- Email: [email protected]
Approaches to improving breast cancer diagnosis using a high resolution, breast specific gamma camera
2004 Workshop on the Nuclear Radiology of Breast C:mcer Rome (Italy) October 22-23, 2004
<~PhysicaMedical) 9Vol. XXI, Supplement 1, 2006
Approaches to Improving Breast Cancer Diagnosis Using a High Resolution, Breast Specific Gamma Camera Rachel F Brem MD, Katherine H. Michener BA, Grace Zawistowski BA Department of Radiology, Breast Imaging and Intervention Center, the George Washington University Medical Center, 2150 Pennsylvania Avenue, N~ Washington, DC 20037
Abstract Mammog:~phy remains the gold standard in breast cancer detection, although there remains a challenge for improvement in sensitivity .,f breast cancer detection and diagnosis. Although mammography is the most frequently utilized examination to screen for l:reast cancer, which has resulted in a reduction of breast cancer mortality;, still some cancers are unable to be visualized on m~ mmographic images. Mammography films are interpreted using an anatomic approach. A new approach to breast cancer diag ~osis utilizes a breast specific gamma camera to measure radiotracer uptake of abnormal tissue in the breast in patients with an abnormal mammogram or palpable mass using technetium sestamibi. Images are taken using the same positioning technm ~es as mammography for comparison of both types of images. Multi-Institutional trials using a traditional gamma camera dei:mnstrated potential for this approach. However, the inability of traditional gamma camera intrinsic resolution and non-optimi:'ed breast imaging limited scintimammography. Therefore, a breast specific, high resolution gamma camera was developed :o overcome these limitations. Results cf clinical studies evaluating BSGI are promising and are increasingly being used. Additionally,means for minimally invasive imaging-guided acquisition of tissue are being developed so that biopsies will be available for areas of interest based on radiotracc r uptake. Breast :;t,ecificgamma camera nuclear imaging of the breast is a developing and increasingly utilized approach to improving breast car o.~r detection and diagnosis. KuYwoiu);: Breast specific gamma imaging, scintimammography, Breast cancer. Breast c~ncer is the most c o m m o n cancer and the secot~d most c o m m o n cause o f cancer death in w o m e n in the United States [1]. Worldwide, breast cxtcer is an important healthcare issue that is respondble for significant morbidity, mortality and heah:hcare dollars spent. W h e n breast cancer is diagno+:ed in its earliest stages it is a curable disease with a 5-year survival o f greater than 90% [1]. M a m m o g r a p h y is the m o s t widely utilized means o f screening for breast cancer, which results in up to 440/0 reduction in breast cancer mortality. However, mammo~4raphy is an imperfect examination with 10-15% of breast cancers not visible on m a m m o graphic examination [2, 3, 4]. Additionally, o f suspicious m a m m o g r a p h i c findings which undergo biopsy, onl'," 20-30% result in a diagnosis o f cancer as m a m m o g r a p h y can only identify abnormalities and not definitively differentiate benign from malignant findings Therefore, there are two basic challenges to improve breast cancer diagnosis. Firstly, to improve the sensitivity o f breast cancer detection in order to detect more cancers and secondly, to improve the specificity o f breast cancer diagnosis in order to improve the ability to differentiate benign from malignant lesions and thereby decrease the n u m b e r o f biopsies performed for benign lesions. Both m a m m o g r a p h y , which is the most c o m m o n imaging modality for detection breast cancer, and
ultrasound, the second most c o m m o n imaging modality utilized in breast cancer diagnosis, are based on the anatomic appearance of the breast. However, the concept o f utilizing a physiologic approach to improve the sensitivity and specificity o f breast cancer diagnosis was proposed. The incidental observation was made that some w o m e n undergoing cardiac evaluation with technetium sestamibi were found to have focal radiotracer uptake in their breast due to a focus of breast cancer. As a result o f this observation a multi-institutional trial was initiated which d e m o n s t r a t e d that s c i n t i m a m m o g r a p h y had potential to improve breast cancer diagnosis with a sensitivity and specificity o f 75.4% and 82.7% [5]. However, in this study a traditional g a m m a camera was used which had limited resolution o f approximately 1 cm as well as design limitations for optimally imaging the breast. In fact, sub-centimeter, non-palpable breast cancers had a sensitivity o f only approximately 20%. However, prominsing results of the study d e m o n s t r a t e d that there was no statistically significant difference in cancer detection in w o m e n with dense versus w o m e n with non-dense breasts. M a m m o g r a p h i c detection of breast cancer is markedly hindered by the presence o f dense breasts as both dense breast tissue and breast cancer can have a similar m a m m o g r a p h i c appearance and thereby markedly diminishes the ability to identify
