Project Funding Details

Background: Most women with early-stage breast cancer have the option of breast conserving therapy (BCT), which involves a partial mastectomy for the complete removal of the primary breast lesion with tumor-free margins, followed by radiotherapy. Numerous studies have shown that the presence of tumor within 1-2 mm of the surgical margin is strongly correlated with the risk of local tumor recurrence. The first step in margin analysis is gross examination of the specimen by the surgeon, which leads to incorrect diagnoses in 25% of cases. Other methods of analyzing margin status, including standard histopathology, frozen section analysis, "touch prep," and ultrasound, have several deficiencies in terms of diagnostic accuracy and/or analysis time that prevent efficient management of breast cancers. An automated method that can assess the resected specimen for clear margins within 1-2 mm of the surface with high sensitivity and specificity will assure complete removal in a single procedure and significantly improve management of the disease. Optical methods and, in particular, Raman spectroscopy (RS), have the capability to perform real-time, automated diagnosis in breast tissue. Our lab has demonstrated that RS can rapidly classify ex vivo breast tissues into four histopathological categories with 99% accuracy. In its conventional form, however, RS cannot interrogate tissue to the depth of 1-2 mm as needed for breast margin assessment, but spatially offset Raman spectroscopy (SORS) has the potential to work around this limitation. With this technique, we can detect tumor-like signatures up to 1-2 mm deep and thus assure tumor-negative margins for improved management of the disease. Here we propose to develop SORS for use in evaluating surgical margin status of partial mastectomy specimens in real time in the operating room.

Objective/Hypothesis: Given the number of women who require a second breast cancer removal operation due to tumor-positive margins, and the limitations of techniques currently available to examine margins, there is a strong need for a device that can quickly, accurately, and non-invasively evaluate the margins of the resected specimen while the patient is in the operating room. In this proposal, we hypothesize that SORS can be used to assess the margin status of a resected breast specimen in real time in the OR with high sensitivity and specificity. The main objectives of this proposal, then, are to characterize the depth sensitivity of SORS in soft tissues and to design systems to implement its use for assessing breast cancer surgical margins.

Specific Aims: To achieve the stated objectives, the following specific aims are proposed. Aim 1. Characterize SORS for soft tissues, using tissue phantoms, breast tissue samples, and Monte Carlo simulations. Aim 2. Evaluate the performance of SORS in normal and malignant breast tissues in vitro. Aim 3. Design, build, and test a surface SORS device for clinical implementation.

Study Design: Existing fiber optic RS probes will first be used to characterize the relationship between source detector separation and depth of interrogation using tissue phantoms as well as portions of human breast tissue samples. This relationship will also be simulated with a Monte Carlo model, and the ability of SORS to discriminate layers in soft tissues will be determined. These results will allow us to estimate the source-detector separation needed for an interrogation of 1-2 mm in depth for intact breast tissues. A preliminary one-dimensional SORS fiber array with separations estimated from above will be developed and evaluated for use on normal and malignant breast tissue samples in vitro. The ability of this probe design to gather spectra from the desired depths, with high signal to noise, and in a reasonable time frame will be evaluated. Finally, a larger array SORS device for determining the margin status on entire resected breast specimens will be formulated and tested in the operating room on at least 30 patients. Specimens will be evaluated one surface at a time and will be compared in depth with detailed histopathology, which will serve as the gold standard. A probabilistic, multivariate statistical algorithm will be developed to predict whether the margins are positive or negative for cancer based on the acquired spectra.

Innovation: The proposed project is highly innovative in a number of aspects. It will be the first reported use of SORS for anything but detecting bone or calcifications through soft tissue; it will be the first reported use of Raman in general to examine a large surface area in a short time; and it will be the first reported use of Raman for the application of evaluating surgical margins in lumpectomy specimens following BCT.

Impact: The proposed research will have a significant impact on the management of breast cancer patients. The proposed tool will ensure complete removal of the tumor in a single procedure, eliminating the need for reoperation, and it will do so in a quicker, more accurate manner compared with currently available techniques, resulting in more efficient care and reduced costs for patients and the health care system.

