Iku Yamaga

Visualization of tumor-related blood vessels in human breast by photoacoustic imaging system with a hemispherical detector array

Noninvasive measurement of the distribution and oxygenation state of hemoglobin (Hb) inside the tissue is strongly required to analyze the tumor-associated vasculatures. We developed a photoacoustic imaging (PAI) system with a hemispherical-shaped detector array (HDA). Here, we show that PAI system with HDA revealed finer vasculature, more detailed blood-vessel branching structures, and more detailed morphological vessel characteristics compared with MRI by the use of breast shape deformation of MRI to PAI and their fused image. Morphologically abnormal peritumoral blood vessel features, including centripetal photoacoustic signals and disruption or narrowing of vessel signals, were observed and intratumoral signals were detected by PAI in breast cancer tissues as a result of the clinical study of 22 malignant cases. Interestingly, it was also possible to analyze anticancer treatment-driven changes in vascular morphological features and function, such as improvement of intratumoral blood perfusion and relevant changes in intravascular hemoglobin saturation of oxygen. This clinical study indicated that PAI appears to be a promising tool for noninvasive analysis of human blood vessels and may contribute to improve cancer diagnosis.

Clinical Report on the First Prototype of a Photoacoustic Tomography System with Dual Illumination for Breast Cancer Imaging.

Photoacoustic tomography is a recently developed imaging modality that can provide high spatial-resolution images of hemoglobin distribution in tissues such as the breast. Because breast cancer is an angiogenesis-dependent type of malignancy, we evaluated the clinical acceptability of breast tissue images produced using our first prototype photoacoustic mammography (PAM) system in patients with known cancer. Post-excisionally, histological sections of the tumors were stained immunohistochemically (IHC) for CD31 (an endothelial marker) and carbonic anhydrase IX (CAIX) (a marker of hypoxia). Whole-slide scanning and image analyses were used to evaluate the tumor microvessel distribution pattern and to calculate the total vascular perimeter (TVP)/area for each lesion. In this clinical study, 42 lesions were primarily scanned using PAM preoperatively, three of which were reported to be benign and were excluded from statistical analysis. Images were produced for 29 out of 39 cancers (visibility rate = 74.4%) at the median depth of 26.5 (3.25–51.2) mm. Age, menopausal status, body mass index, history of neoadjuvant treatment, clinical stage and histological tumor angiogenesis markers did not seem to affect the visibility. The oxygen saturation level in all of the measured lesions was lower than in the subcutaneous counterpart vessels (Wilcoxon test, p value<0.001), as well as in the counterpart contralateral normal breast region of interest (ROI) (Wilcoxon test, p value = 0.001). Although the oxygen saturation level was not statistically significant between CAIX-positive vs. -negative cases, lesional TVP/area showed a positive correlation with the oxygen saturation level only in the group that had received therapy before PAM. In conclusion, the vascular and oxygenation data obtained by PAM have great potential for identifying functional features of breast tumors.

Photoacoustic mammography capable of simultaneously acquiring photoacoustic and ultrasound images

Abstract. We have constructed a prototype photoacoustic mammography system (PAM-02) capable of simultaneously acquiring photoacoustic (PA) and ultrasound (US) images. Each PA, US, and fused PA/US image can be acquired over a wide area of the breast using the scanning module of a US transducer, a PA detector, and optical prisms. The resolution of the PA images exhibits improvement from 2 to 1 mm compared to images acquired using our previous prototype. The maximum scan area of PAM-02 is 90 mm along the horizontal axis and 150 mm along the vertical axis. In a phantom experiment, the available depth was at least 45 mm. A representative example of the application of the PAM-02 prototype in clinical research at Kyoto University is presented and shows S-factor images, which are considered an approximation parameter related to hemoglobin saturation of tumor-related blood vessels. We confirmed the applicability of the system for anatomical and biological research.

