Hopes Revived for Targeting the "Undruggable" RAS Family

OncologyLive, January 2014, Volume 15, Issue 1

The members of the RAS oncogene family are central cogs in many different cell-signaling pathways, coordinate a variety of important cellular processes, and are highly mutated in a number of different cancers, including several with extremely poor prognosis.

The members of the RAS oncogene family are central cogs in many different cell-signaling pathways, coordinate a variety of important cellular processes, and are highly mutated in a number of different cancers, including several with extremely poor prognosis. This makes them extremely attractive targets for anticancer therapy. However, over the course of more than a quarter of a century, the search for small-molecule inhibitors of RAS family members has proved fruitless. While a number of agents that indirectly target RAS have been developed, they have had limited success in patients with RAS mutations. Now, new strategies are offering hope that RAS may, in fact, be “druggable” after all.

The RAS Superfamily: A Molecular Switch

The RAS superfamily is a group of more than 150 genes that act as central nodes in cellular signaling networks, connecting a variety of upstream signals to an even larger array of downstream effector pathways to regulate cellular processes such as cell proliferation, survival, and death. The three “classical” RAS family members—KRAS, HRAS, and NRAS—sit on the inner face of cell membranes and cycle between two states in which they are either bound by guanosine diphosphate (GDP; the “off” state) or guanosine triphosphate (GTP; the “on” state). This switch is regulated by guanine nucleotide exchange factors (GEFs) and GTPase activating proteins (GAPs), which promote the formation of GDP-Ras and GTP-Ras, respectively. In the “on” state, they engage downstream effector proteins, including Raf kinase, phosphatidylinositol 3-kinase (PI3K), and RalGDS (guanine nucleotide dissociation stimulator), and transduce signals from cell membrane receptors such as the epidermal growth factor receptor (EGFR).

A Stuck Switch Leads to Cancer

As might be expected, a disruption to the delicate balance of Ras signaling is frequently implicated in the development of cancer. In fact, the RAS family are among the most commonly mutated genes in human cancer, particularly KRAS, which is mutated in 25% to 30% of all tumors and in up to 90% of pancreatic adenocarcinomas (Table). In both pancreatic and colorectal cancers, KRAS mutation is detected early in the progression of the disease, increases with the evolution of the tumor, and is strongly correlated with worse prognosis and increased aggressiveness of the tumor.

Table. Role of KRAS Mutations in Various Cancers

Tumor Type

Reported Prevalence of KRAS Mutations1-6

Significance for Prognosis or Therapeutic Response

Biliary tract

35%

  • Predicts reduced survival rates7

Bladder

4%

  • No correlation reported8

Breast

4%

  • Predictive of grade 3 tumors9

Cervical

8%

  • Predicts significantly worse recurrence-free survival rates and associated with formation of distant metastases10

Colorectal

30%-40%

  • Predicts poor prognosis and increased aggressiveness
  • Multiple studies demonstrate reduced response to EGFR TKIs
  • NCCN guidelines recommend testing for KRAS mutation—only patients with wild-type KRAS should receive EGFR inhibitors
  • FDA-approved Therascreen KRAS PCR Kit for KRAS mutation testing1,2

Endothelium

15%

  • No data

Hematopoietic malignancies

5%-27%

  • Unclear1

Liver

6%

  • No data

Lung

NSCLC

- Large cell carcinoma

- Squamous cell carcinoma

- Adenocarcinoma

Small cell

16%

21%

6%

25-30%

0%

  • Associated with inferior progression-free survival and response rates
  • Reduced response to EGFR TKIs and certain forms of chemotherapy1,2,11,12

Pancreatic

- Ductal adenocarcinoma

- Endocrine tumor

85-95% (almost 100% in advanced cancer)

1%

  • Predicts poor prognosis and increased aggressiveness (depending on KRAS mutation type)
  • Reduced response to EGFR TKIs1,2

Ovarian

14%

  • Significant association with mucinous histology, well-differentiated tumors, and positive progesterone expression13
  • May predict response to EGFR TKIs14

Prostate

8%

  • No data

Thyroid

2%-9%

  • Predicts poor survival15

Tumor types in bold are those in which most KRAS-directed research has been conducted.

