Effects of antihistamine and anti-inflammatory medication use on risk of specific glioma histologies.

Abstract

Several studies have shown a decrease in glioma risk associated with a personal history of allergic conditions and the medications used to treat the symptoms. However, few studies have been able to examine risk within histological subgroups of glioma. Case-control data from M. D. Anderson Cancer Center and University of California, San Francisco were pooled to conduct the analysis stratified by histological subtype. A risk prediction model considering inflammation-related variables and antihistamine use was built using logistic regression. Of the subtypes examined, long-term antihistamine use was associated with increased risk of anaplastic gliomas, especially when length of use was considered in conjunction with history of asthma or allergy. Anaplastic cases with no history of asthma or allergy were 2.94 times more likely than controls to report antihistamine use for 10 years or more; whereas anaplastic cases with a history of asthma or allergy were 2.34 times more likely than controls to report antihistamine use for 10 years or more. Furthermore, anti-inflammatory medication use was associated with a protective effect against glioblastoma (OR = 0.80; 95% CI: 0.65, 0.99), especially among individuals with no history of asthma or allergies. No statistically significant effects of anti-inflammatory drugs or antihistamines were evident for the other histological subtypes. Thus, modulation of the immune system by the use of common drugs, such as antihistamines or nonsteroidal anti-inflammatory drugs, may contribute to the development of certain types of brain tumors.

Full text

Page 1

1

SHORT REPORT
Effects of antihistamine and anti-inflammatory medication use on risk of specific
glioma histologies
Michael E. Scheurer, PhD1,2, E. Susan Amirian, PhD2, Stacy L. Davlin, MPH2, Terri
Rice, MPH3, Margaret Wrensch, PhD3, Melissa L. Bondy, PhD4
1Department of Pediatrics, Baylor College of Medicine, Houston, TX
2Dan L. Duncan Cancer Center, Baylor College of Medicine, Houston, TX
3Department of Neurological Surgery, University of California, San Francisco, San
Francisco, CA
4Department of Epidemiology, The University of Texas M. D. Anderson Cancer Center,
Houston, TX

Correspondence to: Michael E. Scheurer, PhD, Baylor College of Medicine, One Baylor
Plaza, MS: BCM305, Houston, TX 77030, USA, Tel.: 713-798-5547, Fax: 713-798-
8711, E-mail: scheurer@bcm.edu

Running Title: Antihistamines, NSAIDs, & glioma subtype
Key words: glioma, glioblastoma, epidemiology, histamine antagonists, anti-
inflammatory agents
Abbreviations: NSAID: non-steroidal anti-inflammatory drug; MDACC: The University
of Texas M. D. Anderson Cancer Center; UCSF: University of California, San Francisco;
GBM: glioblastoma; RDD: random-digit dialing; OR: odds ratio; CI: confidence interval
Page 1 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 2

2

ABSTRACT
Several studies have shown a decrease in glioma risk associated with a personal history
of allergic conditions and the medications used to treat the symptoms. However, few
studies have been able to examine risk within histological subgroups of glioma. Case-
control data from M. D. Anderson Cancer Center and University of California, San
Francisco were pooled to conduct the analysis stratified by histological subtype. A risk
prediction model considering inflammation-related variables and antihistamine use was
built using logistic regression. Of the subtypes examined, long-term antihistamine use
was associated with increased risk of anaplastic gliomas, especially when length of use
was considered in conjunction with history of asthma or allergy. Anaplastic cases with no
history of asthma or allergy were 2.94 times more likely than controls to report
antihistamine use for 10 years or more; while anaplastic cases with a history of asthma or
allergy were 2.34 times more likely than controls to report antihistamine use for 10 years
or more. Furthermore, anti-inflammatory medication use was associated with a protective
effect against glioblastoma (OR=0.80; 95% CI: 0.65, 0.99), especially among individuals
with no history of asthma or allergies. No statistically significant effects of anti-
inflammatory drugs or antihistamines were evident for the other histological subtypes.
Thus, modulation of the immune system by the use of common drugs, such as
antihistamines or non-steroidal anti-inflammatory drugs, may contribute to the
development of certain types of brain tumors.

Page 2 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 3

3

INTRODUCTION
Gliomas account for approximately three-fourths of all primary malignant brain tumors,
and, despite extensive study into their etiology, there are few established risk factors.1
Epidemiological studies of gliomas are difficult to conduct both because gliomas are rare,
making recruitment difficult for large studies of risk factors, and because the
classification of “glioma” is a broad grouping that includes several component
histologies. These histologic types exhibit characteristic racial/ethnic and gender
distributions, and may have distinct molecular profiles, according to recent gene
expression studies.2, 3 If the subtypes are actually etiologically heterogeneous, real
differences in risk factors may be missed by combining subtypes in epidemiologic
studies.4 For these reasons, pooled studies from multiple research centers are necessary
to acquire enough cases to provide analyses of more detailed case groups, especially
when examining rare exposures.

