Tooth Dosimetry

Electron paramagnetic resonance (EPR) spectroscopy has been applied to perform retrospective radiation biodosimetry using extracted samples of calcified tissues, especially tooth enamel, following radiation accidents and exposures resulting from weapon use, testing, and production (Ikeya 1986, Ikeya 1984, Chumak 1998, Tolstykh 2000, Romanyukha 2006). This technology is based on the fact that ionizing radiation generates unpaired electrons proportional to dose and that, in tooth enamel, these unpaired electron species are extremely stable, persisting for thousands of years (Desrosiers 2001). In the natural formation of teeth, carbonate ions (CO32-) are incorporated into biological hydroxyapatite [Ca10(PO4)6(OH)2] during mineralization where they substitute for phosphate and hydroxyl ions. Ionizing radiation generates large numbers of unpaired electron species in irradiated materials, including biologic tissues. Following irradiation, free electrons are captured and carbonate radical centers are created. While radicals generated in soft tissue react very quickly to form non-paramagnetic, EPR-silent species, radicals generated in tooth enamel are very stable, persisting indefinitely at levels that are directly proportional to dose. EPR is a spectroscopic technique that specifically and sensitively detects unpaired electron species. Since the numbers of unpaired electrons produced by ionizing radiation are directly proportional to dose, the ability to detect them in enamel essentially renders teeth a naturally occurring dosimeter in every exposed person. Conventional EPR dosimetry prior to the development of in vivo EPR involved the isolation of tooth enamel from teeth in vitro with measurements made at X-band (~10 GHz) in a specialized remote laboratory, either using a universal calibration curve or using a calibration based on the addition of several known doses. While this technique is well suited for use in limited populations following certain exposures, its use as a tool to perform screening after an event where a large number of people have potentially been exposed to clinically relevant doses is limited by the need to extract the tooth and process it remotely at a specialized laboratory. In order to meet the need for large-scale biodosimetry, likely in concert with complementary biologically-based biodosimetry techniques, the instruments and procedures to perform EPR dosimetry in vivo with intact teeth, in the field, have been developed by the team at Dartmouth Medical School (Williams 2010, Swartz 2006, Swartz 2005, Swartz 2007, Iwasaki 2005a, Iwasaki 2005b, Williams 2007, Demidenko 2007, Brady 1968, Pollock 2010).

Below are some of the finding of our development and testing of EPR dosimetry systems for in vivo measurements of intact teeth at L-band (~1.2 GHz):

  1. In the range of clinically relevant doses, the relationship between the radiation induced EPR signal (RIS) amplitude and dose is linear, as evidenced by the data in linear-dose, which show the peak-to-peak (P2P) EPR amplitude measured as a function of dose for a set of 6 different intact natural molar teeth, ranging from 0 to 30 Gy, that were measured in vivo after being inserted into a gap in the dentition of a volunteer (Demidenko 2007). (Note that at the time these data were obtained five minutes of data acquisition was used for each set of measurements while we now obtain the 3 sets in one five minute period)

    Linear Dose Reponse and Effects of Averaging - No Averaging Linear Dose Reponse and Effects of Averaging - No Averaging

    In vivo dose response for inserted molar teeth. (left) Three independent measurements were acquired, each requiring 5 minutes of data acquisition. (right) The measurements can be averaged to reduce the standard error of dose prediction (SEP), in this case to <50 cGy.

  2. High intensity light, whitening bleach, or the presence of teeth with amalgam does not affect the measurements in adjacent teeth (nor in the tooth with amalgam, if the reduction in enamel is taken into account). The teeth with amalgam fillings do not have unusual intrinsic EPR signals and we have been able to measure the irradiation-induced signal from the teeth with amalgam fillings. The teeth with amalgam fillings did not affect the measurements observed on adjacent irradiated teeth (Iwasaki 2005a). Additionally, preliminary studies indicate that high intensity dental light and whitening bleach (3%) treatment does not induce the production of signals in unirradiated teeth or lead to detectable loss of signal in irradiated teeth.

  3. The radiation induced signal in tooth enamel is stable. After 48 hours the magnitude of the signal is constant for an indefinite period (~107 yr. at 25 °C). Within the first 48 hours, there is a small decay observed, (~6-9%) of the total signal, with a predictable decay rate.

