SEARCH

SEARCH BY CITATION

Keywords:

  • meningioma;
  • epidemiology;
  • risk factors;
  • neurosurgery;
  • brain tumor;
  • oral contraceptives;
  • hormone replacement therapy;
  • exogenous hormones

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Hormone Replacement Therapy
  5. Contraceptive Use
  6. Measurement of Tumor Hormone Receptors
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

The decision to commence or continue use of hormone replacement therapy or oral contraceptives in women presumed or known to be diagnosed with intracranial meningioma is a common clinical question in neurosurgery. A review of the English-language literature was undertaken to examine the association between the use of exogenous hormones and meningioma risk. Seven publications were identified, 6 of which met criteria for inclusion. No randomized clinical trial data were available, hence, results were collected from 2 population-based case-control studies, 2 hospital-based case-control studies, 1 nested case-control study drawn from a large national cohort, and 1 retrospective cohort study. At present, there is no statistical evidence of an increased risk of meningioma among users of oral contraceptives. Although not definitive, available data suggest an association between the use of hormone replacement therapy and increased meningioma risk. Further evaluation of exogenous hormone use in women with meningioma is needed with particular attention to stratification by hormone (ie, estrogen and/or progesterone) composition, duration of and age at use as well as tumor receptor subtype. Cancer 2007. © 2007 American Cancer Society.

In 2007, it is estimated that over 100,000 women in the United States will carry a surgically-confirmed diagnosis of intracerebral meningioma, with approximately 9000 new cases diagnosed in women each year.6 These numbers likely represent a lower limit on the number of women with meningioma as many patients presumed to have such a lesion are managed conservatively, i.e. without surgical intervention and pathologic confirmation, and hence are not generally included in national databases that produce such estimates of tumor incidence and prevalence. Common clinical questions raised by female patients with meningioma are 1) is the risk of meningioma incidence or progression altered by the use of exogenous hormones and 2) among those patients undergoing surgical biopsy or resection of a meningioma, should laboratory testing for tumor hormone receptors be undertaken. These clinical questions are driven by evidence of an association between hormones and meningioma risk as suggested by a number of findings including the increased incidence of the disease in women versus men (2:1),6 the presence of estrogen and progesterone receptors on some meningiomas,3, 5, 7, 11, 18, 27 a potential association between breast cancer and meningiomas,10 an association in some studies between exogenous or endogenous hormones and meningioma risk,7, 19, 21, 32, 33 as well as reports that meningiomas may change in size with menstrual cycle phase, as well as with pregnancy and menopausal status.1, 25

These reported associations between hormones and meningioma risk raise the question of whether female patients known or presumed to harbor a meningioma should commence or continue use of exogenous hormones such as oral contraceptives (OC) or hormone replacement therapy (HRT). Although reported rates of HRT use in general have decreased over the past few years particularly after publication of the results of the Women's Health Initiative indicating an increased risk of breast cancer,16, 29 the use of these medications remains relatively widespread. Oral contraceptives remain 1 of the most commonly used medications among females in the general population with over 64% of women in a population-based series of controls reporting ever using such drugs.8 Hence, the advisability of use of these medications is a common question for neurosurgeons and neuro-oncologists who treat such patients: to our knowledge no such review of this topic exists. In addition to evaluating the specific risk of meningioma occurrence or recurrence in a woman considering HRT or OC it is also necessary to assess the overall health risks and benefits potentially associated with the use of such medications including risk of breast and endometrial cancer as well as risk of cardio- and cerebrovascular disease. The current manuscript provides an updated review of the neurosurgical literature with respect to exogenous hormones and meningioma risk and discusses these risks in light of current knowledge of the general risk/benefit ratio associated with these medications in an effort to help guide decisions regarding use of these drugs in women with meningioma.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Hormone Replacement Therapy
  5. Contraceptive Use
  6. Measurement of Tumor Hormone Receptors
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

