Vav2 protein overexpression marks and may predict the aggressive subtype of ductal carcinoma in situ

Background: A subset of patients with ductal carcinoma in situ (DCIS) will develop invasive breast cancer (IBC). To date, there are no effective predictive biomarkers for identifying this subset with worse prognosis whose lesions are essentially indistinguishable histologically from those with favorable outcomes. We hypothesized that measurable parameters that discriminate DCIS from DCIS with concurrent invasion may serve as diagnostic biomarkers (BM) of progressive cancer in situ (CIS).

Results: Using a novel imaging-based method of tissue testing, we measured the relative expression levels of three candidate BM proteins specifically implicated in IBC progression - the insulin-like growth factor I receptor (IGF-IR), Ras-related protein 1 (Rap1), and Vav2 oncoprotein. Protein profiles were compared in 42 histologically normal mammary epithelial samples, 71 CIS (35 without/36 with invasion either on diagnostic biopsy or final surgical excision), and 98 IBC of known estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2) status. The levels of the IGF-IR and Rap1 protein expression were significantly elevated in ER-positive (ER+/PR+/-/HER2 -) DCIS relative to normal epithelium (P <0.0001). The IGF-IR protein expression was also significantly up regulated in HER2-positive (ER+/-/PR+/-/HER2+) DCIS relative to normal epithelium (P = 0.0002). IGF-IR and Rap1 protein expression levels were similar among DCIS patients without or with concurrent invasion. Vav2 upregulation in DCIS relative to normal group was not associated with steroid hormone receptor and HER2 status, but was associated with the presence of concurrent invasion, including microinvasion (invasive foci of less than 1 mm). DCIS with high Vav2 were more than twice as likely to progress to invasive cancers as DCIS with low Vav2 (odds ratio, 2.42; 95% CI, 1.26-4-65; P =0.008). Furthermore, a receiver operating characteristic curve analysis revealed moderate ability of Vav2 protein expression measurements in DCIS to predict the existence of invasion concurrent with DCIS (area under the curve, 0.71; 95% CI, 0.59- 0.84).

Conclusions: Our novel findings hold promise for utilizing Vav2 protein as a predictive BM for differentiating progressive from non-progressive DCIS.