Reproductive hormones, cancers, and conditions in relation to a common genetic variant of luteinizing hormone

D.W. Cramer1,4, K.S.I. Petterson2, R.L. Barbieri1 and I.T. Huhtaniemi3

1 Department of Obstetrics and Gynecology, Brigham and Women's Hospital, Boston, MA, USA, 2 Department of Biotechnology and 3 Department of Physiology, University of Turku, Turku, Finland


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A variant of the ß-subunit of luteinizing hormone (v-LH) is more common among populations also at higher risk for breast and ovarian cancer. To explore the possible relationship between these cancers and v-LH, we examined its frequency in premenopausal women, including 100 with a family history of ovarian cancer, 94 with carcinoma- in-situ of the breast, and 153 age and residence-matched controls. Reproductive histories were assessed and v-LH status measured by immunological assays from plasma drawn during the early follicular phase of cycles. For the entire study population, 283 (81.5%) were wild type; 61 (17.6%) were heterozygous; and three (0.9%) were homozygous for v-LH. Carrier frequency was not elevated among women with a family history of ovarian cancer or personal history of carcinoma-in-situ of the breast compared with controls. Women with the v-LH variant were less likely to report menstrual weight gain or ovarian cysts, more likely to report infertility, and have higher early follicular phase LH concentrations compared with women who were wild type. While there is no evidence from this study that v-LH increases risk for ovarian or breast cancer, we conclude that possession of v-LH may impact on some aspects of reproductive history and LH concentrations.

Key words: breast cancer/genetic polymorphisms/infertility/LH/ovarian cancer


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
An immunological variant of luteinizing hormone (v-LH) has been described and attributed to two point mutations (Trp8Arg and Ile15Thr) in gene coding the LHß-subunit (Pettersson et al., 1992Go; Furui et al., 1994Go; Nilsson et al., 1998Go). The carrier frequency of the v-LHß allele varies widely from 0% to >50%, appearing to be less common among populations from Asia and more common among populations from Northern Europe (Nilsson et al., 1997Go, 1998Go). Whether the variant may affect ovarian (or testicular) function or risk for hormonally-related diseases has not been fully investigated. In particular, based upon the higher incidence of breast and ovarian cancer among Northern Europeans compared with Asians, it is reasonable to ask whether the LH variant is correlated with risk for these cancers. In this study, we sought to investigate the relationship between v-LH and reproductive history and hormones and indirectly assess its possible link to breast and ovarian cancer. We used specimens previously collected in studies of women with a family history of ovarian cancer and women with carcinoma-in-situ of the breast (BCIS).


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Specimens from studies were sought which might both address how v-LH influences reproductive hormones and how it affects risk for breast or ovarian cancer. Since ovarian function is obviously disrupted by ovarian cancer surgery and probably disrupted by adjuvant therapy for breast cancer, it was decided to use banked specimens from two previously completed studies indirectly related to ovarian and breast cancer. A study of women who had primary relatives with ovarian cancer allowed the assessment of ovarian function and the determination of whether the frequency of v-LH was increased among relatives as would be expected if the variant were related to ovarian cancer. A study of women with BCIS who had not had hormonal treatment or chemotherapy allowed assessment of the frequency of the variant and test ovarian function among women with a precursor of breast cancer. Both studies had received approval of the Brigham and Women's Hospital (BWH) Human Subjects Committee.

Study 1 cases were drawn from women with a family history of ovarian cancer recruited through a familial ovarian cancer (FOC) clinic established at the BWH for the purpose of counselling and screening (Muto et al., 1993Go). Methods for this study have been described previously (Cramer et al., 1994Go). Briefly, women with a confirmed family history of at least one primary or two second degree relatives with ovarian cancer were eligible if they were residents of greater Boston, premenopausal, <50 years, and not currently pregnant, breastfeeding, or using oral contraceptives. In the original study, 106 unrelated women with family histories of ovarian cancer were selected and plasma specimens were still available on 101 FOC cases for assessment of LH variant. Study 2 cases were selected from women with ductal BCIS. Cases were <50 years, resided in greater Boston, premenopausal, not pregnant, breastfeeding, or taking exogenous hormones during the preceding 3 months, and had never had chemotherapy, hormonal therapy, or radiation to the pelvis. From this study, there were 94 cases who participated and had plasma specimens still available. In both studies, we selected control women from the general population using annually published lists of residents of Massachusetts. We similarly restricted control selection to women <50 years who were not pregnant, breastfeeding, or using oral contraceptives, and were matched controls to cases by age and precinct of residence. We had the additional requirements that the controls should not have a family history of ovarian cancer or should not have breast cancer or BCIS. Women were randomly sampled to match the age and residency distribution of the FOC or BCIS cases as previously described (Cramer et al., 1994Go; Bohlke et al., 1998Go). From both studies, plasma was available on 154 subjects.

