Extra Copies of Chromosome Affects HER2 Test Results

Women with invasive breast cancer whose tumor cells have extra copies of chromosome 17, where the HER2 gene resides, are more likely to have borderline or "equivocal" results on HER2 testing, a team of Belgian researchers reports. In a study released September 15 in the Journal of Clinical Oncology (JCO), the researchers found that tumors with extra copies of chromosome 17 (a condition called polysomy 17) but not HER2 amplification - that is, extra copies of the HER2 gene - behaved like HER2-negative tumors.

The results of HER2 tests have significant implications for breast cancer patients, including whether they are candidates for the drug trastuzumab (Herceptin), which was specifically developed to treat HER2-positive disease.

"Because polysomy 17 on its own is not associated with HER2 overexpression and because it does not have the same clinicopathologic significance as true HER2 gene amplification, one may wonder whether polysomy 17 tumors benefit from HER2-targeted therapy such as trastuzumab," wrote Dr. Isabelle Vanden Bempt and colleagues from the University Hospital Gasthuisberg in Leuven, Belgium.

The researchers tested tumor samples from 226 patients using the two most common techniques for detecting HER2 - FISH and IHC. They also tested the samples for polysomy 17, finding that it was present in every patient with an equivocal HER2 result (47 of 226 patients), and that it was not associated with characteristics of HER2-positive cancers, such as hormone-receptor negativity or lower disease-free survival.

The authors called for clinical trials to help determine whether women with polysomy 17 tumors can benefit from trastuzumab treatment. In a related editorial in JCO, Dr. Carol Rosenberg from Boston University Medical Center, who lauded the current study, argued that new trials might be premature.

"This finding should influence the interpretation of HER2 test results and could eventually improve selection of patients for HER2-targeted treatment," she wrote. However, "New trials to directly test this question may not be the most efficient approach," she added, noting that review or reanalysis of data from existing trials could determine whether polysomy 17 tumors that lack HER2 are responsive to trastuzumab or lapatinib.

Vaccine Prevents HER2 Tumors in Mice

Researchers at the Karmanos Cancer Institute in Michigan have tested a vaccine that they found to be 100 percent effective at preventing tumors in mice injected with breast cancer cells. The results of their study appeared September 15 in Cancer Research.

The researchers used breast cancer cells that mimic the HER2-positive tumors found in women, which account for 20 to 30 percent of breast cancer cases. Women with HER2-positive tumors can be treated with drugs that target the HER2 receptor, such as trastuzumab (Herceptin) and lapatinib, but in some women these drugs eventually stop working.

The researchers used a panel of four cell types that overexpress HER2 and represent the various prognoses for women with HER2-positive breast cancer: two types that were completely sensitive to targeted drugs, one with initial sensitivity to targeted drugs but eventual resistance, and one with complete resistance to HER2-targeted drugs.

Mice were vaccinated with bacterial DNA engineered to include the gene sequence for a large portion of the HER2 receptor. The researchers used "electrovaccination," where an electric pulse encourages cells to absorb DNA and produce the related protein for presentation to immune cells.

After electrovaccination, the mice were injected with one of the four HER2 breast cancer lines. None of them developed tumors. However, control mice that had been electrovaccinated with a plasmid missing the HER2 DNA sequence developed tumors in every case. After 1 year of follow up, there were no adverse effects from vaccination.

In explaining the mechanisms of resistance to HER2-targeted drugs, the researchers noted that the various cell types they used in the study could co-exist in a single breast tumor, and that selective pressure after eliminating the drug-sensitive cells could cause increased growth of the drug-resistant cells. This would explain the relapse seen in some women.

Other tests in this study revealed aspects of how treatment-resistant cells in the mice were able to abandon HER2 and modify their cell-signaling strategy, employing an "escape mechanism" from HER2-targeted treatment.

"In patients whose tumors are refractory to drug and antibody therapy," the authors stated, "induction of comprehensive immunity by active vaccination will be critical to their long term protection."

