Biomarkers allow doctors to match therapy to patient

Apr 13 2008

Genetic variations ensure that no two people are exactly alike, nor are their cancers.

Researchers now have the tools and the knowledge to predict how individuals will respond to cancer therapy, enabling more precise and effective treatment. At the 2008 Annual Meeting of the American Association for Cancer Research, April 12 – 16, researchers present data on new biomarkers that can predict response to well known treatments for breast cancer, pediatric neuroblastoma, and kidney and non-small cell lung cancer.

CD34bright/CD133neg candidate circulating Endothelial Progenitor Cells (ccEPCs) are a potential biomarker during treatment with sunitinib or bevacizumab: Abstract 4956

Researchers have identified two potential biomarkers that could help doctors monitor the effectiveness of treatment with sunitinib or bevacizumab for kidney and non-small cell lung cancer.

“Our work provides novel data on a potential biomarker for the monitoring of anti-angiogenic drug activity in cancer patients, as well as identifies a cell type that is a potential target for these agents,” said Laura Vroling, M.Sc., a researcher in the Department of Medical Oncology, VU University Medical Center, Amsterdam, The Netherlands.

The vascular endothelial growth factor (VEGF) receptor targeted agents bevacizumab and sunitinib have proven effective against several cancers, such as non-small cell lung cancer, colorectal and kidney cancer, but it is unclear which subset of patients will benefit most from these agents, researchers say. “Therefore, it is of great importance to identify and validate biomarkers for early response or duration of response,” Vroling said.

Vroling and colleagues studied therapy-induced changes in a novel, rare, circulating cell population. They measured these candidate circulating endothelial progenitor cells (ccEPCs) characterized by the markers CD45neg, CD34bright and CD133neg during sunitinib or bevacizumab treatment.

They labeled them “candidate” cells because no data have proven definitively the phenotypic relationship between progenitor and blood-derived endothelial outgrowth cells, Vroling said.

Their study included 23 patients with renal cell cancer and 19 patients with non-small cell lung cancer. The researchers also monitored plasma levels of VEGF. They assessed tumor response with computed tomography scans according to the RECIST criteria.

“This is the first study to assess the kinetics of ccEPCs together with other circulating cells in the peripheral blood of patients with renal cell cancer during the first cycle of sunitinib treatment,” Vroling said.

During a four-week “on” period of treatment with sunitinib, the ccEPC increases paralleled the rise in plasma VEGF levels; they decreased during the two-week “off” period, Vroling reports. Monocytes and hematopoietic progenitor cells (HPCs) demonstrated the opposite pattern, according to Vroling.

“In a preliminary analysis, we found a significant difference in the change of ccEPC numbers and VEGF levels after two weeks of treatment between patients with clinical benefit and progressive disease,” Vroling said. “We also noted that an increase of ccEPCs was indicative of a longer progression-free survival in this small group of patients.”

In the patients with non-small cell lung cancer treated with bevacizumab and erlotinib, ccEPC levels increased, while free plasma VEGF levels decreased. ccEPCs did not rise in a control group treated with erlotinib alone, Vroling said.

“These data suggest that ccEPCs are increased in cancer patients in an anti-angiogenic, treatment-specific way,” she said. Furthermore, this effect does not seem to be related to plasma VEGF levels.

“In our study for the first time the behavior of two CD34bright cell populations, (CD45neg) candidate cEPCs and (CD45dim) HPCs were monitored and showed a different response of both cell populations during sunitinib or bevacizumab therapy. The role of ccEPCs in human tumor angiogenesis and their potential in prediction of treatment outcome of anti-VEGF therapy needs to be addressed in future, larger clinical cohorts,” she said.

Identification of BCAR genes relevant for breast cancer progression and endocrine therapy resistance: Abstract 1582

Dutch researchers report identifying a set of seven genes responsible for drug resistance and aggressiveness in the most common form of breast cancer.

