Article first published online: 11 JAN 2007
Copyright © 2007 American Cancer Society
Volume 109, Issue 4, page 812, 15 February 2007
How to Cite
Pavelka, J. C., Karlan, B. Y. and Li, A. J. (2007), Author reply. Cancer, 109: 812. doi: 10.1002/cncr.22393
- Issue published online: 2 FEB 2007
- Article first published online: 11 JAN 2007
We appreciate and acknowledge the comments of Dr. Mundstedt and his colleagues regarding our findings that correlated obesity with aggressive epithelial ovarian cancer biology.1 Mundstedt et al. argue that our data suggest that the higher incidence of ovarian cancer in obese women leads to an aggressive phenotype and that our findings are limited by the 216 patients who were included in our cohort.
In response to their comments, we would like to point out that we do not assert that the increased incidence of ovarian cancer among obese women supports a more aggressive biology. Current data regarding obesity as a risk factor for ovarian cancer are inconsistent, as we discussed in the article, and our own findings suggest that obesity is an independent poor prognostic factor in women with advanced-stage disease.
Furthermore, we believe the exclusion of nonepithelial histologies, synchronous malignancies, and patients undergoing neoadjuvant chemotherapy strengthens our findings and does not contribute to bias as Mundstedt et al. suggest. In addition, of all patients who were diagnosed with ovarian cancer during the interval of our study and who met the inclusion criteria, only 5 additional patients were excluded because of absent data regarding height and weight.
Munstedt and colleagues are to be congratulated on the robust database that they have assembled. Given that details are not provided for the post-hoc analysis of their large cohort, we cannot support or refute the conclusions suggested in their letter. For example, we have no assessment of the surgical or chemotherapeutic treatments that clearly impact survival. The lack of correlation between stage and body mass index in their data may result from an actual lack of correlation or, possibly, from a difference in the rate of comprehensive surgical staging between our 2 datasets. With regard to the immunohistochemical data, the lack of correlation between estrogen and progesterone receptor status and Ki-67 level with body mass index does not necessarily exclude a difference in tumor biology.
In conclusion, we thank Munstedt and colleagues for their interest in our study. Their own contributions to our knowledge of ovarian cancer epidemiology and biology are not inconsequential. However, their dismissal of our data seems premature, because the nature of the possible interrelations between obesity and cancer is only beginning to be understood. Given a lack of similarly controlled retrospective or prospective series relating obesity to ovarian cancer outcomes, we believe that our data suggest a provocative hypothesis that begs to be investigated through future prospective and translational studies.
James C. Pavelka MD*, Beth Y. Karlan MD*, Andrew J. Li MD*, * Cedars-Sinai Medical Center Women's Cancer Research InstituteSamuel Oschin Comprehensive Cancer Institute Los Angeles, California.