System Description

The digital x-ray and gamma ray detectors are mounted on a conventional upright mammographic unit (Lorad M-III). The digital mammography detector consists of six butted modules arranged in a 2 x 3 array. Each module contains a 2k x 2k CCD bonded to the small end of a 3.3:1 fiber optic taper. The large ends of the six tapers are optically coupled to a common 20 cm x 30 cm Gd2O2S:Tb phosphor. The taper sides are square and polished to achieve intra-module gaps of less than 0.030 mm. The pixel size is 0.046 microns, and a full digital mammogram contains 4224 x 6264 pixels, with 16 bits of intensity resolution per pixel. The inactive border at the patient’s chest wall is 7 mm wide, only slightly more than that of current screen-film mammography systems. Performance of the detector in terms of the spatial frequency dependent detective quantum efficiency has been published elsewhere [Williams 1999].

The dedicated gamma camera of the dual modality breast imaging system consists of 16 tiled position sensitive photomultiplier tubes (PSPMTs), arranged in a 4x4 array. The PSPMT array is optically coupled via acrylic light guides to a 30 x 30 array of NaI(Tl) crystals. Each crystal is 3.0 mm square and 6 mm thick, and the crystal center-to-center spacing is 3.3 mm. Because of the compact design of the PSPMTs and camera housing, the gamma camera is able to image to within eight mm of the patient’s chest wall, a performance comparable to that of x-ray mammography and far superior to that of clinical gamma cameras (tens of cm). The camera uses a high resolution etched tungsten collimator with square apertures. The intrinsic detector spatial resolution is 3.2 mm FWHM, and the camera sensitivity is 300 cpm/_Ci.

Image Acquisition

In a typical dual modality study, the patient is first positioned in a manner similar to that used in conventional mammography, except with substantially reduced breast compression. With the gamma camera pivoted out of the path of the x-ray beam (see Figure 1), a digital mammogram is obtained, and the image appearing on the acquisition monitor is used to verify that the portion of the breast containing the suspicious lesion is approximately centered in the field of view. Next, the gamma camera is swung back into position above the breast and is lowered via a linear translation stage to a position just above the compression paddle (see Figure 2). Following acquisition of the gamma emission image (~10 minute acquisition time), the x-ray and gamma ray images are corrected for translational and rotational offsets and their pixel sizes are re-scaled to a common value using a stored co-registration file. X-ray and gamma ray images are separately intensity scaled for optimal display of the area of the lesion, then the common areas of each are fused into a single dual modality image.

Current Status

The dual modality breast imaging system is now undergoing a small scale (n=40) clinical evaluation at UVa’s Diagnostic Center for Women. Volunteers who are scheduled to undergo breast biopsy are eligible to participate, and are scanned prior to biopsy. We will report on our early experience with the system, and present both phantom and clinical images.


Figure 1. A typical dual modality study: The patient is first positioned in a manner similar to that used in conventional mammography, except with substantially reduced breast compression. The gamma camera is pivoted out of the path of the x-ray beam, a digital mammogram is obtained.


Figure 2. The gamma camera is swung back into position above the breast and is lowered via a linear translation stage to a position just above the compression paddle. acquisition of the gamma emission image is acquired (~10 minute acquisition time).


Figure 3. Mediolateral view of the right breast of a 49 year old woman. Shown are the digital mammogram (lower image), gamma emission image (upper left), and fused image (upper right). The pathology report revealed infiltrating ductal carcinoma.

Reference

MB Williams, PU Simoni, L Smilowitz, M Stanton, WC Phillips, and A Stewart, (1999),
Analysis of the Detective Quantum Efficiency of a Developmental Detector for Digital
Mammography, Med. Phys., 26(11):2273-2285.




Infrared Spectroscopy of Human Cells and Tissue: Detection of Disease

Max Diem, Department of Chemistry and Biochemistry, City University of New York, Hunter College, 695 Park Avenue, New York, NY 10021 (diemhc@aol.com)

During the past decade, several research groups have suggested infrared microspectroscopy as a novel diagnostic tool for pathology and cytology. The underlying assumption for such an application is, of course, that the spectra of normal and diseased tissue exhibit spectral differences. While numerous results have pointed to the fact that normal and cancerous tissues can be distinguished by infrared spectral measurements, we have shown recently that the differences observed between normal and cancerous tissue resulted from different populations of cells exhibiting various levels of activity.

Figure 1

Normal and cancerous tissue could be distinguished in samples from the cervix, prostate, liver, breast and colon. We have developed infrared spectral imaging technology that permits such tissue samples to be analyzed in a totally unsupervised, algorithm-driven approach, using multivariate statistical methods. Shown in the Figure 1 are images of an H&E stained prostate section (left) and a corresponding infrared spectral map (right) obtained from the identical tissue spot prior to staining. All anatomical features of the stained image can be discerned in the infrared spectral image, and stages of disease can be detected..

Figure 2

In order to understand the origins of the observed spectral differences between normal and abnormal tissue, we have carried out mapping experiments of cultured normal human skin fibroblasts and sarcoma cells. In these, up to 400 individual spectra were collected for individual cells, at a spatial resolution as high as 8 x 8 mm2. Spectral differences between the nucleus and the cytoplasm allowed us to monitor the activity level of cells, and gain a detailed understanding of the spectral changes that occur at the different levels of activity. These recent results confirm the earlier observation of spectral changes caused by the stage of cells in the division cycle, and create a sound foundation for the use of infrared microspectroscopy as a diagnostic tool.




Intramuscular and Intratumoral Gene Transfer Using Low Voltage Electric Pulses and Cationic Lipids

D. Scherman, Ph.D.
UMR 7001 CNRS / ENSCP Pierre et Marie Curie University / Aventis Company, Vitry-sur-Seine, France. Email : daniel.scherman@aventis.com or scherman@ext.jussieu.fr

Gene delivery to tumors and to skeletal muscle is a promising strategy for the treatment of cancer. Anti-oncogenic genes such as tumor suppressor p53 can be administered directly by intratumoral injection. Alternatively, genes coding for immunostimulatory cytokines, for antigenic epitopes, or for antiangiogenic factors might be administered either within the tumor, or at distal site of the tumor. Of particular interest for tumor treatment is the systemic secretion of antiangiogenic factors by in vivo transfected skeletal muscle.

Non-viral gene transfer for gene therapy means using a plasmid DNA as gene expression vector, in association or not with a chemical delivery vector or with physical administration technique, is a rapidly expanding field. Cationic lipids are the most commonly used chemical DNA delivery vectors. They self-associate with plasmids and form nanostructures called "lipoplexes". Successful non viral gene transfer requires mastering several steps: preparation, purification and formulation of the therapeutic DNA and synthetic vector, plasmid administration, plasmid access to target cells, then intracellular penetration and nuclear localization. We will present our contribution in different aspects: 1) optimized plasmid for biosafety and bioavailability; 2) effect of plasmid size on lipoplexe structure and on gene delivery efficiency; and 3) plasmid electrotransfer to skeletal muscle and tumor tissue.

Increased gene transfer by minimal size « minicircle » plasmids: in vivo results and structural physico-chemical studies. Minicircles are a new form of supercoiled DNA molecule for non-viral gene transfer which have neither bacterial origin of replication nor antibiotic resistance marker. They are thus smaller and potentially safer than the standard plasmids currently used in gene therapy. They were obtained in E. coli by att site-specific recombination mediated by the phage _ integrase, which was used to excise the expression cassette from the unwanted plasmid sequences. Two minicircles containing the luciferase or b-galactosidase gene under the control of hCMV-IE enhancer/promoter gave 2-5 times more reporter gene activity than the unrecombined plasmid in NIH3T3 cells in rabbit smooth muscle cells. Moreover, injection into mouse cranial tibial muscle, or into human head and neck carcinoma grafted in nude mice resulted in 15-50 fold more reporter gene expression with minicircles than with the unrecombined plasmid or with larger plasmids. Histological analysis in muscle showed there were more transfected myofibers with minicircles than with unrecombined plasmid. Similar results were observed using a pCOR plasmid devoid of prokaryotic replication origin.

Effect of plasmid size on lipoplexe structure and on gene delivery efficiency. In order to further investigate the mechanism of the increase in gene transfer efficiency observed with minicircles, we have explored the physicochemical properties of cationic lipid-DNA particles, with plasmids ranging from 900 to 52,500 base pairs. The morphological and structural features of the lipopolyamine-DNA complexes did not depend on plasmid DNA length. On the other hand, their gene transfer capacity was strongly affected by the size and number of plasmid DNA molecules which were sandwiched between the lipid bilayers. The most effective lipopolyamine-DNA complexes for gene transfer were those containing the largest amount of the shortest plasmid DNA.

Plasmid electrotransfer to skeletal muscle and tumos. We have obtained very efficient plasmid DNA transfer in muscle fibers and tumors using square-wave electric pulses of low field strength (less than 500 V/cm) and of long duration (more than 1 ms). This « electrotransfer » method increases reporter and therapeutic gene expression by several orders of magnitude in various muscles and tumors. We will present recent results concerning the three following aspects.

Mechanism of electrotranfer. A new combination of pulses of various strength and duration leads to similar gene transfer efficiency while delivering less energy to the tissue and leading to lower electro-permabilization. Moreover, these results give insights on the mechanism of DNA entry into electrotransfered cells.

Sustained plasmatic protein secretion. Sustained blood secretion of transgenic "secreted alkaline phophatase" (SeAP) for more than 12 months has been observed, and of Factor IX were observed after a single i.m. electrotransfer. I.M. electrotransfer of a pCMV EPO pCOR plasmid led to an increase of approximately 10- to 100-fold in circulating murine erythropoietin level, as compared to naked DNA alone, and induced a stable and reproducible increase in mouse hematocrit, from 47% up to 80%.

Plasmid electrotransfer into tumors. We have confirmed the potential of long duration/low voltage pulses to transfer DNA to several murine and human tumor models (10-1200 fold increase over values obtained with "naked" DNA injection). An example of tumor growth retardation after electrotransfer of a recently discovered gene will be presented.


Keywords : gene therapy; electric pulses; electrotransfer; tumor; gene transfer




Stereotactic Radiosurgery in the Treatment of Brain Tumors

Joseph F. Emrich, MD, Radiation Oncology, Albany Medical College, Albany NY.


Stereotactic Radiosurgery (SRS) is a non invasive procedure in which a powerful precise beam of radiation is accurately directed to a brain tumor to stop its growth,, while sparing the surrounding vital brain tissue. The successful development of this advance in cancer treatment depends on :

1. High resolution imaging of the tumor and the patient’s neuro-anatomy.
2. Precise localization of the tumor in three dimensional space.
3. A stable, reproducible and accurate delivery system.

Technical advances in anatomical and functional brain imaging, stereotactic localization and radiation delivery systems have brought this treatment to fruition. Advances in treatment planning software design allow the clinician to plan and deliver treatment in a single day. Various anatomic and functional imaging modalities can be registered to stereotactic localization systems to give the clinician as much information as possible about the treatment volume. This has made SRS an effective alternative to the surgical treatment of brain metastases .The procedure is non-invasive and selective. A tumor control rate of 80% can be achieved. Patients treated in this way benefit from an increased survival and an improvement in the quality of their remaining life. It is an effective tool for the palliation of cancer patients.




Screening Mammography: Accomplishments, Expectations, Challenges

Stephen A. Feig, MD, FACR
Mount Sinai School of Medicine, The Mount Sinai Hospital , New York, NY 10029


Statistically significant benefits from screening mammography have been observed in randomized clinical trials (RCT’s). For women aged 50 years and older at entry, these include breast cancer mortality reductions of 23% in the Health Insurance Plan of Greater New York (HIP) trial and 34% in the Two-County Swedish Trial. For women aged 40-49 years at entry, these include mortality reductions of 25% in the HIP trial, 36% in the Malmo, Sweden trial, and 45% in the Gothenburg, Sweden trial. A meta-analysis of all 5 Swedish trials has shown a 29% reduction in breast cancer mortality for women in their 40’s at entry. A non-randomized study in the United Kingdom (UK) has also found statistically significant reductions in breast cancer deaths: 35% for women aged 45-46 years and 27% for those aged 45-64 years at entry.

There are several reasons why these RCT results underestimate benefit for women screened annually today. First, screening intervals in RCT’s were excessively long. Second, not all women invited to screening agreed to participate and some women in the control group obtained screening outside the trials. Third, there has been considerable improvement in mammographic technique since the 1980’s when most trials began. Many studies have shown that improvement in mammographic technique results in higher detection rates, lower interval cancer rates, and earlier stages of disease at detection.

Unlike survival rate studies for cancers detected among women who volunteer for screening, RCT’s are unaffected by lead time bias, length bias, and selection bias. Differences of 1-5 months age between study and control group women in RCT’s are not unexpected. Such differences do not, as claimed in a recent paper, invalidate the benefit proven in RCT’s. With respect to the 40-49 year age group, another issue is whether age should be defined as "age at entry" into the trials or as "age at diagnosis". Because "age at diagnosis" represents a pseudovariable that introduces bias, use of "age at entry" is preferable. However, most breast cancers in the age 40-49 years at entry group were diagnosed prior to age 50.

Benefit for women age 50 and older begins to appear at 5-6 years after screening; whereas, benefit for women ages 40-49 has not been seen until 8-9 years post screening. Although benefit for these younger women is more delayed, these differences between age groups are likely to diminish with shorter screening intervals. Not only is their benefit real, but younger women have a longer normal life expectancy in which experience these benefits.

With the establishment of benefit, the debate regarding screening women ages
40-49 years has shifted to the lower absolute reduction of breast cancer deaths and the higher relative incidence of adverse consequences among women screened in this age group. For women ages 40-49 years:

1) Detection rates are lower due to lower breast cancer incidence, decreased sensitivity of mammography in dense, fibroglandular breasts, and faster breast cancer growth rates.

2) Cancer detection rate/screening recall rate is lower.

3) Biopsy positive predictive values (cancers detected/biopsies performed) are lower.

4) Lives saved through screening vs. lives potentially lost due to radiation risk (benefit/radiation risk) are lower.

5) Cost per year of life expectancy saved is higher. (Cost effectiveness is lower.)

However, differences in benefits and risks for women ages 40-49 versus 50-59 are small. Such changes occur gradually with age rather than abruptly at age 50. Calculated values strongly favor beginning screening mammography at age 40. Nearly 25% of all deaths from breast cancer and 33% of all years of life expectancy lost to breast cancer in the United States occur among women whose breast cancers were found in their 40’s. Even if all United States women began screening at age 40, the increase in total United States health care expenditures would be less than 0.1%. Although ductal carcinoma in situ (DCIS) represents a greater proportion of cancers among younger women, this should be viewed as a benefit rather than a harm from screening.

Quality assurance for mammographic technique and interpretation will increase the benefit and reduce the adverse consequences from screening. Technical quality in the United States has markedly improved due to the American College of Radiology (ACR) Mammography Accreditation Program (MAP) and the Mammography Quality Standards Act (MQSA). Annual screening mammography by age 40 is now recommended by the American Cancer Society (ACS), American Medical Association (AMA), American College of Obstetrics and Gynecology (ACOG), American College of Radiology (ACR), and many other medical organizations. Screening every 1-2 years by age 40 is recommended by the National Cancer Institute (NCI).


References:

1. Feig SA, D’Orsi CJ, Hendrick RE, et. al. American College of Radiology Guidelines for Breast Cancer Screening. AJR 1998; 171:29-33.

2. United Kingdom Trial of Early Detection of Breast Cancer Group: 16-Year mortality from breast cancer in the United Kingdom Trial of Early Detection of Breast Cancer. Lancet 1999; 353:1909-1914.

3. Gotzsche PC, Olsen O: Is screening for breast cancer with mammography advisable? Lancet 2000; 355:129-134.

4. de Koning HJ: Assessment of nationwide screening programs. Lancet 2000; 355:80-81.

5. Feig SA, Hendrick RE: Radiation risk from screening mammography of women aged 40-49 years. Monog Natl Cancer Inst 1997; 22:119-124.

6. Feig SA: Ductal carcinoma in situ (DCIS): Implications for screening mammography. Radiologic Clinics of North America July, 2000; (in press).

7. Feig SA: Age-related accuracy of screening mammography: How should it be measured? Radiology 2000; 214:633-640.

8. Hendrick RE, Bassett LW, Botsco MA, et. al. Mammography Quality Control Manual. Reston, VA: American College of Radiology, 1999, 339 pp.

9. D’Orsi CJ, Bassett LW, Feig SA, et. al. Illustrated Breast Imaging Reporting and Data System (Illustrated BI-RADS), Third Edition. Reston, VA: American College of Radiology, 1998, 262 pp.




Active Immunotherapy with Anti-idiotypic Antibody for Nasopharyngeal Cancer (Poster presentation)

Li Guangcheng, M.D., Xie Lu, M.D., Sun Qubing, M.D., Cancer Research Institute, Hunan Medical University, Chang sha 410078

ABSTRACT According to the immune network hypothesis proposed by Jerne, certain anti-idiotypic antibodies (AB2) express three dimensional shapes which resemble the structure of natural antigens. Here we propose a study of active immunotherapy by AB2 for patients with Nasopharygeal cancer (NPC). Two Ab2, designated 2H4 and 5D3, against two Ab1 (FC2 and HNL5) that recognize NPC associated antigen were generated. They could substitute NPC antigen to induce humoral and cellular immune response against NPC cells in syngeneic mice. Nineteen patients with NPC at stage were chosen for active immunotherapy. They were treated with Alum-precipitated 2H4 or 5D3 accompanying radiotherapy. Nine patients with radiotherapy alone were as control. Both anti-anti-idiotypic antibodies (AB3) and anti-NPC antibodies (AB1’) were increased and human anti-mouse antibodies (HAMA) occurred in nineteen patients of the experimental group; whereas the levels of Ab1’ did not rise in control group. Serum IL-2, INF-_nd TNF-_ level were increased in most patients in experimental group. While in the control group, there was no difference of cytokine level between pretherapy and post-therapy.


Reference

1: Jerne NK, The Nobel Lectures in Immunology. The Nobel Prize for Physiology or
Medicine,1984. The generative grammar of the immune system, Scand J Immunol. 38, 1-9, 1993.

2: Jerne NK, The generative grammar of the immune system., Science, 229, 1057, 1985.

3: Jerne NK, The generative grammar of the immune system. Nobel lecture,, Biosci Rep. 1985 Jun;5(6):439-51.

