Article Text


A rat virus visits the clinic: translating basic discoveries into clinical medicine in the 21st century
  1. C R Boland,
  2. A Goel
  1. Department of Medicine and Comprehensive Cancer Center, University of California San Diego, La Jolla, CA, USA
  1. Correspondence to:
    Dr C R Boland, 4028 Basic Science Building, University of California San Diego, 9500 Gilman Drive, San Diego, CA 92093-0688, USA;

Statistics from

Mutant forms of the KRAS2 gene are present in the serum of patients who have undergone putatively curative surgery for colorectal cancer and may be used to guide novel therapies in the future by identifying those individuals at greatest risk of recurrence

One of the holy grails of biomedical research is to identify markers of occult disease that might lead to early treatment of that disease before the manifestations are overt—and ipso facto incurable. In this issue of Gut, Ryan and colleagues1 from Dublin report that one more application of basic science discovery might be ready for use in the management of patients with colorectal cancer [see page 101]. These investigators have found that mutant forms of the KRAS2 gene are present in the serum of patients who have undergone putatively curative surgery for colorectal cancer. This information can predict tumour recurrences and, by inference, might be used to guide novel therapies in the future by identifying those individuals at greatest risk of recurrence. We have all waited patiently for the fruits of the molecular biology revolution that began nearly two decades ago, and it may be worthwhile to assess the tempo by which these discoveries make their way from the bench to the clinic.

The RAS gene family encodes for a series of at least 50 guanosine triphosphatases (GTPases) which are small proteins involved in the regulation of growth and other biological activities.2 Their link to cancer required a circuitous route through what might have seemed at the time to be obscure basic research. In the 1960s and 1970s, several viruses had been identified that cause sarcomas in rodents and birds. By the late 1970s, it was found that some of these viruses carried single copies of altered genes—viral oncogenes—that were responsible for the transformation process.3,4 By 1980, two members of the RAS family were identified as oncogenes in rats; in fact, the family name for these genes was derived from rat sarcoma.5 The genes had been hijacked from the mammalian genome by the Harvey (HRAS) and Kirsten (KRAS) transforming viruses but in mutant forms that abrogated the GTPase activity of the protein, and converted them from regulators of signal transduction to oncogenic proteins. Finding mutant RAS genes in oncogenic viruses led to a search for mutant forms in the cellular DNA from human tumours. By 1987, two laboratories reported that approximately half of all colon cancers had mutant copies of KRAS.6,7 The technical challenges involved were not trivial, and in one of these papers was a very early application of the polymerase chain reaction (PCR) to solve a problem that was limited by the modest amount of DNA that could be retrieved from a surgically excised tumour sample.6

By 1992, KRAS mutations had been found in the stools of patients with colorectal neoplasia,8 and by 1996, mutant forms of KRAS were reported from the plasma or serum of some cancer patients.9 It had been known for several decades that circulating tumour cells could be seen in the blood of patients undergoing colorectal cancer surgery; interestingly, this did not predict tumour recurrence.10 Before long, groups reported that mutated forms of the tumour suppressor genes APC and p53 were also found in the blood of colorectal cancer patients.11 The attractiveness of KRAS mutations is that the hunt for mutations can be limited, and is clinically more feasible.

What have Ryan et al added to this several decade long saga that stretches from mouse sarcomas to human colon cancers? Theirs is a prospective study in which KRAS mutations were used to predict tumour recurrence. Firstly, even control subjects have wild-type (that is, normal) copies of KRAS in serum. The investigators have focused upon mutant forms of the gene in cancer patients, which is detected by strategic designing of the PCR primers. They demonstrated that 53% of the group of 78 patients who could be studied preoperatively had KRAS mutations in their tumour sample, and that in 76% of these patients a matching mutation was found in serum (which was 41% of all preoperative sera). In the larger group of 94 patients in which there was a putatively curative resection, 64% had KRAS mutations in the tumour. The serum became persistently KRAS mutation positive in the postoperative period in 27% of these 60 patients. The novel finding was that 63% of the mutation positive group developed recurrent cancer whereas only 2% of those whose blood remained negative did so. There was no predictive value for those tumours that did not have a KRAS mutation. One can only speculate on what is occurring in patients with these mutations in the postoperative blood, but do not develop tumour recurrences.

What is the clinical significance of this study? This is a suitably large study, with a prospective design. Persistently positive mutant KRAS in the blood was valuable in predicting tumour recurrence although, as the authors note, not substantially better than using carcinoembryonic antigen. However, a gene based approach has the potential to be made more sensitive. Time will tell. KRAS mutations were also found in the blood of patients with large or villous adenomas but the clinical utility of a blood test for benign lesions is not likely to be high. Moreover, there is no avoiding the issue that about half of colorectal cancers do not harbour KRAS mutations, and the blood test is of no value in these cases. The principal value implied here is that one may be able to detect recurrences earlier, and more sensitively than with other diagnostic modalities, but that remains to be proved.

Perhaps the most illuminating issue is how long it takes for basic science discoveries to be translated to clinical medicine. The use of KRAS for colorectal cancer diagnostics was limited perhaps by those who were discouraged by the fact that only half of colon cancers carry these mutations. As we understand more about the complexity of tumour development, we may discover that a battery of markers will be required to manage our patients, and we may eventually not only accept this limitation but embrace the fact that the mutant KRAS approach may be useful fully for half of our patients at risk of tumour recurrence. These approaches will be more valuable when we finally develop more effective means for treating early signs of recurrence. Optimists will gamely await progress on both fronts.

Mutant forms of the KRAS2 gene are present in the serum of patients who have undergone putatively curative surgery for colorectal cancer and may be used to guide novel therapies in the future by identifying those individuals at greatest risk of recurrence


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles