MORE GENETIC TESTS FOR STAGE II COLORECTAL CANCER

January 20, 2011 — Two more genetic test to predict prognosis after surgery for stage II colon cancer are in the late stages of development: ColoPrint, developed by Agendia; and OncoDefender-CRC, developed by Everist Genomics. Both are scheduled for launch this year, and will be joining a genetic test already marketed for this use — Oncotype DX for colon cancer from Genomic Health, which was launched last year.

The aim of the new tests is to identify colorectal cancer patients who are at high risk for relapse after surgery. These patients might benefit from chemotherapy, whereas patients who are not at high risk for relapse could be spared chemotherapy and its attendant adverse effects. Currently, clinical parameters, such as tumor stage and lymph node involvement, are used for to guide these decisions.

Patients with stage II colorectal cancer have a relatively good prognosis. "About 80% of these patients do well with surgery alone," said Jennifer Obel, MD, from NorthShore University Health System in Evanston, Illinois. "As treating oncologists, we really don't want to give chemotherapy to all of these patients to benefit a few, and expose them all to unnecessary side effects."

Dr. Obel was moderating a presscast organized by the American Society of Clinical Oncology (ASCO) to highlight new research that will be presented at the 2011 Gastrointestinal Cancers Symposium in San Francisco, California. One of the studies highlighted was a validation study for ColoPrint; the other was a study using the OncoDefender-CRC assay.

ColoPrint Study

The data on ColoPrint come from a second independent validation study, which is unique; no other test has undergone this prospectively, said lead investigator Robert Rosenberg, MD, PhD, assistant professor at the University Hospital of the Technical University in Munich, Germany.

The study was conducted in Munich, and involved 135 stage II colon cancer patients. ColoPrint identified 73% of these patients as being low risk, and just 5% experienced a cancer recurrence in 5 years or more, Dr. Rosenberg noted. The remaining 27% of patients were identified as high risk, and 20% experienced a recurrence in the median follow-up of 97 months.

This was highly statistically significant, with a hazard ratio of 4.1, Dr. Rosenberg said. In contrast, using clinical parameters from the ASCO recommendation (e.g., tumor stage, lymph node involvement) was not significant, with a hazard ratio of 2.3. Adding these clinical parameters to the ColoPrint test did not improve the results, he said.

"In this validation study, the performance of ColoPrint seemed to be independent of known clinical factors," Dr. Rosenberg reported.

This is the second independent validation study with ColoPrint, he noted. The first was conducted in Barcelona, Spain (J Clin Oncol. 2010;28[Suppl 15];TPS199), and a third is underway at the University of Texas M.D. Anderson Cancer Center in Houston. In addition, a multinational phase 3 study, the Prospective Analysis of Risk Stratification Using ColoPrint, is underway and is expected to enroll 600 patients. First results from this study are expected this year, he said.

Medscape Medical News asked Dr. Rosenberg how ColoPrint compares with the test already on the market, Oncotype DX. He said that ColoPrint was developed after searching the whole genome for colon cancer relapse-causing genes, and uses 18 such genes. It also identifies patients as high risk or low risk. In contrast, Oncotype DX uses 12 genes and identifies patients as low, intermediate, or high risk.

There are also logistical differences, Dr. Obel said; ColoPrint uses fresh tissue samples, whereas Oncotype DX uses frozen tissue.

Both of these products have followed similar gene signature tests aimed at breast cancer — ColoPrint is developed by the same company that markets MammaPrint (which tests for 70 genes); there is also an Oncotype DX for breast cancer (which tests for 21 genes).

OncoDefender-CRC Study

The other new test in development, OncoDefender-CRC, tests for only 5 genes and is aimed at stage I and II colon caner patients and stage I anal cancer patients. This test is performed on formalin-fixed paraffin-embedded tissue.

Lisa Boardman, MD, from Mayo Clinic in Rochester, Minnesota, and colleagues tested OncoDefender-CRC on 115 patients with stage I or II colorectal cancer. Because it was used on archived tissue, the outcomes were already known. The results showed that the test correctly identified 32 of the 46 cases of recurrence and 38 of the 69 cases of nonrecurrence, for a sensitivity of 70% and a specificity of 55%.

High-risk patients had a significantly higher probability of recurrence within 36 months than low-risk patients (hazard ratio, 2.06; P = .02; positive predictive value, 0.51; negative predictive value, 0.73). In contrast, the National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology were unable to differentiate between these populations (P = .315), Dr. Boardman reported.

In an interview with Medscape Medical News, Peter Lenehan, MD, PhD, chief medical officer of Everist Genomics, said that this test differs from the other 2 in a number of ways. ColoPrint and Oncotype DX both have low positive predictive values, he said. Both are good at identifying patients who are at low risk for relapse and who therefore do not need chemotherapy, but not as good at identifying patients who are at high risk for relapse and who would benefit from chemotherapy and/or aggressive monitoring. In contrast, the OncoDefender-CRC assay is better at identifying patients who are at high risk, and for whom action needs to be taken, which is "clinically more useful for this stage of disease," he said.

In addition, this is the only test that has been investigated in stage I colon cancer patients, Dr. Lenehan explained. These patients have been "neglected" in the past, he said; they do not routinely have any other treatment after surgery. But about 10% to 15% of patients relapse and succumb to the disease, and "our test is the first to identify this subpopulation that may benefit aggressive follow-up and/or chemotherapy," he said.

Both studies were funded by the manufacturers of the genetics tests.

2011 Gastrointestinal Cancers Symposium (GICS). Abstract 437, to be presented January 20, 2011; abstract 358, to be presented January 22, 2011.