Information about demographic variables, medical history and family history was collected by personal interviews. Details of the reproductive history collected included cycle irregularity defined as >10 days variation in cycle length, menstrual pain and symptoms, history of endometriosis or ovarian cysts diagnosed by a physician, and infertility defined as difficulty conceiving lasting for >=2 years. Blood samples were obtained during one of the first 4 days of the menstrual cycle. The blood was drawn in heparinized tubes, separated into plasma, erythrocyte, and buffy coat specimens and stored at -70°C before assay. The hormonal component of this study involved the measurement of gonadotrophins, including follicle stimulating hormone (FSH) and luteinizing hormone (LH), and oestradiol. Oestradiol was measured by a solid phase radioimmunoassay (Coat-A-Count®; Diagnostic Products Corp., Los Angeles, CA, USA) using unextracted plasma and a highly specific oestradiol antibody. LH and FSH were measured using an immunoradiometric assay (Coat-A-Count®) with the World Health Organization (WHO) International Reference Preparations (IRP) for human menopausal gonadotrophin as the standard. Intra- and interassay coefficients of variation were less than 10 and 15% respectively, for all three assays. Between the two study periods the standard for FSH changed from the first to the second IRP leading to lower measured FSH in the second compared to the first study.

Plasma that had been banked from the studies was retrieved and sent to one of the authors (I.T.H.) for determination of v-LH status as previously described (Pettersson et al., 1992Go). LH genotypes were inferred from the ratio of two immunofluorometric assays (DELFIA, Wallac OY, Turku, Finland) with different combinations of monoclonal antibodies (mAb). The reference method (assay 2) used two LHß-specific mAB that recognized wild type and v-LH with similar stoichiometries. In the other assay (assay 1), one of the mAB used recognized only the intact LH{alpha} dimer, but not v-LH. The ratio of LH values measured by assay 1/assay 2 allowed classification of the samples into three categories: (i) >0.9 (normal ratio), the subject had two normal LHß alleles; (ii) 0.2–0.9 (low ratio), the subject was heterozygous for the mutant LHß gene; (iii) <0.15 (`zero' ratio), the subject was homozygous for the mutant LH gene. The intra- and interassay variations of assay 1 and 2 were <4 and 5% respectively, at LH concentrations at and above the lowest standard concentration of 0.6 IU/l of the WHO IRP (80/552). Two subjects from the FOC study (one case and one control) had indeterminate assay results and were excluded from this analysis.

Statistical analyses on hormonal and epidemiological variables were performed using standard SAS programmes for t-tests, {chi}2 or Fisher's exact test, and linear regression analyses (Statistical Analysis System Institute Inc., SAS Campus Dr., Cary, NC, USA). For assessing the effects of several study variables on an outcome variable, generalized linear modelling or multiple linear regression was used.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Table IGo shows the characteristics of the study groups. Subjects from the BCIS study were older than subjects from the FOC study (P < 0.05), and BCIS controls were more likely to be parous compared with BCIS cases (P < 0.05). Subjects did not differ significantly by race or religion. Table IIGo shows the basal hormone concentrations for the four groups. Significant differences in Table IIGo included: lower LH concentrations for FOC cases compared with FOC controls (P < 0.01) previously described (Cramer et al., 1994Go); lower FSH concentrations for the BCIS cases and controls compared with the FOC cases and controls (P < 0.01) representing a change in the reference assay for FSH between the FOC and the BCIS study; and higher LH for the BCIS cases and controls compared with the FOC cases and controls (P < 0.05) reflecting the older ages of the BCIS subjects. Oestradiol concentrations did not differ significantly among groups.