Address correspondence: Rachel E Brem, M. D., Breast Imaging and Interventional Center, Department of Radiology,The George Washington University, 2150 Pennsylvania Ave, NW, Washington DC 20037 (USA). Phone: 202-741-3031, Fax 202-741-3029. E-mail: rt,l :[email protected] 2. Nuclear Medicine Perspective
17
Rachel E Brem, Katherine H. Michener: Approaches to Improving Breast Cancer Diagnosis Using a High Resolution breast cancer. The finding that cancers detected scintimammographically were not impacted by the presence of dense breast tissue demonstrated the potential to utilize this fundamentally different approach to improving one of the greatest challenges in breast cancer diagnosis, that is, cancer detection in w o m e n with dense breasts. Although the limitations of the instrumentation did not allow for the detection of smaller breasts cancer, the results of the multi-institutional study demonstrated sufficient proof of principle to further pursue this approach to breast cancer diagnosis, that is a physiologic approach, as compared to the anatomic approach. Additionally, as a result of these findings, the United States Food and Drug Administration approved the use of tech-
Fro. 1A. High resolution, breast-specific gamma camera; lB. Patient being imaged in the camera in projections comparable to mammography 18
netium sestamibi and stated: , S c i n t i m a m m o g r a p h y is indicated for planar imaging as a second line diagnostic tool after m a m m o g r a p h y to assist in the evaluation of breast lesions in patients with an abnormal m a m m o g r a m or a palpable mass~. A dedicated, high-resolution, breast specific gamma camera was developed that utilizes a molecular approach to breast cancer diagnosis while overcoming the limitations of using a traditional g a m m a camera to image the breasts. This camera utilizes position sensitive photomultiplier tubes [6], has subcentimeter resolution, and is designed to image the breasts in positions comparable to m a m m o g r a p h y with optimal imaging of the medial and posterior portions of the breasts (Figure 1): Additionally, this camera allows for imaging in orthogonal views such that the precise location of the lesion could be identified within the breast. Initial reports with the high-resolution, breast-specific g a m m a camera demonstrated that the limitations of using a traditional g a m m a camera to image the breast could be overcome and that there were improvements in cancer diagnosis using the dedicated camera [7]. The initial report using molecular breast imaging was designed such that imaging was initially performed with the traditional g a m m a camera and only afterwards, was the high-resolution camera utilized. This resulted in imaging the patients with the highresolution camera at a time when the activity of the radio-tracer was decreasing and not optimal. Even under these circumstances, there was improvem e n t in cancers diagnosis. Studies demonstrated that the sensitivity of breast cancer detection using a dedicated, high-resolution breast specific g a m m a camera resulted in an improvement from 64 tO 79% when compared to a traditional g a m m a camera. Additionally, in the multi-institutional trial that utilized a traditional g a m m a camera, the mean cancer size was 2.2 cm and the smallest cancer detected was 0.7 cm. By way of contrast, the m e a n cancer size in the study investigating the high-resolution, breast-specific g a m m a camera was 1.1 cm with the smallest cancer detected measuring 0.3 m m [7]. Furthermore, 71% of the cancers detected were not palpable. These improvements resulted in an approach which utilizes molecular breast imaging in a manner that is clinically useful and which can be integrated into a clinical breast imaging program as it detected the sub-centimeter cancers, i.e. those with the best prognosis, and that allowed for imaging in positions comparable to m a m m o g r a p h y such that m a m m o g r a p h i c findings could be directly correlated with the findings on molecular breast imaging. As a result of these findings, we chose to evaluate the potential impact of screening high-risk women with the high- resolution g a m m a camera and molecular breast imaging. In our study all w o m e n who were at increased risk for breast cancer with
<(Physica Medica)~ . Vol. XXI, Supplement 1, 2006
Left CC
Left ML(
FIG.2.8 rr I i1invasive ductal carcinoma (arrow) detected in a high risk patient with normal mammogram and physical examination.