One current topic of interest in brain tumor epidemiology is the effect of immune
responses on glioma risk. Recent studies suggest an association of self-reported allergy or
asthma with lower glioma risk.5-15 Furthermore, some studies have reported associations
between glioma risk and use of over-the-counter medications associated with
inflammation, particularly non-steroidal anti-inflammatory drugs (NSAIDs) and
antihistamines.9, 16 These drugs act to alleviate symptoms associated with histamine
release after allergen exposure. We speculated that use of these medications would be
more common among persons prone to allergies and more likely to be associated with a
decreased risk, as has been reported in previous studies.10, 11, 15 We used data from case-
Page 3 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 4

4

control studies on adult glioma in the greater Houston, TX area and the San Francisco
Bay area to examine the relationships between these medications and glioma risk by
histologic subtype.

MATERIAL AND METHODS
Data collected by M.D. Anderson Cancer Center (MDACC) as part of the Harris County
Brain Tumor Study and by the University of California, San Francisco (UCSF) as part of
the Bay Area Adult Glioma Study (series 2 and 3) were pooled to increase sample sizes
in three histologic subgroups: high-grade (ICD-O codes 9440-9442), anaplastic glioma
(ICD-O codes 9401, 9392, 9451, 9380, 9382), and low-grade glioma (ICD-O codes 9450,
9400, 9411, 9420, 9421, 9380, 9382, 9383, 9391, 9506).

MDACC Subjects. Cases (≥18 years) were newly diagnosed between January 2001 and
July 2008 with glioma (ICD-O-3 codes 9380-9481) and resided in the following Texas
counties surrounding Houston: Austin, Brazoria, Chambers, Colorado, Fort Bend,
Galveston, Harris, Jefferson, Liberty, Montgomery, Orange, San Jacinto, Walker, Waller,
and Wharton. Cases were interviewed prior to radiotherapy or chemotherapy, but
typically after surgery. Proxies were interviewed for subjects unable to participate
because of disability or death. For all cases, diagnosis was confirmed by a
neuropathologist at MDACC. Controls, frequency-matched to cases by age (within five
years), sex, and ethnicity (White, Black, Hispanic, other), were English-speaking
residents of the same Texas counties obtained through standard random-digit dialing
(RDD).17, 18 Overall response rates were 77% for cases and 53% for controls.
Page 4 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 5

5


UCSF Subjects. Cases (≥20 years) were newly diagnosed with a histologically-confirmed
glioma (ICD-O-3 codes 9380–9481) between May 1997 to August 1999 (series 2), and
November 2001 to September 2004 (series 3). They resided in six counties around the
San Francisco Bay Area (Alameda, Contra Costa, Marin, San Mateo, San Francisco, and
Santa Clara) and were identified using the Northern California Rapid Case Ascertainment
program. Controls resided in the same six counties as the cases and were identified using
RDD. They were frequency matched to cases on age (within five years), sex, and
race/ethnicity (White, Black, Hispanic, Asian, or other).14, 19 Overall response rates were
75% for cases and 73% for controls.

MDACC Interviews. Structured in-person or telephone interviews using questionnaires
and show-cards were conducted with consenting cases (or proxies) and controls. Neither
cases nor controls were told of the study hypotheses during the interview. Subjects
answered questions about demographic characteristics, personal and family medical
history (including personal history of asthma or allergies), and occupational and
environmental exposures. Information on all medications taken during the past ten years
was also obtained, as previously described.9 The two medications of interest for the
analysis were NSAIDs linked with reduced risk in other cancers, and antihistamines,
often taken for allergies, which have been shown to confer a reduced risk for gliomas.
Briefly, participants were shown or read aloud a list of these drugs (generic and brand
names) and asked whether, at anytime during the 10 years prior to diagnosis (or
interview), they had taken these drugs on a regular basis (i.e. at least weekly) for at least
Page 5 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 6

6

six consecutive months. If the respondent reported regular use, the interviewer asked the
specific name of the drug(s) and the duration of regular use; dosage was not obtained.