EGFR TKIs indicates epidermal growth factor receptor tyrosine kinase inhibitors; NSCLC, non-small cell lung cancer.

1Fernández-Medarde et al. Genes Cancer. 2011;2:344; 2Chetty et al. J Clin Pathol. 2013;66:548; 3Smit et al, Nucleic Acids Res.1988;16(16):7773; 4Almoguera et al. Cell.1988;53(4):549; 5Rodenhuis et al. Cancer Res.1992;52(9 suppl):2665s; 6Suda et al. Cancer Metastasis Rev. 2010;29(1):49; 7Malats et al. J Clin Oncol. 1995;13(7):1679; 8Ouerhani et al. Mol Biol Rep. 2013,40:4109; 9Pereira et al. PLoS One. 2013;8(3):e60576; 10Wegman et al. Int J Gynecol Cancer. 2011;21(1):86;11Sun et al. PLoS One. 2013;8(5):e64816;12Johnson et al. Cancer. 2013;119:356; 13Nodin et al. Diagn Path. 2013;8:106; 14Auner et al. BMC Cancer. 2009;9:111;15Garcia-Rostan et al. J Clin Oncol. 2003;21(17):3226.

Activating mutations in the RAS gene essentially keep the switch in the “on” position by preventing GAPs from associating with the Ras protein correctly and hydrolyzing GTP, so that Ras persists in its active GTP-bound state. In addition to mutations in RAS genes, the proteins can be inappropriately activated in other ways, such as via mutations in molecules that interact biochemically with Ras (including Kit and bcrabl) and, much more rarely, mutations in Ras GAPs (thus far, the only GAP implicated in tumorigenesis is NF-1).

The RAS family has been singled out as exciting potential therapeutic targets since their discovery more than 30 years ago. Indeed, it appeared as though “fixing” the broken switch by restoring Ras enzymatic function should be relatively easy. In fact, directly targeting RAS with pharmaceuticals has proved to be anything but simple, to the extent that RAS was essentially labeled undruggable.

Strategies for Therapeutic Targeting of RAS

The initial focus of research into RAS-targeting agents was on the development of small molecules that could antagonize GTP or mimic GAPs, with the aim of switching off Ras by restoring the ability to form inactive GDP-Ras. However, these approaches ultimately failed for a number of different reasons, in particular because of the high affinity of Ras binding for GTP and the even higher concentrations of GTP in the cell, which made competitive inhibition untenable.

Since reversing the constitutive activation of RAS was not easily achieved, the focus next shifted toward the possibility of blocking the association of RAS with cell membranes, which is essential for its function. This strategy engendered significant interest in the 1990s. Ras proteins undergo a series of posttranslational modifications, including farnesylation and palmitoylation, which allow them to form a stable interaction with cell membranes. Inhibitors of the enzyme farnesyltransferase and farnesyl moiety-containing molecules were developed, with the aim of interfering with Ras farnesylation, and thus preventing membrane binding. These agents progressed through to phase II trials, but their development eventually was halted due to negative results.

Though ultimately unsuccessful, these early trials were very informative in our understanding of the biology of the Ras proteins. One thing that they taught investigators was that the three Ras proteins were not functionally identical and that more specific agents may be more effective. From that point onward, research efforts focused on KRAS, the most frequently mutated of the three proteins.

Research also shifted to more indirect means of targeting RAS, and these have proved significantly more fruitful. Given that many of the upstream activators and downstream effectors of Ras are kinases, these proteins make more ideal targets for small-molecule inhibition. Two of the best-characterized proteins are EGFR (upstream) and Raf kinase (downstream). EGFR has been successfully targeted by the development of small-molecule inhibitors such as gefitinib, which has been used to treat patients with nonsmall cell lung cancer (NSCLC), while agents such as sorafenib and vemurafenib have been developed to target Raf and have proved successful for the treatment of renal cancer and latestage melanoma, among others.

