- Email: [email protected]
Menopausal hormone therapy and risk of lung cancer—Systematic review and meta-analysis
Maturitas 65 (2010) 198–204
Contents lists available at ScienceDirect
Maturitas journal homepage: www.elsevier.com/locate/maturitas
Review
Menopausal hormone therapy and risk of lung cancer—Systematic review and meta-analysis Claudia M. Greiser a , Eberhard M. Greiser a,b , Martina Dören c,∗ a
Epi.Consult GmbH, Musweiler, Germany Institute for Public Health and Nursing Research, Faculty of Health Sciences, Bremen University, Bremen, Germany c Charité - Universitätsmedizin Berlin, Campus Benjamin Franklin, Clinical Research Centre of Women’s Health, Hindenburgdamm 30, D-12200 Berlin, Germany b
a r t i c l e
i n f o
Article history: Received 21 September 2009 Received in revised form 26 November 2009 Accepted 27 November 2009
Keywords: (Epidemiology of) Lung cancer Risk factor Estrogen therapy Menopausal hormone therapy Hormone replacement therapy
a b s t r a c t Objectives: Lung cancer rates increase among women in many regions of the world. To explore whether menopausal hormone therapy (MHT) plays a role. Methods: We conducted a systematic search of the literature and performed meta-analyses of cohort studies (C), case–control studies (CC), randomized controlled trials (RCTs), and cancer registry studies (CR) to analyse the impact of estrogen therapy (ET), estrogen/progestin therapy (EPT) and any hormone therapy (HT) on lung cancer risks. We explored associations between ever-use of therapies and risks, analysed annual changes of risk, and the impact of therapies on histological subtypes. We calculated summary odds ratios, relative risks, 95% confidence intervals (CI; fixed-effects model), and assessed heterogeneity across studies. Eighteen studies were eligible (9 CC, 4 C, 3 RCT, 2 CR). Results: We found a significant increase of risk – 76.2% – in non-smoking women with adenocarcinoma (CI 1.072–2.898) reporting ever-use of HT. Estrogen plus progestin therapy does not change the risk; however, the pooled analysis of 2 RCTs points at an increased risk (RR 1.359; CI 1.031–1.791). Our further results should be interpreted with caution as significances were found in analyses only when smoking and non-smoking women, various hormone regimens, or histological subtypes, respectively, were pooled. Conclusions: Dedicated studies designed to more adequately delineate the role of MHT are necessary to substantiate whether use of MHT is a risk factor for this or other types of lung cancer. © 2009 Elsevier Ireland Ltd. All rights reserved.
Contents 1. 2.
3.
4.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Materials and methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Identification of studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Inclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Data extraction, statistical methods and assessment of homogeneity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Assessment of study quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Study characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Ever-use of HT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Duration of use of HT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Declaration of conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Provenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
198 199 199 199 199 200 200 200 202 202 202 203 203 203 203 203
1. Introduction ∗ Corresponding author. Tel.: +49 30 8445 3227; fax: +49 30 8455 2352. E-mail address: [email protected] (M. Dören). 0378-5122/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.maturitas.2009.11.027
Lung cancer is currently the most common cancer in the world and the leading cause of cancer-related death. Lung can-
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
cer incidence rates either decreased or were stable among males except in Japan. In contrast, lung cancer rates increased among women [1], but apparently may no longer be increasing at least in some populations [2], and may have leveled off in women after increasing for decades [3]. Above all, the environmental risk factor cigarette smoking is relatively most relevant, but also exposure to asbestos, radon, air pollution are important in this context, as are behavioural, genetic and dietary factors [4,5]. Squamous cell carcinoma was the most frequent type of lung cancer observed in the past, and small cell carcinoma was the next most frequent. In the late 1970s, the first evidence of a shift toward a predominance of adenocarcinoma was noted [6] and now adenocarcinoma of the lung is the most common histologic type [7]. Whether endogenous and or exogenous estrogens including menopausal hormone therapy (MHT) are relevant to contribute to the understanding of the epidemiology and tumor biology of lung cancer in women and men is unclear [8,9]. It is not known whether women may have a greater risk of lung cancer than men at the same level of smoking. Hypotheses have been based on hormonally related differences in response to carcinogens, but the evidence appears to be both limited and mixed [10]. The significance of estrogen and progesterone receptor expression in normal and tumor cells, adenocarcinoma, squamous, and small cell carcinoma of the lung is unknown [11–13], as these receptors are expressed in many other normal and tumor cells of other organs. However, sex differences in lung cancer outcome have been reported; lung cancer survival in women is decreased compared with men in studies adjusted for smoking and comorbidities [14,15]. According to the evaluation of the International Agency of Research against Cancer (IARC) published 2007 [16], large randomized trials suggest that risk for lung cancer was slightly but not significantly elevated in users of combined estrogen/progestin hormonal therapy. Results of observational studies produced different results, suggesting reduced risk [17–19], increased risk [20], or no change of risk [21,22]. Use of MHT was associated with decreased survival in one study in women with lung cancer [23]. In order to better delineate the impact of MHT on lung cancer risk we conducted a systematic search of the literature and performed meta-analyses of available evidence provided by cohort studies (C), case–control studies (CC), randomized controlled trials (RCTs), and cancer registry studies (CR) to analyse the impact of various menopausal hormone therapies [estrogen replacement therapy (ET), estrogen/progestin therapy (EPT) and hormone therapy (HT), the latter including any hormone regimen, sometimes unspecified or unknown preparations] on lung cancer risks. We explored associations between ever-use of these types of therapy and risks, analysed annual changes of risk, and potentially different impacts of HT on histological subtypes. 2. Materials and methods 2.1. Identification of studies We performed a computerized search of several databases, including Medline (1 January, 1966–25 July 2008), CANCERLIT, EMBASE, Scopus, the Cochrane Library and the Cochrane Controlled Trials Register. We used the Medical Subject Headings and/or text words ‘hormone replacement therapy’, ‘hormone therapy’, ‘(o)estrogen (replacement) therapy’, ‘estradiol (replacement) therapy’, ‘estrogen and progest* (replacement) therapy’, ‘HRT’, ‘ERT’, ‘HT’, ‘post(-)menopausal estrogens (hormones)’, ‘reproductive hormones’, ‘non-contraceptive hormones (estrogens)’, ‘lung cancer’ or ‘carcinoma’ or ‘neoplasm’ or ‘tumo(u)r’, ‘bronchial cancer’ or ‘carcinoma’ or ‘neoplasm’ or ‘tumo(u)r’, ‘case(–)control study’, ‘cohort study’, ‘cancer registry’ and any of the terms ‘randomized, randomized, controlled and clinical’ in conjunction with ‘trial’ or ‘study’ in
199
multiple combinations where applicable. All studies not conducted in women were a priori excluded. We used snowballing (review of references of identified studies), scrutinized systematic reviews addressing various aspects of HT, checked references of previous systematic searches regarding a related cancer topic [24,25] and of systematic evaluations [16] to potentially identify further studies. Search of editorials, supplements, proceedings, books, abstract books and proceedings of major menopause meetings, respectively, was restricted to the preceding 5 years (2003–July 2008). The titles and abstracts of all potentially relevant publications were examined to determine the relevance of the information; full articles were scrutinized if any potentially relevant information was found in a retrieved abstract. Searches were conducted independently by two reviewers (M.D. and C.M.G.). We did not impose language restrictions. 