DNA repair genes polymorphisms in multiple myeloma: no association with XRCC1 (Arg399Gln) polymorphism, but the XRCC4 (VNTR in intron 3 and G-1394T) and XPD (Lys751Gln) polymorphisms is associated with the disease in Turkish patients

Abstract

Objective: This study aims to investigate the association between the polymorphisms in DNA repair genes (XPD, XRCC1, and XRCC4) and clinical parameters in patients with multiple myeloma (MM), their effects on prognosis and their roles in susceptibility to MM.

Patients and methods: Sixty patients, diagnosed with MM and 70 individuals as the healthy control group were included in the study. Gene polymorphisms were detected with the polymerase chain reaction and/or polymerase chain reaction–restriction fragment length polymorphism method. When the genotype frequencies of XPD (Llys751Gln) and XRCC1 (Arg399Gln) genes were examined in the patient and control groups, no significant difference was detected, while a significant association was found in XRCC4 (VNTR in intron 3 and G-1394T) polymorphisms. A significant association was found in the MM patients group for AA genotype and event-free survival (EFS) in terms of XPD (751) gene polymorphism (P = 0·047). When VNTR intron 3 polymorphism was compared for genotype frequency, DD genotype was found to be significantly low (P = 0·012) in the patient group, whereas GG and TT genotypes were found to be significantly lower in the patient group for the genotype frequency XRCC4 (G-1394T) polymorphism when compared to the control group (P = 0·015, P = 0·010, respectively).

Results: These data provide support for the hypothesis that a common variation in the genes encoding XRCC4 DNA repair proteins may contribute to susceptibility to myeloma. These findings require further validation in independent populations.

Keywords:

• Multiple myeloma
• XRCC1
• XPD
• XRCC4
• Polymorphism

Introduction

Multiple myeloma (MM) is a B-cell lymphoproliferative disease in which, in general, progessive immunoglobulin generation with increased malignant plasma cells in the bone marrow is seen, and is generally characterized with osteolytic lesions, hypercalcemia, anemia, monoclonal protein increase, and renal failure.¹1,2 It has been suggested that radiation (short-term high dose or long-term low dose), viral agents, chronic antigenic stimulus, and environmental factors have a role in MM etiology.3

Two oncogenic pathways are believed to play central role in myeloma pathogenesis. First, multiple trisomies of chromosomes; 3, 5, 7, 9, 11, 15, 19 and 21 are related to disease pathogenesis. Second, primary translocation involving heavy chain locus at 14q32 and a limited number of recurrent locus are present.4 These translocations represent abnormal class switch recombination which normally functions to alter immunglobuline isotype with the maturation of the immune response. Disregulated class switch recombination results in the placing of a number of proto-oncogenes and is believed to play a role in myeloma pathogenesis.5

Minor damages in the DNA molecule are generally cleared by DNA repair genes. Under normal circumstances, high level DNA damages leads to cell death by stimulating apoptosis.6 Xeroderma pigmentosum group D (XDP) gene is a DNA repair gene and is found in the nucleotide excision repair (NER) group. During NER, some chemically altered, mismatched or unsuitable bases are excised from the genome, and bases in correct order are replaced. Transformation of lysine in 751th codon of XDP gene to glutamine alters the response to treatment. NER mechanism mostly participates in DNA damage due to ultraviolet light and polycyclic hydrocarbonates.7 X-ray repair cross-complementing group 1 and 4 (XRCC1 and XRCC4) are DNA repair genes in subgroup of the base excision repair (BER) mechanism. BER mechanism participates in DNA damage due to X rays, oxygen radicals and alkylating agents, while mismatch repair is assigned in reparation of replication failures.8 In XRCC1 base damages and one-stranded DNA breaks, DNA functions by binding to damaged DNA part of poly(ADP-ribose) polymerase and DNA polymerase b, and to carboxyl of Ligase III.9

Polymorphisms causing changes in DNA repair capacity have been known to hold an important place in MM etiology and the response to the treatment. Both karyotypic and evidence of DNA repair abnormalities take place in myeloma pathogenesis, and we selected three genes which are involved in DNA repair as the basis for this study. XRCC4 has also been implicated in myeloma pathogenesis.10 Therefore, we investigated XPD (Lys751Gln), XRCC1 (Arg399Gln) polymorphisms and also XRCC4 (VNTR in intron 3 and G-1394T) for assessment of myeloma pathogenesis and relation to clinical parameters.

