A dosimetric uncertainty analysis for photon-emitting brachytherapy sources: Report of AAPM Task Group No.

138 and GEC-ESTRO Larry A. DeWerd Department of Medical Physics and Accredited Dosimetry Calibration Labor atory, University of Wisconsin, Madison, Wisconsin 53706 Geoffrey S. Ibbott Department of Radiation Physics, M. D. Anderson Cancer Center, Houston, Texas 77030 Ali S. Meigooni Department of Radiation Oncology, Comprehensive Cancer Center of Nevada, Las Vegas, Nevada 89169 Michael G. Mitch Ionizing Radiation Division, National Institute of Standards and Technol ogy, Gaithersburg, Maryland 20899 Mark J. Rivarda Department of Radiation Oncology, Tufts University School of Medicine, B oston, Massachusetts 02111 Kurt E. Stump Santa Maria Radiation Oncology Center, Santa Maria, California 93454 Bruce R. Thomadsen Departments of Medical Physics and Human Oncology, University of Wiscons in, Madison, Wisconsin 53706 Jack L. M. Venselaar Department of Medical Physics and Engineering, Instituut Verbeeten, 5042 SB Tilburg, The Netherlands Received 24 June 2010; revised 6 December 2010; accepted for publicatio n 14 December 2010; published 14 January 2011 This report addresses uncertainties pertaining to brachytherapy single-s ource dosimetry preceding clinical use. The International Organization for Standardization ISO Gui de to the Expression of Uncertainty in Measurement GUM and the National Institute of Standards a nd Technology NIST Technical Note 1297 are taken as reference standards for uncertain ty formalism. Uncertainties in using detectors to measure or utilizing Monte Carlo methods to e stimate brachytherapy dose distributions are provided with discussion of the components intrinsic t o the overall dosimetric assessment. Uncertainties provided are based on published observations a nd cited when available. The uncertainty propagation from the primary calibration standard throug h transfer to the clinic for air-kerma strength is covered rst. Uncertainties in each of the brachythe rapy dosimetry parameters of the TG-43 formalism are then explored, ending with transfer to the cl inic and recommended approaches. Dosimetric uncertainties during treatment delivery are consi dered brie y but are not included in the detailed analysis. For low- and high-energy brachytherap y sources of low dose rate and high dose rate, a combined dosimetric uncertainty 5% k = 1 is estima ted, which is consistent with prior literature estimates. Recommendations are provided for clinic al medical physicists, dosimetry investigators, and source and treatment planning system manufact urers. These recommen-

dations include the use of the GUM and NIST reports, a requirement of co nstancy of manufacturer source design, dosimetry investigator guidelines, provision of the lowes t uncertainty for patient treatment dosimetry, and the establishment of an action level based on d osimetric uncertainty. These recommendations re ect the guidance of the American Association of Physici sts in Medicine AAPM and the Groupe Europen de CuriethrapieEuropean Society for Therapeuti c Radiology and Oncology GEC-ESTRO for their members and may also be used as guidanc e to manufacturers and regulatory agencies in developing good manufacturing practices f or sources used in routine clinical treatments. 2011 American Association of Physicists in Medicine . DOI: 10.1118/1.3533720 Key words: brachytherapy, dosimetry, uncertainty, standards TABLE OF CONTENTS III . MEASUREMENT UNCERTAINTY IN BRACHYTHERAPY DOSIMETRY. . . . . . . . . . . . 785

III.A. Intrinsic measurement uncertainties. . . . . . . . . 785 I. INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . 783 III.A.1. Source activity distribution. . . . . . . . . . . . . 785 III.A.2. Source-to-detector positioning. . . . . . . . . . . II. UNCERTAINTY ESTIMATION METHODS. . . . . 30.00 782 Med. Phys. 38 2, February 2011 784 785 0094-2405/2011/382/782/20/$

2011 Am. Assoc. Phys. Med. 782 783 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unc ertainty recommendations 783 III.B. Dose measurement. . . . . . . . . . . . . . . . . . . . . . 7 86 values. The 2004 AAPM TG-43U1 report considered these uncertainties in a cursory manner.2 Before publication of the III.B.1. Thermoluminescent dosimeters. . . . . . . . . . 7

