Non-Invasive Optical Detection of Breast Cancer
|Institution:||University of California, Irvine|
Bruce Tromberg , Ph.D. -
|Award Cycle:||1996 (Cycle II)||Grant #: 2IB-0183||Award: $50,000|
|Imaging, Biomarkers, & Molecular Pathology>Developing and Improving imaging technologies: better and easier detection|
Initial Award Abstract (1996)
Pathologists routinely examine thin sections of surgically removed tissue in order to diagnose cancer. If similar information could be provided by techniques that allow us to see inside the body, physicians could locate tumors non-invasively. Historically, this reasoning led to the development of "breast diaphanography", a near-infrared (NIR) light imaging method introduced nearly 70 years ago to locate and identify breast cancer. NIR imaging is a simple, low-cost, risk-free procedure which, unlike x-ray mammography, does not employ ionizing radiation. Unfortunately, the initial promise of NIR diaphanography was never realized, primarily due to the effect of intense light scattering. Scattering blurs our ability to find small, buried, light-absorbing tumors much in the same way that a cloud obscures our view of an airplane passing overhead. Distinguishing between absorption and scattering is generally recognized as the principal limitation to detecting breast cancer with light. This problem has puzzled scientists who study how light interacts with highly scattering (turbid) materials for hundreds of years. Recent developments in theory and instrumentation, collectively referred to as "Photon Migration", now enable us to probe and analyze multiply-scattered light signals on an ultra-fast time scale, e.g. a millionth of a millionth of a second. These technological advances provide, for the first time, a simple framework for measuring the exact magnitude of absorption and scattering (i.e. optical properties) in turbid materials. Optical properties can, in turn, be used to locate and identify physiological changes characteristic of tumors. The purpose of this grant is to gather data essential to determining whether NIR breast imaging is truly feasible. We have been involved in the development of Photon Migration (PM) instrumentation and theory for the past six years. Research by our group and many others in the Biomedical Optics community has rejuvenated interest in optical methods for breast cancer detection. The use of ultra-fast light provides new information with the same advantages as conventional diaphanography (e.g. low cost, zero risk, and compact devices). However, in order to demonstrate whether normal and malignant structures can actually be resolved using light, reliable measurements of breast tissue optical properties must be performed in human patients. Few studies have been completed which accurately measure human tissue optical properties, primarily due to the fact that PM is still an experimental technology. As a result, the essence of the proposed work is to use our specially designed, portable, hand-held PM probe to measure optical properties in breast cancer patients. This will provide critical information for determining whether non-invasive "optical mammography" is truly feasible. In addition, by measuring unique tumor optical signatures in patients, we expect to gain important insight into physiological changes associated with malignancy, disease progression, and response to therapy.
Final Report (1998)
Non-invasive methods of determining whether a breast anomaly is benign or cancerous would be of great practical benefit to the many thousands of women who undergo needless surgical biopsies each year. One early attempt to develop such a procedure was "breast diaphanography", a near-infrared (NIR) light imaging technique introduced nearly 70 years ago to locate and identify breast cancer. NIR imaging is a simple, low-cost, risk-free procedure which, unlike x-ray mammography, does not employ ionizing radiation. Unfortunately, the initial promise of NIR diaphanography was never realized, primarily due to the effect of intense light scattering. Scattering blurs the ability to find small, buried, light-absorbing tumors much in the same way that a cloud obscures our view of an airplane passing overhead. This work seeks to provide novel information for the successful application of a new class of NIR optical methods to breast cancer detection that might overcome the scattering problem. These methods, broadly referred to as "photon migration", are able to quantitatively determine normal and malignant breast tissue optical properties and their corresponding physiological parameters in human subjects. Accurate knowledge of optical properties can, in turn, be used to realistically estimate the feasibility of breast tumor optical imaging or "optical mammography". In this work a non-invasive, 7- wavelength diode-laser-based "photon migration" device was developed and optimized for clinical optical property studies. Photon migration measurements were conducted prior to surgical intervention on palpable lesions that were typically 1-2 cm in diameter and 1-2 cm deep. Data were acquired using a hand-held scanning probe placed in 9 separate locations covering a 2 x 2 cm grid mapping the breast surface. The probe incorporates source-detector pairs ranging from 1 to 2.5 cm in separation. Both normal and tumor-containing breasts were studied. Each tissue location was characterized in terms wavelength-dependent absorption and scattering which, in turn, was used to calculate physiological properties. All data were correlated with structural information derived from conventional sources, such as clinical exam, mammography, ultrasound and physical examination of tissue. Results from 18 patients are reported. Pre- and post-menopausal normal breasts exhibit detectable wavelength dependent differences in both absorption and scattering. Scattering is approximately 20% greater in pre-menopausal as compared with post-menopausal women. Tissue scattering parameters for post-menopausal subjects receiving hormone replacement therapy are comparable to non-estrogen replacement cohorts. Absorption in estrogen replacement and pre-menopausal groups is similar. Values range from 1.5 to 1.75-fold greater than post-menopausal patients due to elevated tissue water and hemoglobin levels. Invasive and in-situ ductal carcinomas (ductal carcinomas are the most common form of breast cancer) and benign fibroadenomas exhibit 1.25 to 2-fold higher absorption than normal breast tissue. Within this group, absorption is greatest and scattering is lowest for measurements obtained from sites of invasive cancer. Low tissue scattering values are positively correlated with findings of increased water content and high cellularity. Overall, our data suggest that optical contrast is provided by lesion induced degradation of highly-scattering extracellular matrix components, with concomitant elevations in water and blood at the tumor site. These results suggest that optical methods are quite sensitive to cellular, molecular and structural differences between normal and malignant breast tissues. Further work is necessary to pin down the precise hemodynamic, oxygen consumption, and structural changes that are encoded within optical signatures. As this information becomes available, a new generation of safe, non-ionizing diagnostic tools will become available that could complement existing methods and enhance the overall accuracy, specificity and sensitivity of breast cancer detection. We expect these advances to be particularly important for pre- and peri- menopausal patients who display radiographically-dense breast tissue.
Non-invasive measurements of breast tissue opticalproperties using frequency-domain photon migration
Periodical:Philosophical Transactions of the Royal Society of London Series B: Biological Sciences
Index Medicus: Philos Trans R Soc Lond B Biol Sci
Authors: Tromberg BJ, Coquoz O, Fishkin JB, Pham T, Anderson ER, Butler J, and Cahn M, et al
|Yr: 1997||Vol: 352||Nbr:||Abs:||Pg:661-668|
Frequency-domain photon migration measurements of normal malignant tissue optical properties in a human subject.
Authors: Fishkin JB, Coquoz O, Anderson ER, Brenner M, Tromberg BJ
|Yr: 1997||Vol: 36||Nbr: 1||Abs:||Pg:10-20|