View this table:
[in this window]
[in a new window]
 
Table I. Characteristics of cases and controls in familial ovarian cancer (FOC) and breast carcinoma-in-situ (BCIS) studies
 

View this table:
[in this window]
[in a new window]
 
Table II. Basal hormonal concentrations in study groups
 
Table IIIGo shows the distribution of wild type, heterozygotes, and homozygotes for the LH variant by case or control status or demographic features. The frequency of heterozygosity for the LH variant ranged from 16.0% among both the FOC and the BCIS cases to 20.8% among the FOC controls. The average age at diagnosis for BCIS among the 79 who had wild type LH was 41.7 years compared to 42.8 years for the 15 cases who were carriers for v-LH (not significant). Only three homozygotes were identified, two among the FOC cases and one among the BCIS controls. None of these differences were statistically significant; nor was there any significant difference in the frequency of the v-LH carriers by white versus non-white, Jewish versus non-Jewish, or ever versus never married comparisons. Overall, the frequency of heterozygosity for v-LH was 17.6% and the frequency of homozygosity for v-LH was 0.9% in the entire study population.


View this table:
[in this window]
[in a new window]
 
Table III. LH variants by case/control status and demographic factors
 
Table IVGo examines selected medical and reproductive characteristics of heterozygotes and homozygotes compared to wild types in all controls and in all subjects combined. Attention is focused on those characteristics which differed significantly (P <= 0.05) or were of borderline statistical significance (P <= 0.10). LH variant carriers and homozygotes were less likely to report menstrual weight gain (P < 0.05, among all controls and all subjects), to have a history of ovarian cysts (P = 0.08, among all subjects), to be smokers (P = 0.05 among all controls and P = 0.06 among all subjects), and to have a family history of prostate cancer (P = 0.09, among all controls). The observation that v-LH carriers had a lower frequency of menstrual weight gain and ovarian cysts persisted after adjustment for use of oral contraceptives (data not shown). LH variant carriers and homozygotes were more likely to report a history of infertility (P = 0.02 among all controls, P = 0.06 among all subjects), to have used fertility drugs (P = 0.09 among all subjects), and to have had depression requiring medication or medical consultation (P = 0.07, among all controls). Despite the greater frequency of infertility, there was no significant tendency for v-LH carriers to be nulliparous and the average number of pregnancies was 2.06 for v-LH carriers and 2.11 for wild types (not significant). Presence of the LH variant had no significant influence on whether cycles were likely to be irregular or associated with moderate or severe menstrual pain and whether subjects had endometriosis, diabetes, or a family history of breast cancer (data not shown).


View this table:
[in this window]
[in a new window]
 
Table IV. Selected reproductive, medical, and family history variables by LH variant status
 
Table VGo examines selected quantitative characteristics in heterozygotes compared to wild types. Both among all controls and all subjects, heterozygotes for v-LH compared to wild types did not differ significantly in age at menarche, cycle length, or BMI. Because any hormonal differences by LH variant status may be confounded by age and assay differences between studies pointed out above in connection with Table IIGo, we used generalized linear modelling applied to all subjects to adjust for study, case or control status, age, smoking, and body mass index. The P values adjusted for those variables and associated with v-LH heterozygosity as a predictor were 0.15 for FSH, 0.03 for LH and 0.15 for oestradiol. Thus, women who were heterozygotes for v-LH had significantly higher early follicular phase LH concentrations by the assays done in Boston. Two of the women who were homozygous for v-LH had normal early follicular phase hormone concentrations, while the third had very low hormone concentrations with FSH of 0.47 IU/l, LH of 0.10 IU/l, and an oestradiol concentration of 18.8 pg/ml.


View this table:
[in this window]
[in a new window]
 
Table V. Selected baseline and reproductive characteristics and hormone values in wild type and heterozygotes
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Since first being described immunologically in 1992 (Pettersson et al., 1992Go) and genetically in 1994 (Furui et al., 1994Go), there has been increasing interest in a variant form of LH as it might relate to ovarian function and hormonally-related diseases. Because v-LH is more common among populations that are also at higher risk for ovarian and breast cancers, we thought it reasonable to consider the relationship of these disorders to the variant. Using banked specimens from previously completed studies, we determined the v-LH status of women with a family history of ovarian cancer and women with BCIS. Studying these groups, rather than breast or ovarian cancer patients themselves, allowed us to assess the impact of the variant on basal hormone concentrations which would have been altered by ovarian or breast cancer treatment. In our study, we found no evidence that v-LH was more common among women with a family history of ovarian cancer or with BCIS compared to controls. While this suggests that v-LH is not correlated with increased risk for breast or ovarian cancer, additional studies in other populations of women who themselves have invasive breast cancer or ovarian cancer will be necessary before these associations are considered null.

