ATLANTA -- Duke University Medical Center researchers report they are one step closer to finding biochemical guide markers that might help improve treatment for patients with esophageal cancer, one of the deadliest of cancers. The scientists discovered that certain tell-tale proteins seem to signal the success or failure of chemotherapy, as well as resistance to radiation treatment, in esophageal cancer patients. The finding of such markers could lead to more effective targeting of such aggressive therapies, which in general have not proved successful in clinical trials, the researchers said.
The scientists prepared their findings for presentation Saturday at the annual meeting of the American Society for Clinical Oncology. The researchers evaluated the impact of seven different genes and their protein products on the effectiveness of chemotherapy and radiation treatment for esophageal cancer. Three of the genetic markers were linked to significant differences in treatment response and survival.
One of the proteins can identify patients whose tumors will reject chemotherapy even before it has a chance to work, while a different marker indicates a patient might do well with the therapy. The third marker signifies that resistance to radiation treatment is likely.
Information allowing treatment response to be predicted in esophageal cancer would be highly valued by physicians because few patients respond to chemotherapy or radiation. Five-year survival is less than 10 percent, and about 13,000 patients are diagnosed with esophageal cancer in the United States each year. Right now, doctors have no way of knowing who will or won=t be helped by more aggressive therapies.
"We want to find and treat only those patients who need extra therapy, and only those patients who will respond to it," said Harpole, associate professor of surgery in the Thoracic Oncology Program at Duke Comprehensive Cancer Center. The project was funded by a grant from the Duke Thoracic Oncology Program.
While physicians have traditionally used only surgery to treat the cancer, clinical trials are under way using chemotherapy and radiation therapy before surgery. However, the trials have demonstrated no benefit to esophageal cancer patients from this so-called "tri-modality" therapy.
The Duke researchers, however, suspect that the patient group studied was too undifferentiated. By studying patients= tumors on a molecular level, they are seeking to identify a subset of patients who are more likely to benefit from the aggressive tri-modality therapy.
Most tumor cells have different genes and proteins activated compared to normal cells. In some cases tumors express genes that normal cells don=t, and in others the tumors lack genes normally expressed. A third option is for a gene to be altered, or mutated, in a way that scientists can detect. All of these changes are reflected in the presence, lack, or mutation of the genes= product proteins.
By examining the pattern of these genetic alterations, Harpole said he hopes to provide a more detailed picture of tumors to eventually predict patients= responses to treatment. The ultimate goal is to tailor therapy and improve survival.
"We want to know if we can predict from the initial biopsy sample which patients will respond to therapy and survive," he said.
In their study of 112 patients, the research team analyzed tumor samples for genes and their protein products that studies of other cancers, such as colon cancer, indicated had roles in either resistance or sensitivity to chemotherapy or radiation.
After biopsy, the patients underwent chemotherapy and radiation therapy, followed by surgery to remove any remaining tumor tissue. The 34 patients who responded completely to the chemotherapy and radiation and had no viable tumor left at surgery had an average survival longer than five years.
In analyzing the relationship between markers and patients= response to therapy, Harpole found the three protein markers whose presence or absence significantly impacted patient survival.
The presence of one such marker, P-glycoprotein (Pgp), signaled that the patients would not respond well to chemotherapy. Pgp is primarily responsible for inducing multi-drug resistance, in which the tumors become resistant to many chemotherapy drugs after treatment with just one drug. Pgp effectively pumps the drug out of tumor cells before it has time to kill the cells. Pgp is just one protein implicated in multi-drug resistance. Harpole found that the 88 patients with Pgp survived 20.9 months on average, while the 24 patients without Pgp had an average survival of more than 5 years after diagnosis.
Laboratories across the country are looking into ways to stop Pgp or to circumvent its action. One way is to give patients Pgp modulators -- molecules that block Pgp=s drug pump. Another is to develop drugs that Pgp and related proteins can=t remove from the cell.
A second significant marker was glutathione-s-transferase-pi, or GST-pi, which confers radiation resistance. Radiation kills cells by forming oxygen radicals -- highly reactive oxygen atoms that damage DNA. GST-pi reduces these oxygen atoms so that they are no longer reactive. The 84 patients with GST-pi survived an average of 21 months, while those without it survived 47 months on average.
The third implicated marker was HER-2/neu, which initiates chemical signals telling the cell to multiply. Herceptin, a drug recently approved by the FDA to treat breast cancer patients who have excess HER-2/neu, or other agents might eventually be useful in esophageal cancer, Harpole said. Average survival for the 94 HER-2/neu positive patients was 26.8 months, while that for HER-2/neu negative patients was 14.5 months.
The researchers are continuing to study molecular markers, both in esophageal cancer and in lung cancer -- two diseases with survival rates relatively constant over the last 40 years.
A study evaluating 10 markers in lung cancer patients was published by Harpole in the April issue of Annals of Thoracic Surgery. Five markers were significantly related to survival. A prospective study in lung cancer is under way to verify those results.
The researchers also plan to study the important esophageal cancer markers= effects prospectively -- to see if predictions based upon their new models correspond with reality. This national esophageal trial is a multi-center randomized study in which 250 patients will receive the tri-modality therapy and 250 will receive surgery alone, which is the standard treatment. The markers will be linked again to response to treatment and survival in both groups.