Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Dec 1.
Published in final edited form as: Health Phys. 2011 Dec;101(6):746–753. doi: 10.1097/HP.0b013e31821a4838

A novel phantom model for mouse tumor dose assessment under MV beams

Michael S Gossman 1,#, Indra J Das 2, Subhash C Sharma 3, Jeffrey P Lopez 1, Candace M Howard 4, Pier P Claudio 4
PMCID: PMC3208162  NIHMSID: NIHMS289420  PMID: 22048493

Abstract

Purpose

In order to determine a mouse’s dose accurately and prior to engaging in live mouse radiobiological research, a tissue-equivalent tumor-bearing phantom mouse was constructed and bored to accommodate detectors.

Methods and Materials

Comparisons were made between four different types of radiation detectors, each inserted into the phantom mouse for radiation measurement under a 6 MV linear accelerator beam. Dose detection response from a diode, thermoluminescent dosimeters, metal-oxide semiconductor field-effect transistors were used and compared to that of a reference pin-point ionization chamber. Likewise, a computerized treatment planning system was also directly compared.

Results

Each detector system demonstrated results similar to the dose computed by the therapeutic treatment planning system, although some differences were noted. The average disagreement from a accelerator calibrated output dose prescription in the range of 200–400 cGy were −0.4% ± 0.5σ for the diode, −2.4% ± 2.6σ for the TLD, −2.9% ± 5.0σ for the MOSFET and +1.3% ± 1.4σ for the treatment planning system.

Conclusions

This phantom mouse design is unique, simple, reproducible and therefore recommended as a standard approach to dosimetry for radiobiological mouse studies by means of any of the detectors used in this study. We fully advocate for treatment planning modeling when possible prior to linac-based dose delivery.

Keywords: accelerator, dose, Foxn1, mice, pie cage, radiobiology, x-ray

I. INTRODUCTION

Mice have been used in radiobiological research studies for many years. The original development and use of inbred mice for probing the genetic determinants of resistance and susceptibility to infections and tumors was documented from research by Dr. Clara Lynch(Morse 1978). During the 1920’s, she personally brought a mouse strain into the United States from a laboratory in Lausanne, Switzerland. The original population was two males and seven females which Dr. Lynch kept in a secure shoebox, and stored in her stateroom on board the ship.1 Once at Murphy’s laboratory, they were genetically altered in 1937 with a distinctive phenotype(Morse 1978; Leiter 1993; Chia 2005). Now classified Foxn1nu, these mice have no body hair(Suzuki 2003). These so called nude mice have a marked ability to engraft many different types of tumor cells from other animals, including humans. Xenografting, as it is called, is a common technique used by radiobiologists to test the characteristics of disease growth, inhibiting drugs, and tumor responses to drugs and radiation by the injection of tumors and drugs into nude mice. An illustration of nude mice subjects is presented in Figure 1.

Figure 1.

Figure 1

Open in a new tab

Foxn1nu (nude) mice with xenografted tumors.

Having an immunodeficient research subject is a valuable asset since it removes the single most important degree of freedom; tumor rejection. It is from these studies that novel techniques seen to reduce the size of prostate tumor xenografts have had recent success using drugs injected into mice(Greco 2010).

Radiobiologists have valued this animal model for sustained grafted tumor growth, which permits fractionation considerations in therapeutic models, even for radiation therapy (Allam 1995). Almost all radiobiological experiments have been performed with kilovoltage x-ray machines. This is still the primary means of irradiating mouse subjects (Lo 1993; Gorodetsky 1990). Some research has investigated higher x-ray energy resulting from 137Cs and 60Co radioactive material irradiators or even radionuclides of 90Y(Lazewatsky 2003; Urano 1998; Agarwal 1975). Linear accelerators have rarely been used in such studies (Kumar 2008; Kuroda 1999; Jaffe 1987). This is due to the difficulty for radiobiologists to access use of one, since most medical accelerators approved strictly for human therapy governed by state level radiation control entities. Still, radiation control branches and inspector generals have provided approval for a dual use with immunodepressed nude mice when infectious disease control measures are properly in place. Now that megavoltage modalities are being considered more often, dose rates from x-ray machines and radioactive material at 0.4–2.5 Gy/min may now be escalated to pulsed dose rates at 4–10 Gy/min from accelerators(Stuben 1994).