Patients and Methods

Sixty patients, consecutively diagnosed with MM between 2004 and 2008 in the Section of Hematology at the Department of Internal Medicine in the Faculty of Medicine at Gaziantep University, and 70 individuals as the healthy control group were included in the study. The diagnosis was based on clinical criteria, serum protein electrophoresis, immune electrophoresis and immune fixation of serum and urine, and bone marrow examination. The study was approved by the regional ethics committee, and informed consent was obtained from each participant before interview and blood sampling. DNA isolation was performed using 2 cc of blood sample in tubes with EDTA for the control and from bone marrow smear preparations for the patient group at diagnosis. The control group was composed of individuals with no previous a disease history and with completely normal clinical and laboratory findings. DNA isolation was performed using Invitrogen Pure Link (Catalogue Nr: K1820-01) kits in patient group, and control group. The DNA samples were stored at 20°C. Gene polymorphisms were detected with the polymerase chain reaction and/or polymerase chain reaction–restriction fragment length polymorphism method¹¹11,12 (Table 1).

Table 1. XRCC1 (A399G), XPD (−751), XRCC4 (VNTR in intron 3 and G-1394T) primer sequences, restriction enzymes and amplification conditions (on over night incubation)

Repair genes	Primer sequence	Annealing	Restriction enzyme and conditions
XRCC1 (Arg399Gln)	5′-AGT AGT CTG CTG GCT CTG G-3′, 5′-TCT CCC TTG GTC TCC AAC CT-3	55°C ve 32 cycles	Msp1, 37°C – ON and 2% agarose
XPD (Lys751Gln)	5′-ATC CTG TCC CTA CTG GCC ATT C-3′, 5′-TGT GGA CGT GAC AGT GAG AAA T-3′	60°C ve 32 cycles	Pst1, 37°C – ON and 2·5% agarose
XRCC4 (VNTR in intron 3)	5′-TCC TGT TAC CAT TTC AGT GTT AT-3′, 5′-CAC CTG TGT TCA ATT CCA GCT T-3′	55°C ve 32 cycles	3% agarose
XRCC4 (G-1394T)	5′-GAT GCG AAC TCA AAG ATA CTG A-3′, 5′-TGT AAA GCC AGT ACT CAA ACT T-3′	55°C ve 32 cycles	HincII, 37°C – ON and 2·5% agarose

The analyses of data were performed using the software SPSS for Windows (version 13·0; SPSS, Chicago, IL, USA). The statistical significance of the differences between the patient and the control groups was estimated by logistic regression analysis. Adjusted odds ratios (ORs) were calculated with a logistic regression model that controlled for gender and age and were reported at 95% confidence intervals. Differences in DNA Repair Gene XPD, XRCC1, XRCC4 (VNTR in intron 3 and 1394) allele genotype frequency between the control group and the patient groups were compared using the χ²test and, when needed, Fisher exact test was used. Hardy–Weinberg equation was used to calculate estimated genotype frequency and the experienced genotype frequency.13 Survival probabilities were estimated by the Kaplan–Meier method and differences were compared using the log rank test.14 The Cox stepwise regression analysis was employed to confirm significance of risk factors.15 In multivariate analysis, we used eliminated variables stepwise (backward) by a significance of less than 10%. P values <0·05 were considered to indicate statistical significance.