2.2. Inclusion criteria We included C, CC, RCT and CR, if these publications provided information upon ever-use of any type of HT, risk by duration of use or increase of risk within a given time interval, respectively, of ET, or EPT or HT as defined (C.M.G., M.D. and E.M.G.). Fully published studies, not abstracts, were included if confidence intervals (CI) or standard errors of risk estimates and dates on conduct of the study were provided or if data provided allowed calculation of confidence intervals. In studies with multiple publications from the same population, we included only data from the most recent publication. In the case of double publication, we included only the data sets of the first publication or the one providing the most extractable data. 2.3. Data extraction, statistical methods and assessment of homogeneity Data were extracted by two reviewers (C.M.G. and E.M.G.), in case of different raw data sets theses differences were resolved by discussion to reach consensus. A priori objectives were the association between (i) hormone regimens (ET, EPT and HT) and risk of lung cancer, (ii) the magnitude of ever-use (estimate of a total) and annual risk in pre-specified hormone regimen groups (ET, EPT and HT), and (iii) the potential impact of specified hormone regimens on histological subtypes. All statistical analyses were performed independently by one reviewer (E.M.G.). First, to summarize effects of HT on risk of lung cancer irrespective of duration or dosage, point estimates and CI were used in a fixed-effects model applying the general variance-based method [26] (see Ref. [25] for details). Second, slopes for both individual studies and summary slopes were calculated using inverse variance-weighted least squares estimates in order to estimate summary slopes for calculation of increase of risk per year of use [27]. We examined heterogeneity with two methods. We analysed studies by applying the general variance-based method [26], providing for Cochran’s Q for individual substrata and for various totals of substrata. Additionally, we calculated the proportion of variance in pooled estimates [28] due to heterogeneity in calculating I2 . In case of a paucity of eligible data sets we refrained from doing so [29]. Attributable risks were calculated according to the formula (odds ratio − 1)/odds ratio. All analyses were stratified by type of MHT:ET, EPT, and HT as defined. Where possible, analyses were stratified according to histology. When pooling was done in studies which provided risk estimates for several mutually exclusive histological entities, we regarded these risk estimates as being derived from independent datasets, analogous to different independent studies. Therefore we refer to ‘datasets’ instead of ‘studies’. We used SAS version 9.1 (SAS Institute, Cary, NC, USA) for all analyses. Due to the heterogeneity of study types (both randomized trials and epidemiologic stud-
200
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
Table 1 Included studiesa . Study (alphabetic order)
Region
Study type
ET
Blackman et al. [22] Chen et al. [33] Chlebowski et al. [34] Corrao et al. [35] Elliott and Hannaford [36] Ettinger et al. [37] Hulley et al. [31] Kabat et al. [38] Kreuzer et al. [17] Liu et al. [39] Olsson et al. [40] Persson et al. [21] Persson et al. [21] Pukkala et al. [41] Ramnath et al. [42] Rodriguez et al. [19] Schabath et al. [18] Schwartz et al. [43] Taioli et al. [20]
US Taiwan US Italy UK US US Canada Germany Japan Sweden Sweden Sweden Finland US US US US US
CC CC RCT C CC, nested CC RCT (HERS II) RCT (NBSS) CC C C/HR C/SMR C/SIR C/SIR CC C CC CC CC
Duration of MHT use (and histological subtypes where available)
Ever-use of MHT (and histological subtypes, where available)
EPT
HT
Smoking status (HT only)
ADALLNSCSCSQC
HT
Smoking status (HT only)
SM
NSM
SM
ALL ALL
ALLAD ALL
ALL
NSM
ALL ALL ALL ALL ALL AD AD ALL SC SQC
ALL
ALL ALL
ALL ALL ALLAD
ALL ALL
ALL ALL
AD ADALL
ALL ALL
ALL ALL
ALL ALL ALL ALL ALL
NSC ALL ALL NSC NSC AD
ALL ALL
SC
AD
MHT: menopausal hormone therapy; ET: estrogen replacement therapy; EPT: estrogen/progestin therapy; HT: hormone therapy, the latter including any hormone regimen, sometimes unspecified or unknown preparations; SMR: standardized mortality ratio; SIR: standardized incidence ratio; HR: hazard ratio; cc: case control study; c: cohort study; RCT: randomized controlled trial; S: smoker; NSM: non-smoker; AD: adenocarcinoma; ALL: all subtypes combined; NSC: non-small cell carcinoma; SC: small cell carcinoma; SQC: squamous cell carcinoma; WHI: The Womens’ Health Initiative Randomized Controlled Trial; NBSS: Canadian National Breast Screening Study; HERS II: Heart and Estrogen/Progestin Replacement Study Follow-up. a One study excluded [32] after revision and inclusion Ref. [34]. Table 2 Ever-use of ET by histology. Histology
Authors
Risk ratio (95% CI)
All subtypes
Kreuzer et al. [17] Rodriguez et al. [19] Schabath et al. [18] Total
0.780 (0.490–1.210) 0.760 (0.600–0.940) 0.650 (0.470–0.890)
Summary risk ratio (95% CI)
Q (p)
I2
0.728 (0.610–0.869)
0.681 (0.711)
0
ET: estrogen therapy; CI: confidence interval; Q: Cochran’s Q; I2 : 95% uncertainty interval; p < 0.05 regarded as significant.
ies included) neither QUORUM nor MOOSE guidelines were fully instrumental to design this systematic review with meta-analyses.
was sought after discussion of reasons that may lead to inclusion or exclusion of an individual study.
2.4. Assessment of study quality
3. Results
All reviewers assessed included studies. In the absence of a universal scale for measuring quality of both observational studies and randomized trials, we used criteria developed by the U.S. Preventive Services Task Force [30]. In case of different opinions consensus
3.1. Study characteristics Out of 46 electronically retrieved publications, 13 studies not addressing our research questions were excluded, as were 11 nar-
Table 3 Ever-use of EPT by histologya . Histology
Authors
Risk ratio (95% CI)
All subtypes
Chlebowski et al. [34] Hulley et al. [31] Pukkala et al. [41] Pukkala et al. [41] Rodriguez et al. [19] Schabath et al. [18] Total
1.230 (0.920–1.630) 1.730 (0.930–3.210) 1.200 (0.690–1.900)a 0.750 (0.530–1.000)b 0.760 (0.570–1.010) 0.610 (0.400–0.920)
All subtypes RCTs only
Chlebowski et al. [34] Hulley et al. [31] Total (RCTs)
1.230 (0.920–1.630) 1.730 (0.930–3.210) 2
Summary risk ratio (95% CI)
Q (p)
0.912 (0.784–1.061)
4.062 (0.54)
1.359 (1.031–1.791)
0.726 (0.394)
EPT: estrogen/progestin therapy; CI: confidence interval; Q: Cochran’s Q; I (95% uncertainty interval) not calculated because of Q being too near to degrees of freedom; p < 0.05 regarded as significant; RCT: randomized controlled trial. a Data extracted for long-cycle therapy (addition of a progestin to oral HT every 3rd month). b Data extracted for monthly use of progestin in conjunction with oral HT.
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
201
Table 4 Ever-use of HT by histology. Histology
Authors
Risk ratio (95% CI)
All subtypes
Chen et al. [33] Elliott et al. [36] Ettinger et al. [37] Kreuzer et al. [17] Olsson et al. [40] Persson et al. [21] Rodriguez et al. [19] Schabath et al. [18] Schabath et al. [18] Schabath et al. [18] Total
0.700 (0.530–0.940) 2.000 (0.500–2.100) 0.220 (0.040–1.150) 0.830 (0.640–1.090) 0.730 (0.310–1.430) 1.000 (0.800–1.200) 0.760 (0.620–0.920) 0.380 (0.180–0.790)a 0.740 (0.540–1.020)b 0.380 (0.070–2.120)c
Adenocarcinoma
Chen et al. [33] Kabat et al. [38] Kreuzer et al. [17] Taioli et al. [20] Total
0.760 (0.560–1.030) 1.090 (0.850–1.400) 0.890 (0.660–1.190) 1.700 (1.000–2.800)
Non-small
Chen et al. [33] Ramnath et al. [42] Schabath et al. [18] Schwartz et al. [43] Total
0.710 (0.530–0.950) 0.670 (0.530–0.850) 0.690 (0.500–0.960) 0.810 (0.580–1.130)
Small
Chen et al. [33] Kreuzer et al. [17] Schabath et al. [18] Total
0.510 (0.110–2.330) 0.710 (0.460–1.090) 1.550 (0.690–3.450)
Squamous
Chen et al. [33] Kreuzer et al. [17] Total
0.650 (0.290–1.450) 0.740 (0.510–1.210)
Summary risk ratio (95% CI)
Q (p)
I2
0.796 (0.715–0.887)
13.76 (0.131)
0
0.977 (0.837–1.141)
7.902 (0.048)
n.c.
0.708 (0.613–0.817)
0.861 (0.835)
0
0.821 (0.566–1.189)
3.169 (0.205)
n.c.*
0.723 (0.516–1.014)
0.082 (0.775)
n.c.*
HT: hormone therapy as defined; CI: confidence interval; Q: Cochran’s Q; I2 : 95% uncertainty interval; n.c.: I2 not calculated because of heterogeneity; n.c.*: I2 not calculated because of too few observations [29]. a African-American ethnicity. b Caucasian ethnicity. c Hispanic ethnicity.
Table 5 Ever-use of HT by histology and smoking status. Histology
Smoking status
Authors
Risk ratio (95% CI)
All
NSM
Blackman et al. [22] Chen et al. [33] Kabat et al. [38] Kreuzer et al. [17] Liu et al. [39] Liu et al. [39] Rodriguez et al. [19] Schabath et al. [18] Total
1.400 (0.600–3.300) 0.700 (0.520–0.940) 0.950 (0.580–1.550) 1.050 (0.740–1.490) 1.190 (0.600–2.330)a 2.400 (1.070–5.400)b 0.560 (0.330–0.950) 0.720 (0.370–1.400)
Adenocarcinoma
NSM
Blackman et al. [22] Liu et al. [39] Liu et al. [39] Total
2.000 (0.700–5.600)c 1.230 (0.590–2.580)a 2.710 (1.120–6.580)b
All
SM
Blackman et al. [22] Chen et al. [33] Kabat et al. [38] Kreuzer et al. [17] Rodriguez et al. [19] Schabath et al. [18] Schabath et al. [18] Total
1.000 (0.800–1.400) 0.170 (0.010–2.020) 1.080 (0.910–1.290) 0.730 (0.510–1.030) 0.760 (0.550–1.050) 0.590 (0.380–0.920)d 0.730 (0.460–1.150)e
Summary risk ratio (95% CI)
Q (p)
I2
0.878 (0.741–1.041)
14.33 (0.046)
0
1.762 (1.072–2.898)
1.887 (0.389)
n.c.*
0.919 (0.823–1.027)
13.09 (0.042)
0
HT: hormone therapy as defined; CI: confidence interval; SM: current or ever or former smoker; NSM: never-smoker; Q: Cochran’s Q; I2 : 95% uncertainty interval; n.c.*: not calculated [29]. a Women with natural menopause. b Women with induced menopause. c Use of estrogen replacement therapy. d Current smoker. e Former smoker.
202
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
Table 6 Change of risk ratio per year of HT intake. Histology
Authors
Risk change per year of HT intake (95% CI)
Adenocarcinoma
Kabat et al. [38] Kreuzer et al. [17] Taioli et al. [20] Total
1.037 (1.004–.071) 0.953 (0.902–1.008) 1.299 (0.966–1.746)
Corrao et al. [35] Kreuzer et al. [17] Total
0.986 (0.819–1.188) 0.951 (0.906–0.998)
All
Pooled risk change (95% CI)
Q (p)
I2 (95% uncertainty interval)
1.017 (0.989–1.046)
9.57 (0.008)
0 (0–60)
0.953 (0.910–0.999)
2.05 (0.153)
n.c.
