A Novel Predictive Test for Her-2/EGFR Ab-based Therapeutics

Institution: University of California, San Francisco
Investigator(s): Verena  Kallab , M.D. -
Award Cycle: 2003 (Cycle IX) Grant #: 9FB-0088 Award: $79,600
Award Type: Postdoctoral Fellowship
Research Priorities
Pathogenesis>Outbreak -- how cancer spreads: angiogenesis, invasion, and metastasis

Initial Award Abstract (2003)
Invasion and metastasis are still the greatest challenges to the development of successful breast cancer therapies. There is now proof of the existence of cancer cells that, having detached from the primary breast tumor, circulate in the bloodstream and bone marrow. These circulating tumor cells (CTCs) are suspected to constitute the "seeds" for subsequent growth of additional tumors (called metastasis) in tissues distant from the original breast lesion. Thus, CTCs may represent the “real cancer cells” that need to be characterized at the genetic and molecular levels to better understand the dynamic aspects of metastasis.

Our rationale is that CTCs are unique to each patient, and tumor markers in circulating cells might differ compared to the primary tumor. We plan to, first, isolate CTCs from blood and bone marrow from breast cancer patients using antibody (immunomagnetic capture) and flow cytometry techniques. These methods are expected to distinguish CTCs from normal blood cells. We then plan to amplify the CTC DNA and analyze it for genomic changes (comparative genomic hybridization), especially compared to cells from the primary tumor. Finally, we plan to apply these methods to follow patient CTCs during therapy to see whether specific markers (e.g., ErbB2) associated with tumors change during therapy, and whether specific changes can be associated with tumor progression during treatment or through innate or acquired resistance to therapy.

Our investigation of CTCs has two broad goals. First, the success or failure of adjuvant therapy can be determined only after an observation period of several years. Periodic examination of peripheral blood CTCs during therapy could be informative with respect to the therapeutic approach. Thus, chemotherapy might be tailored to the specific tumor of a patient, follow the course of therapy, and provide insights in to the current status of treatment. CTC-based information might predict those patients at risk for relapse. Secondly, our results could identify genes involved in metastatic progression. This information might spare patients unnecessary treatments and develop new information on tumor molecular targets that might be further studied to interrupt the spread of the disease.

Final Report (2006)
Note: In the second year of this project the focus shifted to a study on the internalization of Her-2. The project title was revised: A Novel Predictive Test for Her-2/EGFR Ab-based Therapeutics

Topic addressed: Flow cytometry is a technology that has found wide application in research as well as in clinical medicine. The Instrument itself consists of a number of sophisticated optics and fluids, electronic detectors and computers, which allows the rapid and quantitative analysis of individual cells and mixed cell populations, based on their size, cell shape and the color of fluorescence. This is specifically important, since a human tumor not only consists of tumor cells, but also of connective tissue (fibroblasts) and blood cells, which need to be excluded from the analysis. Furthermore, single tumor cells in human blood can be detected in this manner based on their specific surface molecules. In this project, antibodies against HER2 were coupled to specific, fluorescent-labeled biomolecules (probes) and incubated with the tumor cells. This incubation step allows the antibodies to bind and internalize into the tumor cells which express the HER2 protein. A chemical treatment (EDTA) of the cells was carried out to distinguish between cell-surface bound and intracellular (internalized) antibodies. Fluorescence intensity of the surface-bound/internalized probes was quantified by flow cytometry. This method was applied to a number of cell lines (immortalized human cancer cells), exhibiting different amounts (i.e., expression levels) of HER2. This procedure was further extended to breast cancer cells obtained from tumors grown in mice (xenografts), a common method to study cancer cells in an environment more closely resembling human disease.

In order to mimic circulating tumor cells (CTCs), cultured tumor cells were “spiked” into whole blood samples. Tumor cell detection was achieved by a technique using magnetic beads that bind to the tumor cells, so they can be isolated with a magnet. This is a relatively fast method and results in acceptable yields. Additional staining with tumor-specific antibodies was performed to isolate tumor cells and fluorescence was measured and quantified with flow cytometry.

Progress towards Specific Aims: Aim 1: Internalization of anti-HER2 antibody in trastuzmab (Herceptin) sensitive/resistant cell lines and xenoggrafts by flow cytometry Cell lines exhibiting high levels of the HER2 protein and known to be responsive to therapy targeting HER2, showed high binding and high intracellular location of the anti-HER2 antibody. Low HER expressing cell lines showed only minimal binding and internalization. In order to confirm these results visually, the same studies were performed using microscopy. This technique was further applied to five trastuzumab resistant cell lines. SKOV3, showed high binding but low intracellular uptake, whereas other resistant cell lines showed reduced binding of the antibody. In order to confirm these results visually, the same studies were performed using microscopy. Similar results were obtained in xenografts, where tumor cells could be identified based on a second, tumor-specific antibody.

Aim 2: Internalization of anti-HER2 antibody in primary human breast cancers by flow cytometry I applied this assay to nine human breast cancers, which were mechanically and chemically dissociated before treating them with the immunofluorescent probes. Three HER2 3+ positive breast cancers showed reduced binding of the anti-HER2 antibody and decreased internalization varied. One HER2 negative tumor showed only minimal binding and uptake. One tumor with intermediate HER2 expression (HER2 2+) showed no binding and internalization. And one HER2 3+ positive tumor showed high binding and internalization.

Aim 3: Internalization of anti-HER2 antibody in circulating tumor cells (CTC) Tumor cells of a cell line exhibiting high levels of HER2, spiked into blood, could be detected and intracellular uptake of the anti-HER2 antibody was quantified. These tumor cells showed similar uptake compared to the corresponding cell line. This experiment was performed to mimic circulating tumor cells in peripheral blood. We were also able to detect CTCs in the blood and bone marrow of breast cancer patients. All CTCs were HER2 negative and showed no binding and uptake of the antibody.

Future Directions After my departure another person in the lab will take over my project in order to evaluate more clinical samples of CTCs in blood and bone marrow. In conjunction with clinical information, we will be able to correlate the relationship of HER2 internalization with disease outcome.

Impact Despite important pharmaceutical advances in treating breast cancer, this disease still remains one of the leading causes of death among women. When a physician decides to administer a certain type of treatment, s/he has no information available whether his patients will benefit. Therefore, breast cancer patients are subjected to a repetitive cycle of debilitating therapies and often face progressively less active and more toxic treatment. What is needed in the clinical management of breast cancer patients is a test, which can accurately assess treatment response and guide the doctors to make the most appropriate therapy decision. In the case of Herceptin, clinical response has been limited, although this drug has been reported to improve the outcome of patients who overexpress the HER2 protein. However, the presence of this protein alone is not sufficient as a prognosticator, since even patients with an excessive amount of HER2 fail to respond.

As stated earlier, the mechanism of how HER2-targeted therapy exerts its anti-tumorigenic effect is not known, but one proposed possibility is that the antibody attaches itself to the cells and gets internalized. Therefore, internalization could serve as a marker of response and could determine how likely a patient is to benefit from this treatment. The results of my studies clearly show that there is indeed a connection between internalization and response to treatment. HER2 over-expressing cells, which are known to be sensitive to anti-HER2 treatment, not only showed high binding, but also efficient internalization of anti-HER liposomes. One cell line, SKOV3, on the other hand failed to internalize despite high levels of HER2. This finding correlates with the fact that these cells are resistant to Herceptin. Therefore, a test based on HER2 internalization could be a useful aid in assessing the clinical responsiveness of a patient receiving Herceptin or some other HER2-targeting agent. It will also save a patient's valuable time and could help to prevent inappropriate therapy selection.