Journal of Cancer Research and Therapeutics

: 2018  |  Volume : 14  |  Issue : 8  |  Page : 46--53

Hypoxia-inducible factor-1α rs11549465 C>T and rs11549467 G>A gene polymorphisms are associated with an increased risk of digestive cancers in Asians

Wu Wang, Bu-Qiang Wu, Guang-Bin Chen, Yong Zhou, Zhao-Hua Li, Jian-Liang Zhang, Yin-Lu Ding, Peng Zhang, Jin-Qing Wang 
 Department of General Surgery, Heping Hospital of Changzhi Medical College, Changzhi 046000, Shanxi Province, P.R. China

Correspondence Address:
Guang-Bin Chen
Department of General Surgery, Heping Hospital of Changzhi Medical College, No. 110, Yan'an South Road, Changzhi 046000, Shanxi Province
P.R. China


Objective: We used a meta-analysis framework to examine the correlation between HIF-1α gene polymorphisms and the susceptibility to digestive cancers. Methods: Cochrane Library Database, EMBASE, MEDLINE, Pubmed, CINAHL, Chinese Biomedical Database and Web of Science were searched without language restrictions to identify relevant case-control studies reporting data on HIF-1α gene polymorphisms in digestive cancers. Data was extracted from the selected studies and meta-analysis was carried out using STATA 12.0 and Comprehensive Meta-analysis 2.0 softwares. Relative risk (RR) and its 95% confidence interval (95%CI) were calculated. A total of 8 eligible case-control studies were included. These 8 studies contained a combined total of 1,276 patients diagnosed with various digestive cancers and 3,392 healthy controls. Two functional HIF-1α polymorphisms (rs11549465 C>T and rs11549467 G>A) were examined in these 8 studies. Results: Our findings demonstrated that both rs11549465 C>T and rs11549467 G>A HIF-1α polymorphisms conferred significantly increased risk of digestive cancers. However, ethnicity-stratified analysis revealed that HIF-1α rs11549465 C>T and rs11549467 G>A polymorphisms were associated with an elevated risk of digestive cancer in Asians, but not in Caucasians. These two polymorphisms also conferred different degrees of susceptibility to various digestive cancer types. Conclusion: Our meta-analysis suggests that HIF-1α rs11549465 C>T and rs11549467 G>A polymorphisms influence the pathogenesis of digestive cancers in Asians.

How to cite this article:
Wang W, Wu BQ, Chen GB, Zhou Y, Li ZH, Zhang JL, Ding YL, Zhang P, Wang JQ. Hypoxia-inducible factor-1α rs11549465 C>T and rs11549467 G>A gene polymorphisms are associated with an increased risk of digestive cancers in Asians.J Can Res Ther 2018;14:46-53

How to cite this URL:
Wang W, Wu BQ, Chen GB, Zhou Y, Li ZH, Zhang JL, Ding YL, Zhang P, Wang JQ. Hypoxia-inducible factor-1α rs11549465 C>T and rs11549467 G>A gene polymorphisms are associated with an increased risk of digestive cancers in Asians. J Can Res Ther [serial online] 2018 [cited 2019 Jul 23 ];14:46-53
Available from:

Full Text


Cancers arising within the gastrointestinal tract and accessory digestive organs are grouped into a broader category of digestive cancers. The most common types of digestive cancers are colorectal cancer, pancreatic cancer, esophageal cancer, gastric cancer, and primary liver or intrahepatic biliary cancer, ranked in order based on their incidence rates. In 2013, approximately 290,200 cases of digestive cancers were newly diagnosed, and they accounted for 144,570 deaths worldwide.[1] Although the specific types of digestive cancers show varying incidence rates and trends, depending on the geographical variations, digestive cancers collectively continued to rank among the top five cancers in the world based on data collected in the past three decades.[2] The etiology of digestive cancer is a result of a complex interaction of both genetic and environmental factors.[3],[4] Epidemiological studies have identified multiple risk factors including gender, physical inactivity, excessive alcohol consumption, obesity, and cigarette smoking, which confer increased susceptibility to digestive cancers.[1],[5] Previous studies have identified several important molecular mechanisms activated in digestive cancers and, in particular, the abnormal expression of hypoxia-inducible factor 1 (HIF-1) tightly correlates with the origin and progression of these cancers.[6],[7]