A search of the literature was undertaken by using the terms “meningioma”, “hormone replacement therapy”, “oral contraceptive”, and “exogenous hormones” since the start of electronic databases until December 31, 2006. Seven references4, 9, 15, 19, 21, 30, 32 were identified with six4, 9, 15, 19, 21, 32 providing numeric risk estimates. Study designs varied with a nested case-control study drawn from a large cohort study,19 2 hospital-based case-control studies,15, 21 2 population-based case-control studies,9, 32 and 1 retrospective cohort design.4

Hormone Replacement Therapy

  1. Top of page
  2. Abstract
  3. Methods
  4. Hormone Replacement Therapy
  5. Contraceptive Use
  6. Measurement of Tumor Hormone Receptors
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

Table 1 presents the studies that examine hormone replacement therapy use and meningioma risk. Of the 6 studies that provide risk estimates, 3 report a significant increase in risk while 3 report no evidence of an association. The first study to report a positive association utilized data from a case-control study nested within the Nurse's Health Cohort Study (NHS) including 125 cases of meningioma confirmed during 1,213,522 persons years of follow-up.19 The study participants were female registered nurses aged 30 to 50 years of age at time of enrollment and who completed biennial questionnaires between 1976 and 1996. In these data the relative risk of meningioma associated with current hormone use for pre-menopausal women was 2.48 (95% CI, 1.29, 4.77) when compared with postmenopausal women who had never used hormones. For postmenopausal women who were current hormone users the relative risk was 1.86 (95% CI, 1.07, 3.24). No excess risk was associated with past hormone use. The second study32 to report a positive association between HRT use and meningioma risk comes from the Swedish component of the Interphone study, a population-based case-control study initially designed to examine the relationship between cell phone use and brain tumor risk. This recently published study includes 1 of the larger number of cases, all of whom were diagnosed between 2000–2002. The findings from this group are notable for a significant effect associated with any use of HRT (OR, 1.7, 95%CI, 1.0, 2.8). The results from an additional study were recently published in abstract form at the Society of Neuro-Oncology meetings.4 This retrospective cohort study from the Mayo Clinic Jacksonville patient database between 1993 and 2003 identified 398,889 female patients of whom 17,907 were either current or past users of HRT. A total of 1410 of these women had a diagnosis of either a symptomatic or incidentally discovered meningioma; 156 of these patients were confirmed to be either current or past users of HRT. A multivariate analysis, adjusted for age, confirms the positive NHS and Interphone findings (OR, 2.2, 95% CI, 1.9–2.6) of an association between HRT use and meningioma risk.

Table 1. Hormone Replacement Therapy and Risk of Meningioma
StudyNo. of casesStudy designComparison*Odds ratio (95% CI)
  • CI indicates confidence interval.

  • *

    Baseline category includes women who are postmenopausal and have never used hormone replacement therapy.

  • Adjusted for age and education.

  • Adjusted for age, residential area, education, and parity.

  • §

    Adjusted for matching factors, marital status, and education.

  • Relative risk and adjusted for age and body mass index.

  • Adjusted for age.

Custer et al., 2006143Population-based case/controlPast0.7 (0.4, 1.3)
   Current1.0 (0.5, 2.2)
Lee et al., 2006219Hospital-based case/controlEver0.7 (0.4, 1.2)
Wigertz et al., 2006178Population-based case/controlEver1.7 (1.0, 2.8)
Hatch et al., 2005151Hospital-based case/controlEver§0.84 (0.50, 1.39)
Jhawar et al., 2003125Nested case/control within cohortCurrent premenopausal2.48 (1.29, 4.77)
   Current postmenopausal1.86 (1.07, 3.24)
   Past postmenopausal1.01 (0.49, 2.1)
Blitshteyn et al., 20041410Retrospective cohortEver2.2 (1.9, 2.6)

Three studies present risk estimates (Table 1) indicating no association between meningioma risk and HRT use and include 2 hospital-based15, 21 and 1 population-based case-control study.9 A additional population-based case-control study30 coordinated by the International Agency for Research on Cancer (IARC) that included 331 meningioma cases from 6 of 8 study centers and diagnosed between 1980–1991 states that their results do “not support the hypothesis of an association between meningioma and female hormones”, however, no risk estimates are given within the body of the manuscript.