a normal m a m m o g r a m and a normal physical examinatioa were eligible. This included w o m e n with dense and non-dense breasts. W o m e n at increased risk wet'(: defined as those with risk factors equal to the ri,'~k necessary to be included in the breast cancer prevention trials evaluating Tamoxifen and Raloxifer., i.e. estrogen receptor modulators. In our study, 2/94 patients with normal m a m m o g r a m s and ph}s [cal examination were found to have occult, oth,::rwise undetectable breast cancer (Figure 2). This i:; in contrast to 2-6 cancers for 1,000 screening maim mograms [8]. This study d e m o n s t r a t e d a 100% se~:sitivity, 84.8% specificity and 100% negative predictive value. Undoubtedly, the 100% sensitivity is due to the small sample size in this study. Howeve~ these findings demonstrate the potential to use raolecular breast imaging to improve breast cancer d5 agnosis as well as to diagnose breast cancer earlie:=. Additional, larger and multi-institutional studies a~'e n e e d e d to better define the sensitivity, specificitT, positive predictive value and negative predictive v:~lue of this developing technology. In ord,-'.r to integrate molecular breast imaging into clinical breast imaging, there needs to be an approacl- to tissue acquisition o f areas o f increased radiotracer uptake in order to determine pathologically w h e t h e r the uptake is due to cancer. As it is clear that not all significant lesions detected by molecular breast imaging can be visualized mammographically and therefore a direct approach to minimally invasive biopsy o f these lesions is needed. In collaboration with Drs. Stan Majewski, Benjamin Welch and colleagues at the Jefferson National Accelerator L a b o r a t o r y w e have modified a stereot actic biopsy table to allow for direct tissue sampling of lesions that are detected only with g a m m a imaging. This minimally invasive approach to determining w h e t h e r a region o f increased radiotracer u ptake is due to cancer or a benign cause o f increased tracer uptake such as fi brocystic change
or fat necrosis is necessary for direct integration into clinical practice. We are currently investigating direct biopsy from stereotactically obtained g a m m a images with the high resolution, breastspecific g a m m a camera. Preliminary data in phantoms demonstrates accurate targeting and specimen acquisition using this approach. In summary, molecular breast imaging is a potentially powerful approach to improving breast cancer detection. Additional studies are needed to further understand this technology. Furthermore, direct biopsy of areas of increased radiotracer uptake are being developed to allow for the complete integration of this approach to clinical practice. REFERENCES
[1] [2]
[3] [4] [5]
American Cancer Society Breast Cancer Facts and Figures. 2005. http://www.cancer.org/ National Institutes of Health Consensus Development Conference Statement. Jan 21-23, 1997, 103. Breast cancer screening for women ages 40-49. http://consensus. nih.gov Sickles E A. Mammographic features of early breast cancer. Am J Roentgenol 1984: 143; 461-464. Niloff P H, Sheiner N M. False-negative mammograms in patients with breast cancer. CanJ Surg 1981: 24; 50-52. KhalkhaliI, Villanueva-Meyer J, Edell S L, Connolly J L, Schnitt S J, Baum J K, Houlihan M J, Jenkins R M, Haber S B. Diagnotic accuracy of 99m Tc-sestamibi breast imaging: multicenter trial results. J Nucl Med 2000: 41; 1973-1979.
[6]
[7]
Majewski S, Curran E, Keppel D, Kross B, Pulumbo A, Popov V, Weisenberger A G, Welch B, Wojcik R, Williams M B, Goode A R, More M, Zang G. Optimization of dedicated scintimammography procedure using detector prototypes and compressible phantoms. Nuclear Science Symposium Conference Record, 2000 IEEE; 3: 22/62-22/66. Brem R F, Schoonjans J M, Kieper D A, Majewski S. High-resolution scintimammography: a pilot study. J Nucl Med 2002: 43; 909-915.
[8]
Gur D, SumkinJ H, Hardesty L A, Clearfield R J, Cohen C S, Ganott M A, Hakim C M, Harris K M, Poller W R, Shah R, Wallace L R Rockette H E. Recall and detection rates in screening mammography. Cancer 2004: 100; 1590-1594.