HT: hormone therapy as defined; CI: confidence interval; Q: Cochran’s Q; I2 : 95% uncertainty interval; n.c.: not calculated [29].
rative reviews and one editorial. The remaining 22 studies were further scrutinized; four studies with no extractable data had to be excluded, and two studies providing first reports of a given cohort study with short follow-up (n = 16 eligible studies). It should be noted that two RCTs [31,32] were not identified by electronic but by hand searches of the largest published menopausal hormone therapy trials known to have reported cancer data. The 18 studies finally included in this review (16 + 2 = 18) reflect a broad diversity of study designs regarding variables of analysis (Table 1). Most studies do not provide data on smoking status of participating women or discriminate between different histological diagnoses. Furthermore, only in a minority of studies, the type of HT is described clearly as ET or EPT. Only five studies provide risk estimates by duration of MHT. We excluded two studies [44,45] because another included analysis of the same cohort study provided a longer follow-up [21]. As requested by the reviewers, data from the Women’s Health Initiative (WHI) randomized controlled trial (estrogen plus progestin arm), published 10 October 2009, and therefore outside the time frame for literature search as indicated above, were additionally extracted and included [34]. Consequently, one former publication on WHI health risks, initially included though not dedicated to the subject of lung cancer because it provided some data on lung cancer in the comment section [32], was excluded. 3.2. Ever-use of HT Ever-use of ET is associated with a significant 27% decrease (rounded) of risk in women irrespective of smoking history when all histologies are pooled (Table 2; Cochran’s Q suggests homogeneity). Estrogen plus progestin therapy (six data sets extracted from five studies) does not change the risk; however, the pooled analysis of the only two RCTs available points at an increased risk in the presence of homogeneity for Q (Table 3). When all hormone therapies as defined are pooled as well as all histological subtypes, there is a reduction of risk by 20% (rounded) in a data set suggesting homogeneity (odds ratio 0.796; 95% CI 0.715–0.887). A larger significant decrease (−29%) is found in the analysis of four studies addressing non-small cell cancers with no suggestion of heterogeneity (odds
ratio 0.708; 95% CI 0.613–0.817). Analyses for other histological entities did not yield significant findings (Table 4). There is a significant increase of lung cancer risk in non-smoking women with adenocarcinoma (odds ratio 1.76; 95% CI 1.072–2.898), based on three data sets from two studies with overall 815 cases of lung cancer, Cochran’s Q does not indicate heterogeneity (Table 5). 3.3. Duration of use of HT Four studies have analysed the impact of duration of HT use on lung cancer risk by histological classification. In two studies with all histologies combined (lower panel) there is a small but significant decrease of risk (4.7% per year of HT use; studies appear to be homogenous), whereas the risk for adenocarcinoma appears to be slightly increased, though not significant (OR 1.017; 95% CI 0.989–1.046). However, the latter analysis shows major heterogeneity of studies (Table 6). A further differentiation according to smoking status, based on few eligible trials (n = 3) (Table 7) did not provide significant findings. 4. Discussion We found a significant increase of lung cancer risk in non-smoking women reporting ever-use of HT diagnosed with adenocarcinoma. This is a relevant finding as the attributable risk is 76% (Table 5). Furthermore, data from placebo-controlled randomized trials [31,34] comparing a specific estrogen and progestin therapy indicate an increased risk of lung cancer (histologies reported only specified in Ref. [34]) in hormone-users, representing a heterogenous group of women including never, past and current smokers (Table 3). In one RCT the risk for adenocarcinoma was increased, but this was not significant [34]. In the latest evaluation of the IARC [16] only a fraction of the studies we found eligible for extraction and analyses were briefly reviewed; specific subtypes of lung cancer were not considered. The relative incidence of adenocarcinoma of the lung has been steadily increasing in the recent past, obviously to a greater extent in the US; in the US age-adjusted incidence rates captured 1992–1998, the proportion of this histologic type was 49.4% of all
Table 7 Change of risk ratio per year of HT intake by smoking status. Histology
Authors
Risk change per year of HT intake (95% CI)
Non-smoking women Kabat et al. [38] All Kreuzer et al. [17] Summary
0.989 (0.877–1.115) 0.982 (0.919–1.049)
Smoking women Blackman et al. [22] All Kabat et al. [38] Kreuzer et al. [17] Summary
1.019 (0.980–1.059) 1.042 (1.002–1.084) 0.931 (0.874–0.991)
Pooled risk change (95% CI)
Q (p)
I2 (95% uncertainty interval)
0.984 (0.928–1.043)
0.04 (0.844)
n.c.*
1.014 (0.989–1.039)
9.22 (0.010)
0 (0–60)
HT: hormone therapy as defined; CI: confidence interval; Q: Cochran’s Q; I2 : 95% uncertainty interval; n.c.*: not calculated [29].
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
lung cancers in white women (Table 1 [5]). The significant increase of lung cancer risk of the adenomatous type in non-smoking women is of major importance from a public health point of view, as both in the US as e.g. in Germany this histologic entity is the relatively most common subtype in lung cancer patients. In the Munich Clinical Cancer Registry out of 6.696 female patients with lung cancer 43.8% were adenocarcinomas, the most frequent histologic entity in women (Hölzel, 2008; personal communication), compared to 25.5% of 17.896 lung cancer cases in males. In contrast to these figures, Ginsberg and co-authors [46] report for both genders a prevalence of 40% in the US. The most important risk factors for adenocarcinoma are cigarette smoking and exposure to environmental tobacco smoke [5]. The relevance of exogenous hormones for this entity has not been studied systematically. The overwhelming impact of cigarette smoking as a cause of lung cancer imposes challenges to detecting the role that other lifestyle factors, including MHT, may play in the development of lung cancer. In this context, it is noteworthy that the risk of lung cancer doubled in the HT group, evaluated in the largest randomized placebo-controlled trial of MHT until today, reporting health outcomes at a mean of 2.4 years of follow-up after EPT was stopped [32]. In this trial, most deaths during the postintervention phase were cancer-related in both treatment groups, but only to a smaller extent due to the pre-specified outcomes breast, colorectal, endometrial, or ovarian cancer. Among the other cancers, most were lung cancers. In the initial report of this study [47], lung cancer incidence was not different in women assigned to placebo or EPT (hazard ratio 1.04; 95% CI, 0.71–1.53). A recently published post hoc analysis [34] of this trial [47] confirmed the finding of an increased mortality due to lung cancer in the presence of a nonsignificantly increased incidence for this disease. We do not know whether increased expression of ER- might have a role in the pathogenesis of lung cancer, especially adenocarcinomas, as hypothesized [48], this issue was beyond the scope of our analyses. Our study has a number of limitations: firstly, our other results, except for those regarding lung cancer risk in non-smoking women reporting ever-use of MHT and diagnosed with adenocarcinoma, reported in the previous section should be interpreted with great caution as significances were found in analyses only when smoking and non-smoking women, various hormone regimens, or histological subtypes, respectively, were pooled. There are too few studies providing consistently risk estimates for both smokers and non-smokers; there are no studies providing risk indicators for adenocarcinoma in smoking women. Secondly, there is a paucity of studies to allow assessment of at least defined groups of hormone regimens such as ET and EPT. Thirdly, included studies did not provide extractable data on passive smoking, a further potential risk factor for lung cancer. Likewise, nutritional factors, especially the consumption of red meat in non-smokers [49,50] could not be considered, as could not air pollution and radiation. It is beyond the scope of this analysis to elucidate whether the increase of MHT use for many years [51] has contributed to the changing pattern of histologic subtypes of lung cancer in females. Likewise, whether the recent decrease of use in the US and several other countries [52] may have an impact in the future is also unknown. In conclusion our results suggest that ever-use of HT in nonsmoking women may increase the risk of adenocarcinoma of the lung. Furthermore, data from RCTs suggest that EPT increases the mortality of lung cancer. MHT may be regarded as a potential risk factor for this type of lung cancer, pointing at a potential for prevention. However, compared to other cancers affecting females, the association between lung cancer and use of menopausal hormone therapy clearly warrants more and specifically designed studies to arrive at more definite conclusions.