Abstract P1-02-01: Evaluation of second-generation photoacoustic mammography in detecting the breast cancer vasculature and hypoxic status; a preliminary study

Background: Functional imaging of tumor vasculature and oxygenation status is essential for monitoring the therapeutic response to the manipulation of abnormal vasculature. Moreover, it could be also applicable for the detection and risk assessment of breast lesion with borderline malignancy since hypoxia and angiogenesis is known to be associated with the malignant potential of precursor lesion of solid tumor. Photoacoustic mammography (PAM) is a novel optical imaging technology that can visualize the hemoglobin distribution and its oxygen saturation (SO2) noninvasively. We have previously reported a promising clinical result of a prototype model of PAM (Canon Inc., Tokyo, Japan) in breast cancer patients. However, the improvement of spatial resolution and the identification of signal origin are still big challenges when considering its application for clinical settings. Materials and methods: We developed the second-generation model of PAM (PAM-02). This instrument has achieved the improved spatial resolution (1.3mm) and enhanced detectability by carrying a high-sensitive detector. Moreover, it is equipped with B-mode ultrasound, which enables us to identify the tumor location in PAM images more precisely. The distribution of hemoglobin within breast tissue carrying solid tumor was evaluated by using PAM-02 under the approval of the ethics committee in Kyoto University Hospital, Japan. Contralateral breast without tumor was also evaluated as a control if possible. Calculated SO2 from photoacoustic (PA) signals were illustrated by using color scale. Results: Seventeen breast lesions from 15 patients were analyzed including 4 ductal carcinoma in situ (DCIS), 12 invasive ductal carcinoma (IDC) and one usual ductal hyperplasia. Tumor locations were successfully identified in 14 out of 17 lesions (82.3%) by B-mode ultrasound imaging. The location of 3 lesions undetectable by B-mode ultrasound imaging could be identified by comparing with corresponding MRI images. B-mode ultrasound imaging made it easy to distinguish intra-tumoral PA signals from peri-tumoral PA signals. Intra-tumoral PA signals were detectable in 68.7% of malignant lesions (11 out of 16 lesions). Peri-tumoral PA signals, which were suggested to be from feeding vessels, were detectable in 81.3% of malignant lesion (13 out of 16 lesions). In the case of benign UDH, PA signals were not detected in either intra- or peri-tumoral region. Intra-tumoral SO2 was estimated to be lower than peri-tumoral SO2 in malignant lesion. While peri-tumoral PA signals were often described as continuous vasculature, intra-tumoral PA signals often showed the spotty patterns. In addition, PA signal density was relatively higher in DCIS compared with IDC. These findings was supposed to reflect the decreased hemoglobin perfusion within solid structure of breast cancer. The minimum detectable lesion was DCIS with a diameter of 8mm. Conclusion: Improved spatial resolution and combination with B-mode ultrasound imaging facilitate the region-specific evaluation of PAM imaging. PAM-02 was supposed to be feasible for evaluating the hypoxic status within small breast tumor and its microenvironment. Citation Format: Masahiro Kawashima, Iku Yamaga, Masae Torii, Mariko Tokiwa, Fakhrejahani Elham, Masako Kataoka, Shotaro Kanao, Masahiro Takada, Yasufumi Asao, Tsuyoshi Shiina, Masakazu Toi. Evaluation of second-generation photoacoustic mammography in detecting the breast cancer vasculature and hypoxic status; a preliminary study [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P1-02-01.

Vascular branching point counts using photoacoustic imaging in the superficial layer of the breast: A potential biomarker for breast cancer

This study aimed to identify the characteristics of the vascular network in the superficial subcutaneous layer of the breast and to analyze differences between breasts with cancer and contralateral unaffected breasts using vessel branching points (VBPs) detected by three-dimensional photoacoustic imaging with a hemispherical detector array. In 22 patients with unilateral breast cancer, the average VBP counts to a depth of 7 mm below the skin surface were significantly greater in breasts with cancer than in the contralateral unaffected breasts (p < 0.01). The ratio of the VBP count in the breasts with cancer to that in the contralateral breasts was significantly increased in patients with a high histologic grade (p = 0.03), those with estrogen receptor-negative disease (p < 0.01), and those with highly proliferative disease (p < 0.01). These preliminary findings indicate that a higher number of VBPs in the superficial subcutaneous layer of the breast might be a biomarker for primary breast cancer.

Abstract P4-01-10: Development of photoacoustic vascular imaging system for breast cancer

Background: Tumor angiogenesis and hypoxia are associated with breast cancer growth and metastasis. Photoacoustic (PA) tomography is an optical imaging technology that visualizes distribution and oxygenation status of hemoglobin with high spatial resolution. Initially we developed a photoacoustic mammography (PAM) having a flat-shaped scanning detector that could detect breast tumors. Nevertheless, the flat-shaped detector array has the drawback of a limited view. Here we developed a novel PAM system with a hemispherical-shaped detector array (HDA), which enables us to identify microvasculatures non-invasively and allow the collection of nearly spatially isotropic three-dimensional reconstructed image of blood vessels. This non-invasive vascular imaging system may be able to characterize tumor angiogenesis and analyze the status of microcirculation. The aim of this study was to analyze the imaging findings of tumor-related vasculature in breast cancer patients. Patients and method: A PAM system with HDA has been generated in a cooperation project between Canon Inc., Japan, and Kyoto University. Twenty-two primary breast cancer patients, including 5 patients with non-invasive cancer and 17 patients with invasive cancer, diagnosed between December 2014 and December 2015 underwent the PAM imaging analysis. We also applied the breast deformation algorithm from the breast shape in a MRI image to that in a PA image in order to create a fusion image of the two modalities for the analysis. Features of peri- and intra-tumoral vasculature, and their oxygenation status were evaluated. The study protocol was approved by the institutional review board at Kyoto University Hospital (UMIN000012251). All patients provided informed consent to participate in this study. Results: The abnormal peri-tumoral vasculature was detected in 86% of all non-invasive and invasive disease cases. In invasive cancer cases, most tumor-related blood vessels were centripetally directed toward the tumor, and 93% of centripetal blood vessels appeared to be disrupted or rapidly narrowed at the tumor boundary. The centripetal blood vessel structure was frequently observed in invasive cancer compared with non-invasive cancer (61% vs 35%). PA images before and after preoperative chemotherapy were obtained in one case, where intra-tumoral blood vessels became finer after chemotherapy, reflecting normalization of intra-tumoral microcirculation induced by chemotherapy. Conclusions: A PAM system with HDA has provided a high-resolution vascular images of primary breast cancers. The morphological differences of peri-tumoral vasculature were observed between invasive disease and non-invasive disease. These results suggest the potential of PA imaging as a non-invasive tool to analyze tumor vasculature of human breast cancers and maybe be helpful for breast cancer diagnosis. (Acknowledgements) This work was partially supported by the Innovative Techno-Hub for Integrated Medical Bio-imaging Project of the Special Coordination Funds for Promoting Science and Technology from the Ministry of Education, Culture, Sports, Science, and Technology, Japan. Citation Format: Toi M, Asao Y, Takada M, Kataoka M, Endo T, Kawashima M, Yamaga I, Nakayama Y, Tokiwa M, Fakhrejahani E, Torii M, Kawaguchi-Sakita N, Kanao S, Matsumoto Y, Yagi T, Sakurai T, Togashi K, Shiina T. Development of photoacoustic vascular imaging system for breast cancer [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P4-01-10.

