569P - GIV as a novel prognostic marker in stage II colon cancer

Date 29 September 2014
Event ESMO 2014
Session Poster Display session
Topics Colon Cancer
Translational Research
Presenter Peter Gibbs
Citation Annals of Oncology (2014) 25 (suppl_4): iv167-iv209. 10.1093/annonc/mdu333
Authors P. Gibbs1, P. Ghosh2, P. Waring3, B. Lafleur4, A. Muranyi4, S. Singh4, P. Brunhoeber4, J. Tie1, B. Tran1, J. Desai1, R. Martinez5, K. Janssen6, A. Goel7, S. Hu8, V. Teichgräber8, U. Rohr8, R. Ridder4, K. Shanmugam4
  • 1Medical Oncology, Walter and Eliza Hall Institute, 3052 - Parkville/AU
  • 2Department Of Medicine, Gastroenterology, University of California, San Diego, 92093 - La Jolla/US
  • 3Department Of Pathology, University of Melbourne, Melbourne/AU
  • 4Research, Ventana Medical Systems, Inc., 85755 - Tucson/US
  • 5Department Of Gastroenterology, University of Alicante General Hospital, 03010 - Alicante/ES
  • 6Department Of Surgery, Technical University Munich, 81675 - München/DE
  • 7Baylor Research Institute And Charles A. Sammons Cancer Center, Baylor University Medical Center, 75246 - Huston/US
  • 8Diagnostics Division, F. Hoffmann-La Roche Ltd, 4070 - Basel/CH

Abstract

Aim

There is an evident unmet need to better define recurrence risk for patients with stage II colon cancer (CC), particularly those with mismatch repair proficient (pMMR) tumors, to determine the subgroup of patients that may benefit from adjuvant chemotherapy. GIV/Girdin is a novel metastasis associated protein that triggers tumor cell invasion by enhancing PI3K/AKT signaling downstream of multiple oncogenic receptors. We explored the potential of GIV as a prognostic marker in stage II CC.

Methods

A MMR antibody panel and a GIV specific antibody were developed by Ventana and evaluated by immunohistochemistry (IHC) on a cohort of stage II CC from Melbourne (n = 192), enriched for patients with recurrent disease (n = 44, 25%). Log rank test was used to assess the association of GIV IHC expression status with the recurrence risk. Evaluation of alternative GIV scoring algorithms, combining staining intensity and percent staining, and statistical predictive modeling including the analysis of distant recurrence free survival (DRFS) was undertaken for the chemo-naïve cases, stratified by MMR and AJCC tumor (T) status. The association between GIV IHC status and standard histopathologic criteria was also assessed.

Results

The distribution of deficient MMR (dMMR) vs pMMR cases were 20.8% and 79.2%, respectively. Within the chemo-naïve population (n = 103), expected associations between pathologic features and DRFS were observed, including pMMR vs dMMR (HR 4.12, p = 0.052), T4 vs T3 (HR 2.39, p = 0.055) and lymphovascular invasion (LVI) vs no LVI (HR 2.62, p = 0.021). GIV positivity, defined as >10% of tumor cells stained or any staining of 3+ intensity, was present in 45 (44.0%) of pMMR chemo-naïve cases. For T3 pMMR chemo-naïve cases (n = 91), GIV positivity was associated with significantly reduced DRFS (HR 2.49, p = 0.022). Adding LVI as an additional parameter to the algorithm for this group increased the HR to 4.74 (95% CI 1.90-11.85).

Conclusions

GIV IHC expression status has been found to be associated with T3 disease, pMMR, and DRFS in an initial cohort of stage II CC. This exploratory diagnostic algorithm currently undergoes validation in two independent clinical stage II CC cohorts, with promising preliminary data.

Disclosure

P. Waring: received research grant funding from F. Hoffmann-La Roche; B. Lafleur, A. Muranyi, S. Singh and P. Brunhoeber: Employee of Ventana Medical Systems, Inc., a member of the Roche Group; S. Hu, V. Teichgräber, U. Rohr, R. Ridder:. and K. Shanmugam: Employee of F. Hoffmann-La Roche Ltd. All other authors have declared no conflicts of interest.