4. Jerne NK., Idiotypic networks and other preconceived ideas., Immunol Rev. 79, 5, 1984.

5. Jerne NK, Roland J, Cazenave PA., Recurrent idiotopes and internal images, EMBO J.




X-Ray Optics for Better Diagnostic Imaging

Carolyn A. MacDonald and Walter M. Gibson
Center for X-ray Optics
University at Albany, Albany, NY 12222

Medical imaging was one of the earliest applications of x radiation. The announcements of the discovery of x rays included a radiograph of a human hand, and within a few months, x radiation was being used to diagnose broken limbs on the other side of the Atlantic. One hundred years after its first use, despite the advent of ultrasound and magnetic resonance imaging, x-ray imaging is still the most widely used medical imaging modality. Its low cost and ease of use are critical for mass screening, for example in cancer detection. While it is hoped that molecular detection and intervention will one day provide a more effective treatment, currently, mammography has proven to be the best means of reducing breast cancer mortality. The relative simplicity of the required apparatus and the speed with which the image is obtained (and the resultant relatively low cost) make x-ray imaging an extremely important diagnostic tool.

Notwithstanding its relative age, x-ray science and technology has undergone a renaissance during the past few decades. Led by the development of dedicated synchrotron sources, a veritable alphabet soup of new x-ray analysis tools was spawned. The desire to apply these tools in laboratory settings sparked a surge of development in x-ray sources, detectors, and especially optics. The greatly increased efficiency with which x-rays can now be manipulated has the potential for significantly improving the quality of diagnostic radiology.

The effectiveness of an image to be used in diagnoses depends on its contrast and resolution. For example, the detection of breast cancer from a mammogram relies upon the observation of either contrast between the normal tissue and the only slightly denser cancerous lesion or observation of the existence of clusters of small, irregularly shaped microcalcifications. This requires both good contrast and good resolution. Further, contrast is significantly degraded by the presence of photons that have been Compton scattered out of the primary beam. These scattered photons create a background fog that should be removed. For mammography to be feasible for a screening modality it must also be inexpensive, rapid, and require extremely low radiation dose. In practice, mammographic imaging is often limited by quality assurance issues, which can be favorably addressed by digital processing.

Polycapillary x-ray optics, invented in the mid-eighties, provide an innovative new way to control x-ray beams. Such optics provide extremely efficient scatter rejection, while allowing image magnification, demagnification, and shaping to match with the newly developing digital x-ray detectors. Contrast enhancements of factors of 2-3 have been measured at energies from 20-40 keV. Magnifying optics also improve system resolution. Improvements in system modulation transfer function were measured for all frequencies out to nearly twice the original system resolution. Increases below the system resolution are most important because clinically relevant structures generally occupy lower spatial frequencies.

In addition, polycapillary collimating optics can be used to create a beam of sufficient intensity for monochromatization from a conventional source by diffraction from a single crystal. Monochromatic parallel beam imaging produces high subject contrast, high resolution, and low patient dose. , Measured contrast enhancement at 8 keV was a factor of 5 relative to the polychromatic case, in good agreement with theoretical values. At 17.5 keV, monochromatic subject contrast was a factor of 2 times greater than the conventional polychromatic contrast. An additional factor of two increase in contrast is expected from the removal of scatter obtained from using a parallel beam air gap or second polycapillary optic.


References

W.C. Roentgen, "On a New Form of Radiation," Nature, 53, 274-6, January 23, 1896, English translation from Sitzungsberichte der Würzburger Physik-medic. Gesellschaft, 1895.

E. Trevert, Something about X Rays for Everybody, Bubier Publishing Co., Lynn, Mass, 1896, reprinted by Medical Physics Publishing, Madison, WI.

Strategies for Managing the Breast Cancer Research Program: A Report to the U.S. Army Medical Research and Development Command, 1993.

J. M. Boone and J. A. Seibert, "A Comparison of Mono- and Poly-energetic X-ray beam Performance for Radiographic and Fluoroscopic Imaging." Med. Phys. 21(12), 1853, 1994.

R. J. Jennings, P. W. Quinn, R. M. Gagne and T. R. Fewell, "Evaluation of X-ray Sources for Mammography," Physics of Medical Imaging, SPIE Vol. 1896, 1993, pp. 268-259.