Proteomic profiling of endometrioid endometrial cancer reveals differential expression of hormone receptors and MAPK signaling proteins in obese versus non-obese patients

• Published in: Oncotarget Vol. 8; no. 63; pp. 106989 - 107001
• Main Authors: Mauland, Karen Klepsland, Ju, Zhenlin, Tangen, Ingvild Løberg, Berg, Anna, Kalland, Karl-Henning, Øyan, Anne Margrete, Bjørge, Line, Westin, Shannon N, Krakstad, Camilla, Trovik, Jone, Mills, Gordon B, Hoivik, Erling A, Johanna Werner, Henrica Maria
• Format: Journal Article
• Language: English
• Published: United States Impact Journals LLC 05.12.2017
• Subjects: Research Paper; endometrial cancer; hormone receptors; biomarkers; obesity; MAPK signaling
• ISSN: 1949-2553; 1949-2553
• DOI: 10.18632/oncotarget.22203

Endometrial cancer development is strongly linked to obesity, but knowledge regarding the influence of excess weight on endometrial tumor signaling pathways remains scarce. We therefore analyzed reverse phase protein array (RPPA) data for obesity-related protein expression patterns, using one training (n=272) and two test cohorts (n=68; n=178) of well-annotated samples from women treated for endometrioid endometrial cancer. Gene expression profiling and immunohistochemistry were used for cross-platform validation. Body mass index (BMI) was significantly correlated with progesterone receptor (PR) expression and a hormone receptor protein signature, across all cohorts. In two of the cohorts, BMI was negatively correlated with RTK- and MAPK-pathway activation, particularly phosphorylated MAPK T202 Y204 (p-MAPK) level. Using stepwise selection modelling, a BMI-associated protein signature, including phosphorylated estrogen receptor α S118 (p-ERα) and p-MAPK, was identified. In the subset of FIGO stage 1, grade 1-2 tumors, obese patients (BMI≥30) had better survival compared to non-obese patients in the two cohorts with longest follow-up time (p=0.042, p=0.058). Non-obese patients had higher p-MAPK levels, whereas obese patients had higher p-ERα levels and enrichment of gene signatures related to estrogen signaling, inflammation, immune signaling and hypoxia. In subgroup analysis of non-obese patients with FIGO stage 1 tumors, low PI3K-activation was associated with reduced survival (p=0.002, training cohort). In conclusion, increasing BMI is associated with increased PR and p-ERα levels and reduced MAPK signaling, both in all patients and in subsets with predicted excellent prognosis. The MAPK-pathway represents a potential therapeutic target in non-obese patients with low stage and low grade tumors.