Interleukin-1 Receptor Antagonist Gene Polymorphism in Epithelial Ovarian Cancer

Ovarian cancer is the leading cause of death of all of the gynecological malignancies (1). The overall 5-year survival rate remains poor, despite significant improvements in surgical treatment and chemotherapy (2, 3).

More than 75% of all patients are classified as being FIGO¹ stage III and IV at the time of diagnosis (4).

Over the last decade, ovarian cancer screening using the serum marker CA-125 and transvaginal ultrasound examination has been investigated, but none of these methods fulfills the necessary criteria to be considered as an effective screening test (5). Therefore, new predictive and prognostic factors for ovarian cancer are needed to inprove clinical management.

Unlike gene mutations, polymorphisms in specific genes are common genetic events. It is estimated that there are approximately 200,000 single-nucleotide polymorphisms within the coding regions of the 80,000 human genes (6, 7). These could be either a single-bp substitution of one nucleotide for another or a variable repeated number of a short, repetitive DNA sequences (8).

These variations may affect the rate of gene transcription, the stability of the mRNA, or the quantity and activity of the resulting protein.

The prevalence and severity of a number of disorders will be influenced by certain polymorphic genes possessing specific alleles (7).

The polymorphic gene that encodes the IL-1 RA seems to play an important role in the development of different diseases (8).

IL-1 RA belongs to the IL-1 family, which consists of three linked genes mapping within a 430-kb region of the long arm of chromosome 2 in humans, encoding the secreted glycoproteins IL-1 α, IL-1 β, and IL-1 RA. All three molecules bind to IL-1 receptors (9, 10, 11). IL-1α and IL-1β are potent proinflammatory cytokines, whereas IL-1 RA as an anti-inflammatory cytokine competes with IL-1α and IL-1β in binding to IL-1 receptors without intrinsic effect (12, 13).

Polymorphisms have been reported in all three genes (12). The polymorphism of IL-1α, IL-1β, and IL-1 RA produces alterations of IL-1α, IL-1β, and IL-1 RA protein expression (14, 15, 16, 17, 18, 19), and it may have crucial effects on oncogenic processes (20).

The role of IL-1 RA in ovarian carcinogenesis is currently being investigated. IL-1 increases the growth of ovarian carcinoma cell lines (21) and tumor proliferation (22). Furthermore, different studies reported an association of IL-1 gene polymorphisms with gastric, pancreatic, and cervical cancer (23, 24, 25, 26).

Based on these data, we analyzed the influence of IL-RA polymorphism on the prevalence of ovarian cancer and its correlation with established clinical prognostic factors.

This case-control, mono-institution study was reviewed and approved by the Clinical Review Board and Ethics Committee of the University Hospital, Charité, Berlin, Germany.

Patients with histologically confirmed ovarian cancer were allocated to this trial. Written informed consent was provided by each patient. Borderline ovarian tumors are different tumor entities and therefore were excluded from this study (27).

Germ-line mutations do not fluctuate with age, and the control group consisted of similarly aged women with no history of cancer (4). The characteristics of the control group are summarized in Table 1. None of these women had previous oophorectomy. To avoid confounding by genetic admixture, only white Caucasian women were enrolled in this study.

Two ml of blood were drawn from the antecubital vein, and DNA was extracted using the Qiagen System (QIAamp DNA Blood Midi Kit, Hilden, Germany).

DNA was stored at 4°C until analyzed. A genomic DNA fragment was amplified by the PCR to determine IL-1 RA genotypes.

Oligonucleotide primers flanking the 86-bp repeat region in intron 2 of IL-1 RA were applied. The sequence of the forward primer was 5′-CTCAGCAACACTCCTAT-3′. The reverse primer sequence was 5′-TCCTGGTCTGCAGGTAA-3′ (28).

PCR conditions included an initial denaturing step at 94°C for 5 min, followed by 30 cycles of 94°C for 1 min, 60°C for 1 min, and 70°C for 2 min, and a final extension at 70°C for 4 min.

Using this PCR strategy, the common allele (allele 1) generated a 410-bp band (including four copies of an 86-bp repeat). The uncommon alleles generated a 240-bp band (two copies of the same repeat; allele 2) a 500-bp band (five copies of the same repeat; allele 3), and a 325-bp band (allele 4).

PCR products were resolved on a 3% agarose gel and stained with SYBR Green I (FMC Bio Products Europe, Vallensbaek Strand, Denmark).

SPSS (version 10.0; SPSS Inc., Chicago, IL) for Windows statistical software was used for statistical analyses. Differences in the frequencies of IL-1 RA alleles in the groups were analyzed by the two-tailed Fisher’s exact test. All Ps were two-tailed. P < 0.05 was considered statistically significant. Odds ratios with 95% confidence intervals were calculated.

From June 2001 to April 2003, 179 women with ovarian cancer who were treated in the University Hospital, Charité were enrolled into this study.

All patients with a histologically confirmed diagnosis and written informed consent were included.

DNA extraction was not possible because of technical reasons in two cases. Fifteen patients were excluded from this study because of different histological diagnosis (borderline tumor of the ovary, colon carcinoma). Overall, 162 women with ovarian cancer and 121 women with no history of cancer were analyzed.