203
Declaration of conflict of interest M.D. participated in a phase III study of a drug tested for prevention of osteoporosis in postmenopausal women, sponsored by Pfizer, USA; M.D. is member of an advisory board for Women’s health issues of the Federal Centre for Health Information (Germany; http://www.frauengesundheitsportal.de/?uid= 08c61583f2c66abc96042e584af7b9fa&id=Seite1294). Contributors For the first and second version of the manuscript: C.M. Greiser: Conducted the literature review, contributed to the conception of the analysis, the interpretation of results and to the drafting of the manuscript. E.M. Greiser: Contributed to the conception of the analysis, performed the statistical analyses (first and revised version), and contributed to the interpretation of results. M. Dören: Conceptualized and supervised the study, participated in the literature review was involved in the interpretation of the data, writing of the manuscript (first and revised version) and critical review, and arranged for funding. All authors have seen and approved the final version and the revised version of the manuscript. Berlin, 27 November, 2009 M. Dören, on behalf of all authors. Funding This study was conducted at the 1 Institute for Public Health and Nursing Research, Faculty of Health Sciences, Bremen University; 2 Epi.Consult GmbH, Bremen; and the 3 Charité - Universitätsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany. Provenance Not commissioned but externally peer reviewed. References [1] Ernster VL. Impact of tobacco on women’s health. In: Samet JM, Yoon S-Y, editors. Women and the tobacco epidemic: challenges for the 21st century. Geneva, Switzerland: World Health Organization; 2001. p. 1–16. [2] Espey DK, Wu X-C, Swan J, et al. Annual report to the nation on the status of cancer, 1975–2004, featuring cancer in American Indians and Alaska Natives. Cancer 2007;110(10):2119–52. [3] Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin 2005;55(1):10–30. [4] Kamangar F, Dores GM, Anderson WF. Patterns of cancer incidence, mortality, and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol 2006;24(14):2137–50. [5] Alberg AJ, Samet JM. Epidemiology of lung cancer. Chest 2003;123(1 Suppl.):21–49. [6] Charloux A, Quoix E, Wolkove N, Small D, Pauli G, Kreisman H. The increasing incidence of lung adenocarcinoma: reality or artefact? A review of the epidemiology of lung adenocarcinoma. Int J Epidemiol 1997;26(1):14–23. [7] Wingo PA, Ries LA, Giovino GA, et al. Annual report to the nation on the status of cancer, 1973–1996, with a special section on lung cancer and tobacco smoking. J Natl Cancer Inst 1999;91(8):675–90. [8] Gasperino J, Rom WN. Gender and lung cancer. Clin Lung Cancer 2004;5(6):353–9. [9] Belani CP, Marts S, Schiller J, Socinski MA. Women and lung cancer: epidemiology, tumor biology, and emerging trends in clinical research. Lung Cancer 2007;55(1):5–23. [10] US Department of Health, and Human Services (US-DHSS). Women and smoking: a report of the Surgeon General. Rockville, MD: US Department of Health and Human Services; 2001. [11] Chaudhuri PK, Thomas PA, Walker MJ, Briele HA, Das Gupta TK, Beattie CW. Steroid receptors in human lung cancer cytosols. Cancer Lett 1982;16(3):327–32. [12] Mollerup S, Jorgensen K, Berge G, Haugen A. Expression of estrogen receptors ␣ and  in human lung tissue and cell lines. Lung Cancer 2002;37(2): 153–9. [13] Cagle PT, Mody DR, Schwartz MR. Estrogen and progesterone receptors in bronchogenic carcinoma. Cancer Res 1990;50(20):6632–5.
204
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
[14] Palomares MR, Sayre JW, Shekar KC, Lillington LM, Chlebowski RT. Gender influence on weight loss pattern and survival of non-small cell lung carcinoma patients. Cancer 1996;78(10):2119–26. [15] Wisnivesky JP, Halm EA. Sex differences in lung cancer survival: do tumors behave differently in elderly women? J Clin Oncol 2007;25(13):1705–12. [16] IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Combined estrogen-progestogen contraceptives and combined estrogenprogestogen menopausal therapy, vol. 91, Lyon: France; 2007. http://monographs.iarc.fr/ENG/Monographs/vol91/mono91.pdf (last accessed on 21.09.09). [17] Kreuzer M, Gerken M, Heinrich J, Kreienbrock L, Wichmann H-E. Hormonal factors and risk of lung cancer among women? Int J Epidemiol 2003;32(2):263–71. [18] Schabath MB, Wu X, Vassilopoulou-Sellin R, Vaporciyan AA, Spitz MR. Hormone replacement therapy and lung cancer risk: a case–control analysis. Clin Cancer Res 2004;10(1 Pt 1):113–23. [19] Rodriguez C, Spencer Feigelson F, Deka A, et al. Postmenopausal hormone therapy and lung cancer risk in the cancer Prevention Study II Nutrition Cohort. Cancer Epidemiol Biomarkers Prev 2008;17(3):655–60. [20] Taioli E, Wynder El. Endocrine factors and adenocarcinoma of the lung in women. J Natl Cancer Inst 1994;86(11):869–70. [21] Persson I, Yuen J, Bergkvist L, Schairer C. Cancer incidence and mortality in women receiving estrogen and estrogen–progestin replacement therapy—long-term follow-up of a Swedish cohort. Int J Cancer 1996; 67(3):327–32. [22] Blackman JA, Coogan PF, Rosenberg L, et al. Estrogen replacement therapy and risk of lung cancer. Pharmacoepidemiol Drug Safety 2002;11(7):561–7. [23] Ganti AK, Sahmoun AE, Panwalker AW, Tendulkar KK, Potti A. Hormone replacement therapy is associated with decreased survival in women with lung cancer. J Clin Oncol 2006;24(1):59–63. [24] Greiser CM, Greiser EM, Dören M. Menopausal hormone therapy and risk of breast cancer: a meta-analysis of epidemiological studies and randomized controlled trials. Human Reprod Update 2005;11(6):561–7. [25] Greiser CM, Greiser EM, Dören M. Menopausal hormone therapy and risk of ovarian cancer—systematic review and meta-analysis. Hum Reprod Update 2007;13(5):4534–63. [26] Petitti DB. Statistical methods in meta-analysis. In: Pettiti DB, editor. Metaanalysis, decision analysis, and cost-effectiveness analysis. 2nd ed. New York: Oxford University Press; 2000, 63–7, 111–3, 213–24. [27] Berlin JA, Longnecker MP, Greenland S. Meta-analysis of epidemiologic dose–response data. Epidemiology 1993;4(3):218–28. [28] Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327(7414):557–60. [29] Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21(11):1539–58. [30] Harris RP, Helfand M, Woolf SH, et al. Current methods of the U.S. Preventive Services Task Force. A review of the process. Am J Prev Med 2001;20(3 Suppl.):21–35. [31] Hulley S, Furberg C, Barrett-Connor E, et al. Noncardiocascular disease outcomes during 6.8 years of hormone therapy. JAMA 2002;288(1):58–66. [32] Heiss G, Wallace R, Anderson GL, et al. Health risks and benefits 3 years after stopping randomized treatment with estrogen and progestin. JAMA 2008;299(9):1036045. [33] Chen K-Y, Hsiao C-F, Chang G-C, et al. Hormone replacement therapy and lung cancer risk in Chinese. Cancer 2007;110(8):1768–75.
[34] Chlebowski RT, Schwartz AG, Wakelee H, et al. Oestrogen plus progestin and lung cancer in postmenopausal women (Women’s Health Initiative trial): a post hoc analysis of a randomised controlled trial. Lancet 2009;374:1243–51. [35] Corrao G, Zambon A, Conti V, et al. Menopause hormone replacement therapy and cancer risk: an Italian record linkage investigation. Ann Oncol 2008;19(1):150–5. [36] Elliott AM, Hannaford PC. Use of exogenous hormones by women and lung cancer: evidence from the Royal College of General Practitioners’ Oral Contraception Study. Contraception 2006;73(4):331–5. [37] Ettinger B, Friedman GD, Bush T, Quesenberry CP. Reduced mortality associated with long-term postmenopausal estrogen therapy. Obstet Gynecol 1996;87(1):6–12. [38] Kabat GC, Miller AB, Rohan TE. Reproductive and hormonal factors and risk of lung cancer in women: a prospective cohort study. Int J Cancer 2007;120(10):2214–20. [39] Liu Y, Inoue M, Sobue T, Tsugane S. Reproductive factors, hormone use and the risk of lung cancer among middle-aged never-smoking Japanese women: a large scale-population-based cohort study. Int J Cancer 2005;117(4): 662–6. [40] Olsson H, Bladström A, Ingvar C. Are smoking-associated cancers prevented or postponed in women using hormone replacement therapy? Obstet Gynecol 2003;102(3):565–70. [41] Pukkala E, Tulenheimo-Silfvast A, Leminen A. Incidence of cancer among women using long versus monthly cycle hormone replacement therapy, Finland 1994–1997. Cancer Causes Control 2001;12(2):111–5. [42] Ramnath N, Menezes RJ, Loewen G, et al. Hormone replacement therapy as a risk factor for non-small cell lung cancer: results of a case–control study. Oncology 2007;73(5–6):305–10. [43] Schwartz AG, Wenzlaff AS, Prysak GM, et al. Reproductive factors, hormone use, estrogen receptor expression and risk of non-small-cell lung cancer in women. J Clin Oncol 2007;25(36):5785–92. [44] Adami H-O, Persson I, Hoover R, Schairer C, Bergkvist L. Risk of cancer in women receiving hormone replacement therapy. Int J Cancer 1989;44(5):833–9. [45] Schairer C, Adami H-O, Hoover R, Persson I. Cause-specific mortality in women receiving hormone replacement therapy. Epidemiology 1997;8(1):59–65. [46] Ginsberg RJ, Vokes EE, Rosenzweig K. Non-small cell lung cancer. In: DeVita VT, Hellman S, Rosenberg SA, editors. Cancer—principles and practice of oncology. 6th ed. 2001. p. 925–83. [47] Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002;288(3):321–33. [48] Omoto Y, Kobayahi Y, Nishida K, et al. Expression, function, and clinical implications of the estrogen receptor beta in human lung cancers. Biochem Biophys Res Commun 2001;285(2):340–7. [49] Alavanja MC, Field RW, Sinha R, et al. Lung cancer risk and red meat consumption among Iowa women. Lung Cancer 2001;34(1):37–46. [50] De Stefani E, Brennan P, Boffetta P, et al. Diet and adenocarcinoma of the lung: a case–control study in Uruguay. Lung Cancer 2002;35(1):43–51. [51] Hersh AL, Stefanick ML, Stafford RS. National use of postmenopausal hormone therapy. JAMA 2004;291(1):47–53. [52] Du Y, Dören M, Melchert H-UW, Scheidt-Nave C, Knopf H. Differences in menopausal hormone therapy use among women in Germany between 1998 and 2003. BMC Women’s Health 2004;7:19.
