Targeting the KRAS signaling pathway in cancer

The KRAS oncogene is one of the most frequently mutated oncogenes in human cancers. When mutated it can drive the development of several cancers including:

• Pancreatic cancer: KRAS mutations are found in approximately 90% of pancreatic ductal adenocarcinomas (PDAC), the most common type of pancreatic cancer. This makes it a hallmark feature of this deadly disease.
• Colorectal cancer: Around 40-50% of colorectal cancers (CRC) have KRAS mutations. These mutations are often associated with resistance to certain targeted therapies like EGFR inhibitors.
• Lung cancer: KRAS mutations occur in about 25-30% of non-small cell lung cancers (NSCLC), particularly in adenocarcinomas.

Previously I wrote about a new class of anti-cancer drugs that target KRAS. A common activating mutation is G12C in which the glycine amino acid (G) at position 12 in the protein has been mutated to cysteine (C). For years KRAS was considered undruggable because of its small size and limited surface area for a drug to bind. But advances in rational drug design led to inhibitors such as adagrasib and sotorasib which are used against cancers (e.g. colon and lung cancer) that possess the KRASG12C mutation. This particular mutation is present in approximately 13–15% of NSCLC cases, with a variety of different changes in the protein accounting for the other KRAS mutations. Unfortunately the tumors quite often develop resistance against these drugs.

The blog post described a New England Journal of Medicine (NEJM) study investigating resistance mutations that arise in KRAS G12C tumors during treatment with the drug adagrasib. The study found various resistance mechanisms in 38 patients (27 with NSCLC, 10 with CRC, 1 with appendiceal cancer), including:

• Acquired KRAS Alterations: Additional mutations in KRAS (e.g., G12D/R/V/W, G13D, Q61H) or amplification of the KRAS G12C allele.
• Bypass Mechanisms: Activation of parallel or downstream pathways through mutations in genes like NRAS, BRAF, MAP2K1, RET, or oncogenic fusions involving ALK, RET, BRAF, RAF1, FGFR3. Loss of function mutations in NF1 and PTEN were also seen.

The bypass resistance mutations hint at the possible existence of parallel pathways that can stimulate oncogenic signaling even when the KRAS pathway is blocked by the drug.

A new mini-review in NEJM discussed two new papers investigating the architecture of the KRAS oncogenic signaling pathways and the implications for cancer treatment. In particular, one paper explored the importance of the multiple pathways downstream of KRAS (Figure 1A) by focusing on the RAF-MEK-ERK pathway (Figure 1B). The latter is a crucial signaling cascade within cells that plays a vital role in regulating a wide range of cellular processes including:

• Cell growth and proliferation: It promotes the division and multiplication of cells.
• Cell survival: It helps cells resist programmed cell death (apoptosis).
• Cell migration: It influences the movement of cells.

All of the above are necessary for tumor development, and so not surprisingly hyperactivation of RAF-MEK-ERK can contribute to cancer.

Typically, the RAF-MEK-ERK pathway is activated by external signals, such as growth factors, binding to receptor tyrosine kinases (RTKs) on the cell surface. These tyrosine kinases phosphorylate adaptor proteins that recruit KRAS activating proteins to the plasma membrane which turn on KRAS. Active KRAS activates a family of serine/threonine kinases called RAF (RAF1, BRAF, ARAF). Activated RAF then phosphorylates and activates another set of kinases called MEK1 and MEK2 . MEK1/2, in turn, phosphorylates and activates ERK1 and ERK2, also known as MAPK1 and MAPK3 (Figure 1B). Activated ERK translocates to the nucleus, where it phosphorylates various transcription factors that regulate the expression of genes involved in the cellular processes mentioned above. Oncogenic mutations in KRAS obviate the need for a growth factor input, and hyperactivate this kinase cascade. 

One big question is whether KRAS signals in a promiscuous fashion to multiple downstream pathways (Figure 1A) or selectively activates one primary pathway (i.e. RAF-MEK-ERK, Figure 1B) during cancer progression. A paper by Jeffrey Klomp and colleagues indicate that mutant KRAS primarily signals through the RAF-MEK-ERK pathway, challenging the previous model of multiple effector pathways.

More specifically, these researchers found that the transcriptional output of oncogenic KRAS in pancreatic cancer models was almost entirely recapitulated by targeting the RAF pathway. Constitutively-activated ERK, a downstream component of this pathway, reversed the anti-proliferative effects of KRAS inhibitors, while activated AKT (downstream of PI3K in a different KRAS-activated pathway) had minimal impact.

In other words, scientists took advantage of epistasis experiments in which two different perturbations were combined  to elucidate the pathway ordering and architecture. For example, if the RAF-MEK-ERK pathway is dominant, then the oncogenic impact of activated KRAS should be substantially attenuated by the complete inhibition of ERK which is downstream of KRAS.

However, there was also evidence for compensatory pathways. KRAS-RAF-MEK-ERK inhibition led to increased RHO signaling and the upregulation of other kinases which partially circumvented the inhibition via signaling along parallel pathways.

The reality is that clinical trials of KRAS inhibitors have shown only modest benefits so far, and the development of drug resistance mutations are a particular problem that reduces the efficacy of the drug treatments. The work described above suggests a strategy of multinodal inhibition focusing on the RAF-MEK-ERK pathway perhaps targeting more than one protein in the pathway to ensure durable inhibition. Alternatively, the existence of parallel pathways would argue that there are limits to this more focused approach, and that instead we should target multiple effector pathways immediately downstream of KRAS (Figure 1A).

An additional caveat is the possibility of parallel pathways not including KRAS (e.g. RHO). Another concern is that the study used a cell line which may be different from actual pancreatic cancers in the body. Regardless, scientists continue to make progress and u understand the molecular mechanisms that underlie various aspects of tumorgenesis, and this understanding will aid in the development of new more effective cancer treatments.

Figure 1. Promiscuous versus selective signaling immediately downstream of KRAS in tumorigenesis. Panel A shows a model in which KRAS activates a number of downstream effector pathways in a promiscuous fashion that together produce the physiological effects of cell proliferation and survival necessary for cancer development. Panel B shows an alternative model in which KRAS selectively exerts its primary effect via signaling through RAF, MEK, and ERK. The ERK kinase then is able to turn on a large number of genes which together are essential for tumor progression. In the selective model (B), inhibiting ERK is expected to have a significant effect on the tumor, whereas less so in the promiscuous model (A).