In order to determine the dose to a mouse accurately and prior to engaging in live mouse radiobiological research, a tissue-equivalent tumor-bearing phantom mouse was constructed and bored to accommodate detectors. Comparisons were made between four different types of radiation detectors, each inserted into the phantom mouse for radiation measurement (Yorke 2005; Metcalfe 1993; Butson 1996; Quach 2000). The dose levels obtained by each instrument were determined relative to the calibrated x-ray output of a 6 MV accelerator. Detectors were placed individually and consecutively within the mock-up mouse subject and placed in a pie cage duplicating radiobiology research procedures for actual measurement of absorbed dose. An intercomparison of detector-determined dose is presented along with a computerized tomography (CT) based computerized treatment simulation plan.

II. MATERIALS AND METHODS

II. A. Phantom mouse, pie cage and detectors

Braintree Scientific, Inc. (Braintree, MA) pie cage model MPC with filter model MPC-TP was chosen. Shown in Figure 2, the acrylic mouse unit is 21.5 cm in diameter and 7.5 cm in height; individual chambers are 5 cm (base) and 9 cm (length). The circular cages secure up to 11 mice in any of 12 wedge-shaped chambers. The notched removable lid can be dialed to any of the ventilated pie sliced chambers, making it easy to load mice through the single lid opening. This clinical pie cage is widely used by radiobiology researchers and is therefore the most suitable for conducting this research.

Figure 2.

Figure 2

Open in a new tab

Brain Tree Scientific Model pie cage No. MPC with MPC-TP filter.

It is difficult to measure absorbed doses to mice in vivo. It is even more difficult to measure such doses to mouse tumors in situ. Here, we have chosen a phantom replica mouse to quantify the dose received to living mice. The phantom mouse consisted of a white, rubber polycarbonate material with a 7 cm body length and proportioned to be identical to that of the living mice shown in Figure 1. The elemental composition is estimated as 85.6% amorphous carbon and 14.4% hydrogen, constituting a butyl solid of tissue-like density. Figure 3 shows the mouse bored out to accommodate various sized detectors. The detectors used in this research are presented to scale in Figure 4.

Figure 3.

Figure 3

Open in a new tab

Phantom mouse with bolus tumor simulating in vivo xenografted tumor and ionization chamber inserted.

Figure 4.

Figure 4

Open in a new tab

Detectors used in the measurement of dose from left to right: at top - ionization chamber and MOSFET; at bottom-diode and TLDs.

As shown in Figure 3, an ionization chamber was inserted into the back and pushed superiorly to the location of the left flank of the phantom mouse, where the sensitive volume appears flush with the outer skin layer. To imitate the resulting tumor growth on living mice, a bolus was cut and placed directly on the flank area, directly abutting the underlying detector’s sensitive volume of each detector. The position of the detector thimble was easily seen through the transparent yellowish bolus material as illustrated (Figure 3). Radiation Products and Design, Inc. (RPD) (Albertville, MN) tissue bolus prosthesis model 486–305 was used to simulate flanked nude mouse disease(Greco 2009). With original factory dimensions 30 cm2 × 0.5 cm, the rubber mold was cut down to 1 cm2 × 0.5 cm to resemble the typical tumor-size of a xenografted mouse used in our tumor biology studies.

The experiments were run consecutively for each detector used in this study. Each detector was specifically chosen based on size and importance for involvement in this study as observed in documented research (Stern 2009). A PTW (Freiburg, Germany) model TN31014 miniature thimble-type ionization chamber was used as shown in position in Figure 3, having tip length 6.925 mm, width 3.4 mm and sensitive volume 0.015 cm3. The ion chamber point of measurement was determined as the radius, located 1.7 mm from the outer thimble wall. The ion chamber was connected to CNMC Company, Inc. (Nashville, TN) model 206 with 200 nC module model 206–110 for measurements. The chamber was equilibrated to nominally +300 V at the center-pin.

The Sun Nuclear (Melbourne, FL) Isorad-p™ diode model 1163000–1 was used, having diameter 7.1 mm×29.5 mm long with an 8.3 mm distance from tip to approximate point of measurement within the die. The diode point of measurement was determined as the radius, located 3.6 mm from the outer wall. The p-type diode was connected to a Nuclear Associates (Carle Place, NY) electrometer model 37–720 for measurements. The only requirement on the diode selected was to insure that it contained enough buildup so that the point of measurement was beyond the buildup region to maximum dose registered. Insuring this was the case eliminates the majority of the electron contamination produced by the high energy photon beams. In this research, a constant field size (largest) and a constant source-to-surface distance was set. A field size correction factor was needed for the diode. This value was determined in prior to be CF(FS)=1.04, and was included in the calculation of dose to cross-calibrate the diode’s response. With the source-to-surface distance near to 100 cm, the correction factor for it was determined to be CF(SSD)=1.00.