These genes, some of which are novel, could provide therapeutic targets for personalized treatment of breast cancer and, possibly, for prevention of disease, they say.

These genes are found in estrogen receptor-positive (ER+) breast cancer, which investigators say is usually more treatable than breast cancer that is not fueled by estrogen. The majority of human breast tumors are ER+ and most respond to anti-estrogen therapy, such as tamoxifen. But in advanced disease, only half of ER+ breast cancer is initially sensitive to tamoxifen, and in the majority of those patients, the disease eventually becomes resistant to drug therapy. Understanding the molecular basis of this drug resistance was the focus of the Dutch study.

A research team from Erasmus Medical Center in Rotterdam hypothesized that cell proliferation in the absence of estrogen and in the presence of tamoxifen would be regulated at a molecular level by specific genes. To identify these genes involved in cell growth, they used a functional screen based on insertion mutagenesis with mouse retroviruses. This unique approach in human solid tumor cells took more than a decade of painstaking work.

Some of the seven genes the researchers found – which they have included in a new family dubbed “Breast Cancer Anti-estrogen Resistance” (BCAR) – are already associated with cancer development in general but others are novel. Few of the seven (AKT1, AKT2, BCAR1, BCAR3, EGFR, GRB7, and TRERF1) had been linked to resistance to tamoxifen.

“We set out to define the molecular mechanisms underlying anti-estrogen resistance and this provided a collection of BCAR genes which are representative of the escape pathways that these breast cancer cells take in our laboratory studies,” said the study's lead investigator, Lambert Dorssers, Ph.D., a cell biologist at the Department of Pathology of the Erasmus Medical Center, in Rotterdam.

“We have also shown that the majority of the BCAR genes identified are linked to clinical breast cancer, suggesting that their signaling pathways contribute to the progression of breast cancer and to tamoxifen therapy resistance,” he said.

Using the functional screen, the investigators looked at surgically removed primary tumors from 413 patients who had not yet received systemic therapy and found three genes – AKT2, EGFR, and TRERF1 – that were independently associated with tumor aggressiveness.

In other studies of breast tumor samples from recurrent patients treated with tamoxifen, they had found five genes – BCAR3, ERBB2, GRB7, TLE3, and TRERF1 – that were associated with progression-free survival, depending on the level of their expression. Some of these genes were confirmed by the current study. “For example, we found that high levels of GRB7 are associated with a quicker relapse,” Dorssers said.

The researchers are now defining how these genes function within breast cancer cells in order to pinpoint possible targets for treatment. Expression of the genes could also be used to “classify patients for more intensive or alternative adjuvant treatment, or to suggest specific treatments in the advanced state of ER-positive disease,” he said.

Response to preventive HER2/neu peptide (E75) vaccine based on HER2/neu status: Abstract 2545

A HER2 peptide E75 vaccine reduced mortality in patients with HER2-positive breast cancer by half, according to Texas researchers.

In particular, patients with low-expressing HER2 tumors exhibited better response, not only immunologically, but clinically, with decreased breast cancer recurrence and no mortality following vaccination, report researchers from Brooke Army Medical Center in San Antonio, Texas.

“The fact that HER2 low-expressors responded so favorably not only underscores the difference in mechanism between the vaccine vs. antibody therapy like trastuzumab, but also offers the hope of additional adjuvant therapy to the largest subset of breast cancer patients if proven in the upcoming phase III trial,” said Linda C. Benavides, M.D., a resident in general surgery at Brooke Army Medical Center.

HER2, a source of immunogenic peptides, is over-expressed in approximately 25 to 30 percent of patients with early stage breast cancer.

The CVDP has conducted clinical trials with the HER2 E75-peptide vaccine in lymph node–positive and lymph node–negative patients with breast cancer who demonstrated varying levels of HER2 expression.

