Abstract 3164
Background
We built a risk classification model for resected SqCLC (R-SqCLC) by combining clinicopathological predictors to discriminate patients' (pts) prognosis (Pilotto JTO 2015), we externally validated this model in a pts’ cohort of > 1,300 R-SqCLC (Bria WCLC 2016) and we performed an integrated multi-platforms genome analysis of prognostic outliers to identify potentially druggable modulators as the PI3K/RICTOR-mTORC2 axis (Pilotto WCLC 2016). Here, we validate our molecular findings and we enhance our rationale with in vitro studies.
Methods
Next Generation Sequencing (NGS) analysis of somatic mutations (SM) and copy number alterations (SCNA) was performed (Ion AmpliSeq Lung & Colon Cancer Panel: 22 genes, a SqCLC customized targeted NGS panel: 20 genes, the commercial Ion AmpliSeq Comprehensive Cancer Panel: 409 genes). In vitro experiments were performed using the SqCLC cell line H-1703 (Rictor amplified-6 copies). PF-05212384 [PI3K/mTOR inhibitor (inh)], AZD2014 (mTORC1/2 inh), MK-2206 (panAkt inh), everolimus (mTOR inh) and chemotherapeutic drugs (Docetaxel, Gemcitabine) were tested. Cell viability was assessed by crystal violet assay and the half maximal inhibitory concentration (IC50) was estimated to evaluate drug efficacy.
Results
Main results of overall 97 pts (Training/Validation: 60/37) are presented in the Table.Table:
1640P
| Gene | Training Set [%] | Validation Set [%] | |
|---|---|---|---|
| SM | TP53 | 53 [88.3] | 27 [72.9] |
| TIE1 | 4 [6.7] | 2 [5.4] | |
| PTEN | 6 [10] | 4 [10.8] | |
| PIK3CA | 3 [5] | 3 [8.1] | |
| SCNA Gains | RICTOR | 13 [35.1] | 14 [23.3] |
| PIK3CA | 17 [45.9] | 26 [43.3] | |
| FRS2 | 8 [21.6] | 7 [11.7] | |
| FGFR1 | 14 [37.8] | 18 [30] | |
| SCNA Losses | PTEN | 5 [13.5] | 19 [31.7] |
| TSC2 | 8 [21.6] | 7 [11.7] |
The in vitro results support a significant inhibition of H-1703 cells proliferation by Gemcitabine, Docetaxel, PF-05212384, MK-2206 and AZD2014 with IC50 values of 0.4 nM, 0.45 nM, 10 nM, 66 nM, 110 nM, respectively. Everolimus was not effective.
Conclusions
Our multi-step genomic analysis performed in almost 100 R-SqCLC pts allowed us to identify altered pathways with a biological impact in SqCLC oncogenesis, as the PI3K/RICTOR-mTORC2 axis. Moreover, our in vitro results justify pursuing mTOR inhibition, focusing on mTORC2 complex, in RICTOR-aberrant tumors.
Clinical trial identification
Not Applicable
Legal entity responsible for the study
Emilio Bria
Funding
My First AIRC (Associazione Italiana per la Ricerca sul Cancro) Grant (MFAG) project no. 14282. Young Investigational Award of the International Association for the Study of Lung Cancer (IASLC) 2015
Disclosure
All authors have declared no conflicts of interest.