UCSF Interviews. In-person or telephone interviews with cases (or proxies) and controls
used a structured questionnaire and show cards. The questionnaire asked extensive
information about family and personal medical history including asthma and allergies,
demographics, occupational history, and environmental exposures, as previously
described.14 The interviewer also asked for information on all medication use, including
antihistamines and NSAIDs, as detailed elsewhere.16 Briefly, subjects were shown a list
of drugs including generic and brand names and asked whether, during the past ten years
(or ten years prior to diagnosis for cases), they had taken a total of 600 or more pills of
any of these drugs. If so, the respondent was then asked how many times per day, week,
month, or year the drug had been taken during those ten years. They were also asked the
year of first use and total number of months or years of use for each category of
medication; however, specific dosages were not requested.

Statistical Analyses. Unconditional logistic regression was used to estimate odds ratios
(ORs) and 95% confidence intervals (CIs). Whether the medication was ever used
regularly, as well as the duration of use (<10 years vs. ≥10 years), were examined in
separate models. Variables that were significantly related to case status in univariable
analysis were evaluated for inclusion in the final multivariable model, which was also
adjusted for the frequency-matched variables (age, gender, and race/ethnicity). Reference
categories were those who reported no regular use of that medication. Proxy-reported
Page 6 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 7

7

data did not appreciably alter our findings and were not excluded from the analyses. To
be considered a confounder, the characteristic had to be associated with the outcome in
the unexposed, associated with the exposure among the controls, and must have changed
the point estimate by more than 10% after adjustment. (All analyses were performed
using Stata version 10.0 (Stata Corp, College Station, TX). All p-values are two-sided
(α=0.05).

RESULTS
In the pooled dataset, we obtained complete interview data from 1,534 controls and 1,339
cases, including 294 (22%) proxy-reports of which 34 were from MDACC (8.6% of
MDACC cases), 134 were from UCSF series 2 (33.4% of series 2 cases), and 126 were
from series 3 (23.2% of series 3 cases). Overall, controls were more highly educated,
more likely to have used alcohol and tobacco, and more likely to report a positive history
of the inflammation-related variables than cases (Table 1). These differences were also
consistent within each of the three datasets, MDACC and UCSF series 2 and 3 [not
shown]. However, there was some heterogeneity present across the three series. MDACC
cases (mean: 49.8 years ± 0.68) were significantly younger than UCSF cases [means:
55.8 ± 0.78 and 54.6 ± 0.68 for series 2 and 3, respectively] (p<0.001). Furthermore,
MDACC cases were significantly more likely to have attended at least some college and
to have a positive history of cigarette smoking. Thus, to account for such differences in
the distributions of certain demographics between the three study series, we adjusted all
logistic regression models for series.

Page 7 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 8

8

Antihistamine use was significantly associated with a 64% increased risk (OR = 1.64;
95% CI: 1.03, 2.61) for anaplastic tumors (WHO Grade III), adjusting for relevant
covariates (Table 2). NSAID use was significantly associated with a 20% reduction in
risk (OR = 0.80; 95 % CI: 0.65, 0.99) of glioblastoma [GBM; WHO Grade IV] (Table 2).
History of asthma/allergy was protective against risk across all histologic grades and
overall. Among individuals with no history of asthma/allergy, participants who reported
≥10 years of antihistamine use had a significant 2.94-fold increase in risk of anaplastic
tumors (95 % CI: 1.04, 8.34) [Table 3]. A similar association was also observed among
those participants with a history of asthma/allergy (OR = 2.34; 95% CI: 1.20, 4.57).
Using NSAIDs for <10 years was significantly protective against overall glioma risk,
among those with no history of asthma/allergy (Table 3).

To ensure that our results did not differ substantially in a particular study series, we
conducted a post-hoc analysis in which we ran the logistic regression models within each
series. While there was a certain degree of heterogeneity present, the confidence intervals
for the point estimates of interest were largely overlapping, which indicates that these
measures of association were not notably different from each other.

DISCUSSION
We found that cases with anaplastic glioma were more than two-times as likely as
controls to report regular long-term use of antihistamines, regardless of asthma/allergy
history. Among individuals with no asthma/allergy history, short-term NSAID use (<10
years) was inversely associated with glioma across all subtypes and overall (significant
Page 8 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 9

9

for GBMs and all cases). Because glioma subtypes are different in terms of clinical
course and are also likely to be etiologically distinct, examining them separately is vital
to identifying relevant risk factors for each.4 Thus, pooled analyses, such as this one, are
necessary for obtaining an adequate sample size of each tumor type.