19
<~PhysicaMedical) 9Vol. XXI, Supplement 1, 2006
Approaches to Improving Breast Cancer Diagnosis Using a High Resolution, Breast Specific Gamma Camera Rachel F Brem MD, Katherine H. Michener BA, Grace Zawistowski BA Department of Radiology, Breast Imaging and Intervention Center, the George Washington University Medical Center, 2150 Pennsylvania Avenue, N~ Washington, DC 20037
Abstract Mammog:~phy remains the gold standard in breast cancer detection, although there remains a challenge for improvement in sensitivity .,f breast cancer detection and diagnosis. Although mammography is the most frequently utilized examination to screen for l:reast cancer, which has resulted in a reduction of breast cancer mortality;, still some cancers are unable to be visualized on m~ mmographic images. Mammography films are interpreted using an anatomic approach. A new approach to breast cancer diag ~osis utilizes a breast specific gamma camera to measure radiotracer uptake of abnormal tissue in the breast in patients with an abnormal mammogram or palpable mass using technetium sestamibi. Images are taken using the same positioning technm ~es as mammography for comparison of both types of images. Multi-Institutional trials using a traditional gamma camera dei:mnstrated potential for this approach. However, the inability of traditional gamma camera intrinsic resolution and non-optimi:'ed breast imaging limited scintimammography. Therefore, a breast specific, high resolution gamma camera was developed :o overcome these limitations. Results cf clinical studies evaluating BSGI are promising and are increasingly being used. Additionally,means for minimally invasive imaging-guided acquisition of tissue are being developed so that biopsies will be available for areas of interest based on radiotracc r uptake. Breast :;t,ecificgamma camera nuclear imaging of the breast is a developing and increasingly utilized approach to improving breast car o.~r detection and diagnosis. KuYwoiu);: Breast specific gamma imaging, scintimammography, Breast cancer. Breast c~ncer is the most c o m m o n cancer and the secot~d most c o m m o n cause o f cancer death in w o m e n in the United States [1]. Worldwide, breast cxtcer is an important healthcare issue that is respondble for significant morbidity, mortality and heah:hcare dollars spent. W h e n breast cancer is diagno+:ed in its earliest stages it is a curable disease with a 5-year survival o f greater than 90% [1]. M a m m o g r a p h y is the m o s t widely utilized means o f screening for breast cancer, which results in up to 440/0 reduction in breast cancer mortality. However, mammo~4raphy is an imperfect examination with 10-15% of breast cancers not visible on m a m m o graphic examination [2, 3, 4]. Additionally, o f suspicious m a m m o g r a p h i c findings which undergo biopsy, onl'," 20-30% result in a diagnosis o f cancer as m a m m o g r a p h y can only identify abnormalities and not definitively differentiate benign from malignant findings Therefore, there are two basic challenges to improve breast cancer diagnosis. Firstly, to improve the sensitivity o f breast cancer detection in order to detect more cancers and secondly, to improve the specificity o f breast cancer diagnosis in order to improve the ability to differentiate benign from malignant lesions and thereby decrease the n u m b e r o f biopsies performed for benign lesions. Both m a m m o g r a p h y , which is the most c o m m o n imaging modality for detection breast cancer, and
ultrasound, the second most c o m m o n imaging modality utilized in breast cancer diagnosis, are based on the anatomic appearance of the breast. However, the concept o f utilizing a physiologic approach to improve the sensitivity and specificity o f breast cancer diagnosis was proposed. The incidental observation was made that some w o m e n undergoing cardiac evaluation with technetium sestamibi were found to have focal radiotracer uptake in their breast due to a focus of breast cancer. As a result o f this observation a multi-institutional trial was initiated which d e m o n s t r a t e d that s c i n t i m a m m o g r a p h y had potential to improve breast cancer diagnosis with a sensitivity and specificity o f 75.4% and 82.7% [5]. However, in this study a traditional g a m m a camera was used which had limited resolution o f approximately 1 cm as well as design limitations for optimally imaging the breast. In fact, sub-centimeter, non-palpable breast cancers had a sensitivity o f only approximately 20%. However, prominsing results of the study d e m o n s t r a t e d that there was no statistically significant difference in cancer detection in w o m e n with dense versus w o m e n with non-dense breasts. M a m m o g r a p h i c detection of breast cancer is markedly hindered by the presence o f dense breasts as both dense breast tissue and breast cancer can have a similar m a m m o g r a p h i c appearance and thereby markedly diminishes the ability to identify