    Stability of Radiation-Induced Signal

    Signal amplitude with time course. After 48 hours, following a predictable decay, the magnitude of the signal is constant for an indefinitely long period of time, facilitating dose measurements at any time after exposure.

  4. Variations in the volume of teeth can be compensated through the use of correction factors based on external dimensions of the tooth, which can be measured very quickly (Iwasaki 2005b).

    Effects of Correcting for Tooth Size

    (left) Before normalization the dose-response lines for each tooth are different. (right) After normalization for tooth-size, all of the points fall on a single dose response line.

  5. Consistent dose response and promising standard error of dose prediction have been observed for clinically significant dose ranges. Measurements with doses between 0 and 10 Gy were made in several volunteers. We show here two volunteer subjects with inserted irradiated molar teeth who had multiple measurements over a short period of time. Three separate measurements were made for each subject for each dose and the average radiation induced EPR signal was calculated. The in vivo dose responses measured under consistent experimental conditions, but with different subjects and different inserted teeth, had dose responses that were not statistically different according to the F-test (p-value > 0.6). The standard error of dose prediction for these measurements was 1.0 Gy (Williams 2010).

    Consistent Dose Response for Multiple Teeth

    In vivo and in vitro dose response relationships for molar teeth for two volunteer subjects. The calibration lines for the individual volunteers and the calibration for the combined data were not statistically different according to the F-test (p-value = 0.15). Standard errors of dose predictions were 1.15 Gy and 0.82 Gy.

  6. Measurements can be performed in all tooth types, including molars, premolars, canines and incisors, in order to ensure measurement capabilities in the greatest number of subjects. These experiments also are important steps in determining the preferred teeth types to be measured. We have performed dosimetry using all tooth types in normal and irradiated subjects using the existing instruments. Comparable EPR signal amplitudes are achieved across tooth types, though refinement of resonator geometries is necessary to optimize the precision of dose estimates in anterior teeth, especially canines (Williams 2010).

    Measurements of Different Tooth Types

    Spectra recorded for molar and incisor teeth that had been irradiated to equal doses of 30 Gy, each with an appropriate size resonator, have comparable EPR signal amplitudes. The incisor amplitude is 70% of the amplitude for the molar. No size correction has been applied. This indicates that measurements can be made with incisors with sensitivity approaching that achieved with molars.

  7. Measurements performed on patient's teeth in situ following exposure during radiotherapy provide dose estimates that are consistent with the estimates from treatment planning. Measurements were made on teeth in two patients, referred to as Subject PA and Subject PB, whose treatment plan and remaining dentition provided the potential for measuring different doses in the mouths of the individual subjects during the course of radiation therapy (dose-response-vol).

    Dose Response for Two Volunteers

    Dose-response relationship for two irradiated patient volunteers based on average RIS EPR signal amplitudes (n = 3) averaged for native canine tooth measurements. The dose responses from the individual volunteers was not found to be statistically different (p = 0.66). The joint standard error of prediction was 2.4 Gy (1.3 Gy excluding outlier).

    An extensive set of measurements was performed on two lower canine teeth, #22 and #27, of a patient volunteer receiving radiation treatment involving fractionated irradiation of the oral cavity. The treatment was delivered in 31 fractions over a period of 43 days, with the lower canines each receiving volumetric mean doses of 1.0 Gy (SD ±0.1 Gy) per fraction. EPR dosimetry measurements were performed for each tooth four times each week and analyzed to estimate the RIS and dose response. Based on the average of the two measurements on each day, the SEP = 1.8 Gy (canines).

    Dose Response of Two Canines

    Estimated versus expected dose for lower canine teeth (C22 and C27) measured during the course of radiation therapy where each tooth received 1.0 Gy/fraction. Estimated dose is based on an internal calibration and the SEP = 1.8 Gy. Improvements in resonator design and positioning for canine teeth, as well as the demonstrated use effectiveness of more efficient averaging, are expected to increase the precision of dose estimation.

  8. Incident microwave power saturation differences exist between radiation induced EPR signals and native independent signals which could be used to discriminate between these spectral components and increase the precision of dose estimation (power-sat) (Iwasaki 2005b). The practicality of using this difference will be determined as part of the SOW and if shown, will become incorporated into the final procedure.