Contents lists available at ScienceDirect
Maturitas journal homepage: www.elsevier.com/locate/maturitas
Review
Menopausal hormone therapy and risk of lung cancer—Systematic review and meta-analysis Claudia M. Greiser a , Eberhard M. Greiser a,b , Martina Dören c,∗ a
Epi.Consult GmbH, Musweiler, Germany Institute for Public Health and Nursing Research, Faculty of Health Sciences, Bremen University, Bremen, Germany c Charité - Universitätsmedizin Berlin, Campus Benjamin Franklin, Clinical Research Centre of Women’s Health, Hindenburgdamm 30, D-12200 Berlin, Germany b
a r t i c l e
i n f o
Article history: Received 21 September 2009 Received in revised form 26 November 2009 Accepted 27 November 2009
Keywords: (Epidemiology of) Lung cancer Risk factor Estrogen therapy Menopausal hormone therapy Hormone replacement therapy
a b s t r a c t Objectives: Lung cancer rates increase among women in many regions of the world. To explore whether menopausal hormone therapy (MHT) plays a role. Methods: We conducted a systematic search of the literature and performed meta-analyses of cohort studies (C), case–control studies (CC), randomized controlled trials (RCTs), and cancer registry studies (CR) to analyse the impact of estrogen therapy (ET), estrogen/progestin therapy (EPT) and any hormone therapy (HT) on lung cancer risks. We explored associations between ever-use of therapies and risks, analysed annual changes of risk, and the impact of therapies on histological subtypes. We calculated summary odds ratios, relative risks, 95% confidence intervals (CI; fixed-effects model), and assessed heterogeneity across studies. Eighteen studies were eligible (9 CC, 4 C, 3 RCT, 2 CR). Results: We found a significant increase of risk – 76.2% – in non-smoking women with adenocarcinoma (CI 1.072–2.898) reporting ever-use of HT. Estrogen plus progestin therapy does not change the risk; however, the pooled analysis of 2 RCTs points at an increased risk (RR 1.359; CI 1.031–1.791). Our further results should be interpreted with caution as significances were found in analyses only when smoking and non-smoking women, various hormone regimens, or histological subtypes, respectively, were pooled. Conclusions: Dedicated studies designed to more adequately delineate the role of MHT are necessary to substantiate whether use of MHT is a risk factor for this or other types of lung cancer. © 2009 Elsevier Ireland Ltd. All rights reserved.
Contents 1. 2.
3.
4.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Materials and methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Identification of studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Inclusion criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Data extraction, statistical methods and assessment of homogeneity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Assessment of study quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Study characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Ever-use of HT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Duration of use of HT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Declaration of conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Provenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
198 199 199 199 199 200 200 200 202 202 202 203 203 203 203 203
1. Introduction ∗ Corresponding author. Tel.: +49 30 8445 3227; fax: +49 30 8455 2352. E-mail address: [email protected] (M. Dören). 0378-5122/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.maturitas.2009.11.027
Lung cancer is currently the most common cancer in the world and the leading cause of cancer-related death. Lung can-
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
cer incidence rates either decreased or were stable among males except in Japan. In contrast, lung cancer rates increased among women [1], but apparently may no longer be increasing at least in some populations [2], and may have leveled off in women after increasing for decades [3]. Above all, the environmental risk factor cigarette smoking is relatively most relevant, but also exposure to asbestos, radon, air pollution are important in this context, as are behavioural, genetic and dietary factors [4,5]. Squamous cell carcinoma was the most frequent type of lung cancer observed in the past, and small cell carcinoma was the next most frequent. In the late 1970s, the first evidence of a shift toward a predominance of adenocarcinoma was noted [6] and now adenocarcinoma of the lung is the most common histologic type [7]. Whether endogenous and or exogenous estrogens including menopausal hormone therapy (MHT) are relevant to contribute to the understanding of the epidemiology and tumor biology of lung cancer in women and men is unclear [8,9]. It is not known whether women may have a greater risk of lung cancer than men at the same level of smoking. Hypotheses have been based on hormonally related differences in response to carcinogens, but the evidence appears to be both limited and mixed [10]. The significance of estrogen and progesterone receptor expression in normal and tumor cells, adenocarcinoma, squamous, and small cell carcinoma of the lung is unknown [11–13], as these receptors are expressed in many other normal and tumor cells of other organs. However, sex differences in lung cancer outcome have been reported; lung cancer survival in women is decreased compared with men in studies adjusted for smoking and comorbidities [14,15]. According to the evaluation of the International Agency of Research against Cancer (IARC) published 2007 [16], large randomized trials suggest that risk for lung cancer was slightly but not significantly elevated in users of combined estrogen/progestin hormonal therapy. Results of observational studies produced different results, suggesting reduced risk [17–19], increased risk [20], or no change of risk [21,22]. Use of MHT was associated with decreased survival in one study in women with lung cancer [23]. In order to better delineate the impact of MHT on lung cancer risk we conducted a systematic search of the literature and performed meta-analyses of available evidence provided by cohort studies (C), case–control studies (CC), randomized controlled trials (RCTs), and cancer registry studies (CR) to analyse the impact of various menopausal hormone therapies [estrogen replacement therapy (ET), estrogen/progestin therapy (EPT) and hormone therapy (HT), the latter including any hormone regimen, sometimes unspecified or unknown preparations] on lung cancer risks. We explored associations between ever-use of these types of therapy and risks, analysed annual changes of risk, and potentially different impacts of HT on histological subtypes. 2. Materials and methods 2.1. Identification of studies We performed a computerized search of several databases, including Medline (1 January, 1966–25 July 2008), CANCERLIT, EMBASE, Scopus, the Cochrane Library and the Cochrane Controlled Trials Register. We used the Medical Subject Headings and/or text words ‘hormone replacement therapy’, ‘hormone therapy’, ‘(o)estrogen (replacement) therapy’, ‘estradiol (replacement) therapy’, ‘estrogen and progest* (replacement) therapy’, ‘HRT’, ‘ERT’, ‘HT’, ‘post(-)menopausal estrogens (hormones)’, ‘reproductive hormones’, ‘non-contraceptive hormones (estrogens)’, ‘lung cancer’ or ‘carcinoma’ or ‘neoplasm’ or ‘tumo(u)r’, ‘bronchial cancer’ or ‘carcinoma’ or ‘neoplasm’ or ‘tumo(u)r’, ‘case(–)control study’, ‘cohort study’, ‘cancer registry’ and any of the terms ‘randomized, randomized, controlled and clinical’ in conjunction with ‘trial’ or ‘study’ in
199
multiple combinations where applicable. All studies not conducted in women were a priori excluded. We used snowballing (review of references of identified studies), scrutinized systematic reviews addressing various aspects of HT, checked references of previous systematic searches regarding a related cancer topic [24,25] and of systematic evaluations [16] to potentially identify further studies. Search of editorials, supplements, proceedings, books, abstract books and proceedings of major menopause meetings, respectively, was restricted to the preceding 5 years (2003–July 2008). The titles and abstracts of all potentially relevant publications were examined to determine the relevance of the information; full articles were scrutinized if any potentially relevant information was found in a retrieved abstract. Searches were conducted independently by two reviewers (M.D. and C.M.G.). We did not impose language restrictions. 2.2. Inclusion criteria We included C, CC, RCT and CR, if these publications provided information upon ever-use of any type of HT, risk by duration of use or increase of risk within a given time interval, respectively, of ET, or EPT or HT as defined (C.M.G., M.D. and E.M.G.). Fully published studies, not abstracts, were included if confidence intervals (CI) or standard errors of risk estimates and dates on conduct of the study were provided or if data provided allowed calculation of confidence intervals. In studies with multiple publications from the same population, we included only data from the most recent publication. In the case of double publication, we included only the data sets of the first publication or the one providing the most extractable data. 2.3. Data extraction, statistical methods and assessment of homogeneity Data were extracted by two reviewers (C.M.G. and E.M.G.), in case of different raw data sets theses differences were resolved by discussion to reach consensus. A priori objectives were the association between (i) hormone regimens (ET, EPT and HT) and risk of lung cancer, (ii) the magnitude of ever-use (estimate of a total) and annual risk in pre-specified hormone regimen groups (ET, EPT and HT), and (iii) the potential impact of specified hormone regimens on histological subtypes. All statistical analyses were performed independently by one reviewer (E.M.G.). First, to summarize effects of HT on risk of lung cancer irrespective of duration or dosage, point estimates and CI were used in a fixed-effects model applying the general variance-based method [26] (see Ref. [25] for details). Second, slopes for both individual studies and summary slopes were calculated using inverse variance-weighted least squares estimates in order to estimate summary slopes for calculation of increase of risk per year of use [27]. We examined heterogeneity with two methods. We analysed studies by applying the general variance-based method [26], providing for Cochran’s Q for individual substrata and for various totals of substrata. Additionally, we calculated the proportion of variance in pooled estimates [28] due to heterogeneity in calculating I2 . In case of a paucity of eligible data sets we refrained from doing so [29]. Attributable risks were calculated according to the formula (odds ratio − 1)/odds ratio. All analyses were stratified by type of MHT:ET, EPT, and HT as defined. Where possible, analyses were stratified according to histology. When pooling was done in studies which provided risk estimates for several mutually exclusive histological entities, we regarded these risk estimates as being derived from independent datasets, analogous to different independent studies. Therefore we refer to ‘datasets’ instead of ‘studies’. We used SAS version 9.1 (SAS Institute, Cary, NC, USA) for all analyses. Due to the heterogeneity of study types (both randomized trials and epidemiologic stud-
200
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
Table 1 Included studiesa . Study (alphabetic order)
Region
Study type
ET
Blackman et al. [22] Chen et al. [33] Chlebowski et al. [34] Corrao et al. [35] Elliott and Hannaford [36] Ettinger et al. [37] Hulley et al. [31] Kabat et al. [38] Kreuzer et al. [17] Liu et al. [39] Olsson et al. [40] Persson et al. [21] Persson et al. [21] Pukkala et al. [41] Ramnath et al. [42] Rodriguez et al. [19] Schabath et al. [18] Schwartz et al. [43] Taioli et al. [20]
US Taiwan US Italy UK US US Canada Germany Japan Sweden Sweden Sweden Finland US US US US US
CC CC RCT C CC, nested CC RCT (HERS II) RCT (NBSS) CC C C/HR C/SMR C/SIR C/SIR CC C CC CC CC
Duration of MHT use (and histological subtypes where available)
Ever-use of MHT (and histological subtypes, where available)
EPT
HT
Smoking status (HT only)
ADALLNSCSCSQC
HT
Smoking status (HT only)
SM
NSM
SM
ALL ALL
ALLAD ALL
ALL
NSM
ALL ALL ALL ALL ALL AD AD ALL SC SQC
ALL
ALL ALL
ALL ALL ALLAD
ALL ALL
ALL ALL
AD ADALL
ALL ALL
ALL ALL
ALL ALL ALL ALL ALL
NSC ALL ALL NSC NSC AD
ALL ALL
SC
AD
MHT: menopausal hormone therapy; ET: estrogen replacement therapy; EPT: estrogen/progestin therapy; HT: hormone therapy, the latter including any hormone regimen, sometimes unspecified or unknown preparations; SMR: standardized mortality ratio; SIR: standardized incidence ratio; HR: hazard ratio; cc: case control study; c: cohort study; RCT: randomized controlled trial; S: smoker; NSM: non-smoker; AD: adenocarcinoma; ALL: all subtypes combined; NSC: non-small cell carcinoma; SC: small cell carcinoma; SQC: squamous cell carcinoma; WHI: The Womens’ Health Initiative Randomized Controlled Trial; NBSS: Canadian National Breast Screening Study; HERS II: Heart and Estrogen/Progestin Replacement Study Follow-up. a One study excluded [32] after revision and inclusion Ref. [34]. Table 2 Ever-use of ET by histology. Histology
Authors
Risk ratio (95% CI)
All subtypes
Kreuzer et al. [17] Rodriguez et al. [19] Schabath et al. [18] Total
0.780 (0.490–1.210) 0.760 (0.600–0.940) 0.650 (0.470–0.890)
Summary risk ratio (95% CI)
Q (p)
I2
0.728 (0.610–0.869)
0.681 (0.711)
0
ET: estrogen therapy; CI: confidence interval; Q: Cochran’s Q; I2 : 95% uncertainty interval; p < 0.05 regarded as significant.