Hypoxia-inducible factor-1 is a member of the family of HIF transcription factors and is a heterodimeric protein composed of two subunits, HIF-1 α and HIF-1β.[8] The two subunits control a variety cellular pathways involved in embryonic development and in normal physiological processes such as apoptosis, hypoxia response, chemotaxis, and cell proliferation, which are also critical for cell survival, energy metabolism, angiogenesis and progression, and metastasis of tumors.[9],[10] In normal cells, HIF-1α is rapidly degraded under normoxic conditions, preventing its cellular accumulation, but defects in HIF-1α degradation and its over-accumulation is observed in many human cancers including breast, lung, skin, ovarian, and renal cancer, leading to uncontrolled intratumoral hypoxic response, which includes persistent stimulation of angiogenic pathways and increased demand for oxygen supply.[11],[12],[13] Digestive cancers also exhibit HIF-1α dysregulation.[14],[15],[16] In humans, the HIF1 α gene is located on chromosome 14q21-24 and consists of 15 exons and 14 introns.[17] Tumor vascularization is highly dependent on HIF-1α cellular levels, and hypoxia stabilizes HIF-1α, by preventing its degradation.[18] Under normoxia, HIF-1α undergoes posttranslational hydroxylation on Pro402, and Pro564 located within the oxygen-dependent degeneration domain (ODD), promoting ODD binding to von Hippel-Lindau tumor suppressor protein (pVHL).[19] This in-turn leads to rapid degradation of HIF-1 α via pVHL modulated ubiquitin-proteasome pathway.[12] Under hypoxia, however, hydroxylation of Pro564 and Pro402 is inhibited, resulting in HIF-1α protein accumulation, which dimerizes with HIF-1α to reconstitute the HIF1 transcription factor, eventually activating hypoxic response. This pathway is also exploited by several tumors and is important in the progression and development of digestive cancers.[20] Single nucleotide polymorphisms (SNPs) influence normal functioning of cellular pathways and are important genetic factors in cancers with complex inheritance.[21] Two common SNPs in the HIF-1 α gene result in amino acid substitutions in ODD namely rs11549465 C>T and rs11549467 G>A polymorphisms,[22] and are linked to the development of digestive cancers.[20],[23],[24],[25] On the other hand, other studies reported that HIF-1 α genetic variants do not have a significant impact on the development of digestive cancers.[20],[22] Therefore, we performed this meta-analysis to investigate the precise relationship between HIF-1 α genetic polymorphisms and the pathogenesis of digestive cancers.

 Materials and Methods

Literature search strategy

The Cochrane Library Database, MEDLINE, PubMed, EMBASE, Web of Science, the Chinese Biomedical Database, and CINAHL were searched with no language restrictions to identify case-control studies related to HIF-1 α genetic polymorphisms and digestive cancers. The following keywords were used as a highly sensitive search strategy: (“SNP” or “mutation” or “genetic polymorphism” or “variation” or “polymorphism” or “SNP” or “variant”) and (“digestive system neoplasms” or “cancer of digestive system” or “digestive cancer” or “gastric cancer” or “esophageal cancer” or “colorectal cancer” or “intestinal cancer” or “liver cancer” or “pancreatic cancer”) and (“hypoxia-inducible factor-1α” or “HIF-1 α”). A manual search of cross-references was also performed to find other relevant articles. Any disagreements in the search process were resolved through discussion between the authors.

Literature selection criteria

The inclusion criteria for the selection of studies for meta-analysis were: (1) Study must be related to the association between HIF-1 α SNPs and the pathogenesis of digestive cancers; (2) all patients must have pathologically confirmed diagnosis of digestive cancer; (3) studies should provide sufficient data related to the frequencies of HIF-1 α genetic polymorphisms; (4) the genotype distribution must conform to Hardy-Weinberg equilibrium (HWE). The study with the largest sample size or the most recent publication was included if same authors have published several studies with same subjects.