The characteristics of the women with ovarian cancer are summarized in Table 2. The median age of the patients with ovarian cancer was 56 years (range, 26–81 years). Of 162 patients, only 16% were diagnosed in FIGO stage I; 77.2% were in FIGO stage III-IV. The most common histological type was serous-papillary (74.7%). To enhance the relevance of the statistics, the following subgroups were summarized: for statistical analysis, stage I + II and III + IV; for histological grading, grading I + II and grading III + IV. No statistically significant differences were observed between the study and control groups with regard to allele 1/3 (2.4% versus 0.6%; P = 0.3) or homozygous allele (IL-1RA 1/1; 74% versus 63.6%; P = 0.07).

With regard to allele 1/2 heterozygotes, the distribution of genotype frequencies was significantly different between the study and control groups (32.1% versus 14.9%; P < 0.001). Patients who were heterozygous at allele 2 for IL-1 RA (IL-RA 1/2) had a significantly higher risk for ovarian cancer with a calculated odds ratio of 2.7 (95% CI, 1.5–4.9).

The allele frequencies of IL-RA wild-type and polymorphic alleles in the study and control groups are given in Table 3.

We compared all relevant clinical data of the patients with ovarian cancer and allele 2 heterozygote genotype (IL-RA 1/2) with patients of the other genotypes in the cancer group. No statistically significant differences were found in the correlation between IL-1 RA 1/2 polymorphism and FIGO stage (P = 0.3), histological grading (P = 0.22), recurrence status (P = 0.32) ascites volume (≤500 ml versus ≥500 ml; P = 0.8), residual tumor mass (P = 0.13), and age (P = 0.8).

IL-1 RA may have a function in the host immune responses in the local and general environments of gynecological cancers (29). IL-1RA decreases tumor growth and tumor angiogenesis (30).

Polymorphisms exist in all three genes of the IL-1 family (IL-1α, IL-1β, and IL-1RA) and are located very close each to other on the long arm of human chromosome 2q (31, 32). The mutation in one of these genes up-regulates the expression of these genes, e.g., IL-1 RA polymorphism is associated with enhanced IL-1β production in vitro (16). Also, IL-1 RA genotype plays a major role as a modulator in IL-1β release (33).

Allele 2 of IL-1 RA seems to be the critical point in the molecular pathway of different diseases. The two-repeat allele has been associated with different benign diseases [vestibulitis, ulcerative colitis, alopecia areata, psoriasis, and autoimmune conditions (6, 34)] and idiopathic recurrent miscarriage (35). There are new data that indicate an additional role in cancer.

El-Omar et al. (23) and Machado et al. (36) have reported an association of the IL-1 gene cluster polymorphism with enhanced production of IL-1β and gastric cancer. Carriers of IL-1 1B-511T and allele 2 of IL-1 RA homozygotes had an increased risk of developing gastric cancer, with odds ratios of 2.7 (95% CI, 1.5–4.9) and 3.1 (95% CI, 1.5–6.5), respectively. Their findings complement the most widely accepted multistage model of gastric carcinogenesis and underline the fact that host genetic factors may determine why some people infected with Helicobacter pylori develop gastric cancer, whereas others do not (23, 36).

This hypothesis can be transferred and applied to other solid tumors. Recently, we have published that IL-1 RA is also associated with cervical cancer. Allele 2 heterozygotes have a greater risk of developing cervical cancer [P = 0.04 (37, 25)].

In our study, we have demonstrated that allele 2 polymorphism of the IL-RA gene is significantly associated with ovarian cancer. Allelic frequencies were different between patients with ovarian cancer and controls. Patients who are heterozygotes at allele 2 for IL-1 RA (IL-RA 1/2) have a significantly higher risk of ovarian cancer with an odds ratio of 2.7 (95% CI, 1.5–4.9). However, IL-1 RA polymorphism seems to give no information about the prognosis of the patients. We have not observed any association between specific alleles and clinical features such as FIGO stage, histological type, or grading. Nevertheless, the impact of an allele 1/2 polymorphism on progression-free survival and overall survival should be investigated.

Recently, Hefler et al. (38) published a case-control study in which IL-1 gene cluster was genotyped in 94 ovarian cancer patients and 27 patients with borderline ovarian tumors and 134 healthy women.

In contrast to our results, they found no differences in the prevalence of polymorphisms in the IL-1 RA gene between the ovarian cancer and control groups. The distribution of tumor stage of the enrolled patients in our study was different: stage I, 16% versus 40.8%; stage II, 6.8% versus 12.7%; and stage III-IV, 77.2% versus 45.6%. The histological type of ovarian cancer determines tumor biology and is an important prognostic factor. For instance, cytogenetic alterations in ovarian clear cell carcinoma exist and might be the cause of the poorer prognosis of these patients (39). In the study of Hefler and co-workers, the rates of clear cell carcinoma and endometrioid adenocarcinoma were significantly higher compared with our study (11% versus 3.1%, respectively; 30.8% versus 4.9%). This could be an explanation for the different results we have observed.

The possible limitations of studies with small sample size concerning the statistical validity should be kept in mind when interpreting the results of this study. Ovarian cancer is known to be a polygenic disease, and genetic factors must play an additional role in the induction of this malignancy, whereas only some of the genes were identified in ovarian cancer patients. Maybe some of them can mask potential influences of IL-1 RA polymorphism in ovarian cancer. Therefore, the presented data should be confirmed within a panel of polymorphic genes that are associated with ovarian cancer, such as androgen receptor gene (40), MTHFR gene (41), or cyclin D1 gene (3) in a prospective cohort study of patients with allele 1/2 polymorphism concerning the risk for developing ovarian cancer.