86

III.B.2. Radiochromic lm. . . . . . . . . . . . . . . . . . . . TG-43U1 report, estimates of dosimetry uncertainties for III.B.3. Diamonds, diodes, and MOSFETs. . . . . . . brachytherapy were limited. Most investigators using MC

86

87 techniques presented only statistical uncertainties; only reIV. MONTE CARLO UNCERTAINTY IN BRACHYTHERAPY DOSIMETRY. . . . . . . . . 87 cently have other MC uncertainties been examined. IV.A. Source construction. . . . . . . . . . . . . . . . . . . . . 87 In the current report, the uncertainty propagation from the IV.B. Movable components. . . . . . . . . . . . . . . . . . . . 88 primary calibration standard through transfer to the clinic for IV.C. Source emissions. . . . . . . . . . . . . . . . . . . . . . .

7 7 7 7

88 air-kerma strength SK is detailed Fig. 1 . Uncertainties in IV.D. Phantom geometry. . . . . . . . . . . . . . . . . . . . . . 89 each of the brachytherapy dosimetry parameters are then exIV.E. Phantom composition. . . . . . . . . . . . . . . . . . . . plored, and the related uncertainty in applying these param89 IV.F. Radiation transport code. . . . . . . . . . . . . . . . . . eters to a TPS for dose calculation is discussed. Finally, rec-

89 ommended approaches are given. Section II contains detailed IV.G. Interaction and scoring cross sections. . . . . . . explanations of type A and type B uncertainties. The brachyIV.H. Scoring algorithms and uncertainties. . . . . . . .

90

90

therapy dosimetry formalism outlined in the AAPM TG-43 V. UNCERTAINTY IN THE TG-43 DOSIMETRY report series 1995,3 2004,2 and 2007 Ref. 4 is based on FORMALISM PARAMETERS. . . . . . . . . . . . . . . . . limited explanation of the uncertainties involved in the meaV.A. Air-kerma strength. . . . . . . . . . . . . . . . . . . . . . surements or calculations. The 2004 AAPM TG-43U1 report V.A.1. Uncertainty in NIST primary standard for presented a generic uncertainty analysis speci c to calculaLDR low-energy photon-emitting sources. tions of brachytherapy dose distributions. This analysis inV.A.2. Uncertainty in NIST primary standard for cluded dose estimations based on simulations using experiLDR high-energy photon-emitting sources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 91 mental measurements using thermoluminescent dosimeters TLDs and MC methods. These measurement and simulaV.A.3. SK uncertainty for HDR high-energy sources. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 91 tion uncertainty analyses included components toward develV.A.4. Transfer of NIST standard to the ADCLs. 7 92 oping an uncertainty budget. A coverage factor of 2 k = 2 is V.A.5. Transfer of NIST standard from ADCLs recommended for testing and calibration laboratories per the to the clinic. . . . . . . . . . . . . . . . . . . . . . . . . 7 92 International Organization for Standardization ISO 17025 report5 and in general for medicine;6 we also recommend this V.B. Dose-rate constant. . . . . . . . . . . . . . . . . . . . . . 7 7 7

90

90

90

94

V.C. Geometry function. . . . . . . . . . . . . . . . . . . . . . 94 coverage factor for the scope of uncertainties included in the V.D. Radial dose function. . . . . . . . . . . . . . . . . . . . . current report. Thus, a coverage factor of 2 is used in the 95 V.E. 2D anisotropy function. . . . . . . . . . . . . . . . . . . current report unless explicitly described otherwise.

95 The current report is restricted to the determination of V.F. 1D anisotropy function. . . . . . . . . . . . . . . . . . . dose to water in water without consideration of material hetV.G. TPS uncertainties summary. . . . . . . . . . . . . . .