Four studies examined risk by duration of use;9, 15, 21, 32 results from the Swedish Interphone study32 suggested an increase in risk with use for 10 or more years (OR, 1.9, 95%CI, 1.0, 3.8) although no consistent dose-response relation was observed between years of use and risk. The remaining 3 studies9, 21, 32 did not support any association between duration and risk. Only 2 projects examined risk by time since first or last use9, 15 or by age at first use;15 no significant relationships were reported for these sub-categories of HRT exposure, although the power to detect even moderate increases in risk was limited by small sample size. No group specifically examined the use of fertility medication and meningioma risk; 1 group32 examined the association for hormones used for “nongynecologic” reasons such as irregular menses and found no evidence of an increased risk (Table 2).

Table 2. Hormones Given for Gynecologic Problems* and Risk of Meningioma
StudyNo. of casesStudy designComparisonOdds ratio (95% CI)
  • CI indicates confidence interval.

  • *

    Includes bleeding and irregular menstruation.

  • Baseline category includes women who have never used oral contraceptives.

  • Adjusted for age, residential area, education, and parity.

Wigertz et al., 2006178Population-based case/controlEver1.1 (0.6, 1.9)
   ≥3 y of use0.8 (0.2, 2.3)

Contraceptive Use

  1. Top of page
  2. Abstract
  3. Methods
  4. Hormone Replacement Therapy
  5. Contraceptive Use
  6. Measurement of Tumor Hormone Receptors
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

Five studies9, 15, 19, 21, 32 examined the association between oral contraceptives and risk of meningioma (Table 3). With the exception of 1 study,21 no group noted a statistically significant difference in risk for users versus nonusers with both decreased and increased risks reported. The 1 study to report an significant effect included 219 meningioma cases identified from 3 Chicago area hospitals between 1987 and 1992 and reported a protective effect for ever versus never users (OR, 0.5, 95% CI, 0.4, 0.8) and an even more marked protective effect among current versus never users (OR, 0.2, 95% CI 0.0, 0.8). In this same study, among OC users, this effect was more evident among individuals who commenced use of these medications after the age of 19 years and among individuals with less than 3 years of use.

Table 3. Oral Contraceptive Use and Risk of Meningioma
StudyNo. of casesStudy designComparison*Odds ratio (95% CI)
  • CI indicates confidence interval.

  • *

    Baseline category includes women who have never used oral contraceptives.

  • Adjusted for age, race, hospital, menarche, parity, menopausal status, smoking history, thyroid disorders, and radiation treatment.

  • Adjusted for age, residential area, education, and parity.

  • §

    Adjusted for matching factors, marital status, and education.

  • Adjusted for age, body mass index, menopausal status, and post-menopausal hormone use.

Custer et al., 2006143Population-based case/controlPast1.5 (0.8, 2.7)
  Current2.5 (0.5, 12.6)
Lee et al., 2006219Hospital-based case/controlPast0.5 (0.3, 0.8)
  Current0.1 (0.0, 0.5)
Wigertz et al., 2006178Population-based case/controlEver1.0 (0.6, 1.6)
  ≥10 y of use1.1 (0.6, 1.9)
Hatch et al., 2005151Hospital-based case/controlEver§1.02 (0.63, 1.66)
  ≥10 y of use§0.82 (0.39, 1.74)
Jhawar et al., 2003125CohortPast0.76 (0.51, 1.14)
  Current1.34 (0.18, 9.96)

In addition to oral contraceptives, the Swedish Interphone32 group examined whether the use of other (non-oral) hormone contraceptives (Table 4) such as subdermal implants, injections, and hormonal intrauterine devices (all of which in Sweden contain only progestin related compounds) and noted a relative risk of meningioma of 1.5 (95% CI, 0.9, 2.6). For use of 10 years or more, the risk estimate increased to 2.7 (95% CI, 0.9, 7.5).

