•  Oncogene: also known as a carcinogenic gene, is a class of genes capable of transforming normal cells into cancerous cells. In general terms, an oncogene is a proto-oncogene that has become functionally abnormal.

•  Proto-oncogene: a gene that exists naturally in normal cells, generally related to the regulation of cell growth and proliferation.

•  A fully functional proto-oncogene does not lead to cancer; it only becomes a carcinogenic oncogene upon mutation.

•  KRAS: KRAS belongs to the Rat Sarcoma Viral Oncogene family (RAS), with the ability to hydrolyze GTP into GDP. KRAS mutations occur in about 25% of all cancers, making it the most prevalent oncogenic driver. It was first identified in lung cancer cells back in 1982...

Under normal physiological conditions, KRAS oscillates between an inactive GDP-bound state and an active GTP-bound state, transducing signals from outside the cell to the inside. Upon activation of Receptor Tyrosine Kinases (RTKs), Guanine nucleotide Exchange Factors (GEFs) bind with KRAS, facilitating the replacement of bound GDP with GTP, switching KRAS to its active state. Meanwhile, GAPs maintain KRAS in an inactive state by strengthening the bond between GDP and KRAS.

Among all KRAS mutations, G12 mutations are the most prominent (81%)^[2]. The G12C mutation inhibits the binding between GAP and KRAS, thereby suppressing GTP hydrolysis and locking the G12C mutated KRAS in its active state. Active KRAS induces signal transduction via the MAPK and PI3K pathways, promoting cell proliferation, growth, and survival, thereby facilitating tumor development^[1][2].

The identification of small molecule inhibitors needs the presence of appropriate binding pockets on the protein's surface. Therefore, proteins lacking the pocket are deemed 'undruggable,' which requies innovative therapeutic targeting strategies^[3]. Thus, KRAS has long been viewed as an "undruggable" target.

In 2013, based on the chemical library, the K. Shokat’s team identified compounds capable of covalently and selectively binding with KRAS^G12C-GDP. X-ray crystallography revealed the binding of these compounds to the switch 2 pocket located near the mutated cysteine residue. This revelation undeniably shattered the long-standing belief of KRAS being "undruggable".

Following this breakthrough, multiple collaborations and optimization research led to the development of several potential clinical drugs. In May 2021, AMG510 (Sotorasib), the first inhibitor targeting KRAS^G12C, received FDA approval. In December 2022, a second KRAS^G12C inhibitor, MRTX849 (Adagrasib), was approved by the FDA. Both are employed in the treatment of Non-Small Cell Lung Cancer (NSCLC) patients with KRAS^G12C mutations.

Small molecule inhibitors can directly target the KRAS^G12C mutation by interacting with the Switch II Pocket (S-IIP). These inhibitors selectively bind to and stabilize RAS when it's in the GDP-bound state. It eventually results in reduced signal transmission, especially through the RAF-MEK-ERK/MAP pathway. This process effectively impedes tumor progression.

However, accompanying the promising developments is the troubling emergence of drug resistance.

In the same year, Koga, T., and others identified secondary KRAS mutations causing resistance to Sotorasib and Adagrasib^[5]. Moreover, research revealed that in cases of resistance, even with full inhibition of KRAS, KRAS can maintain signal transmission via the MAPK and/or PI3K pathways by activating bypass signaling pathways^[6]. This undoubtedly adds difficulty to cancer treatments.

Currently, researchers are trying to overcome KRAS targeted therapy resistance through combinatorial drug strategies. For example, SHP2 inhibitors can suppress KRAS^G12C GTPase activity and enhance the effects of G12C inhibitors, which are currently being trialed in combination with Sotorasib (NCT04185883) and other KRAS^G12C inhibitors.

