Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Associations between obesity, smoking and lymph node status at breast cancer diagnosis in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial

  • Amelia Smith ,

    Roles Methodology, Writing – original draft, Writing – review & editing

    smitha25@tcd.ie

    Affiliation Dept. Pharmacology and Therapeutics, Trinity Centre for Health Sciences, St. James’ Hospital, Dublin, Ireland

    ⨯
  • Maeve Mullooly,

    Roles Conceptualization, Methodology, Writing – review & editing

    Affiliations Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, Maryland, United States of America, Cancer Prevention Fellowship Program, Division of Cancer Prevention, National Cancer Institute, Bethesda, Maryland, United States of America, Division of Population Health Sciences, Royal College of Surgeons in Ireland, Dublin, Ireland

    ⨯
  • Laura Murphy,

    Roles Conceptualization, Methodology, Project administration, Supervision

    Affiliation Division of Population Health Sciences, Royal College of Surgeons in Ireland, Dublin, Ireland

    ⨯
  • Thomas Ian Barron,

    Roles Conceptualization, Data curation, Funding acquisition

    Affiliations Dept. Pharmacology and Therapeutics, Trinity Centre for Health Sciences, St. James’ Hospital, Dublin, Ireland, Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, United States of America

    ⨯
  • Kathleen Bennett

    Roles Funding acquisition, Methodology, Supervision, Writing – review & editing

    Affiliation Division of Population Health Sciences, Royal College of Surgeons in Ireland, Dublin, Ireland

    ⨯

Abstract

Introduction

There is evidence suggesting that smoking and obesity prior to a breast cancer diagnosis is associated with poorer outcomes. In this study, we investigate the associations between smoking and obesity prior to a breast cancer diagnosis and the presence of lymph node metastases at diagnosis.

Methods

Women with stage I-III breast cancer (n = 3,304) were identified from the National Cancer Institute’s Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial. Univariable and multivariable log-binomial models were used to estimate relative risks (RR) and 95% confidence intervals (CIs) for associations between lymph node positive breast cancer and; i) smoking, and ii) obesity prior to diagnosis.

Results

Pre-diagnostic smoking/obesity was not associated with lymph node metastasis at diagnosis in multivariable analyses; (RR 0.82, 95%CI 0.61, 1.10) and (RR 0.95, 95% CI 0.81, 1.12), respectively.

Conclusion

Obesity and smoking information was recorded a number of years prior to breast cancer diagnosis, therefore these findings should to be replicated in a larger cohort of women, with more detailed smoking and obesity information.

Citation: Smith A, Mullooly M, Murphy L, Barron TI, Bennett K (2018) Associations between obesity, smoking and lymph node status at breast cancer diagnosis in the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. PLoS ONE 13(8): e0202291. https://doi.org/10.1371/journal.pone.0202291

Editor: Scott M. Langevin, University of Cincinnati College of Medicine, UNITED STATES

Received: February 27, 2018; Accepted: July 31, 2018; Published: August 29, 2018

Copyright: © 2018 Smith et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Data are from the the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial and are owned by the Cancer Data Access System. Detailed instructions on how to request the data can be found in the following link: https://biometry.nci.nih.gov/app_media/cdas-bioshare/documents/studies/plco/instructions/instructions_plco_data_only.44dc9e57c3fd.pdf. Proposals for accessing the data can be made here: https://biometry.nci.nih.gov/cdas/learn/plco/instructions/. The instructions provided are how the authors accessed the data. The authors did not have any data access privileges that others would not have.

Funding: This work was supported by the Irish Cancer Society Collaborative Cancer Research Centre BREAST-PREDICT (https://www.cancer.ie/) [CCRC13GAL to K Bennett, TI Barron]. L Murphy and A Smith’s positions are funded by the Irish Cancer Society Collaborative Cancer Research Centre BREAST-PREDICT [CCRC13GAL]. K Bennett’s [RLA/2015/1579] position was funded by the Health Research Board Ireland (http://www.hrb.ie/home/). The Health Research Board Ireland and the Irish Cancer Society had no role in the study design, collection, analysis, and interpretation of data, writing of the report, or the decision to submit for publication.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Obesity and smoking in women prior to a breast cancer diagnosis has been shown to be associated with disease recurrence and reduced breast cancer-specific survival [1,2]. It is well established that cigarette smoking is linked to cancers such as lung and stomach, and although cigarette smoke contains mammary carcinogens, the link to breast cancer is less certain [3]. In a meta-analysis of 10 studies investigating the association between smoking at diagnosis and breast cancer specific mortality, current smokers had a statistically significant 33% increase in breast cancer specific mortality (hazard ratio (HR) 1.33, 95% confidence intervals (CI) 1.12, 1.58) compared to never smokers [1].

Obesity at breast cancer diagnosis has been linked to aggressive tumours and an overall worse prognosis[2]. In a meta-analysis of 22 studies investigating pre-diagnostic body mass index (BMI) and breast cancer specific-mortality, obese women were 35% more likely to die of their breast cancer, when compared to women with a normal pre-diagnostic BMI (Relative risk (RR) 1.35, 95% CI 1.24, 1.47)[4]. Overweight and obese women have increased levels of; insulin-like growth factor (IGF), TNF-α, IL-6, and VEGF, which can promote angiogenesis, tumour growth, metastasis and cell survival [5]. Both smoking and obesity can induce a pro-inflammatory environment[6]. Inflammation has been shown to possess a variety of tumorigenic properties; enhancing cancer cell survival, promoting angiogenesis, and promoting metastatic dissemination [7]. Lymph node (LN) status has been shown to be an important prognostic factor in several breast cancer subtypes, and can influence important clinical decisions such as the potential therapeutic options, subsequent quality of life and medical costs [8]. Additionally, predictive models have been developed for to predict LN metastasis but these are not well integrated into clinical practice and often do not include lifestyle exposures [9,10]. As such, it is vital to determine the possible effects of lifestyle factors on LN status at breast cancer diagnosis. Therefore, we aimed to examine the associations between; i) pre-diagnostic obesity; and ii) pre-diagnostic smoking; and the presence of LN metastasis at breast cancer diagnosis among women in the US PLCO population[11]. In the analyses of pre-diagnostic smoking and presence of LN metastasis, we also investigated the presence of effect modification by pre-diagnostic obesity.

Methods

Setting and data collection

This study was carried out using data from the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. The PLCO trial was a multicentre, randomized clinical trial designed to evaluate the effect of screening for prostate, lung, colorectal, and ovarian cancer on cancer-specific mortality in over 150,000 individuals in the United States (US). The PLCO study protocol was approved by the Institutional Review Board of the National Cancer Institute and all participating institutions. All participants provided written informed consent upon enrolment. PLCO has the following five ClinicalTrials.org registration numbers: NCT00002540 (Prostate), NCT01696968 (Lung), NCT01696981 (Colorectal), NCT01696994 (Ovarian), and NCT00339495 (EEMS), and details of the methodology of the PLCO trial have been described previously [12]. Participants completed a baseline questionnaire (BQ) at trial entry and an annual update questionnaire requesting details of cancer diagnoses within the previous year. All self-reported cancers were confirmed through medical record review by trained abstractors.

Participants

For this study, we included post-menopausal women who completed the PLCO BQ that were diagnosed with stage I–III breast cancer between 1993 and 2009 (n = 3,304 women). Women were excluded if they were missing information concerning smoking status, BMI, LN status or had a previous invasive cancer diagnosis. Women were followed until time of death, loss to follow-up, December 31st 2009, or a maximum of 13 years; whichever occurred earliest. Information regarding smoking status and BMI were obtained from the BQ. Smoking status was categorised as current, former, or never smoker. BMI was categorised as; <18.5 kg/m2 (underweight), 18.5–24.9 kg/m2 (normal weight), 25–29.9 kg/m2 (overweight) and > = 30 kg/m2 (obese). The median time between completion of the BQ and diagnosis of breast cancer was 5.9 years (min; 6 days, max; 15.9 years).

Outcomes and covariates

Women were identified as LN positive if they had a pathologic LN status of pN1, pN2, or pN3 (see S1 Table). Based on prior literature, we included important predictors of LN status in multivariable models [13]. For example, pre-diagnostic exposure to non-steroidal anti-inflammatory drugs (NSAIDs) is linked to reduced likelihood of presenting with lymph node positive breast cancer at diagnosis [14]. Tumour characteristics such as grade and hormone receptor status are also liked to lymph node status at breast cancer diagnosis [15,16]. Information regarding the following breast tumour characteristics was obtained from medical records: tumour stage (I, IIA, IIB, IIIA, or IIIB); tumour grade (1, 2, or 3/4); tumour size (cm), tumour morphology (lobular, tubular, ductal, other), estrogen receptor (ER) status (positive, negative, unspecified), progesterone receptor (PR) status (positive, negative, unspecified), and human epidermal growth factor receptor 2 (HER2) status (positive, negative, unspecified). Additional data obtained from the BQ included; smoking status (current, former, never), use of NSAIDs (ibuprofen and/or aspirin; yes or no). An indicator of comorbidity (score 0–3) was developed based on reporting of 1 or more of the following conditions at baseline: heart attack, diabetes, hepatitis, stroke, arthritis.

Statistical analysis

LN status was tabulated and differences in the proportion of women with/without LN metastasis were compared for obesity and smoking levels using univariate Poisson regression. Univariable and multivariable log-binomial models were used to estimate relative risks (RR) and 95% confidence intervals (CI) for associations between LN positive breast cancer and;

  1. pre-diagnostic obesity
  2. pre-diagnostic smoking (current smoker at time of BQ)
  3. ever pre-diagnostic smoking (current and former smokers at BQ)

The covariates described were identified for inclusion in the multivariable model based on prior knowledge of clinical and demographic predictors of positive lymph node status. In analyses ii) and iii), the presence of effect modification by obesity was assessed by inclusion of an interaction term in the multivariable model.

We also carried out a post-hoc analysis, whereby the association between LN positive breast cancer and i) pre-diagnostic obesity, ii) pre-diagnostic smoking (current smoker at time of BQ), and iii) ever pre-diagnostic smoking (current and former smokers at BQ) was stratified by time from BQ to breast cancer diagnosis (less than 3 years, 3 years or greater).

Results

Participant characteristics

We identified 3,304 women with stage I-III breast cancer. Of these, 3,011 (91%) women had complete information on pre-diagnostic smoking status, pre-diagnostic obesity status and lymph node status at diagnosis (Fig 1). The characteristics of these women are presented in Table 1. The median age at diagnosis in the cohort was 67 (interquartile range; 63, 73) and the majority of women were white non-Hispanic. In univariable analysis, women with lymph node positive status (n = 743, 24.7%) were significantly more likely to have larger tumours and a higher tumour stage and grade and were also significantly more likely to be PR and HER2 positive (Table 1).

thumbnail
Download:
Fig 1. Flow diagram showing study cohort selection.

https://doi.org/10.1371/journal.pone.0202291.g001

thumbnail
Download:
Table 1. Characteristics of women selected for inclusion in study cohort.

https://doi.org/10.1371/journal.pone.0202291.t001

Pre-diagnostic obesity and smoking, and lymph node status at diagnosis

Relative risks for associations between pre-diagnostic smoking, obesity, and lymph node status are presented in Table 2. No significant association was found between pre-diagnostic obesity and lymph node status at diagnosis, in univariable (RR 1.07, 95% CI 0.93, 1.23) or multivariable analyses (RR 0.95, 95% CI 0.81, 1.12) (Table 2).

thumbnail
Download:
Table 2. Univariable and multivariable RR’s and 95% CI’s for associations between pre-diagnostic obesity, smoking and lymph node positive breast cancer.

https://doi.org/10.1371/journal.pone.0202291.t002

No significant association was found between pre-diagnostic smoking and lymph node status at diagnosis, in univariable (RR 0.82, 95% CI 0.64, 1.06) and multivariable adjusted analyses (RR 0.82, 95% CI 0.61, 1.10). In analyses of effect modification by obesity in pre-diagnostic smokers, the multivariable relative risk remained non-significant (RR 0.98, 95%CI 0.54, 1.78).

In analyses whereby current and former smokers were grouped into ‘ever’ smokers, results remained the same; no associations were observed between smoking and lymph node status in univariable (RR 0.91, 95% CI 0.80, 1.03) or multivariable analyses (RR 0.95, 95% CI 0.83, 1.10). In analyses of effect modification by obesity in pre-diagnostic ever smokers, the multivariable relative risk remained non-significant (RR 0.94, 95%CI 0.74, 1.19) (Table 2).

Pre-diagnostic obesity and smoking, and lymph node status at diagnosis, stratified by time from BQ to breast cancer diagnosis

Relative risks for associations between pre-diagnostic smoking, obesity, and lymph node status, when stratified by time from BQ to breast cancer diagnosis are presented in Table 3. No significant association was found between pre-diagnostic obesity or smoking and lymph node status at diagnosis, when stratified by time from BQ (less than 3 years, 3 years or greater) to breast cancer diagnosis.

thumbnail
Download:
Table 3. Univariable and multivariable RR’s and 95% CI’s for associations between pre-diagnostic obesity, smoking and lymph node positive breast cancer, by time from BQ to diagnosis.

https://doi.org/10.1371/journal.pone.0202291.t003

Discussion

In this study of 3,011 post-menopausal women with stage I-III breast cancer in the PLCO cohort, pre-diagnostic smoking and/or obesity was not significantly associated with positive lymph node status at diagnosis.

Cigarette smoke has been shown to induce cyclooxygenase (COX-2) expression and increase prostaglandin E2 (PGE2) release[17]. Induction of these inflammatory pathways has been linked to the stimulation of lymphangiogensis in in-vitro and in-vivo breast cancer models [18]. While no association between pre-diagnostic smoking and LN metastasis was observed in this study, this may have been due to the small sample size and the recording of smoking status may predate breast cancer diagnosis by a number of years.

Obesity is a pro-inflammatory condition [19] and chronic inflammation can lead to induction of tumorigenic pathways[20]. In postmenopausal women the majority of estrogen is produced in the peripheral adipose tissue, and is increased in women who are overweight and obese [2,21]. Estrogen is converted to estradiol, which is involved in key cellular processes in breast cancer[22]. To our knowledge, our study is the first to investigate the interaction between pre-diagnostic smoking and obesity and LN status at diagnosis in breast cancer, and due to evidence supporting a pro-inflammatory and tumorigenic role of these lifestyle factors, we suggest that this be explored further.

This study has a number of strengths including reliable histological information and comprehensive information on baseline characteristics of the cohort. However, information regarding smoking and BMI was self-reported, and completion of this questionnaire may have preceded breast cancer diagnosis by many years. Therefore, the obesity and smoking status of women may have changed between BQ and breast cancer diagnosis. While the multivariable adjusted analyses took into account a number of important confounders, residual confounding is possible. It is important to consider the fact that the BQ was self-reported, and while tumour-specific variables were obtained from medical records, it is possible that some patient characteristics such as BMI, NSAID use, and comorbidity status were inaccurately recorded. The small sample size in this study limited the number of subgroup analyses performed. For example, it would be interesting to investigate the effect of pre-diagnostic smoking and/or obesity on the number of positive nodes at diagnosis. Utada et al. have shown that increasing from two to three positive LNs is associated with poorer patient outcomes in women with breast cancer [23]. In this subset of the PLCO data we did not have access to mortality information, and could not assess the association between smoking and/or obesity on breast cancer mortality.

This study suggests that smoking and/or obesity before breast cancer diagnosis was not associated with lymph node metastasis. However, given the evidence to suggest a role for these lifestyle factors in the development and progression of breast cancer, a larger study, with more detailed information on history of smoking and obesity, is warranted.

Supporting information

S1 Table. Pathologic TNM stage for primary breast cancer from AJCC cancer staging manual 5th edition.

https://doi.org/10.1371/journal.pone.0202291.s001

(DOCX)

Acknowledgments

We would like to thank the National Cancer Institute for providing access to the data upon which this study was based. The interpretation and reporting of these data are the responsibility of the authors and should in no way be seen as the official policy or interpretation of the National Cancer Institute, nor the study funders. We also thank Professor Kala Visvanathan, Johns Hopkins Bloomberg School of Public Health, for assistance in the development of the study proposal.

References

  1. 1. Bérubé S, Lemieux J, Moore L, Maunsell E, Brisson J. Smoking at time of diagnosis and breast cancer-specific survival: new findings and systematic review with meta-analysis. Breast Cancer Res. 2014;16: R42. pmid:24745601
  2. 2. Porter GA, Inglis KM, Wood LA, Veugelers PJ. Effect of Obesity on Presentation of Breast Cancer. Ann Surg Oncol. 2006;13: 327–332. pmid:16485153
  3. 3. Reynolds P. Smoking and Breast Cancer. J Mammary Gland Biol Neoplasia. 2013;18: 15–23. pmid:23179580
  4. 4. Chan DSM, Vieira AR, Aune D, Bandera EV, Greenwood DC, McTiernan A, et al. Body mass index and survival in women with breast cancer-systematic literature review and meta-analysis of 82 follow-up studies. Ann Oncol Off J Eur Soc Med Oncol. 2014;25: 1901–1914. pmid:24769692
  5. 5. Goodwin PJ, Ennis M, Pritchard KI, Trudeau ME, Koo J, Madarnas Y, et al. Fasting insulin and outcome in early-stage breast cancer: results of a prospective cohort study. J Clin Oncol Off J Am Soc Clin Oncol. 2002;20: 42–51. pmid:11773152
  6. 6. Bennett JM, Glaser R, Andridge RR, Peng J, Malarkey WB, Kiecolt-Glaser JK. Long lasting effects of smoking: Breast cancer survivors’ inflammatory responses to acute stress differ by smoking history. Psychoneuroendocrinology. 2013;38: 179–187. pmid:22727479
  7. 7. Mantovani A, Allavena P, Sica A, Balkwill F. Cancer-related inflammation. Nature. 2008;454: 436–444. pmid:18650914
  8. 8. Howland NK, Driver TD, Sedrak MP, Wen X, Dong W, Hatch S, et al. Lymph node involvement in immunohistochemistry-based molecular classifications of breast cancer. J Surg Res. 2013;185: 697–703. pmid:24095025
  9. 9. Qiu S-Q, Aarnink M, van Maaren MC, Dorrius MD, Bhattacharya A, Veltman J, et al. Validation and update of a lymph node metastasis prediction model for breast cancer. Eur J Surg Oncol J Eur Soc Surg Oncol Br Assoc Surg Oncol. 2018;44: 700–707. pmid:29449047
  10. 10. Qiu S-Q, Zeng H-C, Zhang F, Chen C, Huang W-H, Pleijhuis RG, et al. A nomogram to predict the probability of axillary lymph node metastasis in early breast cancer patients with positive axillary ultrasound. Sci Rep. 2016;6: 21196. pmid:26875677
  11. 11. Cianfrocca M, Goldstein LJ. Prognostic and Predictive Factors in Early-Stage Breast Cancer. The Oncologist. 2004;9: 606–616. pmid:15561805
  12. 12. Prorok PC, Andriole GL, Bresalier RS, Buys SS, Chia D, David Crawford E, et al. Design of the prostate, lung, colorectal and ovarian (PLCO) cancer screening trial. Control Clin Trials. 2000;21: 273S–309S. pmid:11189684
  13. 13. Abdessalam SF, Zervos EE, Prasad M, Farrar WB, Yee LD, Walker MJ, et al. Predictors of positive axillary lymph nodes after sentinel lymph node biopsy in breast cancer. Am J Surg. 2001;182: 316–320. pmid:11720662
  14. 14. Barron TI, Flahavan EM, Sharp L, Bennett K, Visvanathan K. Recent Prediagnostic Aspirin Use, Lymph Node Involvement, and 5-Year Mortality in Women with Stage I–III Breast Cancer: A Nationwide Population-Based Cohort Study. Cancer Res. 2014;74: 4065–4077. pmid:25085874
  15. 15. Mattes MD, Bhatia JK, Metzger D, Ashamalla H, Katsoulakis E. Breast Cancer Subtype as a Predictor of Lymph Node Metastasis according to the SEER Registry. J Breast Cancer. 2015;18: 143–148. pmid:26155290
  16. 16. Majid S, Rydén L, Manjer J. Predictive factors for sentinel node metastases in primary invasive breast cancer: a population-based cohort study of 2552 consecutive patients. World J Surg Oncol. 2018;16. pmid:29530065
  17. 17. Huang R-Y, Chen GG. Cigarette smoking, cyclooxygenase-2 pathway and cancer. Biochim Biophys Acta BBA—Rev Cancer. 2011;1815: 158–169. pmid:21147199
  18. 18. Nandi P, Girish GV, Majumder M, Xin X, Tutunea-Fatan E, Lala PK. PGE2 promotes breast cancer-associated lymphangiogenesis by activation of EP4 receptor on lymphatic endothelial cells. BMC Cancer. 2017;17: 11. pmid:28056899
  19. 19. Das UN. Is obesity an inflammatory condition? Nutr Burbank Los Angel Cty Calif. 2001;17: 953–966.
  20. 20. van Kruijsdijk RCM, van der Wall E, Visseren FLJ. Obesity and Cancer: The Role of Dysfunctional Adipose Tissue. Cancer Epidemiol Prev Biomark. 2009;18: 2569–2578. pmid:19755644
  21. 21. McTiernan A, Rajan KB, Tworoger SS, Irwin M, Bernstein L, Baumgartner R, et al. Adiposity and sex hormones in postmenopausal breast cancer survivors. J Clin Oncol Off J Am Soc Clin Oncol. 2003;21: 1961–1966. pmid:12743149
  22. 22. Feigelson HS, Henderson BE. Future possibilities in the prevention of breast cancer: role of genetic variation in breast cancer prevention. Breast Cancer Res BCR. 2000;2: 277–282. pmid:11250721
  23. 23. Utada Y, Kasumi F, Yoshimoto M, Tada T, Saitoh M, Takahashi K, et al. The Location of Positive Nodes Partly Influences the Prognostic Value of the Number of Positive Nodes in Breast Cancer Patients. Jpn J Clin Oncol. 1999;29: 63–67. pmid:10089945