Although NSAID use has been found to be protective against various cancers in several
previous studies,20 very few have examined its impact specifically on the different
histological subtypes of glioma.1, 16 A previous study examining a subset of the
glioblastoma multiforme cases included in this pooled analysis found a significant
inverse association with regular anti-inflammatory use.16 This study was the first to
examine NSAID use in relation to glioblastoma risk. The current analysis, which includes
more than three times as many cases, corroborates the findings of the previous study and
contributes the additional knowledge that the strongest protective effect of NSAIDs
against GBM may be among those with no history of asthma or allergies.

Results from previous studies on antihistamine use and glioma risk have been
inconsistent. One multinational study reported a 30% reduction in risk of adult glioma
with antihistamine use.11 Two related studies found small non-significant reductions in
glioma risk associated with any antihistamine use by participants reporting a history of
atopic conditions.12, 15 Recently, Scheurer et al. found an overall increase in glioma risk
among long-term users of antihistamines in a subset of the MDACC patients included in
this analysis; that report prompted the current pooled analysis.9

Page 9 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 10

10

These previous reports could differ from the current findings for three reasons. First, the
definition of long-term antihistamine use is inconsistent between studies. Some reported
ever/never use without considering total length of use. Second, differences in
formulations of these drugs between different populations could lead to exposures to
varying doses. Approval of new over-the-counter drugs differs by country, thus
formulations tend to be different not only over time but also geographically. Third, the
heterogeneity of tumor types included in these studies could have also affected the
results. All previous reports included all subtypes of gliomas in their analyses; whereas,
the current pooled analysis examined histological subtypes separately to allow the
differences in effects by histologic group to be detected.

We believe this is the first report to investigate the association between adult glioma and
antihistamine use accounting for different histologic subtypes of glioma. By pooling two
similar datasets, we were able to obtain risk estimates stratified by histology, which
showed a strong effect only among the anaplastic tumors. While the use of self-reported
medication information is a limitation of our study, there is no other valid source for this
information, as most of these drugs are nonprescription. We were also unable to collect
dosage data and reason for antihistamine use. Self-reported dosage information would
likely be unreliable given that the exposures of interest occurred years before the
interview date. While antihistamines can be used for indications other than alleviation of
allergy symptoms (i.e. sleep aids), the use of these drugs as a result of an undiagnosed
glioma is unlikely. Antihistamines are not usually taken for the relief of common brain
tumor symptoms (e.g., seizures, drowsiness, and vision changes, etc.).
Page 10 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 11

11


Although a growing body of literature points to a role of the immune system in
prevention or promotion of gliomas,1, 21 continued work is still needed in the
identification of specific immunomodulatory medications, like antihistamines, as risk
factors for glioma. Many of these medications readily cross the blood-brain barrier and
act as selective inverse agonists to receptors found on the surface of glial cells. Histamine
H1 receptors can stimulate proliferation and trigger numerous inflammation pathways
known to be important in carcinogenesis, such as NF-κB and arachidonic acid.22, 23
Furthermore, compounds that interfere with normal H1 receptor activity can result in
immune-stimulatory or -suppressive behavior, as well as affect cell proliferation and
stability.24 For example, a recent report showed that melanoma cells rich in H1 receptors
undergo intensive DNA damage and apoptosis in the presence of antihistamines.25 These
mechanisms of cellular damage could be relevant in glial cells, which also exhibit high
concentrations of H1 receptors.

The present findings support an increased risk for glioma, specifically anaplastic tumors,
in adults who used antihistamines for ≥10 years. However, small sample sizes in some
groups after stratification lessened the precision of our point estimates. While there is
biological plausibility behind our results, more laboratory evidence related to
antihistamine use and glioma development would enhance support for our findings.
Furthermore, the specificity of the effect to anaplastic tumors needs further biological
explanation that is unavailable with our current knowledge about the biology and
pharmacology of these medications in glial cells.
Page 11 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 12

12

ACKNOWLEDGMENTS
This research was supported in part by grants from the National Cancer Institute (grants
R01CA070917, R01CA052689, P50CA097257, K07CA131505) and the Jean Lodgsdon
Epidemiology Grant from the National Brain Tumor Foundation. Work at University of
California, San Francisco (UCSF) has also been supported by grants from the National
Brain Tumor Foundation, the UCSF Lewis Chair in Brain Tumor Research, and by
donations from families and friends of J. Berardi, H. Glaser and E. Olsen.

The authors would like to acknowledge the contributions of the following individuals to
the overall brain tumor research programs at M. D. Anderson Cancer Center: Georgina
Armstrong (research manager) and Phyllis Adatto (project director); and University of
California, San Francisco: John Wiencke, PhD (co-investigator) and Lucie McCoy, MPH
(study coordinator).