    Power Saturation for Irradiated and Non-Irradiated Teeth

    The relationship between signal amplitude and microwave power and modulation amplitude for (left) irradiated tooth (3000 cGy) and (right) non-irradiated teeth.

  9. Averaging can be used effectively to improve the precision of dose estimates. In addition to random noise, sources of error, including those from electromagnetic interactions and positioning of the resonator can be addressed through repeated independent measurements. This is demonstrated by the data included above in linear-dose, which shows the effect of averaging three such measurements on the SEP for an in vivo tooth measurement.

  10. The tooth dosimetry system is field deployable. Throughout development, we have maintained focus on the field deployment of the technology, the need for high throughput, and use by minimally trained operators. We now have succeeded in having an operational deployable EPR dosimetry system, which we have recently utilized in non-laboratory sites. Progress toward this goal was augmented by concurrent developments supported by a contract with DARPA to develop an EPR tooth dosimetry system suitable for measurements in primates at a remote site. Field deployable systems, based on small permanent magnet designs, have been developed and prototypes are currently being tested (magnets, transportable and firestation). These magnets range from an <0.5 kg intraoral magnet to miniaturized dipole systems, with magnets similar to the existing clinical magnet but reduced in size from its 50 cm pole separation to 24 cm and 17 cm in two different versions. Because they are based on permanent magnets, there are no special power or cooling requirements needed for their operation. The electronics for EPR detection and sweeping of the magnetic field can be powered using standard 120 V AC supplied through the grid or from a generator or through the use of batteries. In order to have a field deployable magnet available as soon as possible, we are focusing on the 17 cm dipole magnet. In the proposed research-developmental phase (Years 1 and 2), we will both work on further improvements in this magnet design and in determining if even smaller designs can be made operational with a satisfactory level of performance and in a timely manner. These results will inform the choice of the magnet to be used for the development of the FDA-compliable prototype.

    Magnets for EPR Dosimetry

    Permanent magnet designs and prototypes developed for transportable EPR dosimetry systems.

    Transporable Dosimetry System

    Fully transportable versions of the in vivo EPR spectrometer, including a self-contained compact electronics and display unit and one of several small permanent magnets, have been assembled and their performances validated in remote locations.

    Transportable Dosimetry System at a Firestation

    Two remote exercises employing a transportable version of the dosimetry system have been performed at the Hanover, NH firehouse and in a field in a tent at local Cancer Center fundraiser. The measurements were performed using the transportable dipole magnet. Measurements at the fundraiser were performed with power from a portable generator.

  11. Anthropomorphic mouth model development. In order to facilitate more rapid development of instrumental and measurement techniques to improve dosimetric precision, especially in the face of the limited number of volunteer subjects suitable for inclusion in the study and the complexities of scheduling their participation, we have designed and constructed a series of anthropomorphic mouth models that accurately simulate the conditions of in vivo measurements. These models incorporate sets of neighboring, and opposing, natural teeth set in a dental casting material with high water content. The dental cast simulates the anatomy and the high water content of the material reproducing the RF characteristics of the oral cavity. In addition to the high water content of the casting material, free saline is applied to the model to mimic the presence of saliva. Details of the mouth model and its use are shown in mouth-model and a comparison of the RIS amplitude acquired in vivo and in the model using the same teeth are shown in mouth-model-results.

    Anthropomorphic Mouth Model

    Fabrication and use of the anthropomorphic mouth model to simulate in vivo conditions.

    Dose Response using the Mouth Model

    Comparison of in vivo dose response to that achieved in a mouth model. In both cases only one tooth was irradiated and the same irradiated tooth was used for both conditions.

    The ability of the mouth model to simulate the RF characteristics was also assessed by measuring the quality factor (Q) and RF frequency (fRF) of the resonator installed in the model and volunteer subjects. Using a resonator with a 10 mm detection loop, the free-space Q and fRF were 520 and 1.189 GHz, respectively. In the mouth model these values dropped to 242 ± 3 and 1.1806 ± 0.0002 GHz. Averaging across human subjects, these values were measured to be 240 ± 10 and 1.185 ± 0.001 GHz. The comparability of these values is indicative of consistent RF characteristics of the system, which lead to comparable losses and sensitivities of the EPR measurements.

  12. Resonator development. Major efforts have included the fabrication and testing of resonators with reduced detection loop diameters (~6 mm) for measurements of anterior teeth (e.g. canines, lingual incisor surfaces), design and fabrication refinements to reduce the presence of baseline distortion and microphonic noise, and refinement of resonators to sample multiple teeth simultaneously. Especially for resonators with smaller detection loops, the presence of baseline distortion remains a major source of noise and potential dosimetric error. Efforts continue to eliminate this source of non-random noise, including increased use of non-magnetic components, alternative configurations of the resonator electrical components, and refined mechanical fabrication techniques. An example of a resonator designed to sample multiple teeth simultaneously is shown in side-loop. This resonator is designed to measure opposing molar teeth simultaneously, without the use of the resonator positioning molds that have been employed previously. Using the transportable dipole system and the associated patient/magnet/resonator positioning assembly, the resonator can be rapidly placed on the teeth of interest for measurements.

    Side-loop Resonator

    "Side-loop" resonator designed to simultaneously measure opposing molar teeth.

  13. Systematic Optimization of Resonators. Finite Element Analysis (FEA) using Ansoft High Frequency Structure Simulator (HFSS) facilitates rapid analysis of design parameters to guide resonator development sim). HFSS models incorporate material properties and geometries to calculate B-field distributions and EPR sensitivity. Such simulations have quantitatively shown that by increasing the separation between the tooth surface and the detection loop, a more uniform magnetic field is established over the tooth volume with decreased sensitivity to variation in position (Pollock 2010).

    Field Map for L-Band EPR Resonator Decay of EPR Signal vs. Distance from Resonator

    (left) A field map of |Hxy| is shown when the distance between the loop and molars is 1.25 mm. (right) HFSS results are validated against experimental data during testing of optimized resonators. Here, comparison of the normalized signal amplitude versus detector/tooth separation for a specific resonator design predicted by HFSS and validated by experimental measurements for a single isolated molar.

  14. Discrimination between unirradiated subjects and TBI patients irradiated with prescribed doses of 2 Gy was achieved using <5 min data collection periods. Preliminary feasibility study of the use of upper incisor measurements for screening was performed using cohorts of 6 TBI patients and unirradiated subjects using the deployable dipole dosimetry system, a resonator specifically designed for incisors, and standard EPR acquisition parameters. Each measurement was completed in <5 min, and consisted of 3 independent sets of 20 3 sec. scans with replacement of the resonator between sets. A 75% increase in signal amplitude was observed for the TBI subjects, relative to the measurements in normal subjects of the same surface (p < 0.02, Student's t-test, two-tailed, unpaired).

    Results Comparing Normal Subjects to TBI Patients

    Radiation induced signal (RIS) EPR amplitude measured for incisor teeth in group of unirradiated "normal" subjects (n=6) and a group that had received total body irradiation with prescribed dose of 2 Gy (n=6). Measurements were made on the front and back surfaces for normal subjects and the front surface for TBI patients. A significant difference was observed between signals recorded from the normal and TBI subjects (p < 0.02), while no such difference was observed between opposite surfaces in the normal subjects. These data indicate promise for the use of incisor measurements for dose estimation and, especially, screening. Data are expressed as the mean ± SE.

Conclusions from Previous Experimental Results

Recent developments in vivo EPR have made it feasible to make the measurements in teeth in situ, greatly expanding the potential for using this approach for immediate screening and dose estimation following radiation exposures. In vivo measurements can be made in molar, premolar, canine, and incisors teeth. Dose calibrations appear to vary systematically by type of tooth but are not unique to individuals after size is taken into account. Dose response curves have been generated for volunteers with inserted molar teeth and for patients using intact canine teeth. Using the current methodology, instrument, and averaged measurements standard errors of dose prediction of 50-100 cGy have been achieved.

Developments to reduce the standard errors are proposed, focusing on geometric optimization of the resonators, detector positioning techniques, and optimal data averaging approaches. Developments are also underway toward automation and miniaturization of the instrumentation for field use by non-expert operators. It seems likely that in vivo EPR dosimetry techniques will play an important role in retrospective dosimetry and screening following exposures involving large numbers of individuals.