Results

In the study, 34 (57%) of the cases were men, and 26 (43%) of them were women. The average age was 59 years (36–80 years). In the control group there were 37 (53%) men, and 33 (47%) women out of 70 healthy volunteers. The average age of the control group was 41 years (20–66 years). Results relating to XPD, XRCC1, XRCC4 gene polymorphisms of the 60 patients diagnosed with MM are shown in Table 1. Paraprotein types detected in MM patient group, stage (according to Durie–Salmon classification), Eastern Cooperative Oncology Group (ECOG) performance scale, demographic and other clinical characteristics of the patients are seen in Table 2. Eight of the patients who were assessed died (infection, heart failure, cerebrovascular accident, gastrointestinal and central nervous system bleeding) shortly after their hospitalization, and were not included in the survival assessment.

Table 2. Clinical characteristics and treatment regimens of multiple myeloma patients

		Multiple myeloma		Healthy control
		Median	n* (%)	n* (%)
Age		59 (36–80)		41 (20–66)
Gender	Female/male		26/34 (43/57)	33/37 (47/53)
Ig subtypes	κ/λ		44/16 (73/27)
	G/A		36/24 (60/40)
	Light chain		10 (17)
Stage (Salmon–Durie)	II/III		19/41 (32/68)
	A/B		45/15 (75/25)
IPI	I		16 (27)
	II/III		23/21 (38/35)
ECOG	>1		14 (23)
Hemoglobin	g/dl	9·8 (6·0–14·7)
Leukocytes	μl	6770 (1620–15160)
Trombocytes	10^3/μl	167 (26–700)
C-reactive protein	mg/dl	6·65 (2·9–70)
LDH	IU/l	386 (164–2540)
b2-microglobulin	mg/l	4·2 (1·7–27)
Albumin	g/l	3·3 (2·1–4·7)
Treatment (n = 52)	1st line (VAD, MP)		47/5
	2nd line (ASCT/T, BODEC, TD)		17/17/13
	3rd line (ASCT/T, BODEC, TD)		6/15/3
Mortality			18 (30)
Median follow-up (months)		30·2 (6·6–85·2)

Note: *n = 60.

VAD, vincristine, adriamycin and dexamethasone; MP, melphalan, prednisolone; BODEC, bortezomib, dexamethasone, cyclosphosphamide; TD, thalidomide, dexamethasone; ASCT, autologous stem cell transplantation; ECOG, performance evaluation at diagnosis; LDH, lactic dehydrogenase; IPI, International prognostic index.

When XPD (Lys751Gln) gene polymorphism results were compared in both control (P = 0·124) and patient group (P = 0·233) a deviation from Hardy Weinberg Equilibrium (HWE) was not detected. When XRCC1 (Arg399Gln) gene polymorphism results were compared in both patient (P = 0·0011) and control group (P = 0·009) a deviation from HWE was detected. When XRCC4 (VNTR in intron 3) gene polymorphism results were compared a deviation from HWE in control group (P = 0·002) was not detected while in patient group (P = 0·002) a deviation from HWE was detected. When XRCC4 (G-1394T) gene polymorphism results were compared in control group (P = 0·043) there was a deviation from HWE, but a deviation was not detected in patient group (P = 0·497). When genotype frequencies obtained from DNA repair gene polymorphisms in MM patient and control group were compared, a significant correlation in XDP (Lys751Gln) and XRCC1 (Arg399Gln) polymorphism results was not found while such a relationship was detected in XRCC4 (VNTR intron 3 and G-1394T) polymorphisms (Table 3).

Table 3. Comparison of genotype of XPD, XRCC1, XRCC4 gene polymorphisms between patients with multiple myeloma and healthy controls