95

95

erogeneities, interseed attenuation, patient scatter conditions, VI. RECOMMENDATIONS. . . . . . . . . . . . . . . . . . . . . 96 or other clinically relevant advancements upon the AAPM VI.A. General uncertainty. . . . . . . . . . . . . . . . . . . . . . TG-43 dose calculation formalism.7 Speci c commercial VI.B. Clinical medical physicists. . . . . . . . . . . . . . . . equipment, instruments, and materials are described in the VI.B.1. SK and TPS data entry. . . . . . . . . . . . . . . . . current report to more fully illustrate the necessary experiVI.B.2. Treatment planning system developments. mental procedures. Such identi cation does not imply recVI.B.3. Clinical dosimetric uncertainties. . . . . . . . . ommendation or endorsement by either the AAPM, ESTRO, VI.C. Dosimetry investigators. . . . . . . . . . . . . . . . . . or the U.S. National Institute of Standards and Technology VI.D. Source and TPS manufacturers. . . . . . . . . . . .

96

96

96

96

97

97

98

NIST , nor does it imply that the material or equipment VII. SUMMARY AND COMPARISON TO identi ed is necessarily the best available for these purposes. EXISTING WRITTEN STANDARDS. . . . . . . . . . These recommendations re ect the guidance of the AAPM 7

98

and GEC-ESTRO for their members and may also be used as I. INTRODUCTION guidance to manufacturers and regulatory agencies in developing good manufacturing practices for sources used in rouThis report addresses uncertainties pertaining to photontine clinical treatments. As these recommendations are made emitting brachytherapy source calibrations and source dojointly by the AAPM and ESTRO standing brachytherapy simetry. In the American Association of Physicists in Medicine AAPM TG-40 report,1 the desired level of accuracy committee, the GEC-ESTRO, some of the speci cally mentioned U.S. agencies, organizations, and standard and precision is provided for treatment delivery. It is generlaboratories should be interpreted in the context of the arally assumed that brachytherapy uncertainties are larger than rangements in other countries where applicable. In particular, those in external beam applications. One objective of the other primary standards laboratories, such as the current report is to quantify the uncertainties involved in Physikalisch-Technische Bundesanstalt PTB in Braunschbrachytherapy so a greater understanding can be achieved. weig, Germany, the National Physical Laboratory NPL in The uncertainty values of brachytherapy apply to both the the United Kingdom, and the Laboratoire National Henri experimentally measured and Monte Carlo MC -estimated Medical Physics, Vol. 38, No. 2, February 2011 784 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry un certainty recommendations 784 FIG. 1. Brachytherapy source dosimetry data chain, highlighting the uncertainty values k = 1 and how they combine to increase the overall dosimetric uncertainty for the U.S. The low-E and high-E refer to low- and high-energy phot on-emitting sources, respectively, and are representative of both LDR and HDR brachytherapy sources. The symbols and notation in this gure are in accordanc e with the 2004 AAPM TG-43U1 report. Symbols such as EXPL, MCL, and CONL represent the active lengths used by the experimental investigators, Mo nte Carlo simulation investigators, and the consensus value, respectively. Following the ow chart, manufacturers rst create sources and follow the AAPM 2004 CLA subcommittee recommendations for initial source calibrations by sending sources to a primary standards laboratory e.g., NIST then to the seco ndary standards laboratories e.g., ADCLs and experimental dosimetry investigator s . The AAPM and GEC-ESTRO then prepare candidate and consensus dos imetry parameters to serve as reference datasets for widespread and uniform clinical implementation. Clinical medical physicists should use these da ta whenever available and assure proper entry and QA for commissioning in their TPS. At the upper-left, calibration intercomparisons are performed to ensu re manufacturers are in agreement with the primary and secondary standards laboratories. When the clinical medical physicist orders sources for treating a patient, sources are calibrated on site using equipment calibrated at a secondar y standards laboratory or ADCL with direct traceability to a primary standards lab