ies included) neither QUORUM nor MOOSE guidelines were fully instrumental to design this systematic review with meta-analyses.
was sought after discussion of reasons that may lead to inclusion or exclusion of an individual study.
2.4. Assessment of study quality
3. Results
All reviewers assessed included studies. In the absence of a universal scale for measuring quality of both observational studies and randomized trials, we used criteria developed by the U.S. Preventive Services Task Force [30]. In case of different opinions consensus
3.1. Study characteristics Out of 46 electronically retrieved publications, 13 studies not addressing our research questions were excluded, as were 11 nar-
Table 3 Ever-use of EPT by histologya . Histology
Authors
Risk ratio (95% CI)
All subtypes
Chlebowski et al. [34] Hulley et al. [31] Pukkala et al. [41] Pukkala et al. [41] Rodriguez et al. [19] Schabath et al. [18] Total
1.230 (0.920–1.630) 1.730 (0.930–3.210) 1.200 (0.690–1.900)a 0.750 (0.530–1.000)b 0.760 (0.570–1.010) 0.610 (0.400–0.920)
All subtypes RCTs only
Chlebowski et al. [34] Hulley et al. [31] Total (RCTs)
1.230 (0.920–1.630) 1.730 (0.930–3.210) 2
Summary risk ratio (95% CI)
Q (p)
0.912 (0.784–1.061)
4.062 (0.54)
1.359 (1.031–1.791)
0.726 (0.394)
EPT: estrogen/progestin therapy; CI: confidence interval; Q: Cochran’s Q; I (95% uncertainty interval) not calculated because of Q being too near to degrees of freedom; p < 0.05 regarded as significant; RCT: randomized controlled trial. a Data extracted for long-cycle therapy (addition of a progestin to oral HT every 3rd month). b Data extracted for monthly use of progestin in conjunction with oral HT.
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
201
Table 4 Ever-use of HT by histology. Histology
Authors
Risk ratio (95% CI)
All subtypes
Chen et al. [33] Elliott et al. [36] Ettinger et al. [37] Kreuzer et al. [17] Olsson et al. [40] Persson et al. [21] Rodriguez et al. [19] Schabath et al. [18] Schabath et al. [18] Schabath et al. [18] Total
0.700 (0.530–0.940) 2.000 (0.500–2.100) 0.220 (0.040–1.150) 0.830 (0.640–1.090) 0.730 (0.310–1.430) 1.000 (0.800–1.200) 0.760 (0.620–0.920) 0.380 (0.180–0.790)a 0.740 (0.540–1.020)b 0.380 (0.070–2.120)c
Adenocarcinoma
Chen et al. [33] Kabat et al. [38] Kreuzer et al. [17] Taioli et al. [20] Total
0.760 (0.560–1.030) 1.090 (0.850–1.400) 0.890 (0.660–1.190) 1.700 (1.000–2.800)
Non-small
Chen et al. [33] Ramnath et al. [42] Schabath et al. [18] Schwartz et al. [43] Total
0.710 (0.530–0.950) 0.670 (0.530–0.850) 0.690 (0.500–0.960) 0.810 (0.580–1.130)
Small
Chen et al. [33] Kreuzer et al. [17] Schabath et al. [18] Total
0.510 (0.110–2.330) 0.710 (0.460–1.090) 1.550 (0.690–3.450)
Squamous
Chen et al. [33] Kreuzer et al. [17] Total
0.650 (0.290–1.450) 0.740 (0.510–1.210)
Summary risk ratio (95% CI)
Q (p)
I2
0.796 (0.715–0.887)
13.76 (0.131)
0
0.977 (0.837–1.141)
7.902 (0.048)
n.c.
0.708 (0.613–0.817)
0.861 (0.835)
0
0.821 (0.566–1.189)
3.169 (0.205)
n.c.*
0.723 (0.516–1.014)
0.082 (0.775)
n.c.*
HT: hormone therapy as defined; CI: confidence interval; Q: Cochran’s Q; I2 : 95% uncertainty interval; n.c.: I2 not calculated because of heterogeneity; n.c.*: I2 not calculated because of too few observations [29]. a African-American ethnicity. b Caucasian ethnicity. c Hispanic ethnicity.
Table 5 Ever-use of HT by histology and smoking status. Histology
Smoking status
Authors
Risk ratio (95% CI)
All
NSM
Blackman et al. [22] Chen et al. [33] Kabat et al. [38] Kreuzer et al. [17] Liu et al. [39] Liu et al. [39] Rodriguez et al. [19] Schabath et al. [18] Total
1.400 (0.600–3.300) 0.700 (0.520–0.940) 0.950 (0.580–1.550) 1.050 (0.740–1.490) 1.190 (0.600–2.330)a 2.400 (1.070–5.400)b 0.560 (0.330–0.950) 0.720 (0.370–1.400)
Adenocarcinoma
NSM
Blackman et al. [22] Liu et al. [39] Liu et al. [39] Total
2.000 (0.700–5.600)c 1.230 (0.590–2.580)a 2.710 (1.120–6.580)b
All
SM
Blackman et al. [22] Chen et al. [33] Kabat et al. [38] Kreuzer et al. [17] Rodriguez et al. [19] Schabath et al. [18] Schabath et al. [18] Total
1.000 (0.800–1.400) 0.170 (0.010–2.020) 1.080 (0.910–1.290) 0.730 (0.510–1.030) 0.760 (0.550–1.050) 0.590 (0.380–0.920)d 0.730 (0.460–1.150)e
Summary risk ratio (95% CI)
Q (p)
I2
0.878 (0.741–1.041)
14.33 (0.046)
0
1.762 (1.072–2.898)
1.887 (0.389)
n.c.*
0.919 (0.823–1.027)
13.09 (0.042)
0
HT: hormone therapy as defined; CI: confidence interval; SM: current or ever or former smoker; NSM: never-smoker; Q: Cochran’s Q; I2 : 95% uncertainty interval; n.c.*: not calculated [29]. a Women with natural menopause. b Women with induced menopause. c Use of estrogen replacement therapy. d Current smoker. e Former smoker.
202
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
Table 6 Change of risk ratio per year of HT intake. Histology
Authors
Risk change per year of HT intake (95% CI)
Adenocarcinoma
Kabat et al. [38] Kreuzer et al. [17] Taioli et al. [20] Total
1.037 (1.004–.071) 0.953 (0.902–1.008) 1.299 (0.966–1.746)
Corrao et al. [35] Kreuzer et al. [17] Total
0.986 (0.819–1.188) 0.951 (0.906–0.998)
All
Pooled risk change (95% CI)
Q (p)
I2 (95% uncertainty interval)
1.017 (0.989–1.046)
9.57 (0.008)
0 (0–60)
0.953 (0.910–0.999)
2.05 (0.153)
n.c.