The authors also thank Kenneth Aldape and Tarik Tihan for pathology review. The UCSF
researchers thank the pathology departments of Alexian, Alta Bates, Brookside,
California Pacific, Doctors Pinole, Eden, El Camino, Good Samaritan, Highland, John
Muir, Kaiser Redwood City, Kaiser San Francisco, Kaiser Santa Teresa, Los Gatos, Los
Medanos, Marin General, Merrithew, Mills Peninsula, Mt. Diablo Hospital, Mt. Zion,
Naval Hospital, O'Connor, Ralph K Davies, Saint Louise, San Francisco General, San
Jose, San Leandro, San Mateo County, San Ramon Valley, Santa Clara Valley, Sequoia,
Seton, St. Francis, St. Luke’s, St. Rose, Stanford, Summit, UC San Francisco, Valley
Page 12 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 13

13

Livermore, Veterans Palo Alto, Veterans SF, and Washington Hospitals and Medical
Centers for providing tumor specimens for review.
Page 13 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 14

14

References

1. Bondy ML, Scheurer ME, Malmer B, Barnholtz-Sloan JS, Davis FG, Il'yasova
D, Kruchko C, McCarthy BJ, Rajaraman P, Schwartzbaum JA, Sadetzki S, Schlehofer B,
et al. Brain tumor epidemiology: consensus from the Brain Tumor Epidemiology
Consortium. Cancer 2008;113:1953-68.
2. Godard S, Getz G, Delorenzi M, Farmer P, Kobayashi H, Desbaillets I, Nozaki
M, Diserens AC, Hamou MF, Dietrich PY, Regli L, Janzer RC, et al. Classification of
human astrocytic gliomas on the basis of gene expression: a correlated group of genes
with angiogenic activity emerges as a strong predictor of subtypes. Cancer Res
2003;63:6613-25.
3. Marucci G, Morandi L, Magrini E, Farnedi A, Franceschi E, Miglio R, Calo D,
Pession A, Foschini MP, Eusebi V. Gene expression profiling in glioblastoma and
immunohistochemical evaluation of IGFBP-2 and CDC20. Virchows Arch 2008;453:599-
609.
4. Davis FG, Malmer BS, Aldape K, Barnholtz-Sloan JS, Bondy ML, Brannstrom
T, Bruner JM, Burger PC, Collins VP, Inskip PD, Kruchko C, McCarthy BJ, et al. Issues
of diagnostic review in brain tumor studies: from the Brain Tumor Epidemiology
Consortium. Cancer Epidemiol Biomarkers Prev 2008;17:484-9.
5. Brenner AV, Linet MS, Fine HA, Shapiro WR, Selker RG, Black PM, Inskip
PD. History of allergies and autoimmune diseases and risk of brain tumors in adults. Int J
Cancer 2002;99:252-9.
Page 14 of 20
John Wiley & Sons, Inc.
International Journal of Cancer

Page 15

15

6. Cicuttini FM, Hurley SF, Forbes A, Donnan GA, Salzberg M, Giles GG,
McNeil JJ. Association of adult glioma with medical conditions, family and reproductive
history. Int J Cancer 1997;71:203-7.
7. Hochberg F, Toniolo P, Cole P, Salcman M. Nonoccupational risk indicators of
glioblastoma in adults. J Neurooncol 1990;8:55-60.
8. Ryan P, Lee MW, North B, McMichael AJ. Risk factors for tumors of the brain
and meninges: results from the Adelaide Adult Brain Tumor Study. Int J Cancer
1992;51:20-7.
9. Scheurer ME, El-Zein R, Thompson PA, Aldape KD, Levin VA, Gilbert MR,
Weinberg JS, Bondy ML. Long-term anti-inflammatory and antihistamine medication use
and adult glioma risk. Cancer Epidemiol Biomarkers Prev 2008;17:1277-81.
10. Schlehofer B, Blettner M, Becker N, Martinsohn C, Wahrendorf J. Medical
risk factors and the development of brain tumors. Cancer 1992;69:2541-7.
11. Schlehofer B, Blettner M, Preston-Martin S, Niehoff D, Wahrendorf J, Arslan
A, Ahlbom A, Choi WN, Giles GG, Howe GR, Little J, Menegoz F, et al. Role of
medical history in brain tumour development. Results from the international adult brain
tumour study. Int J Cancer 1999;82:155-60.
12. Schoemaker MJ, Swerdlow AJ, Hepworth SJ, McKinney PA, van Tongeren
M, Muir KR. History of allergies and risk of glioma in adults. Int J Cancer
2006;119:2165-72.
Page 15 of 20
John Wiley & Sons, Inc.
International Journal of Cancer