Earlier Detection: Improving the Chances for a Cure

Until breast cancer can be prevented, detection at the earliest time possible remains an essential goal. Over the period 1988—1995, the percentage of invasive breast cancer discovered at its earliest stage (Stage I) improved almost 30% percent, from about 35% to 45%. We hope that the application of CBCRP–supported research findings can contribute to the continued improvement in this trend.

Research Conclusions

Developing and Improving Imaging Technologies: Better and Easier Detection

To achieve the goal of improving chances for a cure, cancer must be found as early as possible, when the tumor is still very small and is most likely not to have produced metastasizing cells. Much of CBCRP's funding of early detection technology research has been aimed at:

  1. Improving the sensitivity (capability of finding very small tumors) and sensitivity (capability of distinguishing between true cancers and other breast or image anomalies) of existing technologies.
  2. Adapting existing technologies specifically for breast imaging.
  3. Improving the clinical application of existing technologies.
  4. Bringing new technologies closer to the point of clinical application.

William Moses, Ph.D. of the Lawrence Berkeley National Laboratory has been working to adapt an existing technology that uses gamma rays (Positron Emission Tomography or PET) to be more suitable for breast and axilla imaging. His original goal was to develop an instrument with higher efficiency (up to 30 times greater), finer spatial resolution (0.9 mm), and lower cost (possibly by a factor of 10) than conventional PET cameras currently in use for body imaging. To date, most of the originally proposed work has been completed:

  1. Many details of the detector module concept have been worked out.
  2. Several prototype modules have been constructed and tested, and all parts necessary to complete production are in hand.
  3. The necessary electronics have been fabricated and tested (although some work is still necessary on the custom integrated circuit and one of the four types of readout circuits).
  4. Preliminary versions of the software necessary to operate the camera, acquire data, calibrate the system, and reconstruct images from the data have been developed.

Additional funding (from another source) has been obtained to complete the device.

Manbir Singh, Ph.D. of the University of Southern California was funded (“High Resolution Breast Gamma Emission Imaging System”) to design a novel breast cancer detection system to give a three–dimensional image of small lesions in the breast after injection of a trace amount of a radiolabeled pharmaceutical. Previous studies in this field have relied upon two–dimensional “projection images” which, though limited in their accuracy, can convey very useful physiologic information to help physicians differentiate malignant from benign lesions. With the proposed three–dimensional imaging system, it was hoped that detection of smaller lesions with a volume of .05 to 0.1 cc would be possible. Experimental three–dimensional imaging studies of a test–object simulating radioactive uptake in lesions and a background region in the breast have been conducted with an array of CZT detectors representing a small–scaled version of the proposed system. Results confirm the computer simulation studies and suggest that in an imaging time ranging from 20 minutes to 1 hour, accumulation of the radiolabeled pharmaceutical in lesions of volume 0.1 to 0.5cc would be detectable in a clinical situation. This limit is a factor of 2–4 times better than achievable with current clinical gamma imaging systems and would represent a significant improvement in the ability to distinguish malignant from benign lesions.

Orhan Nalcioglu, Ph.D. from the University of California, Irvine finished a 2–year IDEA project concerned with the MRI Detection of Breast Cancer Blood Vessels. The goal of this work was to develop advanced chemical agents to detect breast cancer blood vessels using Magnetic Resonance Imaging (MRI). This would allow more advanced diagnosis and visualization of breast cancer compared to traditional mammography. Dr. Nalcioglu developed an MRI contrast agent that contained protamine, which is a simple, low molecular weight protein derived from salmon sperm. It is normally used to treat patients that have been overdosed with the anticoagulant heparin following cardiopulmonary bypass. Protamine binds to and neutralizes the activity of heparin. Dr. Nalcioglu attached protamine to an MRI contrast agent, and demonstrated an excellent detection of experimental breast tumors in rats. Apparently, the blood supply to breast tumors contains heparin exposed to the circulation, and results from this project indicate that neutralization of heparin at these sites could have a positive detection or therapeutic value.