Sicel Technologies, Inc. (Morrisville, NC) metal-oxide semiconductor field-effect transistor (MOSFET) model OneDose™ was the third detector type used. Each transistor was 6 mm wide×33 mm long × 0.9 mm thick. The MOSFET point of measurement was determined as the half-thickness, located 0.5 mm from the outer wall. Three were used in this study for constancy verification in the use of the device.

Finally, three Quantaflux, LLC (Dayton, OH) thermoluminescent dosimeter (TLD) chips were used; model TLD-100 including Harshaw Chemical Company (Cleveland, OH) LiF: Mg, Ti powder. Each ribbon had square dimensions (3.2 mm)2×0.15 mm thick. The TLD point of measurement was determined as the half-thickness, located 0.1 mm from the outer wall. Likewise, multiple TLD detectors provided constancy results in the use of this particular detector.

While the TLDs were provided and processed by the University of Wisconsin-Madison Accredited Dosimetry Calibration Laboratory (ADCL) (Madison, WI) with National Institute of Standards and Technology (NIST) (Gaithersburg, MD) traceability, the other three device types were cross-calibrated against an independent ADCL calibrated dosimeter system. Again, each detector was placed into the mouse consecutively during the experiment, such that it resides directly underneath the tumor bolus. Once inserted, the phantom mouse and contained detector were placed into one section of the mouse cage and covered by the cage’s rotating lid.

The remainder of the phantom setup included the necessary backscatter and build-up to duplicate treatment geometry. One CIRS, Inc. (Norfolk, VA) Plastic Water™ model PW-4050 phantom plate was chosen for backscatter. The plate was 5 cm thick and had a surface area of 40×40 cm2. It was positioned directly underneath the pie cage. For buildup anterior to the cage, the RPD tissue bolus prosthesis model 486-305 was again used having dimensions (30 cm)2×0.5 cm and in combination with model 486-310 having dimensions (30 cm)2×1.0 cm for a total of 1.5 cm depth equivalence. More dose uniformity resulted from the use of such bolus material, since off-axis horn effects from linear accelerator treatments are greatly reduced when x-rays traverse deeper in a tissue.

We have studied the efficiency of a medical accelerator to yield a prescribed dose of radiation to a phantom replica of a nude mouse bearing an artificial bolus tumor. The dose levels were determined knowing the x-ray output of a 6 MV machine and from experimental measurements using four different detector systems (Yorke 2005; Metcalfe 1993; Butson 1996; Quach 2000). Detectors were placed consecutively within the mock-up mouse subject and placed in a pie cage duplicating radiobiology research procedures for the absorbed dose measurement. An intercomparison of detector-determined dose, along with a computerized tomography (CT) based computerized treatment simulation plan is presented.

II. B. CT acquisition

Scanning was obtained using a General Electric (GE) Lightspeed RT scanner (Fairfield, CT). The technique for scanning included 120 kVp x-rays, 150 mA current at a 1950 ms scan time conducted in helical mode. A 50 cm diameter circular field of view was used with a couch increment of 2.5 mm/slice. Once the scan was reconstructed, all 121 slices within the set were transferred to a treatment-planning computer.

Computerized tomography was conducted with the backscatter material, pie cage and bolus material in position. The entire phantom set was placed on the CT couch and aligned by lasers such that the pie cage was centered in the beam and at a distance of 100 cm at the anterior surface. For CT acquisition only, no detector was inserted into the replica mouse. Instead, a substitute plastic rod was inserted. This step insured no dose was given to the MOSFET or TLD during scanning, since both are sensitive enough to register unwanted dose during this imaging process. The plastic material was knowingly identifiable once the images were processed, and would then allow us to use the known dimensions of each device to contour and reconstruct each in the treatment planning system.

II. C. Computerized treatment simulation

Varian Medical Systems, Inc. (Palo Alto, CA) model Eclipse External Beam Planning Software version 8.1.20 was used to model the simulated dose distribution. Immediately following scan import, each slice was visually examined for artifacts. Special care was taken to identify and contour the pie cage on each slice, since much of the cage volume consists of air. The phantom mouse and bolus tumor was also 3-dimensionally contoured. Special care was taken to create new structures duplicating the dimensions of the four detectors used; ionization chamber, diode, MOSFET and TLD.