oratory e.g., NIST according to AAPM 2008 LEBSC recommendations. The patient-speci c source strength SK is entered into the TPS, and clinical treatment planning and treatment delivery are performed. Becquerel LNHB in France perform brachytherapy source ned as a measure of the reproducibility of the result. A calibrations, each measurement system having an associated stable instrument capable of making high-precision measureuncertainty budget. It should be noted that many of these ments is desired since it can be calibrated to provide an acuncertainties affect source parameters before use in the clinic curate result. Uncertainty determination takes into account and the clinical medical physicist has no control over them. measurement or calculation variations, including all of the precisions of the measurements or calculations and their effects on the results. Thus, uncertainty is a part of every meaII. UNCERTAINTY ESTIMATION METHODS surement or calculation. The hardest part of uncertainty deUncertainty is a useful and important concept for quantitermination is to account for all possible in uences. The tatively determining the accuracy of measurements and caluncertainty can be thought of as a de ning interval, which is culations. Uncertainty analysis is different from the outdated believed to contain the true value of a quantity with a certain method of random and systematic errors. The terms accuracy level of con dence. For a coverage factor of 2 see above , and precision are still maintained but with slightly different the true value of the quantity is believed to lie within the de nitions. Accuracy is de ned as the proximity of the result uncertainty interval with a 95% level of con dence. to the conventional true value albeit unknown and is an indication of the correctness of the result. Precision is deThe present-day approach to evaluating uncertainty in Medical Physics, Vol. 38, No. 2, February 2011 785 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unce rtainty recommendations 785 characterized by the measurement result y is approximately measurements is based on that recommended by the Comit International des Poids et Msures CIPM in 1981.8 The normal, then y uc gives an interval within which the true CIPM recommendations included grouping uncertainties into value is believe d to lie with a 68% level of con dence. two categories type A and type B, to be explained below , as Normally, th e symbol U is used to express the expanded well as the methods used to combine uncertainty compouncertainty; how ever, to avoid confusion with the unit U for nents. This brief CIPM document was expanded by an ISO air-kerma streng th, this AAPM/GEC-ESTRO report uses the symbol V for thi s quantity. An expanded uncertainty V = kuc, working group into the Guide to the Expression of Uncer-

overage factor, is typically reported and is tainty in Measurement GUM , rst published in 1993 and subsequently updated in 2010.9 This formal method of asthe combined uncertainty, not at each stage of sessing, evaluating, and reporting uncertainties in measuressuming an approximately normal distribu= 2 de nes an interval with a 95% level of ments was presented in a succinct fashion in NIST Technical Note 1297, Guidelines for Evaluating and Expressing the 3uc k = 3 de nes an interval with a level Uncertainty of NIST Measurement Results 1994 .10 The When there is limited data and thus uc main points of this Technical Note relevant to the current of freedom, k = t factor is determined from on.9,10 report are summarized below. Components of measurement uncertainty may be classied into two types, namely, those evaluated by statistical NCERTAINTY IN methods type A and those evaluated by other means type SIMETRY B . In the past, type A and type B uncertainties were comnumber of uncertainties involved in brachymonly referred to as random and systematic errors more y measurements. These measurements are properly uncertainties , respectively. The use of the term erd at research facilities outside the clinic. ror is discouraged in uncertainty analyses since it implies a igators should propose methods to quantify mistake or refers to the difference between the measured ainties and specify them in their publications. value of a quantity and the true value, which is unknown. For example, what might be considered as an error by one domeasurement uncertainties simetry investigator could be considered an uncertainty by another investigator. Speci cally, investigator 1 might assign cteristics of the source and devices used for a large uncertainty to the dimensions of internal source comrements include knowledge of the source ponents without having rst-hand knowledge of source conion and source-to-detector positioning. struction or the ability to open the capsule. Investigator 2 stics contribute to dosimetric uncertainties, might question the values used by investigator 1, considering he model of source and detector. them erroneous, having opened the capsule and measured the dimensions of the internal components. If the true value was activity distribution known, there would be no need to perform the measurement. ty in source activity distribution on the inter-

where k is the c applied only to an evaluation. A tion, V = 2uc k con dence, and V = of con dence 99%. has few degrees the t distributi

III. MEASUREMENT U BRACHYTHERAPY DO There are a therapy dosimetr usually performe Dosimetry invest all these uncert

III.A. Intrinsic Inherent chara dosimetric measu activity distribut These characteri often speci c to t III.A.1. Source An uncertain

