240P - IL10 locus as a biomarker of melanoma survival

Date 28 September 2014
Event ESMO 2014
Session Poster Display session
Topics Melanoma and other Skin Tumours
Translational Research
Presenter Justin Rendleman
Citation Annals of Oncology (2014) 25 (suppl_4): iv58-iv84. 10.1093/annonc/mdu326
Authors J. Rendleman1, C. Adaniel1, E. Kern2, N. Fleming2, M. Krogsgaard1, D. Polsky1, R. Berman3, R. Shapiro3, A. Pavlick4, Y. Shao1, I. Osman1, T. Kirchhoff1
  • 1Nyu Cancer Institute, New York University Langone Medical Center, 10016 - New York/US
  • 2Nyu Cancer Institute, New York University Langone Medical Center, New York/US
  • 3Dermatology, New York University Langone Medical Center, 10016 - New York/US
  • 4Hematology-oncology, New York University Langone Medical Center, New York/US

Abstract

Aim

The 5-year survival rate for melanoma is higher than 90%, however there is a large degree of variability among patients. While clinicopathological variables provide a general prognostic value, there remains an urgent need for more personalized biomarkers to better predict survival outcomes. Immune related molecules, such as interleukins and cytokines, have been shown to contribute to melanoma progression via modulation of tumor immunogenicity, hence suggesting their potential as novel prognostic predictors.

Methods

We have investigated germline variants tagging immunomodulatory genes for their impact on melanoma survival. This included the genotyping of 93 SNPs tagging 43 genes in a population of 1,022 patients. Using the clinical and pathological data available for these patients, we have developed a multivariate Cox model to assess the association between SNPs and recurrence-free and overall survival.

Results

Multiple SNPs tagging a locus on 1q32.1 were associated with overall survival. This region contains four interleukin genes: IL10, IL19, IL20, and IL24. The strongest association was observed for rs3024493 located in a regulatory region of IL10 (P = 0.00069). In the same locus we have also replicated an association with overall survival for rs1800896, which has been suggested previously in small scale studies. In addition, using the data of 251 metastatic melanoma tumors from The Cancer Genome Atlas (TCGA) we have shown that IL10 expression strongly correlates with overall survival (P = 0.0015).

Conclusions

Our findings suggest that the genetic variants in immunomodulatory genes may have a possible clinical utility as novel prognostic biomarkers. Most importantly, for IL10 our study not only validates previous associations with melanoma survival, but also identifies a novel independent SNP (rs3024493) conferring a more significant association effect with overall survival. The identification of multiple independent association signals located in regulatory regions of IL10, and the correlation of IL10 expression with clinical outcome observed in TCGA patients, suggest this locus as a putative prognostic marker of melanoma survival.

Disclosure

All authors have declared no conflicts of interest.