HT: hormone therapy as defined; CI: confidence interval; Q: Cochran’s Q; I2 : 95% uncertainty interval; n.c.: not calculated [29].
rative reviews and one editorial. The remaining 22 studies were further scrutinized; four studies with no extractable data had to be excluded, and two studies providing first reports of a given cohort study with short follow-up (n = 16 eligible studies). It should be noted that two RCTs [31,32] were not identified by electronic but by hand searches of the largest published menopausal hormone therapy trials known to have reported cancer data. The 18 studies finally included in this review (16 + 2 = 18) reflect a broad diversity of study designs regarding variables of analysis (Table 1). Most studies do not provide data on smoking status of participating women or discriminate between different histological diagnoses. Furthermore, only in a minority of studies, the type of HT is described clearly as ET or EPT. Only five studies provide risk estimates by duration of MHT. We excluded two studies [44,45] because another included analysis of the same cohort study provided a longer follow-up [21]. As requested by the reviewers, data from the Women’s Health Initiative (WHI) randomized controlled trial (estrogen plus progestin arm), published 10 October 2009, and therefore outside the time frame for literature search as indicated above, were additionally extracted and included [34]. Consequently, one former publication on WHI health risks, initially included though not dedicated to the subject of lung cancer because it provided some data on lung cancer in the comment section [32], was excluded. 3.2. Ever-use of HT Ever-use of ET is associated with a significant 27% decrease (rounded) of risk in women irrespective of smoking history when all histologies are pooled (Table 2; Cochran’s Q suggests homogeneity). Estrogen plus progestin therapy (six data sets extracted from five studies) does not change the risk; however, the pooled analysis of the only two RCTs available points at an increased risk in the presence of homogeneity for Q (Table 3). When all hormone therapies as defined are pooled as well as all histological subtypes, there is a reduction of risk by 20% (rounded) in a data set suggesting homogeneity (odds ratio 0.796; 95% CI 0.715–0.887). A larger significant decrease (−29%) is found in the analysis of four studies addressing non-small cell cancers with no suggestion of heterogeneity (odds
ratio 0.708; 95% CI 0.613–0.817). Analyses for other histological entities did not yield significant findings (Table 4). There is a significant increase of lung cancer risk in non-smoking women with adenocarcinoma (odds ratio 1.76; 95% CI 1.072–2.898), based on three data sets from two studies with overall 815 cases of lung cancer, Cochran’s Q does not indicate heterogeneity (Table 5). 3.3. Duration of use of HT Four studies have analysed the impact of duration of HT use on lung cancer risk by histological classification. In two studies with all histologies combined (lower panel) there is a small but significant decrease of risk (4.7% per year of HT use; studies appear to be homogenous), whereas the risk for adenocarcinoma appears to be slightly increased, though not significant (OR 1.017; 95% CI 0.989–1.046). However, the latter analysis shows major heterogeneity of studies (Table 6). A further differentiation according to smoking status, based on few eligible trials (n = 3) (Table 7) did not provide significant findings. 4. Discussion We found a significant increase of lung cancer risk in non-smoking women reporting ever-use of HT diagnosed with adenocarcinoma. This is a relevant finding as the attributable risk is 76% (Table 5). Furthermore, data from placebo-controlled randomized trials [31,34] comparing a specific estrogen and progestin therapy indicate an increased risk of lung cancer (histologies reported only specified in Ref. [34]) in hormone-users, representing a heterogenous group of women including never, past and current smokers (Table 3). In one RCT the risk for adenocarcinoma was increased, but this was not significant [34]. In the latest evaluation of the IARC [16] only a fraction of the studies we found eligible for extraction and analyses were briefly reviewed; specific subtypes of lung cancer were not considered. The relative incidence of adenocarcinoma of the lung has been steadily increasing in the recent past, obviously to a greater extent in the US; in the US age-adjusted incidence rates captured 1992–1998, the proportion of this histologic type was 49.4% of all
Table 7 Change of risk ratio per year of HT intake by smoking status. Histology
Authors
Risk change per year of HT intake (95% CI)
Non-smoking women Kabat et al. [38] All Kreuzer et al. [17] Summary
0.989 (0.877–1.115) 0.982 (0.919–1.049)
Smoking women Blackman et al. [22] All Kabat et al. [38] Kreuzer et al. [17] Summary
1.019 (0.980–1.059) 1.042 (1.002–1.084) 0.931 (0.874–0.991)
Pooled risk change (95% CI)
Q (p)
I2 (95% uncertainty interval)
0.984 (0.928–1.043)
0.04 (0.844)
n.c.*
1.014 (0.989–1.039)
9.22 (0.010)
0 (0–60)
HT: hormone therapy as defined; CI: confidence interval; Q: Cochran’s Q; I2 : 95% uncertainty interval; n.c.*: not calculated [29].
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
lung cancers in white women (Table 1 [5]). The significant increase of lung cancer risk of the adenomatous type in non-smoking women is of major importance from a public health point of view, as both in the US as e.g. in Germany this histologic entity is the relatively most common subtype in lung cancer patients. In the Munich Clinical Cancer Registry out of 6.696 female patients with lung cancer 43.8% were adenocarcinomas, the most frequent histologic entity in women (Hölzel, 2008; personal communication), compared to 25.5% of 17.896 lung cancer cases in males. In contrast to these figures, Ginsberg and co-authors [46] report for both genders a prevalence of 40% in the US. The most important risk factors for adenocarcinoma are cigarette smoking and exposure to environmental tobacco smoke [5]. The relevance of exogenous hormones for this entity has not been studied systematically. The overwhelming impact of cigarette smoking as a cause of lung cancer imposes challenges to detecting the role that other lifestyle factors, including MHT, may play in the development of lung cancer. In this context, it is noteworthy that the risk of lung cancer doubled in the HT group, evaluated in the largest randomized placebo-controlled trial of MHT until today, reporting health outcomes at a mean of 2.4 years of follow-up after EPT was stopped [32]. In this trial, most deaths during the postintervention phase were cancer-related in both treatment groups, but only to a smaller extent due to the pre-specified outcomes breast, colorectal, endometrial, or ovarian cancer. Among the other cancers, most were lung cancers. In the initial report of this study [47], lung cancer incidence was not different in women assigned to placebo or EPT (hazard ratio 1.04; 95% CI, 0.71–1.53). A recently published post hoc analysis [34] of this trial [47] confirmed the finding of an increased mortality due to lung cancer in the presence of a nonsignificantly increased incidence for this disease. We do not know whether increased expression of ER- might have a role in the pathogenesis of lung cancer, especially adenocarcinomas, as hypothesized [48], this issue was beyond the scope of our analyses. Our study has a number of limitations: firstly, our other results, except for those regarding lung cancer risk in non-smoking women reporting ever-use of MHT and diagnosed with adenocarcinoma, reported in the previous section should be interpreted with great caution as significances were found in analyses only when smoking and non-smoking women, various hormone regimens, or histological subtypes, respectively, were pooled. There are too few studies providing consistently risk estimates for both smokers and non-smokers; there are no studies providing risk indicators for adenocarcinoma in smoking women. Secondly, there is a paucity of studies to allow assessment of at least defined groups of hormone regimens such as ET and EPT. Thirdly, included studies did not provide extractable data on passive smoking, a further potential risk factor for lung cancer. Likewise, nutritional factors, especially the consumption of red meat in non-smokers [49,50] could not be considered, as could not air pollution and radiation. It is beyond the scope of this analysis to elucidate whether the increase of MHT use for many years [51] has contributed to the changing pattern of histologic subtypes of lung cancer in females. Likewise, whether the recent decrease of use in the US and several other countries [52] may have an impact in the future is also unknown. In conclusion our results suggest that ever-use of HT in nonsmoking women may increase the risk of adenocarcinoma of the lung. Furthermore, data from RCTs suggest that EPT increases the mortality of lung cancer. MHT may be regarded as a potential risk factor for this type of lung cancer, pointing at a potential for prevention. However, compared to other cancers affecting females, the association between lung cancer and use of menopausal hormone therapy clearly warrants more and specifically designed studies to arrive at more definite conclusions.
