Browsing by Subject "Fluorescence microscopy"

With the widespread adoption of mammograms for early breast cancer detection in high income countries (HICs), modern research has principally pivoted towards a focus on reducing overtreatment of patients, particularly those with early stage breast cancer. There are numerous examples of clinical practices and regulations that reflect this shift, including changes in screening recommendations, patient monitoring, re-excision guidelines, and genetic testing, all of which seek to reduce unnecessary intervention without compromising patient outcomes.

Despite changing guidelines, there remains a distinct lack of technologies to reduce overtreatment while ensuring the best possible outcome for patients. One such example is Breast Conserving Surgery (BCS) followed by radiation therapy. There is a wide-range of re-excision rates reported in the literature, but most groups report that 20-40% of patients undergo at least one re-excision. Taking additional shavings during BCS, new guidelines dictating relationships between margin status after BCS and re-excision, and radiation therapy all strive to maximize removal of residual tumor cells with as few surgeries as possible in patients with a new breast cancer diagnosis. However, secondary cancers from radiation therapy, the potential for cancer dissemination as a result of re-excision surgeries, and the burgeoning costs of repeat visits and interventions to an already depleted health care system necessitate new and innovative solutions to improve health outcomes while reducing health expenditures.

While seeking to improve patient experience in HICs, many women in low to middle income countries (LMICs) are unable to access adequate screening and life-saving treatments. Even those who manage to receive a proper diagnostic test are all too frequently lost to follow up care, leading to a disproportionately high burden of breast cancer mortality in LMICs.

The goal of the work presented here is to reduce overtreatment in HICs while simultaneously minimizing access barriers to screening and treatment for women in LMICs through developing a rapid and low-cost molecular diagnostic platform for breast cancer. In HICs, this diagnostic platform could be deployed at two points in the breast cancer care cascade: 1) during diagnostic biopsy to ensure adequate lesion sampling at the point-of-diagnosis, and 2) during intraoperative margin assessment as a ‘see-and-treat’ paradigm that utilizes a single agent to guide surgical resection and to treat residual disease during surgery. In LMICs where limited access to tissue processing equipment and a pathologist often render histological examination of tissue impossible, the diagnostic platform could be used to cheaply and robustly diagnose tissue at the point-of-care.

Three specific aims were proposed to develop the diagnostic platform. The first aim was to demonstrate a single-agent see-and-treat paradigm in pre-clinical models of breast cancer using a fluorescent tracer across all subtypes of breast cancer. The diagnostic piece began by showing that a fluorescently-tethered Hsp90 inhibitor (HS-27), made up of an Hsp90 inhibitor previously used in clinical trials tethered to a fluorescein isothiocyanate (FITC) derivative, is taken up by breast cancer cells in vitro regardless of receptor subtype, and that blocking the ATP binding pocket of Hsp90 leads to reduced HS-27 fluorescence, confirming that fluorescence is a result of HS-27 bound to its target. The in vitro study was expanded to define the therapeutic potential of HS-27 by demonstrating degradation of Hsp90 client proteins consistent with Hsp90 inhibition, and reduction of cellular metabolism, confirming protein degradation led to downstream effects on its signaling pathway. To round out the therapy component of our ‘see-and-treat’ paradigm, HS-27 treatment was found to reduce cell proliferation rates across breast cancer receptor subtypes.

The diagnostic component was moved into animals using a dorsal skinfold window chamber model to interrogate HS-27 uptake in vivo in the context of tumors and their surrounding microenvironment. As expected, HS-27 uptake was significantly greater in tumor window chambers than in non-tumor controls. Utilizing a fluorescent glucose analog to examine glucose uptake levels in tumors as a surrogate for aggressive disease showed that HS-27 strongly correlated (R2 = 0.96) with glucose uptake, suggesting surface Hsp90 expression is upregulated in aggressive glycolytic tumors.

To finish aim 1, HS-27 staining was performed on tumors ex vivo, achieving comparable contrast to in vivo agent administration, providing a path towards translating HS-27 to ex vivo clinical use. A small ex vivo pilot clinical study in patients undergoing diagnostic biopsy revealed a significant correlation between HS-27 uptake and the percentage of tumor present in the sample, providing first proof-of-principle of our HS-27 fluorescence-based diagnostic platform in patients. HS-27 was first imaged in biopsies in order to enroll patients across different receptor subtypes rather than at surgery where the majority of patients have estrogen receptor positive (ER+) disease. To summarize, aim 1 demonstrated a ‘see-and-treat’ paradigm in pre-clinical models of breast cancer, and provided a path towards moving HS-27 into the clinic.

With proof-of-principle patient results revealing that HS-27 may be a feasible diagnostic tool, the focus of aim 2 transitioned towards optimizing the imaging system and protocol. The ex vivo imaging strategy was optimized to minimize non-specific HS-27 uptake in preclinical models. Imaging parameters were fully vetted in a clinical study designed to interrogate HS-27 uptake in patients with breast cancer or benign conditions, as well as in a disease-free population. A high-resolution microendoscope (HRME) designed to image FITC fluorescence in a pre-clinical biopsy model was used to investigate how time between tissue excision and imaging, agent incubation time, and agent dose affect the specificity of HS-27 based diagnostics. For these experiments, a modified version of HS-27 with a 100-fold reduction in Hsp90 affinity, called HS-217, was used to establish non-specific fluorophore uptake. Calculating the ratio of HS-27 fluorescence to HS-217 fluorescence provided a ‘specificity ratio’ that was maximized with a post-excision window up to 10-minutes, 1-minute incubation time, and 100 µM dose.

The optimized protocol was then tested in 37 patients undergoing ultrasound-guided core needle biopsy and in 6 disease-free patients undergoing breast reduction mammoplasty. HS-27 uptake was significantly greater in tumor samples than mammoplasty control samples. Interestingly, HS-27 uptake was similar in tumor and benign lesion samples on average, however, examining the distribution of fluorescence across the biopsy reveals different staining patterns between tumor and benign lesions. Concurrent with the finding in aim 1 that HS-27 levels are elevated in aggressive tumors, HS-27 strongly and inversely correlated with the presence of tumor infiltrating lymphocytes, a positive prognostic marker in Her2+ and triple negative breast cancers. Additionally, leveraging both intensity and spatial patterns to generate a Gaussian support vector machine classifier allowed for accurate classification of tumor, benign lesion, and mammoplasty samples. Classification of tumor vs benign lesions resulted in an area under the receiver operating characteristic curve (AUC) of 0.93 with a sensitivity of 82% and specificity of 100%. Classification of tumor vs mammoplasty samples resulted in an AUC of 0.96 with a sensitivity of 86% and specificity of 100%.

So far, HS-27 uptake has been shown to be specific to tumor over non-tumor tissues, increased HS-27 fluorescence was suggestive of an aggressive tumor phenotype, and ex vivo HS-27 imaging accurately distinguished tumor from both benign and normal breast tissue. Two limitations of the imaging system used in aims 1 and 2 were: 1) the requirement to place the HRME probe in contact with the tissue, potentially causing artificial changes in signal due to pressure differences during probe placement, and 2) the small field of view, which prohibited translation to samples larger than 1-2 cm. Thus, the goal of aim 3 was to develop a wide-field, non-contact imaging system to demonstrate feasibility of translating ex vivo HS-27 imaging to multiple points in the breast cancer care cascade.

We have previously developed a Pocket colposcope for cervical pre-cancer detection and have recently completed construction and testing of an alpha prototype. The colposcope contains a 5 MP camera and white and green light emitting diodes (LEDs) on the tip. It weighs 1 pound, and interfaces with a phone, tablet, or computer, which provides power to the device and enables image capture. The Pocket colposcope, which will now be referred to as a Pocket mammascope is well-suited for breast margin imaging with the ability to, survey breast tumor margins as large as 10 -cm2 in a few snapshots, while maintaining the ability to image a cluster of tumor cells on a length scale of several microns.

The Pocket colposcope was modified into a Pocket mammoscope to perform fluorescence imaging through the addition of a collar with excitation LEDs and a bandpass filter for fluorescence collection. A series of bench tests show that the Pocket mammoscope can perform fluorescence imaging in a wide-field mode with a diagonal field of view of 3.25 cm (compared to 750 µm with the HRME) at a resolution of 25 µm (compared to ~4 µm with the HRME), and high-resolution mode with a diagonal field of view of 1.25 cm and resolution of 12 µm. The two imaging modes are easily navigated between through the use of a simple slider mechanism. The Pocket mammoscope was next used to image HS-27 fluorescence across in vivo and ex vivo models, with comparable results to our previous imaging systems. Additionally, the optimized ex vivo imaging protocol from aim 2 was used to shown to be compatible with the Pocket mammoscope in a cohort of patients undergoing standard-of-care ultrasound-guided core needle biopsy, and that that Pocket mammoscope is capable of imaging an entire biopsy in a single snapshot. Proof-of-concept translation to intraoperative margin assessment utilizing a window chamber model, similar to aim 1, validated that the Pocket mammoscope could image HS-27 both systemically and topically delivered to a tumor.

