Date Published: December 18, 2009
Publisher: Public Library of Science
Author(s): Paloma Cejas, Miriam López-Gómez, Cristina Aguayo, Rosario Madero, Javier de Castro Carpeño, Cristóbal Belda-Iniesta, Jorge Barriuso, Víctor Moreno García, Javier Larrauri, Rocío López, Enrique Casado, Manuel Gonzalez-Barón, Jaime Feliu, Alfons Navarro. http://doi.org/10.1371/journal.pone.0008199
Abstract: KRAS mutations in colorectal cancer primary tumors predict resistance to anti-Epidermal Growth Factor Receptor (EGFR) monoclonal antibody therapy in patients with metastatic colorectal cancer, and thus represent a true indicator of EGFR pathway activation status.
Partial Text: Colorectal cancer (CRC) is one of the most common malignancies and one of the leading causes of cancer-related death in developed countries . Distant metastasis is the main cause of death in CRC patients. Depending on the stage of the primary tumor, liver metastases occur in 20% to 70% of patients, and lung metastases in 10% to 20% of patients . Surgical resection remains the only potentially curative option for patients with metastatic CRC. However, curative resection is possible in less than 25% of patients with stage IV disease , and less than 5% of patients with unresectable metastatic CRC are alive after 5 years. Major efforts are being made to improve the prognosis for patients with metastatic CRC, especially in the development of new therapeutic strategies. The Epidermal Growth Factor Receptor (EGFR) signalling pathway has become a key target for therapeutic intervention because two monoclonal antibodies directed against EGFR have become important tools in the management of advanced disease: cetuximab and panitumumab , . EGFR activates proliferative and antiapoptotic signalling pathways, such as the phosphatidylinositol 3′ kinase/Akt and Ras/Raf/mitogen-activated protein kinase (MAPK) pathways . Aberrant activation of the EGFR pathway in CRC could be caused by either EGFR overexpression or mutational activation of downstream elements of the EGFR pathway .
The prognosis for patients with metastatic CRC has improved since the introduction of novel therapeutic agents such as anti-EGFR antibodies . This therapeutic success highlights the importance of counteracting the EGFR pathway to control advanced disease. Aberrant activation of the EGFR pathway sometimes occurs by mutational activation of KRAS. Recent clinical trials have shown that the presence of activating mutations in KRAS identifies patients who are non-responders to cetuximab  or panitumumab . In fact, based on these results, mutational analysis of KRAS is now recommended by The US Food and Drug Administration (FDA) prior to cetuximab or panitumumab administration to patients with metastatic CRC. Mutational analysis of KRAS is commonly performed on the primary tumor because it is often the only available tissue. Moreover, the value of performing this analysis on the primary tumor is further supported by evidence that KRAS point mutations occur early in CRC carcinogenesis . However, recent data have demonstrated an increased frequency of mutations in lymph node metastases as compared with their related primary tumors . With this in mind, the potential need for rebiopsy and analysis of KRAS in the metastases has been suggested . Other recent studies have reported a high degree of concordance in KRAS mutational status between primary tumors and their related liver metastases , , . The results from our series of 110 patients are in complete agreement with these reports, both in the frequency of KRAS mutations in primary tumors, which was 34% in our study, and in the high degree of concordance between primary tumors and their related metastases. All of these findings confirm that analysis of KRAS mutational status in the primary tumor is an adequate surrogate marker of KRAS mutational status in metastases. Our study includes an analysis of both liver and lung metastases. As far as we know, this is the first study to analyse primary colorectal tumors and their related lung metastases and correlated KRAS mutational status based on both clinicopathological features and survival data. The liver and the lung are both common sites of CRC metastases. Secondary to their respective anatomical blood vessel distribution, lung metastases are more common with rectal cancers and liver metastases are more frequent with colon cancers. However, some colon cancer patients experience lung metastases without evidence of previous liver metastases. This unexplained anatomical pattern of metastasis, observed with increasing frequency due to more accurate diagnoses based on highly efficient CT scans , suggests a peculiar biological mechanism of carcinogenesis or a special susceptibility of the lung parenchyma to tumors in these patients. Our results demonstrate a higher percentage of KRAS mutations in primary tumors of patients with lung metastases as compared with those with liver metastases (59% vs 32%). Moreover, when we stratified our results based on the primary tumor site, only tumors of colonic origin had a significantly higher frequency of KRAS mutations in primary tumors of patients with lung metastases. By contrast, rectal primary tumors showed a similar frequency of KRAS mutations in patients with lung metastases than in those with liver metastases. To further analyze this finding we studied an independent series of primary tumors from patients with unresectable metastatic CRC and examined the relationship between KRAS mutational status in the primary tumor and the site of tumor metastasis. In these patients, KRAS mutational status in the primary tumor was previously analysed for clinical purposes prior to cetuximab administration. Statistical analysis of the whole series (230 patients) revealed that KRAS is more frequently mutated in the primary tumors of patients with lung metastases compared with patients with liver metastases (57% vs 35%, P = 0.006). When the results are stratified according to the site of the primary tumor, the analysis we performed including all 230 patients confirmed the results we obtained with the smaller sample of 110 patients: that the significantly increased frequency of KRAS mutations in primary tumors in patients with lung metastases was restricted to those primary tumors that originated in the colon. This finding suggests that EGFR pathway activation may allow colonic tumor cells to nest preferentially in the lung parenchyma avoiding an initial step of liver metastasis. A previous study that evaluated KRAS status in primary colorectal tumors and non-matched liver and lung metastases showed concordant results with our study and revealed a higher incidence of KRAS mutations in lung metastases than in liver metastases (57% vs 50%) . However, in that study, a major difference in KRAS activation between the primary and metastatic tumors could have been masked by the absence of related primary and metastatic samples. This finding has clinical importance because it may allow the identification of patients who are more likely to develop lung metastases based on KRAS analysis of the primary tumor. These patients should potentially receive a more thorough clinical workup, including a thorax scan, which is not always included in the standard clinical workup for CRC.