Most women diagnosed with early-stage breast cancer prefer to preserve their breast while completely removing the tumor. This is possible with what is referred to as breast conserving therapy (BCT), in which the primary breast tumor is removed via a lumpectomy, which is followed by radiation therapy. This procedure relies on the surgeon's ability to remove the entire tumor, leaving no cancer cells behind for future return of the disease. To ensure complete removal, a small amount (1-2 mm) of normal tissue surrounding the tumor must be removed as well. When no tumor cells are found in the outer layers of the removed tissue, it is said to have tumor-negative margins. Lumpectomy patients with negative margins have a statistically identical chance of the tumor coming back as total mastectomy patients, for whom the entire breast is removed. Currently, surgeons rely on simply looking at the removed tissue to determine if they have removed all of the tumor while still in the operating room, but their assessment is incorrect in at least 25% of cases. The treatment ultimately relies on a pathologist to view the removed tissue under a microscope to determine if the edges of the tissue are clear of tumor cells. This is usually done in the week after the surgery is completed, and if tumor cells are found at the margins of the tissue that was removed, the patient is called back for a second surgery. Thus, it would be extremely useful, with respect to time, money, and treatment of the patient, if a method could be developed that would determine if the edges of the removed tissue are clear of tumor cells while the patient is still in the operating room, so that a repeat surgery is not needed and the patient's long-term prognosis is improved.

This proposal addresses this important problem and provides a possible solution using light. Light has been used by many researchers for the detection of cancers, including a technique known as Raman spectroscopy (RS). In RS, when light enters the tissue, it collides with the molecules inside and transfers some of its energy to the molecules. If one looks at the light coming back out, the change in its energy tells us something about what kinds of molecules the light collided with. Studying the intensity of this energy change is referred to as RS. Since there are molecules that are different in normal versus cancerous tissues, many researchers, including the PI, have shown that RS can be used for cancer detection. In fact, the PI has shown that RS can correctly identify breast tissue as normal or one of three different kinds of breast tumors 99% of the time. Unfortunately, standard Raman instruments cannot look deeply enough into the removed breast specimens to make sure that there is 1-2 mm of normal tissue around the removed tumor. In this Idea Award we propose to implement RS in a new way, wherein RS signals can be detected from a variety of depths up to the 1-2 mm required for this application. This method is called "spatially offset Raman spectroscopy," or SORS, and involves the fiber optic sending light to the tissue (source fiber) being physically separated from the fiber optic collecting the Raman light (detector fiber), such that signal from deeper layers can be collected. SORS has never been applied for soft tissues or for detecting cancer. However, with the ability of RS to detect tumors 99% correctly and the ability of SORS to look deeper into tissue, SORS has the potential to be extremely useful for verifying that tumors removed in BCT are completely out with a negative margin of sufficient size while the patient is still in the operating room.

This proposal seeks to develop the use of SORS for measuring the entire surface of the removed breast tissue following lumpectomies, and thus determining whether the surgeon has removed the entire tumor and enough surrounding normal tissue. These measurements will be made in the operating room, and they will provide instant feedback to the surgeon so that a second surgery down the line will not be necessary. The first steps of this process will use tissue-mimicking materials to verify that we can obtain Raman signal from deeper layers of soft tissue (unlike tissues such as bone) and to determine how far apart the source fiber and the detection fibers need to be so that we can see at least 1-2 mm into the tissue. These results will be used to design a fiber arrangement for SORS that can be used to study breast tissue samples and verify that SORS is able to differentiate between normal and cancerous breast tissues. Finally, the fiber device will be tested on surgically removed breast samples in the operating room and compared to the standard pathology to determine the ultimate success of the technique for this application. The proposed research will help ensure that tumors are completely removed in a single surgery when patients elect for lumpectomy (BCT), eliminating the significant anxiety, time, and costs associated with reoperations. The impact of this research is high and immediate within the practical needs of today's breast cancer treatment. We anticipate that at the end of the proposal period, the system and method will be ready for widespread testing and subsequent FDA approval for routine patient use. Based on the development of other medical applications of light, the time frame for this commercial use is expected to be less than 5 years.