Address correspondence: Rachel E Brem, M. D., Breast Imaging and Interventional Center, Department of Radiology,The George Washington University, 2150 Pennsylvania Ave, NW, Washington DC 20037 (USA). Phone: 202-741-3031, Fax 202-741-3029. E-mail: rt,l :[email protected] 2. Nuclear Medicine Perspective
17
Rachel E Brem, Katherine H. Michener: Approaches to Improving Breast Cancer Diagnosis Using a High Resolution breast cancer. The finding that cancers detected scintimammographically were not impacted by the presence of dense breast tissue demonstrated the potential to utilize this fundamentally different approach to improving one of the greatest challenges in breast cancer diagnosis, that is, cancer detection in w o m e n with dense breasts. Although the limitations of the instrumentation did not allow for the detection of smaller breasts cancer, the results of the multi-institutional study demonstrated sufficient proof of principle to further pursue this approach to breast cancer diagnosis, that is a physiologic approach, as compared to the anatomic approach. Additionally, as a result of these findings, the United States Food and Drug Administration approved the use of tech-
Fro. 1A. High resolution, breast-specific gamma camera; lB. Patient being imaged in the camera in projections comparable to mammography 18
netium sestamibi and stated: , S c i n t i m a m m o g r a p h y is indicated for planar imaging as a second line diagnostic tool after m a m m o g r a p h y to assist in the evaluation of breast lesions in patients with an abnormal m a m m o g r a m or a palpable mass~. A dedicated, high-resolution, breast specific gamma camera was developed that utilizes a molecular approach to breast cancer diagnosis while overcoming the limitations of using a traditional g a m m a camera to image the breasts. This camera utilizes position sensitive photomultiplier tubes [6], has subcentimeter resolution, and is designed to image the breasts in positions comparable to m a m m o g r a p h y with optimal imaging of the medial and posterior portions of the breasts (Figure 1): Additionally, this camera allows for imaging in orthogonal views such that the precise location of the lesion could be identified within the breast. Initial reports with the high-resolution, breast-specific g a m m a camera demonstrated that the limitations of using a traditional g a m m a camera to image the breast could be overcome and that there were improvements in cancer diagnosis using the dedicated camera [7]. The initial report using molecular breast imaging was designed such that imaging was initially performed with the traditional g a m m a camera and only afterwards, was the high-resolution camera utilized. This resulted in imaging the patients with the highresolution camera at a time when the activity of the radio-tracer was decreasing and not optimal. Even under these circumstances, there was improvem e n t in cancers diagnosis. Studies demonstrated that the sensitivity of breast cancer detection using a dedicated, high-resolution breast specific g a m m a camera resulted in an improvement from 64 tO 79% when compared to a traditional g a m m a camera. Additionally, in the multi-institutional trial that utilized a traditional g a m m a camera, the mean cancer size was 2.2 cm and the smallest cancer detected was 0.7 cm. By way of contrast, the m e a n cancer size in the study investigating the high-resolution, breast-specific g a m m a camera was 1.1 cm with the smallest cancer detected measuring 0.3 m m [7]. Furthermore, 71% of the cancers detected were not palpable. These improvements resulted in an approach which utilizes molecular breast imaging in a manner that is clinically useful and which can be integrated into a clinical breast imaging program as it detected the sub-centimeter cancers, i.e. those with the best prognosis, and that allowed for imaging in positions comparable to m a m m o g r a p h y such that m a m m o g r a p h i c findings could be directly correlated with the findings on molecular breast imaging. As a result of these findings, we chose to evaluate the potential impact of screening high-risk women with the high- resolution g a m m a camera and molecular breast imaging. In our study all w o m e n who were at increased risk for breast cancer with
<(Physica Medica)~ . Vol. XXI, Supplement 1, 2006
Left CC
Left ML(
FIG.2.8 rr I i1invasive ductal carcinoma (arrow) detected in a high risk patient with normal mammogram and physical examination.