Table 4. Other (Non-Oral) Hormonal Contraceptive Use and Risk of Meningioma
StudyNo. of casesStudy designComparison*Odds ratio (95% CI)
  • CI indicates confidence interval.

  • *

    Baseline category includes women who have never used oral contraceptives.

  • Adjusted for age, residential area, education, and parity.

Wigertz et al., 2006178Population-based case/controlEver1.5 (0.9, 2.6)
   ≥10 y of use1.7 (0.9, 7.5)

Measurement of Tumor Hormone Receptors

  1. Top of page
  2. Abstract
  3. Methods
  4. Hormone Replacement Therapy
  5. Contraceptive Use
  6. Measurement of Tumor Hormone Receptors
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

Only 1 population-based study has examined the association between exogenous hormone use and meningioma risk by compound type as well as by tumor estrogen and progesterone receptor status.9 In this report, 142 meningioma tumor specimens were available with 2 (1%) expressing estrogen receptors and 130 (92%) expressing progesterone receptors. The study9 found a higher risk estimate for use of HRT preparations with both estrogen and progestin (OR,1.3, 95% CI, 0.6, 2.8) than for compounds containing only estrogen (OR, 0.9, 95% CI, 0.5, 1.6), but both confidence intervals are wide and not statistically significantly different. The study found no association between HRT use and progesterone receptor status but these results are based on only 46 controls and 23 cases. This group, while noting an elevated but not statistically significant association for overall use of OC, did find an increased risk of a meningiomas expressing less rather than more progesterone receptor (OR,3.2, 95% CI, 1.3, 8.0) with OC use.

DISCUSSION

  1. Top of page
  2. Abstract
  3. Methods
  4. Hormone Replacement Therapy
  5. Contraceptive Use
  6. Measurement of Tumor Hormone Receptors
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

These data represent an effort to review the current epidemiologic literature regarding the use of exogenous hormones and risk of meningioma. Only 7 studies on the topic were identified, 6 of which presented numeric risk estimates. With respect to the use of oral contraceptives, none of the publications identified reported a statistically increased risk of meningioma and in fact, 1 study reported a protective effect associated with OC use.21 Of interest, given the generally low rate of estrogen and high rate of progesterone receptors on most meningiomas, 2 reports suggested that meningioma risk was increased by contraceptive preparations including progestins, either alone and in non-oral form32 or in oral forms that also include estrogen,9 although in neither instance were the results statistically significant. Few of the studies that examined OC use were able to perform statistical sub-analyses by exposure categories such as timing of use (i.e. current, present, or ever), age at first and/or last use or duration of use and those that did were severely hampered by small sample size.

Although not all studies concur, the current literature reveals a suggestion that HRT use may be associated with meningioma risk with increased risks found in the nested case-control study drawn from the Nurses Health Cohort (NHS) as well as population-based data from the Swedish component of the Interphone Study and a retrospective cohort drawn from the Mayo Clinic patient population. A strength of the findings derived from the Nurse's Health Study (NHS) Cohort, which found evidence for an association between current use of HRT and meningioma risk in both pre- and post-menopausal subjects, relates to the prospective nature of its data collection and hence the minimization of recall bias associated with exposure reporting after case diagnosis. Data from the Mayo Clinic included both symptomatic and incidental meningioma cases while cases from the NHS are defined as symptomatic. Of note, neither of the hospital based case-control studies15, 21 found any evidence for an increase in risk, raising the question of whether hospital-based controls may be more likely than population-based controls to take HRT. It is of interest, that among those studies that also examined the association between HRT use and risk of other brain tumors such as glioma, that neither reported an association between glioma risk and HRT use.