Data extraction and methodological assessment

With a standardized form, two authors independently extracted the following data from each selected trial systematically: First author, publication year, language, design of study, geographical location, the source of the subjects, sample size, genotyping method, allele frequencies, and evidence of HWE in healthy controls. The quality of the selected studies was measured using critical appraisal skill program (CASP) (!casp-tools-checklists/c18f8). The CASP criteria included 11 aspects: A clear focus (CASP01); an appropriate research question and a research design to the question (CASP02); acceptable method for case recruitment (CASP03); acceptable method for control selection (CASP04); accurate measurement for exposure factors to minimize bias (CASP05); controls with other important confounding factors (CASP06); completeness of the research result (CASP07); accuracy of the research result (CASP08); reliability of the research result (CASP09); applicability of the research result to the local population (CASP10); conformity of the research result with other available evidence (CASP11).

Statistical analysis

All statistical analyses were performed using STATA version 12.0 (Stata Corporation, College Station, TX, USA) and Comprehensive Meta-analysis 2.0 (Biostatic Inc., Englewood, New Jersey, USA). Relative risk (RR) and its 95% confidence interval (CI) (95% CI) were calculated. Z-test was applied to evaluate whether the pooled RRs was statistically significant. Heterogeneity across the studies were evaluated through Cochran's Q-statistic and I2 tests.[26] If Q-test showed a P < 0.05 or I2 test showed a P > 50%, indicating significant heterogeneity, a random-effect model was used, otherwise a fixed-effect model was applied. To investigate the reasons for heterogeneity, subgroup analyses were conducted. Sensitivity analysis was carried out to measure the effect of single studies on the overall results. Funnel plots, classic fail-safe N,[27],[28] and Egger's linear regression test [29] were applied to investigate publication bias. All tests were two-sided, and P < 0.05 was regarded as statistically significant.


Characteristics of selected trials

Following an exhaustive database search, a total of 129 articles were identified. Upon reviewing the titles and abstracts, 67 articles were excluded for various reasons. Full texts of the remaining articles were reviewed systematically, and an additional 54 studies were removed. Finally, 8 case-control trials, involving a pooled total of 1276 digestive cancer patients and 3392 healthy controls, were included in this meta-analysis.[8],[10],[18],[20],[22],[24],[25],[30] The publication year of the selected studies ranged between 2004 and 2011. Qualities of all selected studies are displayed in [Figure 1]. Of the 8 studies, 2 trials were conducted in Caucasians and 6 studies were performed in Asians. Polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP), TaqMan and MicroArray assays were the methods applied for genotyping in the 8 studies. Overall, only two functional polymorphisms (rs11549465 C>T and rs11549467 G>A) were examined in the 8 studies and none of the enrolled studies deviated from the HWE (all P > 0.05). Study characteristics and methodological quality are summarized in [Table 1].{Figure 1}{Table 1}

Meta-analysis results

Our meta-analysis results revealed that both rs11549465 C>T (T allele vs. C allele: RR = 1.36, 95% CI = 1.07–1.74, P= 0.012; CT + TT vs. CC: RR = 1.32, 95% CI = 1.02–1.70, P= 0.034, respectively) and rs11549467 G>A polymorphisms (A allele vs. G allele: RR = 1.96, 95% CI = 1.02–3.76, P= 0.042; GA + AA vs. GG: RR = 1.93, 95% CI = 1.02–3.75, P= 0.041, respectively) showed significant correlation with an increased risk of digestive cancers [Figure 2].{Figure 2}

Subgroup analyses were performed to examine the influence of ethnicity and cancer type on the meta-analysis results. Ethnicity-stratified analysis indicated a significant association of HIF-1 α rs11549465 C>T (T allele vs. C allele: I2 = 55.9%, P= 0.045; Z = 2.08, P= 0.046; CT + TT vs. CC: I2 = 62.1%, P= 0.022; Z = 2.16, P= 0.037, respectively) and rs11549467 G>A polymorphisms (A allele vs. G allele: I2 = 0.00%, P= 0.860; Z = 6.23, P < 0.001; GA + AA vs. GG: I2 = 0.00%, P= 0.848; Z = 6.36, P < 0.001, respectively) with an elevated risk of digestive cancers in Asians, but not in Caucasian population (T allele vs. C allele: I2 = 0.00%, P= 0.427; Z = 1.33, P= 0.184; CT + TT vs. CC: I2 = 0.00%, P= 0.576; Z = 1.03, P= 0.303, respectively, A allele vs. G allele: I2 = 0.00%, P= 0.327; Z = 0.50, P= 0.619; GA + AA vs. GG: I2 = 19.5%, P= 0.265; Z = 0.49, P= 0.622, respectively) [Figure 3]. Further, subgroup analysis based on cancer type revealed that HIF-1 α genetic polymorphisms (rs11549467 G>A) strongly correlated with significantly higher risks of gastric, liver and pancreatic cancers (all P < 0.05) [Figure 4]. However, we could not obtain a definitive relationship between colorectal cancer and HIF-1 α genetic polymorphisms in this study (all P > 0.05).{Figure 3}{Figure 4}

Sensitivity analyses were carried out to assess the impact of each single study on the summary RRs by deleting individual studies. The analysis results suggested that no single study markedly influenced the pooled RRs [Figure 5] indicating statistically stable results of our meta-analysis. Funnel plots suggested no detectable asymmetry [Figure 6] and Egger's test also revealed no significant publication bias (all P > 0.05).{Figure 5}{Figure 6}


In this meta-analysis, we investigated the relationship between HIF-1 α rs11549465 C>T and rs11549467 G>A polymorphisms and digestive cancer susceptibility. Our findings showed that both rs11549465 C>T and rs11549467 G>A SNPs were significantly linked with digestive cancer risk, suggesting that specific variants of HIF-1 α may be important candidate biomarkers for predicting digestive cancer susceptibility. HIF-1 α is rapidly degraded in normal cells under normoxic conditions, but several human cancers contain mutations that stabilize and/or over-express HIF-1 α in presence or absence of intratumoral hypoxia.[13] Moreover, genetic polymorphisms associated with inter-individual diversity significantly influence the clinical course of complex diseases.[31] However, from previous studies and in this study, the exact mechanisms by which HIF-1 α SNPs lead to an increased risk of digestive cancer are still not fully known. HIF-1 α is a subunit of the major transcription factor HIF that mediates cellular response to hypoxia and plays a leading role in angiogenesis, cell proliferation, cell invasion, and metastasis in pathological conditions, such as in aggressive tumors.[18] Therefore, we hypothesize that certain HIF-1 α polymorphisms may be associated with significantly higher transcriptional activities or influence the degradation of HIF-1α protein, resulting in enhanced angiogenesis and elevated risk of digestive cancers. HIF-1 α rs11549465 C>T and rs11549467 G>A polymorphisms create amino acid substitutions from alanine 588 to threonine and from proline 582 to serine, respectively, which influences protein stability and/or HIF-1α ability to associate with accessory proteins important in normoxia/hypoxia response.[32] Our findings are consistent with a study that also suggested that HIF-1 α rs11549467 G>A polymorphism is associated with increased gastric cancer risk in Tibetan population, which was attributed to enhanced transcriptional activity due to structural changes in the protein.[20] Furthermore, Fransén et al. concluded that polymorphic C1772T variant may be a potential target for oncogenic activation of HIF-1 α via C1772T phosphorylation, which significantly enhances the transactivation capacity of HIF1, promoting the growth of tumors.[18]

In consideration of the potential heterogeneity that might affect our results, subgroup analyses were performed based on ethnicity and cancer type. Interestingly, Asians showed a significantly elevated risk of digestive cancers in relation to both polymorphisms studied. However, in case of 1790G/A polymorphism, we failed to find any correlation between the 1790G/A SNP and digestive cancer risk in Caucasians. This outcome can be attributed to different genetic backgrounds, environmental exposure, and risk factors associated with different lifestyles between Asian and Caucasian populations. In the subgroup study based on cancer type, the susceptibility to gastric cancer, liver cancer, and pancreatic cancer was elevated in patients with 1790G/A variation. On the other hand, 1772C/T SNP was associated with an increased risk of colorectal cancer and pancreatic cancer. Taken together, these results reveal significant differences in the capacities of HIF-1 α SNPs to confer susceptibility to different digestive cancer types. However, the RR values in some subgroup analyses were large and lacked statistical power due to heterogeneity. Meta-regression analyses also showed that ethnicity may be the main source of heterogeneity. In conclusion, our results support several previous studies that reported the association between HIF-1 α genetic and increased susceptibility to the development and progression of digestive cancers, suggesting HIF-1 α SNPs could be valuable biomarkers in predicting susceptibility to digestive cancers. However, we also observed the difference in susceptibility based on ethnicity.