Abstract P4-03-03: Detection of the tumor vasculature and the hypoxic status of breast lesions using second-generation photoacoustic mammography: An exploratory study

Background: Tumor angiogenesis and hypoxia are associated with breast cancer growth and metastasis. Photoacoustic mammography (PAM) non-invasively visualizes hemoglobin distribution inside the breast by detecting thermoelastic waves from hemoglobin generated by the irradiation of a near-infrared laser pulse. Oxygen saturation (SO2) can be calculated using photoacoustic (PA) signals obtained by two laser pulses of different wavelengths. We further improved the spatial resolution of PAM by approximately 1 mm and enhanced detectability by using a high-sensitivity detector. This new PAM technique can obtain both PAM images and ultrasonography (US) images simultaneously. The aim of this study was to explore the clinical usefulness of this PAM technique. Patients and methods: Women who had breast lesions were eligible for this study. The participants9 lesions were measured using the new PAM technique before they began treatment. The PAM images were evaluated by 5 physicians. First, the lesions were identified using only the PAM images. Second, we used US or contrast-enhanced magnetic resonance images (CE-MRI) to identify the locations of the lesions. Next, we evaluated the photoacoustic (PA) signals based on their locations. Peri-tumoral PA signals were defined as linear signals that congregated in the peri-tumoral area, boundary PA signals were defined as peri-tumoral signals that were disrupted at the lesion9s boundaries, and intra-tumoral PA signals were defined as any significant PA signals inside the tumor. SO2 was illustrated using a color scale. The study protocol was approved by the institutional review board at Kyoto University Hospital, Japan (UMIN000007464). Results: PAM was performed on 48 breast lesions in 45 patients, including 36 invasive carcinoma lesions, 8 ductal carcinoma in situ (DCIS) lesions, and 4 benign lesions. Evaluations of PA signals according to the locations of the lesion, with confirmation from US or CE-MRI, were successfully performed for 38 lesions. Peri-tumoral PA signals were detected in 33 lesions (87%), disrupted boundary PA signals were detected in 30 lesions (79%), and intra-tumoral PA signals were detected in 25 lesions (66%). The detection rates for peri-tumoral, boundary and intra-tumoral PA signals were 94%, 87%, and 65% for invasive carcinoma, and 60%, 40%, and 80% for DCIS, respectively. Intra-tumoral PA signals tended to be weaker than peri-tumoral PA signals in invasive carcinoma lesions, and they often displayed a spotty rather than a linear shape. Intra-tumoral PA signals were observed to have lower SO2 levels than peri-tumoral PA signals in 95% of invasive carcinoma lesions and in 75% of DCIS lesions. Although peri-tumoral and boundary PA signals were also detected in a 38-mm fibroadenoma, the intra-tumoral PA signals displayed a diffuse pattern. Conclusions: We demonstrated that high spatial resolution and use in combination with US and CE-MRI facilitate the region-specific evaluation of PAM imaging. PAM could become a useful tool for the evaluation of the hypoxic status of tumors by enhancing its sensitivity. Citation Format: Takada M, Kawashima M, Kataoka M, Kanao S, Yamaga I, Torii M, Tokiwa M, Fakhrejahani E, Sakurai T, Asao Y, Haga H, Shiina T, Togashi K, Toi M. Detection of the tumor vasculature and the hypoxic status of breast lesions using second-generation photoacoustic mammography: An exploratory study. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P4-03-03.

Abstract P3-01-02: Clinical, anatomical and histological characteristics of breast lesions visualized by photoacoustic mammography; first clinical study in CK project