The Varian Anisotropic Analytical Algorithm (AAA) version 8.6.15 was commissioned with heterogeneity correction for density enabled in the planning software, specifically for beam data corresponding to a Trilogy upgraded Varian model 21EX particle accelerator at photon energy 6 MV. An output calibration was performed prior to experimentation on the particle accelerator according to the AAPM Task Group No. 51 protocol (Almond 1999). The calibration geometry included a source-to-surface distance of 98.5 cm in water and was defined by 10×10 cm2 radiation field size. The effective point of measurement for a reference ionization chamber at the depth of maximum dose was nominally 1.5 cm, such that the output determined at the axis of rotation for the machine (100 cm) was precisely 1.00 cGy for a 100 monitor unit delivery.

The radiation field was aimed anterior to the phantom set. The central axis of the beam was designed to pass directly through the center of the pie cage. The rotating canopy of the cage was set to the beam isocenter, located exactly 100 cm from the accelerator source. The resulting source-to-surface (SSD) distance at the bolus was 98.5 cm. The radiation field was defined by 40×40 cm2 jaw collimation, which easily envelopes the phantom, where beam properties of flatness and symmetry are ideal. The digitally reconstructed radiograph (DRR) illustrated in Figure 5 indicates the phantom set is encompassed within the radiation field.

Figure 5.

Figure 5

Open in a new tab

Digitally reconstructed radiograph of the phantom set in the accelerator field.

A dose calculation point was specifically set within the treatment planning system for the location of the detector. Each detector was positioned directly under the bolus material through the bore created in the mock-up mouse. The smallest possible dose calculation grid of 1.25 mm was assigned for best computational resolution. The computer algorithm was prescribed to receive 3 Gy at a rate of 6 Gy/min. The resulting isodose distribution is shown for the axial slice containing the phantom mouse in Figure 6.

Figure 6.

Figure 6

Open in a new tab

Axial slice view with simulation software indicating the mathematically calculated dose levels in the pie cage.

II. D. Radiation measurement

As previously discussed, the phantom system used was identical between simulation in the CT scanner and radiation measurement under the linear accelerator. A total of 1.5 cm bolus build-up material was placed on top of the pie cage (having phantom mouse and detector in place). Both were then laid on top of a 5 cm Plastic Water™ slab for sufficient backscatter. Beam geometry with the phantom placed on the couch of the accelerator was precisely the same as planned in simulation software. Again, lasers verified the position of the pie cage to beam isocenter and with its centroid in-line with the central axis of the beam. The accelerator jaws were fully opened and programmed to deliver a dose at a rate of 6 Gy/min using 6 MV x-rays, for timer settings of 200 MU, 300 MU and 400 MU corresponding to calibrated doses of 2 Gy (200 cGy), 3 Gy (300 cGy) and 4 Gy (400cGy).

III. RESULTS AND DISCUSSION

The density of the mouse phantom mouse was determined to be very similar to tissue, with Hounsfield Units of +135 ± 30. This concludes that the mouse density corresponded to a range between adipose tissue (−43 ± 50 HU) with an electron density of 3.17×1023 g/cm3 and density 0.970 g/cm3 to that of liver (+124 ± 50 HU) with an electron density of 3.516×1023 g/cm3 and density 1.070 g/cm3. Therefore, the phantom mouse chosen was estimated to be indistinguishable from a real mouse subject.

The analysis of the axial isodose plot in Figure 6 shows an overall general symmetric shape to the dose distribution laterally. The isodose distribution was unique between wedged areas that contain the phantom mouse and those which did not. Therefore, the phantom mouse was seen to attenuate the beam. While the 100% isodose line passed directly through the axis of the phantom mouse on the right, on the left the same isodose line existed more posteriorly against the lower pie cage plate. Having identified the physical density of the phantom mouse to be similar to that of tissue, the attenuation effect on the beam that resulted was consistent.