In conclusion, this work set out to provide a theranostic tool to reduce overtreatment for patients with breast cancer in HICs, and provide a rapid diagnostic test implementable at the point-of-care in LMICs. Towards these goals, aim 1 showed that HS-27 uptake is higher in more aggressive tumors, potentially serving as a prognostic marker delineating which patients require more or less aggressive treatment regimens. Aim 2 found that a Gaussian support vector machine classification scheme based on features from ex vivo HS-27 images accurately distinguishes tumor from both benign conditions and normal breast tissue. Finally, aim 3 demonstrated the feasibility of translating HS-27 to both diagnostic biopsy and intraoperative margin assessment by creating a Pocket mammoscope capable of imaging an entire biopsy and a tumor margin in a few snapshots. Ultimately, this work demonstrates that HS-27 imaging with the Pocket mammoscope is a means for rapid, automated detection of breast cancer, regardless of subtype, which could improve breast cancer management in both HICs and LMICs.

Histopathology is the clinical standard for tissue diagnosis. However, histopathology has several limitations including that it requires tissue processing, which can take 30 minutes or more, and requires a highly trained pathologist to diagnose the tissue. Additionally, the diagnosis is qualitative, and the lack of quantitation leads to possible observer-specific diagnosis. Taken together, it is difficult to diagnose tissue at the point of care using histopathology.

Several clinical situations could benefit from more rapid and automated histological processing, which could reduce the time and the number of steps required between obtaining a fresh tissue specimen and rendering a diagnosis. For example, there is need for rapid detection of residual cancer on the surface of tumor resection specimens during excisional surgeries, which is known as intraoperative tumor margin assessment. Additionally, rapid assessment of biopsy specimens at the point-of-care could enable clinicians to confirm that a suspicious lesion is successfully sampled, thus preventing an unnecessary repeat biopsy procedure. Rapid and low cost histological processing could also be potentially useful in settings lacking the human resources and equipment necessary to perform standard histologic assessment. Lastly, automated interpretation of tissue samples could potentially reduce inter-observer error, particularly in the diagnosis of borderline lesions.

To address these needs, high quality microscopic images of the tissue must be obtained in rapid timeframes, in order for a pathologic assessment to be useful for guiding the intervention. Optical microscopy is a powerful technique to obtain high-resolution images of tissue morphology in real-time at the point of care, without the need for tissue processing. In particular, a number of groups have combined fluorescence microscopy with vital fluorescent stains to visualize micro-anatomical features of thick (i.e. unsectioned or unprocessed) tissue. However, robust methods for segmentation and quantitative analysis of heterogeneous images are essential to enable automated diagnosis. Thus, the goal of this work was to obtain high resolution imaging of tissue morphology through employing fluorescence microscopy and vital fluorescent stains and to develop a quantitative strategy to segment and quantify tissue features in heterogeneous images, such as nuclei and the surrounding stroma, which will enable automated diagnosis of thick tissues.

To achieve these goals, three specific aims were proposed. The first aim was to develop an image processing method that can differentiate nuclei from background tissue heterogeneity and enable automated diagnosis of thick tissue at the point of care. A computational technique called sparse component analysis (SCA) was adapted to isolate features of interest, such as nuclei, from the background. SCA has been used previously in the image processing community for image compression, enhancement, and restoration, but has never been applied to separate distinct tissue types in a heterogeneous image. In combination with a high resolution fluorescence microendoscope (HRME) and a contrast agent acriflavine, the utility of this technique was demonstrated through imaging preclinical sarcoma tumor margins. Acriflavine localizes to the nuclei of cells where it reversibly associates with RNA and DNA. Additionally, acriflavine shows some affinity for collagen and muscle. SCA was adapted to isolate acriflavine positive features or APFs (which correspond to RNA and DNA) from background tissue heterogeneity. The circle transform (CT) was applied to the SCA output to quantify the size and density of overlapping APFs. The sensitivity of the SCA+CT approach to variations in APF size, density and background heterogeneity was demonstrated through simulations. Specifically, SCA+CT achieved the lowest errors for higher contrast ratios and larger APF sizes. When applied to tissue images of excised sarcoma margins, SCA+CT correctly isolated APFs and showed consistently increased density in tumor and tumor + muscle images compared to images containing muscle. Next, variables were quantified from images of resected primary sarcomas and used to optimize a multivariate model. The sensitivity and specificity for differentiating positive from negative ex vivo resected tumor margins was 82% and 75%. The utility of this approach was further tested by imaging the in vivo tumor cavities from 34 mice after resection of a sarcoma with local recurrence as a bench mark. When applied prospectively to images from the tumor cavity, the sensitivity and specificity for differentiating local recurrence was 78% and 82%. The results indicate that SCA+CT can accurately delineate APFs in heterogeneous tissue, which is essential to enable automated and rapid surveillance of tissue pathology.

Two primary challenges were identified in the work in aim 1. First, while SCA can be used to isolate features, such as APFs, from heterogeneous images, its performance is limited by the contrast between APFs and the background. Second, while it is feasible to create mosaics by scanning a sarcoma tumor bed in a mouse, which is on the order of 3-7 mm in any one dimension, it is not feasible to evaluate an entire human surgical margin. Thus, improvements to the microscopic imaging system were made to (1) improve image contrast through rejecting out-of-focus background fluorescence and to (2) increase the field of view (FOV) while maintaining the sub-cellular resolution needed for delineation of nuclei. To address these challenges, a technique called structured illumination microscopy (SIM) was employed in which the entire FOV is illuminated with a defined spatial pattern rather than scanning a focal spot, such as in confocal microscopy.

