NEDLANDS, AustraliaIf the implications of an Australian study are validated in further prospective studies, microsatellite instability (MSI) could potentially predict those patients with Dukes C and possibly Dukes B colorectal cancer who will respond best to chemotherapy.
Patients with MSI appear to derive very good survival benefit from chemotherapy, Hany Elsaleh, MBBS, said at the recent annual meeting of the American Society of Clinical Oncology. Conversely, he reported that p53 mutation in the tumor predicts a very poor response to chemotherapy.
The study, conducted by Dr. Barry Iacopetta and Dr. Elsaleh at the Sir Charles Gairdner Hospital and the University of Western Australia, Nedlands, put several hypotheses on the table for consideration.
One was whether the response to chemotherapy could differ between phenotypes, since it is known that genetic alterations differ within colon carcinomas. Also, since the frequency of these genetic alterations seems to differ between the proximal and distal regions of the colon, could site of origin also determine treatment outcome?
The testing of genetic alterations may allow us to identify patients who benefit from therapy and those who dont, Dr. Elsaleh said.
Microsatellite instability is a disorder of DNA mismatch repair, Dr. Elsaleh explained. This results in changes in the length of repetitive nucleotide sequences scattered throughout the genome.
MSI is found both in hereditary nonpolyposis colorectal cancer and sporadic colorectal cancer, and is found mostly in the proximal colon, he said.
The aims of the study were to identify the prevalence of MSI and p53 mutations in a large consecutive series of single-staged, surgical-margin-negative Dukes C colorectal carcinomas, Dr. Elsaleh said. Improvement in cancer-specific survival was measured, as well as the potential influence of chemotherapy on the disease. Survival was the primary endpoint, and adjuvant chemotherapy status was defined on an intention to treat basis.
Dr. Elsaleh and his colleagues looked at 891 consecutive Dukes C colorectal cancer patients (median age, 67.6 years) over a 14-year period. The median follow-up was 6.5 years. Of the total, 30% (n = 272) received adjuvant chemotherapy consisting of 5-fluorouracil/levamisole. By the end of the study, 55% had died of colon cancer.
In the chemotherapy vs non-chemotherapy group, tumor site, histological grade, p53 mutation, and MSI were balanced. The only major differences between the two groups were age and gender. The average age of those receiving chemotherapy was 60 vs 71 for those not on chemotherapy.
Because of younger age at presentation, slightly more males received chemotherapy than females.
Looking at all the cases, a 10% survival benefit was seen in the chemotherapy group. When MSI and p53 were analyzed, it was apparent that there were differences in the response to chemotherapy depending on these factors. Patients whose tumors showed MSI had a much better response to chemotherapy, compared with those receiving chemotherapy whose tumors were MSI negative (96% survival vs 37% at 5 years).
Dr. Elsalehs view is that previous reports of prognostic value for MSI were primarily due to the predictive effect of MSI for good response to chemotherapy, as there was no survival benefit seen in having MSI in the non-chemotherapy-treated group.
Because MSI was seen primarily in the proximal colon (20% vs 1%), this prompted us to look at the effect of chemotherapy in the proximal group vs the distal group, Dr. Elsaleh said. What they found was that the proximal group had a 25% absolute survival benefit at 5 years with chemotherapy. When the cancer was confined to the distal colon, the chemotherapy resulted in only a marginal 1% benefit at 5 years.
Since MSI is most common in the proximal colon and p53 mutations occur more often in the distal area, Dr. Elsaleh and his colleagues concluded that perhaps the site where the cancer occurs would predict chemotherapy response.
Gender differences too were looked at in the study. Females were found to have a 20% absolute survival benefit from chemotherapy at the 10-year projections compared to males.
So there are potentially differences in the genetic alteration profile of tumors based not only on site but also on gender. This may also contribute to gender differences observed in the response to chemotherapy, Dr. Elsaleh said.
The data on the predictive value of the p53 mutation was also interesting, he said. It did not have prognostic significance when the treatment and nontreatment group were compared. However, as a predictive marker, Dr. Elsaleh said, it can help to identify patients who are not deriving benefit from chemotherapy.
Patients with normal p53 derived a survival benefit, but not those with mutant p53.
A multivariate analysis of the treatment group found that gender, p53, and MSI were independent predictors of survival benefit from chemotherapy.
The prospect of using tumor-related genetic alterations to identify differences in biological response to treatment is attractive, and if this defines distinct entities, then it will assist us to reclassify this disease. This will only be possible when the predictive values of genetic biomarkers are tested in prospective studies, Dr. Elsaleh concluded.