Some limitation of this analysis should be noted. First, our lack of access to the original data from the studies limited further evaluation of potential interactions between other factors and the susceptibility to digestive cancers, such as gene-gene and gene-environment interactions. Second, the P value of the HWE of three enrolled trials was lower than 0.05, indicating that these study populations could not be representative of a broader population. Finally, although all cases and controls of each study were well-defined with similar inclusion criteria, there may be other potential factors that were not taken into account and this may have influenced our results. Despite these limitations, our study had strong advantages. This meta-analysis revealed the correlation between HIF-1 α SNPs and increased risk for various digestive cancers. In addition, the quality of the enrolled studies was satisfactory. Furthermore, substantial cases numbers were extracted from different trials, which increased the statistical power of this analysis. There was no publication bias in the collected data.


The current meta-analysis provides evidence that HIF-1 α rs11549465 C>T and rs11549467 G>A polymorphisms may be involved in the pathogenesis of digestive cancers among Asians. However, future studies with larger sample size are needed to verify our findings, with particular attention to ethnicity differences. More trials on interactions of gene-gene and gene-environment should also be carried out in the future to have a comprehensive understanding of the association between HIF-1 α SNP and digestive cancer risk.


We would like to appreciate the reviewers for their helpful suggestions or comments on this paper.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer J Clin 2013;63:11-30.
2Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61:69-90.
3Pelucchi C, Gallus S, Garavello W, Bosetti C, La Vecchia C. Alcohol and tobacco use, and cancer risk for upper aerodigestive tract and liver. Eur J Cancer Prev 2008;17:340-4.
4Fearon ER. Molecular genetics of colorectal cancer. Annu Rev Pathol 2011;6:479-507.
5Haas SL, Ye W, Löhr JM. Alcohol consumption and digestive tract cancer. Curr Opin Clin Nutr Metab Care 2012;15:457-67.
6Cao D, Hou M, Guan YS, Jiang M, Yang Y, Gou HF. Expression of HIF-1alpha and VEGF in colorectal cancer: Association with clinical outcomes and prognostic implications. BMC Cancer 2009;9:432.
7Xu J, Xu L, Li L, You Q, Cha L. HIF-1α C1772T polymorphism and gastrointestinal tract cancer risk: A meta-analysis and meta-regression analysis. Genet Test Mol Biomarkers 2013;17:918-25.
8Knechtel G, Szkandera J, Stotz M, Hofmann G, Langsenlehner U, Krippl P, et al. Single nucleotide polymorphisms in the hypoxia-inducible factor-1 gene and colorectal cancer risk. Mol Carcinog 2010;49:805-9.
9Majmundar AJ, Wong WJ, Simon MC. Hypoxia-inducible factors and the response to hypoxic stress. Mol Cell 2010;40:294-309.
10Hsiao PC, Chen MK, Su SC, Ueng KC, Chen YC, Hsieh YH, et al. Hypoxia inducible factor-1alpha gene polymorphism G1790A and its interaction with tobacco and alcohol consumptions increase susceptibility to hepatocellular carcinoma. J Surg Oncol 2010;102:163-9.
11Semenza GL. Defining the role of hypoxia-inducible factor 1 in cancer biology and therapeutics. Oncogene 2010;29:625-34.
12Semenza GL. Regulation of cancer cell metabolism by hypoxia-inducible factor 1. Semin Cancer Biol 2009;19:12-6.
13Rey S, Semenza GL. Hypoxia-inducible factor-1-dependent mechanisms of vascularization and vascular remodelling. Cardiovasc Res 2010;86:236-42.
14Zhao XY, Chen TT, Xia L, Guo M, Xu Y, Yue F, et al. Hypoxia inducible factor-1 mediates expression of galectin-1: The potential role in migration/invasion of colorectal cancer cells. Carcinogenesis 2010;31:1367-75.
