• Preserved tissue from the QUASAR study was able to validate  a multiple panel of 7 genes that show the risk of recurrence for stage II colon cancer.  However, it could not prove that another 6 genes could predict who would benefit from treatment with 5-FU and leucovorin over surgery alone.   C3 has previously covered this analysis in depth. Dr. David Kerr in discussing the results emphasized the importance of combining recurrence risk scores with information about tumor stage and MSI status in making decisions about chemotherapy for stage II patients.
• A review of the PETACC-3 randomized clinical trial that compared infusional 5-FU to infusional 5-FU plus irinotecan found that MSI (microsatellite instability) status predicted both relapse free survival at three and five years and overall survival for stage II and III colon cancer.  However, the benefit was much stronger in stage II than in stage III leading Dr. Sabine Tejpar and her team to conclude that there may be biological effects of MSI that happen specifically in stage II.  As a hypothesis, she considered whether MSI prevents or reduces the ability of cancer to move into nearby lymph nodes. In contrast to a previous study that looked at adding irinotecan to bolus 5-FU, the PETACC-3 study found no benefit from irinotecan in stage III colon cancer.
• Prognostic molecular markers were quite different between stage II and stage III colon cancer in an analysis of several different markers in tumor tissue from the PETACC-3 study. While MSI or microsatellite instability was a strong marker of good prognosis in stage II, it lost its value as a prognostic marker completely in stage III.  p53 and the SMAD4 genes had prognostic value for stage III but not for stage II.  A marker previously identified with poor prognosis in stage II disease — loss of heterozygosity at 18q or 18qLOH — lost its prognostic value completely when analysed together with MSI in stage II and had no value in stage III.  In his conclusion Dr. Arnaud Roth questioned whether stage II and III colon cancers are biological different diseases rather than steps in continuous cancer development.

In discussionof the three molecular marker oral presentations at ASCO, Dr. Charles Fuchs pointed out the difference between

• Predictive markers that tell whether a particular treatment will be beneficial for an individual patient or not.
• Prognostic markers that tell how likely cancer is to recur or patients to survive regardless of treatment.

He questioned whether or not the gene analysis from the QUASAR trial is ready to help make clinical decisions about whether or not to treat stage II patients with chemotherapy and called for additional study of benefit from chemo in each of the recurrence risk groups.