Mutations in the phosphatidylinositol-3-kinase pathway predict for antitumor activity of the inhibitor PX-866 whereas oncogenic Ras is a dominant predictor for resistance

Abstract

The novel phosphatidylinositol-3-kinase (PI3K) inhibitor PX-866 was tested against 13 experimental human tumor xenografts derived from cell lines of various tissue origins. Mutant PI3K (PIK3CA) and loss of PTEN activity were sufficient, but not necessary, as predictors of sensitivity to the antitumor activity of the PI3K inhibitor PX-866 in the presence of wild-type Ras, whereas mutant oncogenic Ras was a dominant determinant of resistance, even in tumors with coexisting mutations in PIK3CA. The level of activation of PI3K signaling measured by tumor phosphorylated Ser(473)-Akt was insufficient to predict in vivo antitumor response to PX-866. Reverse-phase protein array revealed that the Ras-dependent downstream targets c-Myc and cyclin B were elevated in cell lines resistant to PX-866 in vivo. Studies using an H-Ras construct to constitutively and preferentially activate the three best-defined downstream targets of Ras, i.e., Raf, RalGDS, and PI3K, showed that mutant Ras mediates resistance through its ability to use multiple pathways for tumorigenesis. The identification of Ras and downstream signaling pathways driving resistance to PI3K inhibition might serve as an important guide for patient selection as inhibitors enter clinical trials and for the development of rational combinations with other molecularly targeted agents.

Figure 1. Effect of PX-866 on cell line derived xenografts

A, Cell line derived xenografts were grown subcutaneously in female SCID mice. Upon reaching 200mm³ the mice were treated with PX-866, 2.5–3.0 mg/kg every other day administered po. At the end of treatment the tumor volume was expressed as a percentage of the increase in the vehicle alone treated tumor volume (T/C%). * p <0.05. Tumor responses were characterized as No response (T/C >70%), Low response (T/C 35–69%), or Antitumor (T/C <35%). The Ras, Raf, PIK3A, LKB1 and PTEN mutation status of the tumors is shown: B, phospho-Ser⁴⁷³-Akt levels measured in representative tumors removed from mice treated with PX-866 2.5–3 mg/kg po at the end of treatment, Images were taken from different fields on the same film.

Figure 2. Protein analysis of sensitive and resistant cell lines

A, Cell lines were analyzed by 52 validated antibodies in a reverse phase protein array (RPPA). Protein levels were quantified and arranged in a heat map, red indicating high expression, black median and green low expression. B, Analysis of expression of components of the PI-3-kinase/Akt pathway and correlation with in vivo antitumor response. The lower panel shows a plot of phospho-Akt levels against in vivo antitumor response. C, Levels of c-Myc and cyclin B proteins showing differences between sensitive and resistant cell lines. * p < 0.05

Figure 3. Signaling and clonogenic potential of HCT-116 H- Ras construct cells

A, Measurement of cellular PI3-kinase activity and Raf actvation by on the left Western blot analysis using phospho-Ser⁴⁷³-Akt and phospho-Ser³³⁸-Raf The panel on the right showns densiometric data from the blots. B, On the left Western blot analysis for cellular cyclin B and c-Myc normalized against actin The panel on the right showns densiometric data from the blots.. C, Colony formation assay performed on H-Ras construct HCT-116 cells treated with 0.5 μM PX-866. * p<0.05 compared to wild type cells.

Figure 4. Apoptosis and in vivo effects of PX-866 on HCT-116 H- Ras construct cells

A. Trypan blue and flow cytometry analysis of annexin positive cells treated with 0.5 μM PX-866. * p<0.05 compared to wild type cells. B, Comparison of final volumes of vehicle (white bars) or PX-866 (black bars) treated tumors, * p<0.05 of treated compared to control.

Figure 5. Signaling in resistant and sensitive lines

Diagram showing the interactions of the cell signaling studied in sensitive and resistant lines. Signaling in PX-866 sensitive tumors comes from an increased reliance on the PI-3-kinase pathway, arising from aberrant activation through growth factors (GF) or mutated components of the pathway itself (*). Tumors with an activated Ras protein show a minimal response to inhibition of the PI-3-kinase pathway due to a shared reliance on alternate signaling pathways including the Raf and RalGDS pathways.