Recently, a study named "Chemical remodeling of a cellular chaperone to target the active state of mutant KRAS" was published in the prestigious journal, Science. they constructed a small molecule, RMC-4998 able to reconfigure the surface of Cyclophilin A (CYPA). This process generates a new conformational interface displaying high selectivity and binding affinity to the activated state of KRAS^G12C. The ensuing CYPA-drug-KRAS^G12C trimeric complex deactivates the carcinogenic signals, leading to tumor regression in a variety of human cancer models.

Currently, trimeric complex inhibitors selectively target active KRAS^G12C or multiple RAS mutants. For example, Compounds like RMC-6291, an optimized version of RMC-4998, are currently undergoing clinical trials (NCT05462717, NCT05379985)^[3].

An abundance of evidence indicates that abnormal KRAS protein functions are vital for the growth and sustenance of cancer. Currently, different strategies to inhibit KRAS signal transduction are under investigation^[7].

•  Direct KRAS Inhibition

Directly inhibiting RAS is the most attractive approach to developing treatments for KRAS mutant cancer, such as the previously mentioned Sotorasib and Adagrasib. Additionally, some studies focus on antisense oligonucleotide (ASO) therapy. For example, Macleod et al. developed a nanoparticle containing small interfering RNA targeting KRAS mutants (AZD4785). AZD4785 demonstrated anti-tumor activity by inhibiting KRAS expression in mouse tumor models bearing either KRAS mutant NSCLC cell line xenografts or patient-derived xenografts^[8].

•  Disruption of Post-Translational Membrane Localization of KRAS

KRAS must be positioned on the inner surface of the cell membrane to be biologically active. Therefore, inhibiting post-translational membrane localization of KRAS is also a viable strategy. Among these, farnesyl transferase inhibitors (FTIs) can block the activity of farnesyltransferase (FTase) by inhibiting the isoprenylation of the CAAX. CAAX is a post-translational modification that promotes KRAS membrane binding to the membrane. FTase inhibitors like Lonafarnib have been approved by the FDA for the treatment of Hutchinson-Gilford Progeria Syndrome. Tipifarnib is also in phase 2 clinical trials for head and neck squamous cell carcinoma^[2][9].

•  Inhibiting KRAS-Mediated Signal Transduction

Due to the challenges associated with directly targeting the KRAS protein, inhibiting KRAS-mediated signal transduction has been a viable approach. BI1701963 is an inhibitor targeting SOS1 (a key protein in the KRAS signaling pathway), and is currently in phase 1 clinical trials. BI1701963 can inhibit the binding between KRAS and GTP. Furthermore, its combination with Trametinib could be used to treat patients carrying KRAS mutations. Additionally, a combination treatment of ARS-1620 and SHP-099 (an SHP2 inhibitor), or combination of MRTX849 and RMC-4550 (an SHP2 inhibitor) can enhance tumor suppression effects. Apart from these, inhibiting downstream RAF-MEK-ERK signaling pathways (such as Trametinib, BVD-523) and PI3K-AKT-mTOR signaling pathways (like BYL-719, MK2206) represent other strategies for inhibiting KRAS^[2][10].

•  Finding Synthetic Lethal Interactors for KRAS

Synthetic lethality occurs when the combined mutations in two or more genes lead to cellular death. Therefore, finding synthetic lethal interactors for mutant KRAS are also a potential approach. Furthermore, oncogenic KRAS is associated with numerous metabolic processes, including induced glucose uptake, re-utilization of glutamine metabolism, and increased autophagy and macropinocytosis. Further elucidation of the link between oncogenic KRAS and cancer-related metabolic dysregulation may yield new anti-KRAS treatments. Indeed, with the success of cancer immunotherapy, combining KRAS-targeted therapy with immunotherapy presents substantial promise as a treatment strategy.

Today, we have introduced the basic theories, mechanisms, drug resistance, and latest research progress of KRAS - a 'nondruggable' hot target. We also reviewed various strategies for KRAS inhibitors. As our understanding of KRAS deepens, we believe that more compounds with higher activity and better bioavailability will emerge.