Correctly computed, 100% of the dose was normalized to the posterior aspect of the tumor bolus. It is conclusive that a calculated absorbed dose of 3 Gy was received by the entirety of the tumor bolus for the given time the beam was on. The planning simulation report concluded that 300 MU were required to yield a dose of 301 cGy. With a machine calibration of 1.00 cGy/MU at the nominal distance of 1.5 cm and at beam isocenter, the dose to the mouse tumor was very nearly the same, even though substantially further away from the isocenter in the setup. With a height of 7.5 cm, the phantom mouse was positioned in it, resting on the lower plate of the pie cage. The bolus on the mouse was only 3 cm from the posterior side of the lower pie cage plate. Thus, with beam isocenter aligned to the anterior surface of the pie cage, the tumor bolus was 4.5 cm further downstream from isocenter. Having the same depth as that of calibration, yet further away from isocenter, the mouse tumor was expected to have lower dose calculated to it. According to a 1/r2 fall-off for intensity, given [(100 cm + 1.5 cm)/(104.5 +1.5 cm)] 2 = 0.917, one would expect approximately an 8.3 % reduction in dose intensity. As this was not the case, the substantial air volume within the pie cage had caused equally as much increase in output. The cause could be found in the secondary electrons emitted in interactions induced in the top of the cage. For 300 MU, the tumor dose was 301 cGyat 6 MV x-ray energy as it is shown in the dose-volume histogram in Figure 7.

Figure 7.

Figure 7

Open in a new tab

Dose-volume histogram for all four detectors and including the mouse and tumor.

For the volume labeled “Tumor”, 100 % of the 3 Gy dose was given to 100 % of the tumor volume. As the 100 % isodose line tracks through the axis of the phantom mouse, the dose to the whole mouse should be substantially less, given the shallow curve representing its volume, labeled “Mouse” in Figure 7. The dose volume histogram (DVH) also predicts similar doses should be received by the various detectors.

Reviewing the doses received by the detectors and depicted in Figure 8, it becomes evident that the measured absorbed doses to the diode, TLDs, MOSFETs and treatment planning system correlate well with the reference ionization chamber. The overall chamber response (n=3) was determined to be reproducible to within a standard deviation of 0.3σ The measurements of the ionization chamber were compared directly to each of the other systems.

Figure 8.

Figure 8

Open in a new tab

Response of each detector type for phantom mouse tumor: Dose (cGy)/MU.

The diode dose detection response (n=3) was reproducible to within 0.5σ It proved the most accurate with mean disagreement from the ion chamber of −0.4%. Less accurate measurements were obtained for the TLD chips and MOSFETs. The resulting precision for these detectors were 2.6σ and 5.0σ respectively. The mean measurement inaccuracy versus the ionization chamber was found to be −2.4% for the TLDs (n=3) and −2.9% for the MOSFET (n=3). The worst amongst all detectors was determined to be the MOSFET, although it was found to be reliable in estimating the dose to the tumor. All detector systems were found to underestimate the dose for all measurements in comparison to the accurate and precise calibrated ionization chamber response. The physical science of detection is different between these devices, due primarily to their design and construction, both of which enable the ionization chamber to be the most sensitive to charge collection than the diode, TLD or MOSFET.

The treatment planning system was found to have reproducibility in its dose determination to within 1.3 σ. Conversely as seen from detector systems, the computer generally overestimated dose by comparison to the ion chamber. The treatment planning system (n=3) accuracy fell within a mean disagreement from the ion chamber of +1.3%.

IV. CONCLUSIONS

This study introduces in vivo dose analysis of a radiobiologically studied mouse tumor, using a replica mouse phantom and tumor, both having a rubber density similar to that of adipose tissue or liver. A tumor volume created from bolus material overlaying a pre-bored cavity accommodating a dosimeter was introduced. With CT acquisition and computerized simulation, dose levels to the mouse structure and to the target phantom tumor were found to be accurate with respect to measurements. Further, commonly used detector systems including diodes, thermoluminescent dosimeters, MOSFETs, and a commercially available treatment planning system were intercompared to reference ionization chamber results. The average disagreement from a accelerator calibrated output dose prescription in the range of 200–400 cGy were −0.4% ± 0.5σ for the diode, −2.4% ± 2.6σ for the TLD, −2.9% ± 5.0 σ for the MOSFET and +1.3% ± 1.4σ for the treatment planning system.