Although these drugs have achieved FDA approval, they have somewhat paradoxically ultimately demonstrated limited activity in tumors that have RAS mutations. It is thought that treatment with EGFR inhibitors drives RAS activation and allows the cancer cell to bypass EGFR signaling, while use of Raf inhibitors in RAS-mutant cells promotes Raf signaling, working against Raf inhibition. KRAS mutations have subsequently been shown to be among the most significant predictors of response to EGFR inhibitors in colorectal, pancreatic, and other forms of cancer. In the case of colorectal cancer, National Comprehensive Cancer Network guidelines recommend testing for KRAS mutation before administering these agents and limiting their use to patients with KRAS wild-type tumors.

Another strategy that has been explored for targeting KRAS-mutant tumors is synthetic lethality, the concept that two mutations that are nonlethal by themselves drive cell death when present simultaneously in a cell. In this case, rather than trying to reverse or bypass the RAS mutation, researchers seek to exploit the RAS-mutant status of the tumor by finding secondary mutations that are synthetically lethal with it, and thus drive cell death. This goal is achieved using RNA interference screens, in which panels of mutations are screened in RAS-mutant cell lines to identify “hits” that cause cell death. Among the numerous “hits” that have been identified via synthetic lethality screening in KRAS-mutant cell lines, the most promising is polo-like kinase 1 (PLK1), a protein involved in cell division. Several PLK1 inhibitors are currently undergoing clinical evaluation, including volasertib (BI6727), which is currently in phase II trials (NCT01121406).

The Importance of Molecular Testing

Currently, mandated KRAS testing is restricted to mutations in codons 12 and 13. Recent trials examining the predictive potential of other KRAS and NRAS mutations in patients with metastatic colorectal cancer suggest that expanding the range of mutations for which tumors are tested could help to improve survival even further following treatment with EGFR inhibitors. The phase II PEAK trial suggested that codons 61, 117, and 146 for both KRAS and NRAS should now be included in molecular testing and could help to screen a further 20% of patients.

Meanwhile, a retrospective biomarker analysis from the phase III PRIME study suggests that inclusion of exons 2, 3, and 4 could add an additional 10% of patients to the pool, and removal of these patients improved survival following administration of the chemotherapy regimen FOLFOX and the EGFR inhibitor panitumumab by almost 6 months. Furthermore, these trials also suggested that administration of EGFR inhibitors in these patients could in fact be detrimental, emphasizing the importance of mutation testing and improved selection of patients for treatment.

Given clinicians’ concerns about payment for molecular testing and the time taken to administer these tests, confirmatory studies are needed prior to any practice-changing recommendations. Indeed, further analyses are going to be applied to the FIRE-3 and CALGB-8045 trials, two phase III trials that are examining various combinations of the FOLFOX and FOLFIRI chemotherapy regimens and the EGFR and vascular endothelial growth factor (VEGF) inhibitors cetuximab and bevacizumab.

Revived Hope for Targeting KRAS

There have been exciting recent preclinical developments in both the direct and indirect targeting of KRAS that have revived hopes of treating KRAS-mutant tumors. Researchers have now found a way to target RalGDS, a downstream signaling pathway of particular promise because it is a Ras-specific effector pathway. RalGDS has proved very difficult to target in the past, but now Hu et al from the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins in Baltimore, Maryland, have identified a druggable target in this pathway: cyclin-dependent kinase 5 (CDK5). This development was reported at the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics in October 2013 (Abstract B263).

CDK5 can be targeted with the pan-CDK inhibitor dinaciclib, and in preclinical models of pancreatic cancer, combined inhibition with dinaciclib and PI3K/Akt and Raf kinase inhibitors (pan-Akt inhibitor MK-2206 and ERK inhibitor SCH7722984) led to a substantial inhibition of tumor growth and reduced number of metastases. The combination of dinaciclib with MK-2206 was particularly potent, decreasing tumor growth by 90% and leading to complete responses in three of 14 mice. The combination of dinaciclib and MK-2206 is currently undergoing phase I clinical testing in patients with pancreatic cancer (NCT01783171).

Several other recent developments have revived the idea of directly targeting RAS for the treatment of cancer. The first discovery returns to the idea of targeting the membrane association of Ras. Photoreceptor cGMP phosphodiesterase delta subunit (PDEδ) is a prenyl-binding protein that interacts with farnesylated KRAS and ensures its correct localization to the cell membrane. A small molecule that inhibits the binding of KRAS to PDEδ was developed, and preclinical studies indicate that this agent inhibits the oncogenic signaling of Ras and suppresses the proliferation of KRAS-mutant pancreatic cancer cell lines.

The second discovery was presented at the AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics (abstract C209). Burns et al from the Vanderbilt School of Medicine in Nashville, Tennessee, were searching for small-molecule inhibitors of the protein son of sevenless (SOS), the primary GEF that activates Ras. Somewhat counterintuitively, the team actually found that small molecules that activated rather than inhibited SOS reduced downstream signaling through mitogen- activated protein kinase (MAPK) and PI3K. These molecules bound to a hydrophobic pocket in the CDC25 domain of SOS, and the team is now working to understand why they cause an inhibition of downstream signaling; however, this could potentially serve as a novel way to target aberrant KRAS signaling in cancer cells.

Finally, a very recent study has demonstrated a novel mechanism by which KRAS mutations in epithelial cells may promote cancer cell invasion and could potentially offer up a novel therapeutic target. Slattum and colleagues at the Huntsman Cancer Institute at the University of Utah in Salt Lake City found that KRAS mutations allow epithelial cells to subvert the process of extrusion (the normal process of epithelial cell turnover, whereby cells are pushed out of the epithelium and undergo cell death). KRAS mutation causes the cells to be extruded into rather than out of the tissue, and the cells are subsequently able to survive and potentially migrate to other areas of the body instead of undergoing cell death. The mutant cells are thought to achieve this feat by removing a signaling lipid, sphingosine 1-phosphate (S1P), which plays a key role in the extrusion process by degrading it via enhanced autophagy. The study went on to show that blocking autophagy allowed proper extrusion and death of KRAS-mutant cells.

Key Research Baines AT, Xu D, Der CJ. Inhibition of Ras for cancer treatment: the search continues. Future Med Chem. 2011;3(14):1787-1808.

Burns M, Sun Q, Daniels R. Approach for targeting Ras with small molecules that activate SOS-mediated nucleotide exchange. Mol Cancer Ther. 2013;12(11 suppl; abstr C209).

Chetty R, Govender D. Gene of the month: KRAS. J Clin Pathol. 2013;66:548-550.

Douillard J-Y, Oliner KS, Siena S, et al. Panitumumab-FOLFOX4 treatment and Ras mutations in colorectal cancer. N Engl J Med. 2013;369:1023-1034.

Fernández-Medarde A, Santos E. Ras in cancer and developmental diseases. Genes Cancer. 2011;2(3):344-358.

Frett B, Wang Y, Li H-Y. Targeting the K-Ras/PDEδ protein—protein interaction: the solution for Ras-driven cancers or just another therapeutic mirage? Chem Med Chem. 2013;8:1620-1622.

Hu C, Dadon T, Chenna V, et al. Combined inhibition of cyclin- dependent kinases (dinaciclib) and AKT (MK-2206) or ERK (SCH772984) dramatically blocks pancreatic tumor growth and metastases in patient-derived orthotopic xenograft models. Mol Cancer Ther. 2013;12(11 suppl; abstr B263).

Schwartzberg LS, Rivera F, Karthaus M, et al. Analysis of KRAS/ NRAS mutations in PEAK: a randomized phase II study of FOLFOX6 plus panitumumab (pmab) or bevacizumab (bev) as first-line treatment (tx) for wild-type (WT) KRAS (exon 2) metastatic colorectal cancer (mCRC). J Clin Oncol. 2013;31(suppl; abstr 3631).

Slattum G, Gu Y, Sabbadini R, Rosenblatt J. Autophagy in oncogenic K-Ras promotes basal extrusion of epithelial cells by degrading S1P [published online December 18, 2013]. Curr Biol. doi: 10.1016/j.cub.2013.11.029.

Wang Y, Kaiser CE, Frett B, Li H-Y. Targeting mutant KRAS for anticancer therapeutics: a review of novel small molecule modulators. J Med Chem. 2013;56: 5219-5230.

Zimmermann G, Papke B, Ismail S, et al. Small molecule inhibitor of KRAS-PDEδ interaction impairs oncogenic KRAS signalling. Nature. 2013;497:638-642.