Testing for Recurrence

Early-stage colorectal cancer patients may benefit from a new test that predicts their risk of recurrence.

Stage 2 colon cancer has a 20 percent risk of recurrence, although that risk varies based on the genetic makeup of the tumor. Unfortunately, until recently, there was no way to determine which patients had a high or low risk of recurrence, which meant most patients who had chemotherapy to prevent a recurrence weren’t likely to benefit from it.

“So a lot of effort has been made over the last couple of years to develop genetic testing to [identify] the patient who is in the category of higher or lower risk [for recurrence],” says Heinz-Josef Lenz, MD, co-leader of the Gastrointestinal Cancers Programs at the USC Norris Comprehensive Cancer Center. Now people with stage 2 colon cancer—in which the tumor has penetrated the bowel wall, but the disease has not spread to surrounding lymph nodes—can undergo a genetic test called Oncotype DX Colon Cancer Assay to help gauge their risk for recurrence. The test may also help physicians make more precise risk assessments when planning post-surgery treatment for these patients.

By using the tumor tissue collected at the time of surgery, the Oncotype DX test can determine if the gene expression profile is associated with an increased risk of cancer recurrence. The gene expression profile can be used to predict how cancer cells will behave. Developed by Genomic Health, Inc., Oncotype DX for colon cancer became commercially available in January.

Oncotype DX first made headlines around 2004 when the test showed it could predict recurrence in patients with early-stage, estrogen receptor-positive breast cancer. “Oncotype DX for colon cancer is similar to Oncotype DX for breast cancer; they use the tumor tissue and identify the significant genetic signature for a number of genes,” Lenz says. Whereas the breast cancer test looks at 21 genes (16 cancer-related genes and five reference genes), the colon cancer test looks at only 12 genes (seven cancer-related and five reference genes).

The Oncotype DX Colon Cancer Assay is based on the work of several studies. Over the past 30 years, more than 1,800 patients with stage 2 and stage 3 colon cancer have been involved in four clinical trials, and the tissue gathered from these trials were later used to find genes that would predict the risk of cancer recurrence. The studies helped investigators create thresholds of recurrence risk into low, intermediate, and high groups. Patients are then assigned a recurrence score, which range from 0 to 100 based on the presence of those genes.

Data from those studies were further examined in another large clinical trial called QUASAR, which included 1,436 patients with stage 2 colon cancer. This trial proved Oncotype DX Colon Cancer Assay was useful for categorizing patients with stage 2 colon cancer into three groups: low-risk, intermediate-risk, and high-risk groups. Three years after surgery, the groups showed a recurrence rate of 12 percent, 18 percent, and 22 percent respectively.

“Here is the problem,” says Lenz. “The spectrum, from 12 percent to 22 percent, is still pretty low. That is not the same for breast cancer, where the low risk is 7 percent and the high risk is 30 percent. For clinical use, we expect a better distinction between high and low risk.”

Although genetic profiles developed in the first four studies were effective in both stage 2 and stage 3 colon cancer patients, Oncotype DX Colon Cancer Assay has not yet been tested in a separate set of patients with stage 3 colon cancer. Genomic Health is currently conducting studies to see if the recurrence score is useful in stage 3 patients.

Another distinction between the breast cancer test and colon cancer test is that the colon cancer assay does not appear to directly predict whether patients will benefit from chemotherapy. One goal of the studies was to establish a treatment score to indicate the benefit of 5-FU and leucovorin chemotherapy by analyzing a separate set of genes. Unfortunately, the treatment score did not prove clinically significant in determining whether patients would benefit. However, when considered along with other clinical and pathological factors, the test may still be useful to patients and doctors in addressing that dilemma.

Researchers and commercial entities are currently working to develop different gene profiles that will help identify which patients will benefit from specific chemotherapy treatment. While chemotherapy is not standard treatment for most patients with stage 2 colon cancer, it is indicated by guidelines issued by the American Society of Clinical Oncology as an option for patients at high risk for recurrence based on factors other than just the stage.

“I suspect that within the next year or so there will be predictive tests available,” says Michael J. O’Connell, MD, associate chairman of the National Surgical Adjuvant Breast and Bowel Project (NSABP), a clinical collaborator for the test.

O’Connell considers the results of the Oncotype DX assay, in regards to recurrence risk, valid and accurate. But he takes other pathologic features of a patient’s disease into consideration along with the recurrence score to further determine which patients are at higher risk for recurrence, and who may be more willing to undergo chemotherapy.

“If…your chance of recurrence is only 5 percent, that means there is a 95 percent probability that you’re already cured, O’Connell says. Knowing that chemotherapy decreases the recurrence rate by only 25 to 30 percent of the overall 5 percent benefit—that translates to about a 1 percent benefit to that patient. Most of these patients and their physicians would deem the possible benefit too small to be worth undergoing chemotherapy, he explains.

“On the other hand, if the risk of recurrence is 30 percent or higher, the proportional benefit from chemotherapy would be a 6 percent improvement in outcome. Most surveys suggest that degree of benefit would be worth it for both the patient and the physician if the patient was otherwise in good condition and a good candidate to receive chemotherapy,” O’Connell says.