a normal m a m m o g r a m and a normal physical examinatioa were eligible. This included w o m e n with dense and non-dense breasts. W o m e n at increased risk wet'(: defined as those with risk factors equal to the ri,'~k necessary to be included in the breast cancer prevention trials evaluating Tamoxifen and Raloxifer., i.e. estrogen receptor modulators. In our study, 2/94 patients with normal m a m m o g r a m s and ph}s [cal examination were found to have occult, oth,::rwise undetectable breast cancer (Figure 2). This i:; in contrast to 2-6 cancers for 1,000 screening maim mograms [8]. This study d e m o n s t r a t e d a 100% se~:sitivity, 84.8% specificity and 100% negative predictive value. Undoubtedly, the 100% sensitivity is due to the small sample size in this study. Howeve~ these findings demonstrate the potential to use raolecular breast imaging to improve breast cancer d5 agnosis as well as to diagnose breast cancer earlie:=. Additional, larger and multi-institutional studies a~'e n e e d e d to better define the sensitivity, specificitT, positive predictive value and negative predictive v:~lue of this developing technology. In ord,-'.r to integrate molecular breast imaging into clinical breast imaging, there needs to be an approacl- to tissue acquisition o f areas o f increased radiotracer uptake in order to determine pathologically w h e t h e r the uptake is due to cancer. As it is clear that not all significant lesions detected by molecular breast imaging can be visualized mammographically and therefore a direct approach to minimally invasive biopsy o f these lesions is needed. In collaboration with Drs. Stan Majewski, Benjamin Welch and colleagues at the Jefferson National Accelerator L a b o r a t o r y w e have modified a stereot actic biopsy table to allow for direct tissue sampling of lesions that are detected only with g a m m a imaging. This minimally invasive approach to determining w h e t h e r a region o f increased radiotracer u ptake is due to cancer or a benign cause o f increased tracer uptake such as fi brocystic change
or fat necrosis is necessary for direct integration into clinical practice. We are currently investigating direct biopsy from stereotactically obtained g a m m a images with the high resolution, breastspecific g a m m a camera. Preliminary data in phantoms demonstrates accurate targeting and specimen acquisition using this approach. In summary, molecular breast imaging is a potentially powerful approach to improving breast cancer detection. Additional studies are needed to further understand this technology. Furthermore, direct biopsy of areas of increased radiotracer uptake are being developed to allow for the complete integration of this approach to clinical practice. REFERENCES
[1] [2]
[3] [4] [5]
American Cancer Society Breast Cancer Facts and Figures. 2005. http://www.cancer.org/ National Institutes of Health Consensus Development Conference Statement. Jan 21-23, 1997, 103. Breast cancer screening for women ages 40-49. http://consensus. nih.gov Sickles E A. Mammographic features of early breast cancer. Am J Roentgenol 1984: 143; 461-464. Niloff P H, Sheiner N M. False-negative mammograms in patients with breast cancer. CanJ Surg 1981: 24; 50-52. KhalkhaliI, Villanueva-Meyer J, Edell S L, Connolly J L, Schnitt S J, Baum J K, Houlihan M J, Jenkins R M, Haber S B. Diagnotic accuracy of 99m Tc-sestamibi breast imaging: multicenter trial results. J Nucl Med 2000: 41; 1973-1979.
[6]
[7]
Majewski S, Curran E, Keppel D, Kross B, Pulumbo A, Popov V, Weisenberger A G, Welch B, Wojcik R, Williams M B, Goode A R, More M, Zang G. Optimization of dedicated scintimammography procedure using detector prototypes and compressible phantoms. Nuclear Science Symposium Conference Record, 2000 IEEE; 3: 22/62-22/66. Brem R F, Schoonjans J M, Kieper D A, Majewski S. High-resolution scintimammography: a pilot study. J Nucl Med 2002: 43; 909-915.
[8]
Gur D, SumkinJ H, Hardesty L A, Clearfield R J, Cohen C S, Ganott M A, Hakim C M, Harris K M, Poller W R, Shah R, Wallace L R Rockette H E. Recall and detection rates in screening mammography. Cancer 2004: 100; 1590-1594.
19