Thus, the second aim was to improve image contrast and increase the FOV through employing wide-field, non-contact structured illumination microscopy and optimize the segmentation algorithm for new imaging modality. Both image contrast and FOV were increased through the development of a wide-field fluorescence SIM system. Clear improvement in image contrast was seen in structured illumination images compared to uniform illumination images. Additionally, the FOV is over 13X larger than the fluorescence microendoscope used in aim 1. Initial segmentation results of SIM images revealed that SCA is unable to segment large numbers of APFs in the tumor images. Because the FOV of the SIM system is over 13X larger than the FOV of the fluorescence microendoscope, dense collections of APFs commonly seen in tumor images could no longer be sparsely represented, and the fundamental sparsity assumption associated with SCA was no longer met. Thus, an algorithm called maximally stable extremal regions (MSER) was investigated as an alternative approach for APF segmentation in SIM images. MSER was able to accurately segment large numbers of APFs in SIM images of tumor tissue. In addition to optimizing MSER for SIM image segmentation, an optimal frequency of the illumination pattern used in SIM was carefully selected because the image signal to noise ratio (SNR) is dependent on the grid frequency. A grid frequency of 31.7 mm-1 led to the highest SNR and lowest percent error associated with MSER segmentation.

Once MSER was optimized for SIM image segmentation and the optimal grid frequency was selected, a quantitative model was developed to diagnose mouse sarcoma tumor margins that were imaged ex vivo with SIM. Tumor margins were stained with acridine orange (AO) in aim 2 because AO was found to stain the sarcoma tissue more brightly than acriflavine. Both acriflavine and AO are intravital dyes, which have been shown to stain nuclei, skeletal muscle, and collagenous stroma. A tissue-type classification model was developed to differentiate localized regions (75x75 µm) of tumor from skeletal muscle and adipose tissue based on the MSER segmentation output. Specifically, a logistic regression model was used to classify each localized region. The logistic regression model yielded an output in terms of probability (0-100%) that tumor was located within each 75x75 µm region. The model performance was tested using a receiver operator characteristic (ROC) curve analysis that revealed 77% sensitivity and 81% specificity. For margin classification, the whole margin image was divided into localized regions and this tissue-type classification model was applied. In a subset of 6 margins (3 negative, 3 positive), it was shown that with a tumor probability threshold of 50%, 8% of all regions from negative margins exceeded this threshold, while over 17% of all regions exceeded the threshold in the positive margins. Thus, 8% of regions in negative margins were considered false positives. These false positive regions are likely due to the high density of APFs present in normal tissues, which clearly demonstrates a challenge in implementing this automatic algorithm based on AO staining alone.

Thus, the third aim was to improve the specificity of the diagnostic model through leveraging other sources of contrast. Modifications were made to the SIM system to enable fluorescence imaging at a variety of wavelengths. Specifically, the SIM system was modified to enabling imaging of red fluorescent protein (RFP) expressing sarcomas, which were used to delineate the location of tumor cells within each image. Initial analysis of AO stained panels confirmed that there was room for improvement in tumor detection, particularly in regards to false positive regions that were negative for RFP. One approach for improving the specificity of the diagnostic model was to investigate using a fluorophore that was more specific to staining tumor. Specifically, tetracycline was selected because it appeared to specifically stain freshly excised tumor tissue in a matter of minutes, and was non-toxic and stable in solution. Results indicated that tetracycline staining has promise for increasing the specificity of tumor detection in SIM images of a preclinical sarcoma model and further investigation is warranted.

In conclusion, this work presents the development of a combination of tools that is capable of automated segmentation and quantification of micro-anatomical images of thick tissue. When compared to the fluorescence microendoscope, wide-field multispectral fluorescence SIM imaging provided improved image contrast, a larger FOV with comparable resolution, and the ability to image a variety of fluorophores. MSER was an appropriate and rapid approach to segment dense collections of APFs from wide-field SIM images. Variables that reflect the morphology of the tissue, such as the density, size, and shape of nuclei and nucleoli, can be used to automatically diagnose SIM images. The clinical utility of SIM imaging and MSER segmentation to detect microscopic residual disease has been demonstrated by imaging excised preclinical sarcoma margins. Ultimately, this work demonstrates that fluorescence imaging of tissue micro-anatomy combined with a specialized algorithm for delineation and quantification of features is a means for rapid, non-destructive and automated detection of microscopic disease, which could improve cancer management in a variety of clinical scenarios.