They conducted a subset analysis of 163 patients with breast cancer enrolled in the E75 vaccine trial to determine whether the level of HER2 expression affected vaccine response. Of 163 patients assessed, 92 underwent vaccination. Within the vaccinated treatment arm, 29 (34 percent) were defined as HER2 over-expressors, and 56 (66 percent) were defined as low-expressors. The 71 patients in the control group included 22 (33 percent) over-expressors and 44 (67 percent) low-expressors. Patients over-expressing HER2 were similar with regard to prognostic and treatment factors, except that a statistically larger number of vaccinated over-expressors had hormone receptor–negative tumors (P = 0.02).

Following vaccination, immunologic responses were similar as measured by delayed-type hypersensitivity reaction; however, patients in the vaccination arm who were low-expressors of HER2 demonstrated an increased number of E75-specific CD8+ T cells when compared with the vaccinated over-expressors.

At a median follow-up of 30 months, disease recurrence rates were similar between HER2 over-expressors in both the vaccine and control groups, with recurrence rates of 18.2 percent and 13.8 percent, respectively. Although these recurrence rates were comparable (P = 0.7), the researchers observed a greater than 50 percent reduction in mortality rate among patients whose disease recurred. Interestingly, recurrence was more substantially reduced for vaccinated patients with low HER2 expression, Benavides says. Vaccinated low-expressors experienced 10.7 percent recurrence, compared with 18.2 percent for participants in the control group. Furthermore, the mortality rate among low-expressors with recurrent disease was 0 percent among vaccinated patients, versus 38 percent among the control group (P=0.08). Taken together these findings may be significant for the greater than 50 percent of breast cancer patients whose tumors fall into the HER2 low-expressing category and who are not eligible for trastuzumab treatment, Benavides concludes.

Overexpression of ODC1 is associated with poor outcome in childhood neuroblastoma and represents an important therapeutic target: Abstract 5832

Australian researchers have identified a potential new target for treatment of neuroblastoma, the most common solid tumor among young children.

The treatment involves inhibiting the production of ornithine decarboxylase (ODC1), a gene driven by the MYCN oncogene that is a powerful predictor of death from this disease. Researchers report that ODC1 inhibition delayed or prevented the development of neuroblastoma in a clinically relevant animal model, suggesting that suppressing ODC1 could be target for treating this cancer.

“This disease, particularly in patients whose tumors carry multiple copies of the MYCN oncogene, has a particularly poor prognosis and new therapies are urgently needed,” said Michelle Haber, Ph.D., professor and executive director of Children's Cancer Institute Australia for Medical Research in Sydney, Australia. “Our findings suggest that MYCN-driven over-expression of ODC1 in this disease, or genetic variations associated with increased expression of the gene, contribute to the aggressive biology of this tumor, and that inhibition of this gene may lead to an important new therapeutic avenue for this disease.”

Researchers observed 209 patients with untreated neuroblastoma. As expected, they confirmed that older age, advanced stage and MYCN amplification were all associated with highly aggressive disease and poor clinical outcomes.

In a subsequent animal study, researchers tested whether inhibiting ODC1 activity with difluoromethylornithine (DFMO), a proven ODC1 inhibitor, would improve treatment of neuroblastoma, when used in combination with conventional chemotherapeutic drugs. They found that the combined

DFMO/chemotherapeutic drug therapy prolonged tumor-free survival by comparison with chemotherapeutic drugs alone, suggesting that targeting this oncogene for suppression is a potentially valuable therapeutic approach.

“We could actually delay, and in some cases block, neuroblastoma formation in our transgenic MYCN mouse model by continuous exposure to DFMO either from birth or following weaning, and found that this delay was associated with depletion of tumoral polyamines,” said Haber.

The next step would be to test DFMO in combination with other chemotherapeutic agents, particularly newer agents that might be most effective in treating drug-resistant disease, and also testing the drug combinations in different models of childhood neuroblastoma. “Because DFMO has been shown to be quite safe for use in humans, we would hope that we can proceed rapidly to clinical trials,” said Haber.