42P - Atopy and cancer risk: Insights from a prospective study on serum specific and total immunoglobulin E

Date 19 December 2015
Event ESMO Asia 2015 Congress
Session Poster presentation 1
Topics Translational Research
Presenter Wahyu Wulaningsih
Citation Annals of Oncology (2015) 26 (suppl_9): 8-15. 10.1093/annonc/mdv518
Authors W. Wulaningsih1, H. Garmo1, L. Holmberg1, H. Malmstrom2, M. Lambe3, N. Hammar2, G. Walldius2, I. Jungner4, M. van Hemelrijck1
  • 1Division Of Cancer Studies, King's College London Guy's Hospital, SE1 9RT - London/GB
  • 2Institute Of Environmental Medicine, Karolinska University Hospital and Karolinska Institute-Huddinge, 17177 - Stockholm/SE
  • 3Department Of Medical Epidemiology And Biostatistics, Karolinska University Hospital and Karolinska Institute-Huddinge, 17177 - Stockholm/SE
  • 4Department Of Medicine, Karolinska University Hospital and Karolinska Institute-Huddinge, 17177 - Stockholm/SE



Atopy refers to specific immunoglobulin E (IgE)-mediated hypersensitivity reactions against certain allergens. Prior findings linking atopy and cancer incidence have been inconsistent, which may be driven by diverse assessment methods. We aimed to utilise serum specific IgE alone and in combination with serum total IgE in studying this association.


From the Apolipoprotein MORtality RISk Study (AMORIS), we selected 8,727 men and women who had measurements of serum specific IgE against inhalant allergens and total IgE between 1992 and 1996. Multivariable Cox regression using age as a timescale was performed to assess the associations of atopy, defined by serum specific IgE ≥35 kU/L, with risk of overall and specific cancers. A test for trend was performed by assigning specific IgE scores (0 to 6) as an ordinal scale. Further adjustment and stratification analysis were performed to account for levels of total IgE.


During a mean follow-up of 16 years, 689 persons were diagnosed with cancer. Atopy showed an inverse association to cancer risk, with a hazard ratio (HR) of 0.82 and 95% confidence intervals (CI) of 0.69-0.99 after adjustment with serum total IgE. A similar trend was seen with specific IgE scores overall (Ptrend = 0.003) and in women (Ptrend = 0.009). Stratification analysis showed that this trend was strongest in those with high level of total IgE (≥100 kU/L), particularly in women. Although no clear association was observed with risk of common site-specific cancers, serum specific IgE scores were inversely associated with melanoma risk in men and women combined, and with risk of female breast and gynaecological cancers combined.


Atopy against inhalant allergens was inversely associated with incident cancer when serum total IgE was taken into account. These findings indicate the importance of further investigations of the role of immunologic processes in the development of cancer.

Clinical trial identification


N. Hammar: employed by the AstraZeneca but the views expressed in this work are his views and not those of AstraZeneca's. All other authors have declared no conflicts of interest.