A Personalised Medicine Resource

Bowel cancer is the second biggest cancer killer in Australia. With the arrival of more chemotherapy options and with the availability of biologic therapies in the recent past, mortality rates are declining, and patients are living longer. Over the past 20 years, survival in metastatic bowel cancer has more than doubled.
The main histologic subtype of bowel cancer is adenocarcinoma. Bowel adenocarcinomas arise through the acquisition of a series of mutations that occur over the space of many years, and results in the evolution of normal epithelial tissue to adenoma to carcinoma to metastasis. In the past two decades, there has been increasing recognition that some somatic mutations may be prognostic or predictive markers for specific therapies available in bowel cancer. These mutations involve genes such as KR​AS, BRAF, PIK3CA, AKT1, SMAD4, PTEN, NRAS, and TGFBR2.

Approximately 36–40% of patients with bowel cancer have tumor-associated KRAS mutations. The concordance between primary tumor and metastases is high, with only 3–7% of the tumors discordant. The majority of the mutations occur at codons 12, 13, and 61 of the KRAS gene. The result of these mutations is constitutive activation of KRAS signaling pathways.

Multiple studies have now shown that patients with tumors harboring mutations in KRAS are unlikely to benefit from anti-EGFR antibody therapy, either as monotherapy or in combination with chemotherapy. Further, in trials of oxaliplatin based chemotherapy, the patients with KRAS mutated tumors appeared to do worse when treated with EGFR antibody therapy combined with an oxaliplatin based chemotherapy compared to the patients treated with an oxaliplatin based treatment alone.


Approximately 8–15% of bowel cancer (CRC) tumors harbor BRAF mutations. The presence of BRAF mutation is significantly associated with right-sided bowel cancers and is associated with decreased overall survival. Several studies have reported that patients with metastatic CRC (mCRC) that harbor BRAF mutations do not respond to anti-EGFR antibody agents cetuximab or panitumumab in the chemotherapy-refractory setting. Based on these findings, BRAF mutations were suggested to be a negative predictor of response to anti-EGFR therapy.

The most frequently reported BRAF mutation is an activating missense mutation in which the amino acid glutamic acid is substituted for valine at amino acid position 600. This mutation is also associated with unresponsiveness to anti-EGFR therapy in wild type KRAS patients with mCRC.

While BRAF V600-mutated melanomas are sensitive to vemurafenib, BRAF V600-mutated CRCs may not be as sensitive. Activation of EGFR in bowel cancer could explain why bowel cancers generally have a lower response to BRAF inhibitors.​


Somatic mutations in AKT1 have been found in <1–6% of all bowel cancer. In bowel cancer, the only AKT1 mutation observed up to this time is the E17K mutation, which has also been observed in other types of cancer. This mutation in the Pleckstrin homology domain alters the ligand binding site and leads to constitutive kinase activity. Preclinical data have shown that the presence of this activating mutation results in cellular transformation in vitro and in vivo. Specific clinical characteristics of colorectal cancer patients harboring AKT1 mutations have yet to be described. AKT1 mutations and PTEN mutations appear to be mutually exclusive. Likewise, AKT1 mutations and PI3K mutations appear to be mutually exclusive.​


Somatic mutations in PIK3CA have been found in 10–30% of bowel cancers.

These mutations usually occur within two "hotspot" areas within exon 9 (the helical domain) and exon 20 (the kinase domain).


PTEN mutations occur in 5–14% of colorectal cancers. PTEN is a tumor suppressor gene, and loss of PTEN results in upregulation of the PI3K/ AKT pathway. PTEN loss of expression is observed with KRAS, BRAF, and PIK3CA mutations.

Germline mutations in PTEN result in PTEN hamartoma tumor syndrome and are associated with increased risk for melanoma, breast, thyroid, endometrial, colorectal, and kidney cancer report that somatic PTEN mutations are weakly prognostic in stage 4 CRC patients. Research into the prognostic and predictive significance of PTEN mutations and other mechanisms for loss of PTEN expression is ongoing.

The role of PTEN loss in response to PI3K and mTOR inhibitors is being explored in clinical trials. Preclinical studies have shown that p110α-specific PI3K inhibitors may be needed for treatment of PTEN-deficient cancers and that pan-PI3K inhibitors such as PX-866 may be effective in PIK3CA-mutated cancers and cancers showing PTEN loss. Likewise, in vitro studies have shown that inactivating mutations in the PTEN gene may confer sensitivity to PI3K-mTOR inhibitors as well as FRAP/mTOR inhibitors.

In retrospective studies, PTEN loss is associated with decreased sensitivity of colorectal cancer tumors to anti-EGFR antibodies. PTEN loss is associated with lack of benefit of the anti-EGFR antibody, cetuximab.


SMAD4 mutations are found in ~10–35% of bowel cancer (CRC) tumors.

In CRC, loss of SMAD4 has been historically thought to be a late event in tumor development with rates of SMAD4 loss of 0%, 8%, 6%, and 22% in stages I-IV CRC, respectively. However, downregulation of SMAD4 is associated with worse survival in stages I–II bowel cancer patients. Loss of SMAD4 protein expression evaluated by immunohistochemistry in stage III (lymph node positive disease) is associated with worse overall and disease-free survival. Low SMAD4 expression may also identify a subset of patients with early recurrence after curative therapy.

SMAD4 mutations or loss of expression may portend a worse prognosis in patients with resected colorectal cancer. In a retrospective analysis of molecular markers in tumors from patients treated on two adjuvant trials (E2284 and INT 0035) for resected stage III or high risk stage II bowel cancer, patients with tumors containing 18q LOH (SMAD4 resides on 18q) had significantly worse disease-free survival than patients with tumors retaining both 18q alleles. However, a retrospective analysis of 2 other adjuvant bowel trials (CALGB 9581 and CALGB 89803) for stage II and III resected bowel cancer did not show that 18q LOH was prognostic for survival. E5202 was designed to prospectively investigate the impact of several molecular markers, including loss of heterozygosity (LOH) of chromosome 18q on recurrence in stage II CRC (NCT00217737). This trial was terminated early, and results from the trial have not yet been reported.​​

About Bowel Cancer-My-Cancer-Genome 770
Knowledge of the genetic mutations associated with different cancer types is increasing rapidly and is already influencing treatment choices.
The My Cancer Genome website is a database of information on genes, genetic mutations and cancer.  It includes information on the latest research and the latest treatments across many cancer types (including bowel cancer).
The website also highlights the benefits of genetic testing especially when knowledge of a particular gene mutation can affect treatment decisions.
For example, RAS testing in bowel cancer patients can help decide whether they are likely to respond to specific targeted therapies such as Erbitux (cetuximab) or Vectibix (panitumumab).
Mutations in other genes such as BRAF, PIK3CA, AKT1, SMAD4, PTEN, NRAS, and TGFBR2F are also relevant for bowel cancer patients and their families.
For further details please visit the My Cancer Genome website.