	Genotype	MM	Healthy control	OR	95%CI	P
		n* (%)	n† (%)
XPD	AA	25 (41·7)	37 (52·9)	0·520	0·157–1·719	0·283
(Arg399Gln)	AC	24 (40·0)	24 (34·3)	1·066	0·310–3·669	0·919
	CC	11 (18·3)	9 (12·8)	0·657	0·252–1·713	0·468
XRCC1	AA	23 (38·3)	25 (35·7)	1·005	0·351–2·879	0·993
(Lys751Gln)	AG	20 (33·3)	24 (34·3)	1·167	0·404–3·374	0·776
	GG	17 (28·4)	21 (30)	1·084	0·507–2·316	0·849
XRCC4	DD	1 (1·7)	10 (14·3)	0·043	0·004–0·504	0·012
(VNTR in intron 3)	DI	36 (60·0)	42 (60)	1·000	0·495–2·022	1·000
	II	23 (38·3)	18 (25·7)	0·390	0·145–1·047	0·062
XRCC4	GG	20 (33·3)	17 (24·3)	4·398	1·338–14·461	0·015
(G-1394T)	GT	27 (45·0)	26 (37·1)	0·722	0·358–1·458	0·377
	TT	13 (21·7)	27 (38·6)	4·481	1·427–14·074	0·010

Note: *n = 60.

†n = 70.

‡OR (95%CI) was adjusted by age and sex.

§Fisher’s Exact Test.

Using International Myeloma Working Group Uniform Response Criteria,16 initial treatment and treatment results of patients whose response evaluations are complete are seen in Table 4. In univariate analyses of MM patients, for male gender 6-years median EFS was found significant, while in patients whose IPI score was I/III, ECOG performance scale was <1/>1 and platelet number was <150×10⁹/l/⩾150×10⁹/l, 6-years OS was found significant (respectively P = 0·044, P = 0·037, P = 0·033) (Fig. 1). While in XPD (Lys751Gln) gene group AA genotype was found insignificant (P = 0·061) in comparison to AC and CC genotypes in terms of 6-years median OS, it was found significant (P = 0·047) (Table 5) (Fig. 2) in terms of EFS. However, in multivariate analyses, a significant correlation was not found in terms of male gender, IPI score, ECOG performance scale, platelet number, XDP (751) AA genotype frequency, OS and EFS. In MM patients, a significant correlation was not detected between XDP, XRCC1, XRCC4 gene polymorphisms and beta2 microglobulin, disease stage, paraprotein type, gender, lactic dehydrogenase, plasma cell ratio, seasonal relation of the disease and leukocyte number. Also there was no association of age with XDP, XRCC1, XRCC4 gene polymorphisms.

Figure 1. Kaplan–Meier plots on overall survival (OS) time according to the type of IPI risk score at diagnosis.

Figure 2. Kaplan–Meier plots on event-free survival (EFS) time according to the type of XPD (Lys751Gln) genotypes.

Table 4. Initial treatment modalities and treatment response of myeloma patients

		n	Median OS mo (6 yr %)	Log rank P value	Median EFS mo (6 yr %)	Log rank P value
All patients		52	65		34·5
Initial treatment	VAD+ASCT	17	73·4		38·8
	VAD or M+BODEC	16	(61)		20·6
	VAD or M+TD	13	65	0·279	33·5	0·324
Initial treatment	CR	15	73·2		62·7
Response	VGPR	9	(88)		57·3
	PR	16	37·9		24·6
	SD+PROG	6	56·3	0·050	14·8	0·005

Note: OS, overall survival; EFS, event-free survival; VAD, vincristine, adriamycin and dexamethasone; MP, melphalan, prednisolone; BODEC, bortezomib, dexamethasone, cyclosphosphamide; TD, thalidomide, dexamethasone; ASCT, autologous stem cell transplantation.

Table 5. Univariate analysis (Log rank test) of XPD, XRCC1, XRCC4 − intron 3 and XRCC4 – 1394 Gene Polymorphisms in the patients with multiple myeloma