203
Declaration of conflict of interest M.D. participated in a phase III study of a drug tested for prevention of osteoporosis in postmenopausal women, sponsored by Pfizer, USA; M.D. is member of an advisory board for Women’s health issues of the Federal Centre for Health Information (Germany; http://www.frauengesundheitsportal.de/?uid= 08c61583f2c66abc96042e584af7b9fa&id=Seite1294). Contributors For the first and second version of the manuscript: C.M. Greiser: Conducted the literature review, contributed to the conception of the analysis, the interpretation of results and to the drafting of the manuscript. E.M. Greiser: Contributed to the conception of the analysis, performed the statistical analyses (first and revised version), and contributed to the interpretation of results. M. Dören: Conceptualized and supervised the study, participated in the literature review was involved in the interpretation of the data, writing of the manuscript (first and revised version) and critical review, and arranged for funding. All authors have seen and approved the final version and the revised version of the manuscript. Berlin, 27 November, 2009 M. Dören, on behalf of all authors. Funding This study was conducted at the 1 Institute for Public Health and Nursing Research, Faculty of Health Sciences, Bremen University; 2 Epi.Consult GmbH, Bremen; and the 3 Charité - Universitätsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany. Provenance Not commissioned but externally peer reviewed. References [1] Ernster VL. Impact of tobacco on women’s health. In: Samet JM, Yoon S-Y, editors. Women and the tobacco epidemic: challenges for the 21st century. Geneva, Switzerland: World Health Organization; 2001. p. 1–16. [2] Espey DK, Wu X-C, Swan J, et al. Annual report to the nation on the status of cancer, 1975–2004, featuring cancer in American Indians and Alaska Natives. Cancer 2007;110(10):2119–52. [3] Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin 2005;55(1):10–30. [4] Kamangar F, Dores GM, Anderson WF. Patterns of cancer incidence, mortality, and prevalence across five continents: defining priorities to reduce cancer disparities in different geographic regions of the world. J Clin Oncol 2006;24(14):2137–50. [5] Alberg AJ, Samet JM. Epidemiology of lung cancer. Chest 2003;123(1 Suppl.):21–49. [6] Charloux A, Quoix E, Wolkove N, Small D, Pauli G, Kreisman H. The increasing incidence of lung adenocarcinoma: reality or artefact? A review of the epidemiology of lung adenocarcinoma. Int J Epidemiol 1997;26(1):14–23. [7] Wingo PA, Ries LA, Giovino GA, et al. Annual report to the nation on the status of cancer, 1973–1996, with a special section on lung cancer and tobacco smoking. J Natl Cancer Inst 1999;91(8):675–90. [8] Gasperino J, Rom WN. Gender and lung cancer. Clin Lung Cancer 2004;5(6):353–9. [9] Belani CP, Marts S, Schiller J, Socinski MA. Women and lung cancer: epidemiology, tumor biology, and emerging trends in clinical research. Lung Cancer 2007;55(1):5–23. [10] US Department of Health, and Human Services (US-DHSS). Women and smoking: a report of the Surgeon General. Rockville, MD: US Department of Health and Human Services; 2001. [11] Chaudhuri PK, Thomas PA, Walker MJ, Briele HA, Das Gupta TK, Beattie CW. Steroid receptors in human lung cancer cytosols. Cancer Lett 1982;16(3):327–32. [12] Mollerup S, Jorgensen K, Berge G, Haugen A. Expression of estrogen receptors ␣ and  in human lung tissue and cell lines. Lung Cancer 2002;37(2): 153–9. [13] Cagle PT, Mody DR, Schwartz MR. Estrogen and progesterone receptors in bronchogenic carcinoma. Cancer Res 1990;50(20):6632–5.
204
C.M. Greiser et al. / Maturitas 65 (2010) 198–204
[14] Palomares MR, Sayre JW, Shekar KC, Lillington LM, Chlebowski RT. Gender influence on weight loss pattern and survival of non-small cell lung carcinoma patients. Cancer 1996;78(10):2119–26. [15] Wisnivesky JP, Halm EA. Sex differences in lung cancer survival: do tumors behave differently in elderly women? J Clin Oncol 2007;25(13):1705–12. [16] IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Combined estrogen-progestogen contraceptives and combined estrogenprogestogen menopausal therapy, vol. 91, Lyon: France; 2007. http://monographs.iarc.fr/ENG/Monographs/vol91/mono91.pdf (last accessed on 21.09.09). [17] Kreuzer M, Gerken M, Heinrich J, Kreienbrock L, Wichmann H-E. Hormonal factors and risk of lung cancer among women? Int J Epidemiol 2003;32(2):263–71. [18] Schabath MB, Wu X, Vassilopoulou-Sellin R, Vaporciyan AA, Spitz MR. Hormone replacement therapy and lung cancer risk: a case–control analysis. Clin Cancer Res 2004;10(1 Pt 1):113–23. [19] Rodriguez C, Spencer Feigelson F, Deka A, et al. Postmenopausal hormone therapy and lung cancer risk in the cancer Prevention Study II Nutrition Cohort. Cancer Epidemiol Biomarkers Prev 2008;17(3):655–60. [20] Taioli E, Wynder El. Endocrine factors and adenocarcinoma of the lung in women. J Natl Cancer Inst 1994;86(11):869–70. [21] Persson I, Yuen J, Bergkvist L, Schairer C. Cancer incidence and mortality in women receiving estrogen and estrogen–progestin replacement therapy—long-term follow-up of a Swedish cohort. Int J Cancer 1996; 67(3):327–32. [22] Blackman JA, Coogan PF, Rosenberg L, et al. Estrogen replacement therapy and risk of lung cancer. Pharmacoepidemiol Drug Safety 2002;11(7):561–7. [23] Ganti AK, Sahmoun AE, Panwalker AW, Tendulkar KK, Potti A. Hormone replacement therapy is associated with decreased survival in women with lung cancer. J Clin Oncol 2006;24(1):59–63. [24] Greiser CM, Greiser EM, Dören M. Menopausal hormone therapy and risk of breast cancer: a meta-analysis of epidemiological studies and randomized controlled trials. Human Reprod Update 2005;11(6):561–7. [25] Greiser CM, Greiser EM, Dören M. Menopausal hormone therapy and risk of ovarian cancer—systematic review and meta-analysis. Hum Reprod Update 2007;13(5):4534–63. [26] Petitti DB. Statistical methods in meta-analysis. In: Pettiti DB, editor. Metaanalysis, decision analysis, and cost-effectiveness analysis. 2nd ed. New York: Oxford University Press; 2000, 63–7, 111–3, 213–24. [27] Berlin JA, Longnecker MP, Greenland S. Meta-analysis of epidemiologic dose–response data. Epidemiology 1993;4(3):218–28. [28] Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327(7414):557–60. [29] Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21(11):1539–58. [30] Harris RP, Helfand M, Woolf SH, et al. Current methods of the U.S. Preventive Services Task Force. A review of the process. Am J Prev Med 2001;20(3 Suppl.):21–35. [31] Hulley S, Furberg C, Barrett-Connor E, et al. Noncardiocascular disease outcomes during 6.8 years of hormone therapy. JAMA 2002;288(1):58–66. [32] Heiss G, Wallace R, Anderson GL, et al. Health risks and benefits 3 years after stopping randomized treatment with estrogen and progestin. JAMA 2008;299(9):1036045. [33] Chen K-Y, Hsiao C-F, Chang G-C, et al. Hormone replacement therapy and lung cancer risk in Chinese. Cancer 2007;110(8):1768–75.
[34] Chlebowski RT, Schwartz AG, Wakelee H, et al. Oestrogen plus progestin and lung cancer in postmenopausal women (Women’s Health Initiative trial): a post hoc analysis of a randomised controlled trial. Lancet 2009;374:1243–51. [35] Corrao G, Zambon A, Conti V, et al. Menopause hormone replacement therapy and cancer risk: an Italian record linkage investigation. Ann Oncol 2008;19(1):150–5. [36] Elliott AM, Hannaford PC. Use of exogenous hormones by women and lung cancer: evidence from the Royal College of General Practitioners’ Oral Contraception Study. Contraception 2006;73(4):331–5. [37] Ettinger B, Friedman GD, Bush T, Quesenberry CP. Reduced mortality associated with long-term postmenopausal estrogen therapy. Obstet Gynecol 1996;87(1):6–12. [38] Kabat GC, Miller AB, Rohan TE. Reproductive and hormonal factors and risk of lung cancer in women: a prospective cohort study. Int J Cancer 2007;120(10):2214–20. [39] Liu Y, Inoue M, Sobue T, Tsugane S. Reproductive factors, hormone use and the risk of lung cancer among middle-aged never-smoking Japanese women: a large scale-population-based cohort study. Int J Cancer 2005;117(4): 662–6. [40] Olsson H, Bladström A, Ingvar C. Are smoking-associated cancers prevented or postponed in women using hormone replacement therapy? Obstet Gynecol 2003;102(3):565–70. [41] Pukkala E, Tulenheimo-Silfvast A, Leminen A. Incidence of cancer among women using long versus monthly cycle hormone replacement therapy, Finland 1994–1997. Cancer Causes Control 2001;12(2):111–5. [42] Ramnath N, Menezes RJ, Loewen G, et al. Hormone replacement therapy as a risk factor for non-small cell lung cancer: results of a case–control study. Oncology 2007;73(5–6):305–10. [43] Schwartz AG, Wenzlaff AS, Prysak GM, et al. Reproductive factors, hormone use, estrogen receptor expression and risk of non-small-cell lung cancer in women. J Clin Oncol 2007;25(36):5785–92. [44] Adami H-O, Persson I, Hoover R, Schairer C, Bergkvist L. Risk of cancer in women receiving hormone replacement therapy. Int J Cancer 1989;44(5):833–9. [45] Schairer C, Adami H-O, Hoover R, Persson I. Cause-specific mortality in women receiving hormone replacement therapy. Epidemiology 1997;8(1):59–65. [46] Ginsberg RJ, Vokes EE, Rosenzweig K. Non-small cell lung cancer. In: DeVita VT, Hellman S, Rosenberg SA, editors. Cancer—principles and practice of oncology. 6th ed. 2001. p. 925–83. [47] Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002;288(3):321–33. [48] Omoto Y, Kobayahi Y, Nishida K, et al. Expression, function, and clinical implications of the estrogen receptor beta in human lung cancers. Biochem Biophys Res Commun 2001;285(2):340–7. [49] Alavanja MC, Field RW, Sinha R, et al. Lung cancer risk and red meat consumption among Iowa women. Lung Cancer 2001;34(1):37–46. [50] De Stefani E, Brennan P, Boffetta P, et al. Diet and adenocarcinoma of the lung: a case–control study in Uruguay. Lung Cancer 2002;35(1):43–51. [51] Hersh AL, Stefanick ML, Stafford RS. National use of postmenopausal hormone therapy. JAMA 2004;291(1):47–53. [52] Du Y, Dören M, Melchert H-UW, Scheidt-Nave C, Knopf H. Differences in menopausal hormone therapy use among women in Germany between 1998 and 2003. BMC Women’s Health 2004;7:19.