We found the overall frequency of heterozygosity for v-LH allele to be 17.6% which matches closely previous estimates of the frequency of v-LH carriers in Northern European populations (Nilsson et al., 1997Go). The high frequency of the variant has been interpreted to indicate that v-LH probably doesn't impact on male or female fertility but published results are discrepant. A study of Japanese women suggested that the variant does occur with higher frequency in women who have ovulatory problems (Takahaski et al., 1999), while another study (Ramanujam et al., 2000Go) found no association between male infertility and LH variants. Among the combined controls in our study, who were selected from the general population rather than infertility clinics, we found a higher frequency of self-reported delay-to-conception associated with v-LH. Medical records were not available to clarify the nature of the fertility problems in women with v-LH. Overall, among v-LH carriers or homozygotes, the proportion of nulliparous women was not significantly increased suggesting that v-LH may be related to subfertility rather than sterility. Of potential interest is the fact that the 16 subjects with v-LH who reported infertility had apparently lower BMI (23.3 kg/m2) compared to the BMI (25.3 kg/m2) of those women with v-LH who did not report infertility (not significant). Although this difference was not significant, it contrasts with nearly identical BMI in the women with and without infertility who were wild type, 23.3 and 23.9 kg/m2 respectively.

There is also some controversy whether v-LH influences menstrual regularity or ovarian conditions such as the polycystic ovarian syndrome (PCOS). An initial report suggested that the frequency of v-LH was increased among women with PCOS, especially those who were obese (Rajkhowa et al., 1995Go). However, this was not confirmed in a subsequent report involving a different population (Tapanainen et al., 1999Go). In a study of women with a variety of menstrual disorders, including PCOS, no association with v-LH was observed (Ramanujam et al., 1999Go). In the current study, we had no medically confirmed cases of PCOS but we believe it is of interest that women with v-LH had no higher occurrence of irregular menses and were apparently less likely to report physician-diagnosed ovarian cysts and menstrual weight gain. The latter finding might be compatible with a greater occurrence of anovulatory cycles suggested by the association we found with infertility.

It is difficult to predict a priori what effects v-LH might have on ovarian function compared to the wild type since its bioactivity appears to be increased in vitro but its half-life in circulation may be shorter (Haavisto et al., 1995Go; Suganuma et al., 1996Go). A previous study suggested that follicular phase oestradiol, testosterone, and sex hormone binding globulin concentrations may be higher in women heterozygous for v-LH (Tapanainen et al., 1999Go). In this study of early follicular phase hormone concentrations, we found that women who were heterozygous for v-LH compared to wild types had higher early follicular phase LH as measured by the assays done in Boston, after adjustment for age, study, case or control status, BMI, and smoking.

Chance must be regarded as a possible explanation for some of our findings including the observation that women with v-LH were apparently more likely to be depressed and less likely to be smokers compared to the women who were wild type. A finding which was not statistically significant was that women heterozygous or homozygous for v-LH were apparently less likely to have a family history of prostate cancer. This observation may be of interest in view of a report that delayed puberty was more likely to occur in boys who carry v-LH (Raivio et al., 1996Go). A conclusion of that report was that v-LH might be involved in the regulation of the growth hormone and insulin-like growth factor I axis during childhood, which may also influence prostate cancer risk (Chan et al., 1998Go).

In summary, among premenopausal women with either a family history of ovarian cancer or BCIS, we found no evidence of a greater frequency of v-LH that might suggest an association with ovarian or breast cancer. However, our study suggests that v-LH may impact on ovarian function. In particular, women who possess v-LH may have sub-fertility, especially those with normal or below normal BMI. Further studies of the effect of v-LH on ovarian (and testicular) function may be of value in sorting out the long-appreciated link between body mass and fertility.


    Acknowledgments
 
This work was supported by Grant 90-BW25 and 9420 from the American Institute for Cancer Research, the Finnish Cancer Foundation, and the Sigrid Juselius Foundation. The technical and programming assistance of Ms Tarja Laiho and Mrs Rebecca Liberman is gratefully acknowledged.