Background: Photoacoustic mammography (PAM) is an optical imaging technique potentially capable of imaging breast vasculature as well as measuring hemoglobin oxygen saturation(SO2) in focal breast lesions. We presented the initial observation on first 20 cases using the first generation PAM (PAM-01) prototype made by Canon Inc.(Tokyo, Japan) (spatial resolution:2mm) in SABCS 2013. Here we report the clinico-pathological characteristics of all cases recruited in the first clinical study in CK project (Kyoto University/Canon joint research project) between Aug 2010 and March 2012. Methods: 57 patients were recruited in IRB approved study at Kyoto University Hospital, Japan. Forty-two breast harboring lesions and when possible contralateral breasts were evaluated by PAM-01. Axial maximum intensity projection (MIP)s were obtained and signals from consecutive MIPs confirmed to be associated with the tumor location in MRI by an expert breast radiologist were considered to be the region of interest (ROI).The same depth was used for ROI in the normal breast as control. Histological sections from the widest area of the lesions were evaluated post-excisional by immunohistochemistry using anti-CD31 as endothelial marker and anti-carbonic anhydrase IX (CA IX) as hypoxia marker. Histological slides were scanned and divided into 1680 squares (0.84x0.84 mm2)(Hamamatsu Inc. Japan) for image analysis. Total vascular perimeter (TVP)-in mm- was calculated for all the histological section by using Image Pro-Plus 7.0 software (Media Cybernetics, USA). Tumor area was measured in mm2. TVP index was calculated as TVP/area. Results: Photoacoustic signal was detected in 30 lesions out of 42 at the depth of 26.8 ± 12.8 mm from which 80% were located superior to nipple. CA IX positive cases in comparison with CA IX negative cases significantly showed higher TVP index (p-value =0.028 Mann-Whitney Test ) suggesting more angiogenic profile of hypoxic tumors. However, lesions without any detectable signal were reported to have only a bigger mass size histologically (26.6 vs. 14.8 mm, p-value 0.28, Mann-Whitney Test) regardless of their TVP index or CA IX expression level. Moreover,SO2 was calculated 70.9% for signals located inside tumors and 85.5% for signals associated with subcutaneous vessels in the same breast (p-value Conclusion: This is the largest clinical study of PAM till today and the correlation between histological profile of hypoxia and tumor microvasculature with PAM signal visibility as well as tissue SO2 seems promising. However, the improvement of techniques and resolution is necessary to develop a more clinically applicable non-invasive functional breast imaging modality for analyzing breast tumor vasculature and hypoxia. Some of these improvements have taken place for the second generation PAM (PAM-02) which is now under a clinical evaluation study. Citation Format: Elham Fakhrejahani, Masae Torii, Yasufumi Asao, Iku Yamaga, Toshiyuki Kitai, Masako Kataoka, Shotaro Kanao, Masahiro Takada, Tsuyoshi Shiina, Masakazu Toi. Clinical, anatomical and histological characteristics of breast lesions visualized by photoacoustic mammography; first clinical study in CK project [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P3-01-02.

Abstract 388: Morphological assessment of tumor-associated vessels: Automated image processing of whole tumor sections in invasive breast cancer.

Proceedings: AACR 104th Annual Meeting 2013; Apr 6-10, 2013; Washington, DC Clinical benefit of anti-angiogenic therapies in different cancers is a result of normalizing the tumor vasculature by these agents. However, tumor vascular imaging modalities in human cancers are in their early evaluating phases. Photoacoustic mammography (PAM) is a newly introduced technique which can image the hemoglobin distribution and oxygenation state inside breast tissue with high spatial resolution. We have previously presented the oxygenation data of our first study in breast cancers by PAM in European Congress of Radiology (ECR) 2011. Here we report the morphological assessment of tumor-associated vessels in 81 histological sections of 20 invasive breast carcinomas. The histological slides were scanned with the magnification of 200 after CD31 immunohistochemical staining. For automated CD31 analysis, 105 jpeg files were extracted from each virtual slide and were changed to binary images with Image Pro-Plus 7.0 software. The optimal hue, saturation and intensity (HSI) of stained structures was selected for each virtual slide. Each binary file was divided into 16 squares of 0.84mm x 0.84 mm size. Total number, area and length of stained structures were calculated in each square. Patients’ age ranged from 36 to 84 years.19 cases were invasive ductal carcinomas and 1 case was invasive lobular carcinoma. Nine cases were luminal A, 9 cases luminal B and 2 cases were HER-2 subtype. For Ki67 scoring ndex,5 cases had <5%, 9 cases between 5-20% and 6 cases had more than 20%. There was a wide heterogeneity among total of 2.95 x10 6 analyzed tumor-associated vessels in 81 histological sections. View this table: Table-1 shows the characteristics and distribution pattern. Additional analysis of shape and pericyte coverage in these vessels is on its way. In conjunction with oxygenation data obtained by PAM,this can be used as a tumor vessel normalization assessment tool which can be used as a biomarker in anti-angiogenic treatments in breast cancers. Citation Format: Elham Fakhrejahani, Yasufumi Asao, Iku Yamaga, Masae Torii, Masahiro Takada, Toshiyuki Kitai, Tomoharu Sugie, Masakazu Toi. Morphological assessment of tumor-associated vessels: Automated image processing of whole tumor sections in invasive breast cancer. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 388. doi:10.1158/1538-7445.AM2013-388

Abstract P2-03-09: Tissue hemoglobin oxygen saturation measured by photoacoustic mammography correlates with microvasculature properties assessed by histological image analysis, a preliminary study

Background: Photoacoustic mammography (PAM) is a new optical imaging technology with the potential of imaging tumor vasculature in breast cancer.The technique is applicable to the measurement of hemoglobin oxygen saturation (SO2).We have previously published an initial clinical result using a prototype machine (Canon Inc.,Tokyo, Japan) in breast cancer. We have also presented the morphological characteristics of tumor vessels analyzed by automated image processing during AACR annual meeting 2013. Here, we report the oxygenation data obtained by PAM in relation with histological assessment of tumor vasculature and hypoxia. Methods: Forty–four breast lesions were evaluated by PAM in this IRB approved prospective study at Kyoto University Hospital, Japan. PAM evaluation was performed on both breasts whenever possible.SO2 was calculated in region of interest after the radiologist confirmed the signal was associated with the tumor location in MRI images. For the normal breast, signals obtained at the same depth of the tumor, were selected. Eighty–one histological sections from 20 available invasive carcinoma tissues at the time of this analysis were selected for immunohistochemical assessment of hypoxia by anti –carbonic anhydrase IX (CA IX) and tumor vasculature image analysis using anti–CD31. Five 0.5 mm2 area of each cancer and 3 area of normal mammary tissue associated with the same lesion were randomly selected from different sections. Total vascular area in each square was calculated by using Image Pro-Plus 7.0 software (Media Cybernetics, USA). Tumor–to–normal vascular area ratio (T/N VA) was calculated for each lesion as an index for tumor blood supply. Results: Patients’ age ranged from 36 to 83 years old. Tumor associated signals were detected by PAM in 18 out of 20 lesions for which tissues were available for histological examination. SO2 in tumor was calculated 70.6% ±13.2 and 83.3% ±10.7 in the normal counterpart. While T/N VA ranged between 0.11 to 1.14, it was almost 3 times lower in lesions with CA IX positive cytoplasmic membrane staining (0.21 vs 0.7, p–value = 0.021 Mann–Whitney Test). Normalized tumor SO2 (tumor SO2/normal counterpart SO2) was significantly lower in the group with lower T/N VA (0.9 vs. 0.8,p–value = 0.045, Student T–test). To better evaluate the accuracy of PAM measurement in calculating SO2,3780 tumor-associated and 2835 normal microvessels were analyzed by image analysis software. Tumor associated vessels had significantly smaller area (p-value<0.001) and vessels with irregular lumens were more frequent in tumor (76.5% vs 19.6% p-value <0.001) compatible with lower SO2 in tumor areas. Conclusion: Although the future result of our ongoing clinical studies of PAM measurement in breast cancer patients are more than necessary, the strong correlation between histological evaluation of hypoxia and angiogenesis with PAM measurement of oxygenation shows the promising prospective for clinical application of this new technology in breast cancer. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P2-03-09.