Abstract A240: Preclinical molecular stratification of human tumors sensitive or resistant to S 49076, a novel MET/FGFR/AXL kinase inhibitor.

Protein kinase inhibitors constitute a class of anticancer agents with demonstrated clinical efficacy. However, the clinical benefit of these agents is often limited to a subset of patients with specific genomic lesions in their tumor cells rendering them sensitive to inhibition of one or more target kinases. It is therefore becoming increasingly important to identify biomarkers of patient stratification to enable personalized care. For those inhibitors targeting just one kinase this stratification strategy is relatively straightforward, although the number of patients who will benefit from these highly selective inhibitors alone is in general severely limited. In the majority of cases, the simultaneous targeting of more than one kinase, possibly by the same drug, will be required for therapeutic efficacy. However, for multi-target kinase drugs, patient stratification is considerably more complex and is becoming a major challenge during clinical development. S 49076 is a novel, potent, ATP-competitive tyrosine kinase inhibitor with activity demonstrated in cell and animal assays against the receptor tyrosine kinases MET, FGFR1/2/3 and AXL, all of which are associated with progression of a wide range of human malignancies. In kinase binding assays, S 49076 also binds to clinically-relevant MET mutated isoforms and 32 other kinases, many of which have also been implicated in cancer pathology. In the present study, the antitumor activity of S 49076 was assessed in a panel of 53 patient-derived tumors of diverse origin growing in three-dimensional in vitro culture. Sixteen (30%) of these tumors were sensitive to S 49076 at 1 μM or less. With the aim to define molecular signatures of sensitivity or resistance, all 53 tumors were characterized for expression, activation and mutation of all of the known target kinases using protein, mRNA and DNA analysis techniques. On an individual basis, no clear relationship between expression level or activity of each target kinase and sensitivity or resistance to S 49076 could be established. However, analysis of microarray data using z-score transformation enabled a clear link to emerge between the sum of the z-scores of all potential target kinases (z-score) and tumor response to S 49076. Moreover, when the presence of RAS or PIK3CA activating mutations was taken into account, the predictive power of the analysis (in particular the positive predictive value for sensitivity) was even more striking. Overall, 91% (29/32) of tumors with a z-score less than 0 were resistant to S 49076, whilst 85% (11/13) of tumors with a z-score greater than 0 and with no known mutations in RAS or PIK3CA were sensitive. Six out of the 8 tumors with a z-score greater than 0 but with mutations in RAS or PIK3CA were resistant. In vivo evaluation of the response to S 49076 of several of the sensitive tumors is in progress in xenograft studies. Together, these data demonstrate the potential power of molecular profiling and, in particular, transcriptomic analyzes in the stratification of patient tumors for treatment with a multi-kinase inhibitor. It is hoped that similar analysis of patient tumors during clinical trials of S 49076 will ultimately contribute to the progress towards a more personalized approach to cancer patient care. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr A240.