For a 300 MU delivery at the particle accelerator, a dose of 300–301 cGy was detected by all four-detector systems within the calculated accuracy and standard deviation. Therefore, in the geometry employed, a general 1:1 correlation was seen between the number of monitor units programmed on a linear accelerator and the dose required at the location of a mouse tumor. These detector types have proven capable of precisely determining the output for mice in a pie cage treatment geometry. The treatment planning system is also a valuable tool, which may serve as a verification to visualize dose levels prior to being administered, where changes may be considered for asymmetries in the dose profile that may be unobserved otherwise.

For higher dose prescriptions of 8 Gy to mice tumors, it is possible to obtain this dose on a linear accelerator calibrated to 1.00 cGy/MU when run at 6 Gy/min for only 80 seconds and thus saving a significant amount of time for live nude mouse exposures conducted identically. We stress that although this set-up resulted in the same dose for the number of monitor units programmed (i.e. 300 MU yielding 300 cGy), future investigations should be experimentally verified with radiation detectors in this manor prior to proceeding, as this may not be true for other geometries, equipment, and targets. This phantom mouse design is unique, simple, reproducible and therefore recommended as a standard approach to dosimetry for radiobiological mouse studies using any of the detectors in this study, while understanding the observed accuracy of each.

We fully advocate our novel consideration of the use of the treatment planning system to model the dose to a mouse for planning purposes, prior to irradiation similarly. The implantation of such dose evaluations prior to dose delivery is a necessary quality assurance step for humans, and as such should be used if available and evaluated for accurate radiobiology setup considerations.

Acknowledgments

We gratefully acknowledge the Marshall University Biochemistry and Microbiology & Surgery Departments for their support. The present studies were supported by National Cancer Institute (NCI) awards CA131395 and CA140024, and in part by National Institutes of Health (NIH)-Centers of Biomedical Research Excellence (COBRA) award 5P20RR020180 and West Virginia IDeA Network of Biomedical Research Excellence (WV-INBRE) award 5P20RR016477 (each to PPC). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institute of Health.

Footnotes

Conflict Statement:

We provide no conflicts of interest. We do fully acknowledge the present studies were supported by National Cancer Institute (NCI) awards CA131395 and CA140024, in part by National Institutes of Health (NIH)-Centers of Biomedical Research Excellence (COBRA) award 5P20RR020180, and in part by West Virginia IDeA Network of Biomedical Research Excellence(WV-INBRE) award 5P20RR016477 (each to PPC).

