Citation: European Journal of Cancer Supplements, Vol 7 No 2, September 2009, Page 17

K. Berns¹, H.M. Horlings², J.B.G. Halfwerk², B.T. Hennessy³, S.C. Linn⁴, M. Hauptmann⁵, G.B. Mills³, M.J. van de Vijver², R. Bernards<sup>1

1</sup>The Netherlands Cancer Institute, Department of Molecular Carcinogenesis, Amsterdam, The Netherlands
²Academic Medical Centre, Department of Pathology, Amsterdam, The Netherlands
³MD Anderson Cancer Center, Department of Systems Biology, Houston, USA
⁴Netherlands Cancer Institute, Department of Medical Oncology, Amsterdam, The Netherlands
⁵Netherlands Cancer Institute, Department of Molecular Biology, Amsterdam, The Netherlands

Background: At present, 1 year of Trastuzumab (Herceptin^®) treatment has become part of the adjuvant treatment of women with early stage HER2-positive breast cancer. However, it is still largely unclear why almost half of the breast cancer patients that over-express HER2 is non-responsive to Trastuzumab based therapy or become resistant to Trastuzumab during treatment. Therefore, clinical questions such as how to optimize patient selection or prevent resistance to Trastuzumab-based therapy still await answers.
We present here a method to identify biomarkers associated with non-responsiveness to Trastuzumab in cell culture and demonstrate that these biomarkers have predictive value in a patient cohort treated with Trastuzumab combination therapy.
Material and Methods: As an unbiased approach to identify genes involved in Trastuzumab resistance, we used a large-scale RNA interference genetic screen in the HER2-overexpressing breast cancer cell line BT-474. We have generated a library of 24,000 shRNA retroviral vectors targeting some 8,000 human genes for suppression by RNA interference and developed a technology to rapidly screen such libraries, named siRNA bar code screening.
Results: Of the 8,000 genes tested, we found that only knock down of PTEN conferred resistance to Trastuzumab. Decreased PTEN expression results in hyper activation of the PI3K pathway. Significantly, activating mutations in the gene encoding the p110a catalytic subunit of PI3K (PIK3CA) have been identified in some 25% of primary breast cancers potentially mimicking the effects of PTEN loss. Indeed, overexpression of the breast cancer-derived mutant PIK3CA (H1047R) also conferred resistance to Trastuzumab in cell culture. These findings are consistent with a major role of the PI3K pathway in the development of resistance to Trastuzumab.
Our cell culture experiments led us to investigate whether PI3K pathway activation is able to predict Trastuzumab resistance in the clinic. In a cohort of 55 patients treated for metastatic breast cancer, activation of the PI3K pathway, as judged by the presence of oncogenic PIK3CA mutations or low PTEN expression, was associated with poor prognosis after Trastuzumab therapy. Furthermore, the combined analysis of PTEN and PIK3CA identified twice as many patients at increased risk for progression compared to PTEN alone.
Conclusions: The present work highlights the central importance of PI3K signalling in risk for progression after Trastuzumab-based therapy, which in turn suggests combination therapeutic strategies to treat Trastuzumab unresponsive breast cancer or to prevent emergence of resistance. We are currently analyzing a cohort of 50 patients who received neo-adjuvant Trastuzumab-based therapy to test whether PI3K pathway activation status validates as a biomarker for response prediction in the neo-adjuvant setting.