H.K. Huang, D.Sc. of the University of California, San Francisco was funded to try to integrate the process of acquiring, storing, communicating, visualizing, managing, and analyzing digital mammography examinations within the operational environment of a university–based diagnostic center and a community–based general–radiologist mammography practice based some distance away. Technically, they found that the image quality of their full field digital mammography system was at least as good, if not better, than a conventional film/screen mammography system, and it took less than seven seconds to transmit a full resolution digital mammogram (40 Mbytes) between the two locations. Also, the image management system requires only two minutes to automatically archive the images after an examination and categorize them in the database for immediate distribution; and, the workstations, coupled with a dual cursor system, can be used for telediagnosis and teleconsultation between a referring physician and an expert mammographer at a different location. In summary, he and his team feel that they have developed and validated a telemammography system that can be used for the real–time off–site management and interpretation of a community–based general radiologist mammography practice by mammography specialists.

Jack Sklansky, Eng.Sc.D. of the Charles R. Drew University of Medicine & Science and his colleagues were funded (“A Neural Network for Case–Based Diagnosis in Digital Mammography”) to help radiologists use large collections of digitized mammograms as aids in determining whether or not to recommend biopsies and to help reduce the number of unnecessary biopsies without increasing the number of missed cancers. He tested the system with a panel of four radiologists, using a database of “regions of interest” (ROIs) in which microcalcifications are visible, and found that the sensitivity and the specificity of each computer–aided radiologist significantly exceeded the sensitivity and specificity of each unaided radiologist and of the computer alone. He concluded that that radiologists interacting with this mapped database of proven mammographic ROIs are likely to achieve significant reductions in the number of unnecessary biopsies and the number of misdiagnosed cancers when interpreting mammograms containing images of microcalcifications.

Daniel Valentino, Ph.D. at the University of California, Los Angeles completed his four–year project (“Digital Imaging for Hospital and Community–Based Mammography”). During the final year, his team completed the development of an infrastructure for high–performance interfaces, networks and servers that enable direct digital mammography units to send images quickly to display workstations, and to store images permanently in archival devices. They also tested a high–speed ATM network (ultrafast networks using asynchronous transfer mode—ATM–technology) which can provide the bandwidth and throughput that may produce images of sufficient quality to satisfy the medical imaging community.

Improving Access to Screening: Reaching Every Woman

Stephen J. McPhee, M.D. of the University of California, San Francisco in his project “Promoting Early Detection Of Breast Cancer Among Vietnamese” sought to increase access to early breast cancer detection services by Vietnamese women through a large community intervention. Although efforts of the California Breast Cancer Early Detection Program served to increase awareness in the comparison community, thus making interpretation of results difficult, women who reported greater exposure to the various intervention elements were significantly more likely to have heard of, had, and plan a clinical breast examination and to have heard of, had, and plan a mammogram than women with lesser exposure.

The investigators conclude that, although the intervention had no beneficial effect in the community at–large, it had a modest positive impact on those women who reported exposure to the intervention.

CBCRP feels that it is important that researchers developing new technology have a broader understanding of the clinical issues associated with new technologies. Many breast cancer researchers have very narrow, specialized understanding of the disease or misunderstand the relationship of their research to breast cancer detection and treatment. Carolyn Kimme–Smith, Ph.D. at the University of California, Los Angeles has completed three years of her “UCLA Biomedical Physics Graduate Training in Breast Cancer”. Two of the students trained in the program have specialized in developing PET and Gamma camera detectors for breast cancer, and two others specialized in Radiation Oncology for Breast Cancer. One student has specialized in Digital Mammography. In addition to this specific training, the program provides a broad base of knowledge to allow graduates to be resources for researchers other than physicists alone, as well as become principal investigators on their own research projects.

Noreen Facione, Ph.D., R.N. of the University of California, San Francisco in a study involving more than 1,000 women in the San Francisco Bay area in community based focus groups, surveys, and interviews, described how women make judgments about self-discovered breast symptoms that might signal breast cancer. The many influences on this judgment process were described in three diverse groups: women who identified themselves as Anglo/White, African American/Black, or Latino/Hispanic. The likelihood of making a judgment to delay a medical visit was correctly predicted in 31% to 66% of the women in these three cultural groups. The most influential of the variables are attitudes and beliefs that are common to the three cultural groups. These included inaccurate beliefs about the incurability of breast cancer and its presenting symptoms, perceptions of constraints on early detection associated with women's roles, immigration concerns and provider relationship issues. A proportion of this volunteer sample (13.1%) reported currently having breast symptoms and only 50% had been examined by a provider. The variables associated with a decision to delay seeking medical evaluation of symptoms correctly predicted actual delay behavior in 87% (52 or 60) of these women.