    Notes
 
4 To whom correspondence should be addressed at: Ob-Gyn Epidemiology Center, Brigham and Women's Hospital, 221 Longwood Avenue, Boston, MA 02115, USA. E-mail: dwcramer{at}rics.bwh.harvard.edu Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Bohlke, K., Cramer, D.W., Trichopoulos, D. et al. (1998) Insulin-like growth factor-I in relation to premenopausal ductal carcinoma in situ of the breast. Epidemiology, 9, 570–573.[ISI][Medline]

Chan, J.M., Stampfer, M.J., Giovannucci, E. et al. (1998) Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study. Science, 279, 563–566.[Abstract/Free Full Text]

Cramer, D.W., Barbieri, R.L., Muto, M.G. et al. (1994) Characteristics of women with a family history of ovarian cancer. II. Follicular phase hormone concentrations. Cancer, 74, 1318–1322.[ISI][Medline]

Furui, K., Suganuma, N., Tsukahara, S.I. et al. (1994) Identification of two point mutations in the gene coding luteinizing hormone (LH) ß subunit, associated with immunologically anomalous LH variants. J. Clin. Endocrinol. Metab., 78, 107–113.[Abstract]

Haavisto, A.M., Pettersson, K., Bergendahl, M. et al. (1995) Occurrence and biological properties of a common genetic variant of luteinizing hormone. J. Clin. Endocrinol. Metab., 80, 1257–1263.[Abstract]

Muto, M.G., Cramer, D.W., Brown, D.L. et al. (1993) Screening for ovarian cancer: the preliminary experience of a familial ovarian cancer center. Gynecol. Oncol., 51, 12–20.[ISI][Medline]

Nilsson, C., Pettersson, K., Millar, R.P. et al. (1997) Worldwide frequency of a common genetic variant of luteinizing hormone: an international collaborative research. Fertil. Steril., 67, 998–1004.[ISI][Medline]

Nilsson, C., Jiang, M., Pettersson, K. et al. (1998) Determination of a common genetic variant of luteinizing hormone using DNA hybridization and immunoassays. Clin. Endocrinol., 49, 369–376.[ISI][Medline]

Pettersson, K., Ding, Y.Q. and Huhtaniemi, I. (1992) An immunologically anomalous luteinizing hormone variant in a healthy woman. J. Clin. Endocrinol. Metab., 74, 164–171.[Abstract]

Raivio, T., Huhtaniemi, I., Anttila, R. et al. (1996) The role of luteinizing hormone ß gene polymorphism in the onset and progression of puberty in healthy boys. J. Clin. Endocrinol. Metab., 81, 3278–3282.[Abstract]

Rajkhowa, M., Talbot, J.A., Jones, P.W. et al. (1995) Prevalence of an immunological LH ß-subunit variant in a UK population of healthy women and women with polycystic ovary syndrome. Clin. Endocrinol., 43, 297–303.[ISI][Medline]

Ramanujam, L.N., Liao, W.X., Roy, A.C. et al. (1999) Association of molecular variants of lutenizing hormone with menstrual disorders. Clin. Endocrinol., 51, 243–246.[ISI][Medline]

Ramanujam, L.N., Liao, W.X., Roy, A.C. et al. (2000) Association of molecular variants of luteinizing hormone with male infertility. Hum. Reprod., 15, 925–928.[Abstract/Free Full Text]

Suganuma, N., Furui, K., Kikkawa, F. et al. (1996) Effects of the mutations (Trp8->Arg and Ile 15->Thr) in human luteinizing hormone (LH) ß-subunit on LH bioactivity in vitro and in vivo. Endocrinology, 137, 831–838.[Abstract]

Takahashi, K., Ozaki, T., Okada, M. et al. (1999) Increased prevalence of luteinizing hormone ß-subunit variant in patients with premature ovarian failure. Fertil. Steril., 71, 96–101.[ISI][Medline]

Tapanainen, J.S., Koivunen, R., Fauser, B.C. et al. (1999) A new contributing factor to polycystic ovary syndrome: the genetic variant of luteinizing hormone. J. Clin. Endocrinol. Metab., 84, 1711–1715.[Abstract/Free Full Text]

Submitted on March 28, 2000; accepted on June 19, 2000.