		n	Median OS mo (6 yr %)	Log rank P value	Median EFS mo (6 yr %)	Log rank P value
All patients		52	65		34·5
Gender	Female/male	21/31	(57)/65	0·908	30·1/48·7	0·028
Age	<65/⩾65	37/15	65/53·2	0·671	34·5/35	0·375
Stage (Salmon–Durie)	II/III	19/33	(81)/56·3	0·148	33·5/48·7	0·155
	A/B	39/13	73·2/65	0·139	40·2/33·5	0·392
IPI (ISS)	I	14	(86)		57·3
	II	19	(56)		33·5
	III	19	56·3	0·044	29·7	0·131
Ig subtypes	κ/λ	38/14	67·1/73·4	0·182	35/20·6	0·300
	G	30	65		35
	A	12	67·1		67·1
	Light chain	10	73·4	0·744	29·4	0·232
ECOG	⩽1/>1	43/9	73·2/53·2	0·037	33·5/48·7	0·452
Platelets (×10^9/l)	<150/⩾150	14/38	37·9/73·2	0·033	24·6/38·8	0·479
LDH (IU/l)	<480/⩾480	41/11	53·2/67·1	0·278	34·5/38·8	0·352
C-reactive protein (mg/l)	<5/⩾5	19/33	37·9/65	0·497	35/38·8	0·820
Initial treatment	VAD+ASCT	17	73·4		38·8
	VAD or M+BODEC	16	(61)		20·6
	VAD or M+TD	13	65	0·279	33·5	0·324
Initial treatment	CR	15	73·2		62·7
Response	VGPR	9	(88)		57·3
	PR	16	37·9		24·6
	SD+PROG	6	56·3	0·050	14·8	0·005
XPD (Lys751Gln)	AA	21	73·2		48·7
	AC	20	37·9		30·1
	CC	11	67·1	0·136	20·6	0·137
	AA/AC, CC	21/31	73·2/56·3	0·061	29·7/47·7	0·047
XRCC1 (Arg399Gln)	AA	19	65		48·7
	AG	18	56·3		27·2
	GG	15	67·1	0·236	35	0·081
	AA/AG, GG	19/33	65/56·3	0·116	29·7/47·7	0·067
XRCC4 – intron 3	DD	1	73·2		62·7
	DI	32	56·3		34·5
	II	19	65	0·980	38·8	0·867
XRCC4 (G-1394T)	GG	19	63		35
	GT	21	56·3		29·7
	TT	12	65	0·238	62·7	0·194

Note: *Median (year).

OS, overall survival; EFS, event-free survival; VAD, vincristine, adriamycin and dexamethasone; MP, melphalan, prednisolone; BODEC, bortezomib, dexamethasone, cyclosphosphamide; TD, thalidomide, dexamethasone; ASCT, autologous stem cell transplantation; ECOG, performance evaluation at diagnosis; LDH, lactic dehydrogenase; IPI; International prognostic index.

Discussion

Correlations between polymorphisms in DNA repair genes and the risk of developing various cancers using meta-analyses has been shown. Defective DNA repair has also been implicated directly as an etiological factor in hematological malignancies.17,18 Our study shows that in MM patients, XDP (Lys751Gln), XRCC1 (Arg399Gln), XRCC4 (VNTRintron 3 and G-1394T) gene polymorphisms functioning in DNA repair systems have no effect on treatment protocol, prognosis, and survival; however, XDP (Lys751Gln) AA genotype is significant in terms of 6-years EFS (Fig. 2, Tables 4 and 5). It was found that DD genotype in XRCC4 (VNTR intron 3) and GG and TT genotype in XRCC4 (G-1394T) are significantly low in the patient group (Table 3).

In some of the studies on XPD (Lys751Gln) polymorphism, it was noted that the XPD Lys allele is associated with an increased risk for lung cancer, melanoma and basal cell carcinoma.19 A correlation between XPD Gln allele and acute myeloblastic leukemia developing following chemotherapy was found, while a correlation between other prognostic factors (age, gender, WBC number, FAB class, response to treatment) and XPD (Lys751Gln) polymorphism was not detected.²⁰20,21 In follicular lymphoma patients, ERCC2 and ERCC1 gene polymorphisms was not associated with the disease; however, it was stated that follicular lymphoma patients who are smoking and XRCC3 gene variations can be related to the disease.22In our study, in 60 MM patients XDP (Lys751Gln) gene distribution was as following: AA: 25 (41·7%), AC: 24 (40%), CC: 11 (18·3%); a deviation from HWE was not detected in control groups. A significant disparity between MM patient group (P = 0·12) and control group (P = 0·22) was not detected in terms of genotype frequency. In our study, in XPD (Lys751Gln) gene group, AA genotype was not found significant (P = 0·061) in comparison with AC and CC genotype in terms of 6-years median OS; however, it was found significant (P = 0·047) in terms of EFS (Table 5, Fig. 2). In this condition, in MM patients, XPD (Lys751Gln) AA genotype can be said to play a protective role. However, in multivariate analyses, a significant correlation was not detected in terms of XPD (751) AA genotype frequency OS and EFS.