California represents a major gateway for Pacific Islanders into the United States and little is known about their health beliefs, practices, and needs. Sora Park Tanjasiri, Dr.P.H. of the University of California, Irvine examined the breast cancer knowledge, attitudes and screening behaviors of two groups of Pacific Islander women — Chamorros (indigenous Guamanians) and Tongans — aged 40 and older in Los Angeles and Orange County. Among her findings was the fact that only 12.9% of Tongans had yearly clinical breast exams and only 10.2% had yearly mammograms. For Chamorros, while 66.2% had yearly clinical breast exams, only 25.3% had yearly mammograms. In addition, many barriers to screening were identified, including cost, language, and lack of knowledge for both populations.

Biomarkers and Novel Screening Approaches: Unmasking the Hidden Signs

Margaret Wrensch, Ph.D. of the University of California, San Francisco studied breast cancer incidence in women with abnormal cytology as determined by using nipple aspirate fluid from a previous study of 6,908 women. Preliminary findings indicate that compared with women who yielded no fluid, women who yielded fluid had relative risks of breast cancer as follows: normal cytology, 1.5; hyperplasia, 2.2; and atypical hyperplasia 2.5. Women from whom they obtained fluid were 1.6 times as likely to develop breast cancer as women who did not yield fluid were (95%CI: 1.3-2.1).

These findings varied somewhat according to the women's family history of breast cancer or previous benign breast biopsy and provide continued strong support for the hypothesis that hyperplasia and atypical hyperplasia diagnosed in nipple aspirate fluid indicate an increased risk of breast cancer.

Research In Progress

Developing and Improving Imaging Technologies: Better and Easier Detection

Laura Esserman, M.D. of the University of California, San Francisco is investigating the factors required to achieve high cost-effectiveness in mammography screening in California, and then developing a plan to promote the provision of the highest quality screening services for the least possible cost. To date she has found, based on California data, that there is indeed a correlation between volume of mammograms read by a radiologist and increased sensitivity and specificity in the radiologist's interpretations. This volume-outcome relationship is at least one factor that promotes efficiency and quality in mammographic services.

Also, preliminary data from randomized controlled trials of mammography in Malmo and Stockholm suggest that sensitivity has remained constant over a 20-year period, cancer to biopsy rates increased, and both interval cancers and false-negative rates have decreased during the same period, demonstrating that increased specificity is learned over time, without impact on sensitivity.

Improving Access to Screening: Reaching Every Woman

Mary Smith, M.S., C.R.C. and Carol D'Onofrio, Dr. P.H. of the Alta Bates Foundation and Northern California Cancer Center, respectively, have undertaken a study of issues intended to increase access to breast screening for women with disabilities. The need for this study was demonstrated in a previous CBCRP–funded pilot study. The work to date has involved the preparation of several large databases necessary for initial analysis and for the creation of a field survey.

Biomarkers and Novel Screening Approaches: Unmasking the Hidden Signs

David Vera, Ph.D. of the University of California, San Diego intends to synthesize and test a molecule that, when injected around a breast tumor, will deliver radioactivity and fluorescence to the first lymph node that receives lymph from the tumor site, thus enabling the surgeon to find this “sentinel” lymph node with the aid of an ultraviolet lamp. To date, he has synthesized the radiolabeled molecule, which can be labeled with technetium–99m and attached mannose to the chain, which permits the lymph node to avidly bind the radioactive chain.

Newly Initiated Research

CBCRP funded five projects in 1999 that address earlier detection. Carolyn Kimme–Smith, Ph.D. of the University of California, Los Angeles will continue to direct a medical physics training program that exposes students to pathology, laboratory, and clinical issues relevant to breast cancer. The remaining four grants relate to biomarkers of breast cancer. First, Jeffrey Smith, Ph.D. will explore a novel molecular approach to detect breast tumor proteases that might be released into the blood. Second, H. Phillip Koeffler, M.D. is funded to use a yeast–based system to detect proteins secreted from breast cancer cells to discover potential biomarkers for future study. Third, David Hoon, M.Sc., Ph.D. will investigate the potential of sampling blood to isolate circulating tumor–specific DNA to test for markers associated with breast cancer. This approach has the potential to substitute for surgical biopsy in order to gain critical information regarding tumor status and possible treatment strategies. Finally, Robert Cardiff, M.D. will study biomarkers in mouse systems to gain information on the relationship of current and potential markers to stage and progression of breast cancer.