VI. REFERENCES

  1. Agarwal JK, Ayappan P, Santhamma AV. Carcinogenic effect of radiation on the mouse embryo; physical aspects of radiation dosimetry. Strahlentherapie. 1975;150(6):618–623. [PubMed] [Google Scholar]
  2. Allam A, Perez LA, Huang P, Taghian A, Azinovic I, Freeman J, Duffy M, Efird J, Suit HD. The effect of the overall treatment time of fractionated irradiation on the tumor control probability of a human soft tissue sarcoma xenograft in nude mice. Int J Radiation Oncology Biol Phys. 1995;32(1):105–111. doi: 10.1016/0360-3016(95)00511-V. [DOI] [PubMed] [Google Scholar]
  3. Almond PR, Biggs PJ, Coursey BM, Hanson WF, Huq MS, Nath R, Rogers DW. AAPM’s TG-51 protocol for clinical reference dosimetry of high-energy photon and electron beams. Med Phys. 1999;26(9):1847–1870. doi: 10.1118/1.598691. [DOI] [PubMed] [Google Scholar]
  4. Butson MJ, Rozenfeld A, Mathur JN, Carolan M, Wong TP, Metcalfe PW. A new radiotherapy dose detector: The MOSFET. Med Phys. 1996;23(5):655–658. doi: 10.1118/1.597702. [DOI] [PubMed] [Google Scholar]
  5. Chia R, Achilli F, Festing MFW, Fisher EMC. The origins and uses of mouse outbred stocks. Nature Genetics. 2005;37:1181–1186. doi: 10.1038/ng1665. [DOI] [PubMed] [Google Scholar]
  6. Gorodetsky R, Mou X, Fisher DR, Taylor JM, Withers HR. Radiation effect in mouse skin: dose fractionation and wound healing. Int J Radiation Oncology Biol Phys. 1990;18(5):1077–1081. doi: 10.1016/0360-3016(90)90443-n. [DOI] [PubMed] [Google Scholar]
  7. Greco A, Benedetto AD, Howard CM, Kelly S, Nande R, Dementieva Y, Miranda M, Brunetti A, Salvatore M, Claudio L, Sarkar D, Dent P, Curiel DT, Fisher PB, Claudio PP. Eradication of therapy-resistant human prostate tumors using an ultrasound guided site-specific cancer terminator virus delivery approach. Molecular Therapy. 2010;18:295–306. doi: 10.1038/mt.2009.252. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Jaffe DR, Williamson JF, Tim G. Ionizing radiation enhances malignant progression of mouse skin tumors. Carcinogenesis. 1987;8(11):1753–1755. doi: 10.1093/carcin/8.11.1753. [DOI] [PubMed] [Google Scholar]
  9. Kumar S, Kolozsvary A, Kohl R, Lu M, Brown S, Kim JH. Radiation-induced skin injury in the animal model of scleroderma: implications for post-radiotherapy fibrosis. Rad Oncology. 2008;3:40. doi: 10.1186/1748-717X-3-40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Kuroda M, Inamura K, Tahara S, Kurabayashi Y, Akagi T, Asaumi J, Togami I, Takemoto M, Honda O, Morioka Y, Kawasaki S, Hiraki Y. A new experimental system for irradiating tumors in mice using a linear accelerator under specific pathogen-free conditions. Acta Med Okayama. 1999;53(3):111–118. doi: 10.18926/AMO/31638. [DOI] [PubMed] [Google Scholar]
  11. Lazewatsky J, Ding Y, Onthank D, Silva P, Solon E, Robinson S. Radiation dose to abdominal organs of the mouse due to 90Y in the urinary bladder. Cancer Biotherapy & Radiopharmaceuticals. 2003;18(3):413–419. doi: 10.1089/108497803322285152. [DOI] [PubMed] [Google Scholar]
  12. Leiter EH. The NOD mouse: a model for analyzing the interplay between heredity and environment in development of autoimmune disease. Institute for Lab Animal Research (ILAR) 1993;35(1) All. [Google Scholar]
  13. Lo Y, Taylor JMG, McBride WH, Withers HR. The effect of fractionated doses of radiation on mouse spinal cord. Int J Radiation Oncology Biol Phys. 1993;27(2):309–317. doi: 10.1016/0360-3016(93)90242-n. [DOI] [PubMed] [Google Scholar]
  14. Metcalfe P, Kron T, Elliott A, Wong T. Dosimetry of 6-MV x-ray beam penumbra. Med Phys. 1993;20(5):1439–1445. doi: 10.1118/1.597107. [DOI] [PubMed] [Google Scholar]
  15. Morse HC., III . National Institute of Allergy and Infectious Diseases. National Institutes of Health; Bethesda, Maryland: 1978. Origins of Inbred Mice. [Google Scholar]
  16. Quach KY, Morales J, Butson MJ, Rosenfeld, Metcalfe PE. Measurement of radiotherapy x-ray skin dose on a chest wall phantom. Med Phys. 2000;27(7):1676–1680. doi: 10.1118/1.599035. [DOI] [PubMed] [Google Scholar]
  17. Stern RL. Dosimetry metrology for IMRT: Part I. Med Phys. 2009;36(6):2756. [Google Scholar]
  18. Stüben G, Budach W, Schick KH, Stuschke M, Stapper N, Muller S, Feldmann A time-saving system for irradiations of experimental tumors. Strahlenther Onkol. 1994;170:36–41. [PubMed] [Google Scholar]
  19. Suzuki N, Hirata M, Kondo S. Traveling stripes on the skin of a mutant mouse. Proceedings of the National Academy of Sciences of the United States of America (PNAS) 2003;100(17):9680–9685. doi: 10.1073/pnas.1731184100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Urano M, Nishimura Y, Kuroda M, Reynolds R. Are hypoxic cells critical for the outcome of fractionated radiotherapy in a slow-growing mouse tumor? Radiotherapy and Oncol. 1998;48:221–228. doi: 10.1016/s0167-8140(98)00010-3. [DOI] [PubMed] [Google Scholar]
  21. Yorke E, Alecu R, Ding L, Fontenla D, Kalend A, Kaurin D, Masterson-McGary ME, Marinello G, Matzen T, Saini A, Shi J, Simon W, Zhu TC, Zhu XR, Rikner G, Nilsson G. AAPM Report No. 87 – Report of Task Group 62 of the Radiation Therapy Committee: Diode in vivo dosimetry for patients receiving external beam radiation therapy. Medical Physics Publishing; Madison, Wisconsin: 2005. [Google Scholar]

RESOURCES