15Nam SY, Ko YS, Jung J, Yoon J, Kim YH, Choi YJ, et al. A hypoxia-dependent upregulation of hypoxia-inducible factor-1 by nuclear factor-κB promotes gastric tumour growth and angiogenesis. Br J Cancer 2011;104:166-74.
16Chen Y, Lu Y, Lu C, Zhang L. Beclin-1 expression is a predictor of clinical outcome in patients with esophageal squamous cell carcinoma and correlated to hypoxia-inducible factor (HIF)-1alpha expression. Pathol Oncol Res 2009;15:487-93.
17Baranello L, Bertozzi D, Fogli MV, Pommier Y, Capranico G. DNA topoisomerase I inhibition by camptothecin induces escape of RNA polymerase II from promoter-proximal pause site, antisense transcription and histone acetylation at the human HIF-1alpha gene locus. Nucleic Acids Res 2010;38:159-71.
18Fransén K, Fenech M, Fredrikson M, Dabrosin C, Söderkvist P. Association between ulcerative growth and hypoxia inducible factor-1alpha polymorphisms in colorectal cancer patients. Mol Carcinog 2006;45:833-40.
19Ouyang W, Torigoe C, Fang H, Xie T, Frucht DM. Anthrax lethal toxin inhibits translation of hypoxia-inducible factor 1α and causes decreased tolerance to hypoxic stress. J Biol Chem 2014;289:4180-90.
20Li K, Zhang Y, Dan Z, Wang Y, Ren ZC. Association of the hypoxia inducible factor-1alpha gene polymorphisms with gastric cancer in Tibetans. Biochem Genet 2009;47:625-34.
21Percy MJ, Mooney SM, McMullin MF, Flores A, Lappin TR, Lee FS. A common polymorphism in the oxygen-dependent degradation (ODD) domain of hypoxia inducible factor-1alpha (HIF-1alpha) does not impair Pro-564 hydroxylation. Mol Cancer 2003;2:31.
22Ling TS, Shi RH, Zhang GX, Zhu H, Yu LZ, Ding XF. Common single nucleotide polymorphism of hypoxia-inducible factor-1alpha and its impact on the clinicopathological features of esophageal squamous cell carcinoma. Chin J Dig Dis 2005;6:155-8.
23Yang X, Zhu HC, Zhang C, Qin Q, Liu J, Xu LP, et al. HIF-1α 1772 C/T and 1790 G/A polymorphisms are significantly associated with higher cancer risk: An updated meta-analysis from 34 case-control studies. PLoS One 2013;8:e80396.
24Kang MJ, Jung SA, Jung JM, Kim SE, Jung HK, Kim TH, et al. Associations between single nucleotide polymorphisms of MMP2, VEGF, and HIF1A genes and the risk of developing colorectal cancer. Anticancer Res 2011;31:575-84.
25Wang X, Liu Y, Ren H, Yuan Z, Li S, Sheng J, et al. Polymorphisms in the hypoxia-inducible factor-1α gene confer susceptibility to pancreatic cancer. Cancer Biol Ther 2011;12:383-7.
26Zintzaras E, Ioannidis JP. HEGESMA: Genome search meta-analysis and heterogeneity testing. Bioinformatics 2005;21:3672-3.
27Sterne JA, Egger M. Funnel plots for detecting bias in meta-analysis: Guidelines on choice of axis. J Clin Epidemiol 2001;54:1046-55.
28Wikstrom EA, Naik S, Lodha N, Cauraugh JH. Balance capabilities after lateral ankle trauma and intervention: A meta-analysis. Med Sci Sports Exerc 2009;41:1287-95.
29Inaba Y, Chen JA, Bergmann SR. Carotid plaque, compared with carotid intima-media thickness, more accurately predicts coronary artery disease events: A meta-analysis. Atherosclerosis 2012;220:128-33.
30Kuwai T, Kitadai Y, Tanaka S, Kuroda T, Ochiumi T, Matsumura S, et al. Single nucleotide polymorphism in the hypoxia-inducible factor-1alpha gene in colorectal carcinoma. Oncol Rep 2004;12:1033-7.
31Rannala B. Finding genes influencing susceptibility to complex diseases in the post-genome era. Am J Pharmacogenomics 2001;1:203-21.
32Ollerenshaw M, Page T, Hammonds J, Demaine A. Polymorphisms in the hypoxia inducible factor-1alpha gene (HIF1A) are associated with the renal cell carcinoma phenotype. Cancer Genet Cytogenet 2004;153:122-6.