Genetic Factors Tied to Blood Cancer Risk

It is still largely unknown what causes the group of diseases called myeloproliferative neoplasms (MPNs), including some rare forms of leukemia, where stem cells that replenish the blood supply begin to grow or differentiate out of control. But, a recent study shows their incidence is five to seven times higher among first-degree relatives of individuals who have been diagnosed with such a disease, providing strong evidence for a genetic link. This analysis, led by researchers from NCI's Division of Cancer Epidemiology and Genetics and collaborators in Sweden, appeared September 15 in Blood.

Using population-based data from Sweden, which has a universal health care system, the researchers identified all patients who had been diagnosed with an MPN between 1958 and 2005. For each patient, four population-based controls matched by sex, year of birth, and county of residence were chosen randomly from the Swedish Population database. From the Swedish Multigenerational Registry, which includes information on parent-offspring relations for all Swedish citizens who were born in 1932 or later, the researchers obtained information on all first-degree relatives (parents, siblings, and offspring) of both patients and controls and linked them to the Swedish Cancer Registry to collect information on incident cancer cases. The study included a total of 11,039 patients with an MPN, 43,550 controls, 24,577 first-degree relatives of patients, and 99,542 first-degree relatives of controls.

The researchers found that the relative risk of developing an MPN among first-degree relatives of patients (versus first-degree relatives of controls) was increased 5.6-fold. The relative risk for specific diseases including polycythemia vera (PV), essential thrombocythemia (ET), and myeloproliferative neoplasm unclassifiable (NOS) was 5.7, 7.4, and 7.5, respectively. Only myelofibrosis (MF) - a disease in which the bone marrow is replaced by fibrous scar tissue - showed no increased risk among first-degree relatives of patients, though the overall incidence of this disease was very rare. Compared to first-degree relatives of controls, first-degree relatives of patients with MPNs had a borderline-significant 1.9-fold increased risk of chronic myeloid leukemia.

"Our results support the theory that there are common, strong, shared susceptibility genes that predispose to PV, ET, MF, and possibly chronic myeloid leukemia," the researchers conclude, noting that this is the largest population-based, case-control study of its kind, and that it "supports the application of gene mapping and candidate gene approaches in high-risk families and case-control studies."

In a related comment in the journal, Dr. Jerry Spivak of Johns Hopkins University concludes that the present study serves to remind us that in MPNs, disease characteristics are "driven as much and possibly more by as-yet-undefined genetic influences as it is by those that have been defined."

Most Cancer Clinical Trials Go Unpublished

Findings from fewer than one in five registered cancer clinical trials are published in peer-reviewed journals, according to a study that appeared September 15 in The Oncologist. This finding, the authors state, raises the concern of publication bias in cancer clinical trials.

Drs. Scott Ramsey and John Scoggins of the Fred Hutchinson Cancer Research Center and the University of Washington found that between 1999 and 2007, only 17.6 percent of cancer-related trials registered with ClinicalTrials.gov (the federal registry for clinical trials of interventions for serious or life-threatening conditions) went on to be published in widely accessible journals listed in the PubMed.gov online database.

The researchers also found evidence of a selection bias on the part of investigators, who are less likely to publish results of a trial that did not meet its endpoints. Though more than 94 percent of the registered, industry-sponsored trials were never published, three-fourths of those in PubMed.gov reported positive results. By comparison, 59 percent of NCI-supported trials were published, half of which reported negative results.

In a related commentary, Dr. James H. Doroshow, director of NCI's Division of Cancer Treatment and Diagnosis, wrote that "the apparent lack of access to the final efficacy and toxicity data…poses multiple scientific and ethical questions." He described an NCI clinical trials database project expected to launch in 2009 that will address this problem by requiring administrative and outcome data for all intervention studies that receive NCI support.

The Oncologist is also considering creating a peer-reviewed, searchable venue for "well-executed trials that fail to meet positive endpoints," wrote Editor-in-Chief Dr. Bruce A. Chabner and Senior Editor Dr. Gregory A. Curt, for trials that are "'negative' in a sense, but valuable nonetheless."