chemotherapy: phenotypic and genotypic analyses.
J Clin Oncol 20(12):2832–2843
34. Uchida K, Hayashi K, Kawakami K, Schneider S, Yochim JM,
Kuramochi H, Takasaki K, Danenberg KD, Danenberg PV (2004)
Loss of heterozygosity at the thymidylate synthase (TS) locus on
chromosome 18 affects tumor response and survival in individuals
heterozygous for a 28-bp polymorphism in the TS gene. Clin
Cancer Res 10(2):433–439. doi:10.1158/1078-0432.CCR-0200-03
35. Wang TL, Diaz LA Jr, Romans K, Bardelli A, Saha S, Galizia G,
Choti M, Donehower R, Parmigiani G, Shih Ie M, Iacobuzio-
Donahue C, Kinzler KW, Vogelstein B, Lengauer C, Velculescu
VE (2004) Digital karyotyping identifies thymidylate synthase
amplification as a mechanism of resistance to 5-fluorouracil in
metastatic colorectal cancer patients. Proc Natl Acad Sci USA
101(9):3089–3094. doi:10.1073/pnas.0308716101
36. Ooyama A, Okayama Y, Takechi T, Sugimoto Y, Oka T,
Fukushima M (2007) Genome-wide screening of loci associated
with drug resistance to 5-fluorouracil-based drugs. Cancer Sci
98(4):577–583. doi:10.1111/j.1349-7006.2007.00424.x
37. Jakobsen A, Nielsen JN, Gyldenkerne N, Lindeberg J (2005)
Thymidylate synthase and methylenetetrahydrofolate reductase
gene polymorphism in normal tissue as predictors of fluorouracil
sensitivity. J Clin Oncol 23(7):1365–1369. doi:10.1200/JCO.
2005.06.219
38. Li WJ, Jiang H, Fang XJ, Ye HL, Liu MH, Liu YW, Chen Q, Zhang
L, Zhang JY, Yuan CL, Zhang QY (2013) Polymorphisms in thy-
midylate synthase and reduced folate carrier (SLC19A1) genes
predict survival outcome in advanced non-small cell lung cancer
patients treated with pemetrexed-based chemotherapy. Oncol Lett
5(4):1165–1170. doi:10.3892/ol.2013.1175ol-05-04-1165
39. Wang X, Wang Y, Cheng J, Ha M (2013) Association of thy-
midylate synthase gene 30-untranslated region polymorphism
with sensitivity of non-small cell lung cancer to pemetrexed
treatment: TS gene polymorphism and pemetrexed sensitivity in
NSCLC. J Biomed Sci 20:5. doi:10.1186/1423-0127-20-51423-
0127-20-5
40. Mandola MV, Stoehlmacher J, Zhang W, Groshen S, Yu MC,
Iqbal S, Lenz HJ, Ladner RD (2004) A 6 bp polymorphism in the
thymidylate synthase gene causes message instability and is
associated with decreased intratumoral TS mRNA levels. Phar-
macogenetics 14(5):319–327
41. Merkelbach-Bruse S, Hans V, Mathiak M, Sanguedolce R,
Alessandro R, Ruschoff J, Buttner R, Houshdaran F, Gullotti L
(2004) Associations between polymorphisms in the thymidylate
synthase gene, the expression of thymidylate synthase mRNA and
the microsatellite instability phenotype of colorectal cancer.
Oncol Rep 11(4):839–843
42. Ulrich CM, Bigler J, Bostick R, Fosdick L, Potter JD (2002)
Thymidylate synthase promoter polymorphism, interaction with
folate intake, and risk of colorectal adenomas. Cancer Res
62(12):3361–3364
43. Graziano F, Kawakami K, Watanabe G, Ruzzo A, Humar B,
Santini D, Catalano V, Ficarelli R, Merriman T, Panunzi S, Testa
Mol Biol Rep
123
Author's personal copy
E, Cascinu S, Bearzi I, Tonini G, Magnani M (2004) Association
of thymidylate synthase polymorphisms with gastric cancer sus-
ceptibility. Int J Cancer 112(6):1010–1014. doi:10.1002/ijc.
20489
44. Kawakami K, Graziano F, Watanabe G, Ruzzo A, Santini D,
Catalano V, Bisonni R, Arduini F, Bearzi I, Cascinu S, Muretto P,
Perrone G, Rabitti C, Giustini L, Tonini G, Pizzagalli F, Magnani
M (2005) Prognostic role of thymidylate synthase polymorphisms
in gastric cancer patients treated with surgery and adjuvant
chemotherapy. Clin Cancer Res 11(10):3778–3783. doi:10.1158/
1078-0432.CCR-04-2428
45. Fernandez-Contreras ME, Sanchez-Hernandez JJ, Gonzalez E,
Herraez B, Dominguez I, Lozano M, Garcia De Paredes ML,
Munoz A, Gamallo C (2009) Combination of polymorphisms
within 50 and 30 untranslated regions of thymidylate synthase
gene modulates survival in 5 fluorouracil-treated colorectal can-
cer patients. Int J Oncol 34(1):219–229. doi:10.3892/ijo_
00000144
Mol Biol Rep
123
Author's personal copy