Tumor Growth and Angiogenesis in Ovarian Cancer Linked to Key Adapter Protein

Ovarian Cancer Tumor Growth Linked to Key Adapter Protein

Results from The Cancer Genome Atlas (TCGA) have shown that alterations in gene copy number (ie, amplifications or deletions) are an important characteristic of high-grade serous ovarian cancers (HGSOC).^1,2

Recently published findings inOncogeneby Christopher Duckworth, BS, et al have begun to characterize the impact of overexpression of a previously established oncogenic driver, GRB2-associated binding protein-2 (GAB2) in the context of ovarian cancers.^3,4They have found that GAB2 overexpression causes upregulation of multiple chemokines leading to tumor angiogenesis, and further, this process can be blocked by cotargeting two important signaling pathways that lie downstream of GAB2.³

In an earlier work, Gavin P. Dunn, MD, PhD, Department of Neurosurgery, Massachusetts General Hospital, Harvard Medical School, et al identifiedGAB2as an oncogenic driver from a sampling of 455 genes amplified in ovarian cancer.⁴In this study, GAB2, a signaling adaptor protein, was capable of potently transforming immortalized ovarian and fallopian tube secretory cells, and researchers demonstrated that cancer cell lines that overexpress GAB2 have enhanced phosphatidylinositol 3 kinase (PI3K) activation and are dependent on GAB2 for survival.⁴They also found that theGAB2gene was amplified in 44% of 562 samples, and that the GAB2 protein was overexpressed in 18% of 326 samples from the TCGA ovarian cancer dataset.

The findings establishedGAB2as an ovarian cancer oncogene, and emphasized the importance of PI3K signaling in ovarian cancer.⁴Activation of the PI3K pathway in ovarian cancer has previously been linked to poor outcome, and it has been suggested that inhibition of multiple signaling pathways may be necessary for targeted therapy in patients whose tumors exhibit this characteristic, referred to as "PI3Kness."^5,6

In the current study, Duckworth et al expanded on the earlier GAB2 findings, by showing that, by suppressing GAB2 expression in ovarian cancer cells (using an inducible small hairpin RNA system), tumor cell proliferation, angiogenesis, and peritoneal tumor growth, could all be effectively suppressed.³Using antibody arrays, they showed that increased levels of multiple chemokines, including CXCL1, CXCL2, and CXCL8, could be detected in the conditioned medium of the GAB2-overexpressing cells; this finding was of particular interest, because the upregulation of at least two of these chemokines (CXCL1 and CXCL8) has been previously associated with poor survival in ovarian cancer. In additional experiments, researchers showed a proangiogenic effect of these chemokines, at least partly dependent on signaling through the CXCR2 receptor, as assessed by endothelial cell tube formation. They also showed that these three chemokines could indeed act as autocrine factors, directly impacting the proliferation and survival of ovarian cancer cells overexpressing GAB2.³

This study also examined the signaling pathways necessary for GAB2-mediated chemokine expression.³Interestingly, the researchers found that chemokine expression was not suppressed with PI3K, PI3K/mTOR, or MEK inhibition, despite PI3K and mitogen-activated protein kinase (MAPK) constituting major signaling pathways that are regulated by GAB2. By comparison, inhibitors of the nuclear factor kappa-B kinase subunit b (IKKb) could effectively suppress chemokine mRNA expression induced by GAB2. Moreover, they found that a combination of IKKb and PI3K/mTOR inhibition could more effectively suppress proliferation and survival in ovarian cancer cells overexpressing GAB2 compared with targeting either of these pathways individually.³

"There are relatively few established targets, and ovarian cancer is genomically complex." said Duckworth, research specialist, Department of Pathology and Laboratory Medicine at the Medical University of South Carolina College of Medicine in an exclusive interview withTargeted Oncology. Duckworth also noted a paucity of developed single agents that have thus far had a meaningful effect on curtailing progression in ovarian cancer. "The most effective way to combat ovarian cancer seems to be utilizing combined targeted therapies, which adds an extra dimension of effort."

Duckworth believes that the current study provides new scientific knowledge suggesting thatGAB2gene amplification in ovarian cancer cells contributes to tumorigenesis by a mechanism involving the pathway-dependent induction of multiple chemokines, which enhance tumor blood vessel formation. In ongoing studies, Duckworth expects to determine whether or not it is the copy number status of theGAB2gene or its protein expression levels that would predict the tumor's vulnerability to inhibition of specific cancer pathways.

Asked how the coinhibition of IKKβ might also be useful to enhance the efficacy of PI3K/mTOR inhibitors in ovarian cancer, Duckworth noted that, as highlighted in the study, GAB2-amplified ovarian cancer cell lines are dependent on GAB2 for PI3K pathway activation.

"They show sensitivity to PI3K inhibition, but not MEK inhibition, which suggests GAB2 overexpression contributes to HGSOCs' PI3Kness," said Duckworth. "Our paper demonstrated that, although GAB2 activates PI3K and MEK effector pathways, their inhibition alone did not suppress the chemokine induction by GAB2 overexpression. Previous studies have indicated that chemokine signaling may enhance cancer cell survival in response to PI3K-pathway targeted therapy. Since we showed the induction of chemokines in GAB2-overexpressing ovarian cancer cells to be dependent on the IKKb pathway, we coinhibited the pathways to mitigate that response. This actually showed a more effective reduction in clonogenic growth than with individual pathway inhibition. One could say that, in a sense, this rationale may enhance the efficacy of the PI3K/mTOR inhibitors by further inhibiting cancer cell proliferation and survival."

Looking forward, from the perspective of targeted therapy, Duckworth believes that a subset of ovarian cancers might benefit from the combined IKKb/PI3K inhibition.

"We know that theGAB2gene is highly amplified in approximately 14% of HGOCs. Previous studies have established that the PI3K and NF-κB pathways were each activated in the majority of ovarian cancers. Therefore, it is important to assess the therapeutic efficacy of the combination treatment in more cancer cell lines—carrying different genetic alterations—to establish which tumor subtypes will be more likely to benefit from such a therapeutic approach." Duckworth said that such assessments will be a component of ongoing studies. "Currently, we are pursuing furtherin-vivodrug treatment studies involving the dual-pathway targeting strategy, to establish whether clinical studies at a higher level are warranted. Our laboratory is also exploring the interaction between GAB2 and other molecular targets in ovarian cancer."

References

1. Cho KR, Shih IeM. Ovarian cancer. Annu Rev Pathol. 2009; 4:287-313.
2. Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian carcinoma.Nature. 2011; 474(7353): 609-615.
3. Duckworth C, Zhang L, Carroll SL, Ethier SP, Cheung HW. Overexpression of GAB2 in ovarian cancer cells promotes tumor growth and angiogenesis by upregulating chemokine expression.Oncogene. 2015. doi: 10.1038/onc.2015.472. [Epub ahead of print]
4. Dunn GP, Cheung HW, Agarwalla PK, et al. In vivo multiplexed interrogation of amplified genes identifies GAB2 as an ovarian cancer oncogene.Proc Natl Acad Sci USA. 2014; 111(3): 1102-1107.
5. Huang J, Zhang L, Greshock J, et al. Frequent genetic abnormalities of the PI3K/AKT pathway in primary ovarian cancer predict patient outcome.Genes Chromosomes Cancer. 2011; 50(8): 606-618.
6. Bast RC Jr, Mills GB. Dissecting "PI3Kness": the complexity of personalized therapy for ovarian cancer.Cancer Discov. 2012; 2(1): 16-18.