Caveats for all of the studies include the small sample sizes and hence the relatively low power to detect even effects of moderate size. For example, even the largest study21 had only about 50% power to detect an odds ratio of 1.6 (ie, a 60% percent increase in risk) or lower. All of the studies examined primarily oral HRT and therefore no data are available on non-oral (ie, transdermal) or nonstandard dose regimes and meningioma risk. As discussed by Jhawar et al.,19 it is difficult to define the effect of including only symptomatic or surgically confirmed meningioma patients in these analyses. If the biological behavior of symptomatic cases is different from that of asymptomatic cases then the total number of cases is decreased as is statistical power. It is also important to note that any associations found with HRT use may be with symptomatic progression rather than occurrence; this topic remains to be explored. To our knowledge no data outside of individual reports exist with respect to hormone replacement therapy or oral contraceptive use and tumor progression.

The prevalence and function of estrogen and progesterone receptors in meningiomas remains a topic of great interest given the possibility of therapeutic options associated with treatment related to these receptors, although the majority of receptor based therapies have to date proved relatively unsuccessful on other than an individual basis.7, 14 In general, meningiomas show a low rate of estrogen receptor expression, but high rates of progesterone receptor; this is part can make statistical detection of an effect difficult as little variation exists for the factor. Furthermore, although most meningiomas appear to possess progesterone receptors, the function or role of these receptors is unclear and large-scale epidemiologic studies to examine the correlation between binding and a meaningful biological response or clinical outcome are limited. The data that do exist are suggestive18, 26 but not definitive and conclusions are further hampered by variable methods of receptor detection. However, results such as these highlight the potential clinical value of accurate knowledge of progesterone receptor status (both overall as well as for the 2 isoforms, alpha and beta) in meningiomas. In this review, only 1 group9 examined these receptors in concert with exogenous hormone use. In their sample of 140 cases, 92% expressed progesterone receptors (versus 1% for estrogen receptors) and found that OC use was associated with an increased risk of a meningioma expressing less rather than more progesterone receptors. This is indeed of interest particularly given reports that the expression of progesterone receptors (either alone or in concert with other predictive variables such as mitotic index) signal a more favorable clinical prognosis.27

In addition to the possible risk of meningioma, decisions regarding the use of oral contraceptives and HRT must also weigh the risks and benefit of other outcomes potentially associated with the use of these medications. The Women's Health Initiative (WHI), a randomized trial, reported an increased risk of breast cancer in post-menopausal women using opposed HRT (ie, estrogen and progestin)29 with observational studies in general agreement with these results. Recently, a 7.2 percent reduction in breast cancer rates in the United States for 2003 was reported, 1 year after millions of women ceased taking HRT in association with published reports of the WHI results,16 suggesting again a possible relation between the 2. Furthermore, results from the WHI also suggest little or no cardiovascular benefit17 and a probable increased risk of stroke and DVT associated with HRT leading some experts to question the use of at least standard HRT for the majority of women given what appear to be a higher rate of risks relative to benefits. A number of consensus statements and review articles have been produced to guide clinicians in selecting patients for whom HRT use may be appropriate;2, 13 these listings would seem to suggest that patients with meningioma may not be candidates for standard HRT use based both on biological mechanism2 as well as possible increased risk of breast cancer.13 At present, it would appear that the risks associated with HRT use should be carefully considered in planning the management of women diagnosed with intra-cranial meningioma. The literature does not suggest any association between the use of oral contraceptives and meningioma risk. The need to further study this complex topic and to better define the risk by exogenous hormone compound type and tumor receptor subtype is evident.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Methods
  4. Hormone Replacement Therapy
  5. Contraceptive Use
  6. Measurement of Tumor Hormone Receptors
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

This work was supported by the Brain Science Foundation, the Meningioma Mommas, and by NIH R01 grants CA109468, CA109461, CA109745, CA108473, and CA109475.

REFERENCES

  1. Top of page
  2. Abstract
  3. Methods
  4. Hormone Replacement Therapy
  5. Contraceptive Use
  6. Measurement of Tumor Hormone Receptors
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES