Cigarette smoking and risk of ovarian cancer: A pooled analysis of 21 case-control studies

Mette T. Faber, Susanne K. Kjær, Christian Dehlendorff, Jenny Chang-Claude, Klaus K. Andersen, Estrid Høgdall, Penelope M. Webb, Susan J. Jordan, Mary Anne Rossing, Jennifer A. Doherty, Galina Lurie, Pamela J. Thompson, Michael E. Carney, Marc T. Goodman, Roberta B. Ness, Francesmary Modugno, Robert P. Edwards, Clareann H. Bunker, Ellen L. Goode, Brooke L. FridleyRobert A. Vierkant, Melissa C. Larson, Joellen Schildkraut, Daniel W. Cramer, Kathryn L. Terry, Allison F. Vitonis, Elisa V. Bandera, Sara H. Olson, Melony King, Urmila Chandran, Lambertus A. Kiemeney, Leon F.A.G. Massuger, Anne M. Van Altena, Sita H. Vermeulen, Louise Brinton, Nicolas Wentzensen, Jolanta Lissowska, Hannah P. Yang, Kirsten B. Moysich, Kunle Odunsi, Karin Kasza, Oluwatosin Odunsi-Akanji, Honglin Song, Paul Pharaoh, Mitul Shah, Alice S. Whittemore, Valerie McGuire, Weiva Sieh, Rebecca Sutphen, Usha Menon, Simon A. Gayther, Susan J. Ramus, Aleksandra Gentry-Maharaj, Celeste Leigh Pearce, Anna H. Wu, Malcolm C. Pike, Harvey A. Risch, Allan Jensen

Research output: Contribution to journalArticlepeer-review

62 Scopus citations


Purpose The majority of previous studies have observed an increased risk of mucinous ovarian tumors associated with cigarette smoking, but the association with other histological types is unclear. In a large pooled analysis, we examined the risk of epithelial ovarian cancer associated with multiple measures of cigarette smoking with a focus on characterizing risks according to tumor behavior and histology. Methods We used data from 21 case-control studies of ovarian cancer (19,066 controls, 11,972 invasive and 2,752 borderline cases). Study-specific odds ratios (OR) and 95 % confidence intervals (CI) were obtained from logistic regression models and combined into a pooled odds ratio using a random effects model. Results Current cigarette smoking increased the risk of invasive mucinous (OR = 1.31; 95 % CI: 1.03-1.65) and borderline mucinous ovarian tumors (OR = 1.83; 95 % CI: 1.39-2.41), while former smoking increased the risk of borderline serous ovarian tumors (OR = 1.30; 95 % CI: 1.12-1.50). For these histological types, consistent dose- response associations were observed. No convincing associations between smoking and risk of invasive serous and endometrioid ovarian cancer were observed, while our results provided some evidence of a decreased risk of invasive clear cell ovarian cancer. Conclusions Our results revealed marked differences in the risk profiles of histological types of ovarian cancer with regard to cigarette smoking, although the magnitude of the observed associations was modest. Our findings, which may reflect different etiologies of the histological types, add to the fact that ovarian cancer is a heterogeneous disease.

Original languageEnglish (US)
Pages (from-to)989-1004
Number of pages16
JournalCancer Causes and Control
Issue number5
StatePublished - May 2013


  • Case-control studies
  • Histological type
  • Ovarian neoplasms
  • Smoking

ASJC Scopus subject areas

  • Oncology
  • Cancer Research


Dive into the research topics of 'Cigarette smoking and risk of ovarian cancer: A pooled analysis of 21 case-control studies'. Together they form a unique fingerprint.

Cite this