In a study in which the relationship between XRCC1 polymorphisms (codon 194, 280, and 399) and acute lymphoblastic leukemia (ALL) risk was investigated, it was detected that only XRCC1 Gln variant allele constitutes a risk for acute lymphoblastic leukemia.23 XRCC1 (A194T allele) polymorphism is associated with an insignificant increase in NHL risk, while this allele was noted to be associated with a decrease in diffuse large B-cell lymphoma. In another study, it was reported that a correlation between XRCC1 (Arg339 Gln) polymorphism and NHL subtypes did not exist.²⁴24,25In our study, in 60 patients diagnosed with MM, XRCC1 (Arg399Gln) gene polymorphism distribution was as following: AA: 23 (38·3%), AG: 20 (33·3%), GG: 17 (28·4%); and in both control (P = 0·009) and patient group (P = 0·0011) a deviation from HWE was detected. As a result of the comparison of MM patient group to control group, a significant disparity between XRCC1 codon (Arg399Gln) gene polymorphisms and patient group was not found in terms of genotype frequency and clinical parameters (Table 5).

In XRCC4 G-1394T gene polymorphism, an increased correlation between T/T genotype and gastric cancer was detected, although it was explained that XRCC4 (247, intron 3 and intron 7) gene polymorphism was not associated with gastric cancer.²⁶26,27In a study by Hayden et al. in which XRCC3, XRCC4, and XRCC5 polymorphisms were investigated and in which 307 MM and 263 control group patients were participated, single nucleotide polymorphisms were studied, and it was noted that A allele (P = 0·0133) in rs963248 numbered single nucleotide polymorphisms in XRCC4 gene, and AA genotype (P = 0·026) increased significantly in patient group in comparison with control group. In this case, they also reported that it increased susceptibility to MM.10 In our study, out of 60 cases, XRCC4 (VNTR intron 3) polymorphism distribution was detected as: DD: 1 (1·7%), DI: 36 (60%), II: 23 (38·3%). While a deviation from HWE in control group and in the MM patient group was not detected. When patient and control group were compared in terms of DD, DI, and II genotypes, it was determined that DD genotype was found to be significantly low (P = 0·012) in patient group in comparison with healthy group. In this case, it can be implied that as DD genotype decreases, susceptibility to MM increases, and protectiveness from MM disappears as DI and II increases.

In our study, in patient group XRRC4 (G-1394T) gene polymorphism distribution was detected as: GG, 20 (33%); GT, 27 (45%); TT, 13 (21·7%). When a healthy control group and MM patient group were compared, in terms of genotype frequency GG and TT genotype were found to be significantly low in patient group in comparison with healthy control group (Table 3). We believe that this decrease in GG and TT genotype may increase the susceptibility to MM, and its high incidence in control group may be shown as a protection-creating situation against tumor.

Conclusion

In this study, XPD (Lys751Gln) and XRCC1 (Arg399Gln) gene polymorphisms functioning in BER and NER DNA repair system in MM patients were investigated. The polymorphisms of these areas were found to be significant, and we believe the genes’ different loci may be related to MM pathogenesis and these different loci should be further investigated. In conclusion, our data contribute to the hypothesis that variations in genes encoding for DNA repair proteins increases the susceptibility to MM. These results need to be replicated in larger numbers of patients to confirm and elucidate their role in myeloma pathogenesis.

This research study was supported by Gaziantep University Medical School Research Council. (Project No. TF 09·03).