Representing each component of uncertainty by an estimponents becomes a systematic uncertainty, mated standard deviation yields the standard uncertainty, u. ll dosimetric measurements. Most brachyFor the ith type A component, ui = si, the statistically estiare assumed to be uniform about the circummated standard deviation is evaluated as the standard deviaong axis due to their cylindrical symmetry. tion of the mean of a series of measurements. For the jth ity the vast majority of sources demonstrate type B component, u j is an estimate of the corresponding 0% in the intensity of emissions about the standard deviation of an assumed probability distribution gh- and low-energy photon emitters. Such e.g., normal, rectangular, or triangular based on scienti c e ected in the statistical uncertainty of meajudgment, experience with instrument behavior, and/or the surements are made at numerous circumferinstrument manufacturer s speci cations. Historical data in around the source, and the results are the form of control charts from a given measurement process hese variations have been demonstrated in may be used to evaluate type B components of uncertainty. formed at NIST.13 The combined standard uncertainty uc represents the estimated standard deviation of a measurement result and is calto-detector positioning culated by taking the square root of the sum-of-the-squares s of uncertainty arise from the relative posiof the type A and type B components. This technique of rce and detector and depend on the phantom combining components of uncertainty, including relevant detector. If TLDs are used, the detector equations such as the Law of Propagation of Uncertainty, is illustrated in Sec. IV C of the TG-43U1 report.2 In the curs, or capsules of powder may lead to difties in the location of the detector relative to rent report, uncertainty propagation is accomplished by add, generally radiochromic lm, has become a ing in quadrature the relative % uncertainties at each step for brachytherapy dose measurements. The

nal substrate co propagating to a therapy sources ference of the l However, in real variations of 2%2 long axis for hi variations are r surements if mea ential positions averaged.11,12 T calibrations per

III.A.2. SourceSeveral type tions of the sou material and the shape rods, chip ferent uncertain the source. Film common detector

of a measurement traceability chain. This is only the case tainty for lm has two components: the posince the measurement equation is a simple product of mea-

positional uncer

sitioning of the lm and the positional uncertainty for relatsured or calculated quantities. If the probability distribution Medical Physics, Vol. 38, No. 2, February 2011 786 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unce rtainty recommendations 786 ing the reading of the optical density to the position in the complete decay of the source is less than 1.5 cGy. Thus, extending the e xposure time for the more distant points canphantom. For measurements of some parameters, such as not be consider ed equivalent. dose-rate constant and radial dose function g r , the source is positioned normal to the detector plane. A type A dosimetric uncertainty in detector distance from the source relative III.B. Dose mea surement to the mean detector distance appears as an uncertainty in There are u nique challenges to measuring radiation dose detector reading. However, a type B uncertainty in the mean in the presence of either a high dose gradient or a very low distance of a group of detectors must be considered in the dose rate LDR , particularly for low-energy photon-emitting analysis of and g r . For measurements of the 2D anisosources. The ma jor consideration is the need for a detector tropy function F r , , the uncertainty in the distance of each with a wide dyn amic range, at energy response, small geodetector from the source must be determined. In addition, the metric dimensio ns, and adequate sensitivity. Radiation meauncertainty in the angle from the source long axis must be surement device s in general use for brachytherapy source considered. Tailor et al.14 determined the uncertainty k = 1 dosimetry are L iF TLDs, radiochromic lms, diamond, diin the mean distance to the detectors in a water phantom to ode, and metaloxide- eld effect transistor MOSFET detecbe 0.09 mm. However, Tailor et al.15 claimed a seed-to-TLD tors. These det ector types are considered below and may be positioning uncertainty of 0.05 mm k = 3 for a 0.3% type B chosen for thei r dynamic dose range, high-spatial resolution, component dosimetric uncertainty at r = 1 cm. More typical feasibility for in vivo dosimetry, approximation to human soft tissue, or relative ease of use. However, the accuracy of values obtained by a routine investigator would fall around the results fro m these detectors is subject to the uncertainties 0.5 mm k = 1 . due to volume a veraging, self-attenuation, and absorbedThe uncertainty in the detector point of measurement vardose sensitivit y. At the small source:detector distances of ies somewhat with the phantom material and related techbrachytherapy,

detector size can in uence self-attenuation nique. If a water tank scanner is used, there is an uncertainty aging. associated with the movement and positioning. A scanning system might display a source-to-detector positioning precision of 0.1 mm. However, typical positioning accuracy of a luminescent dosimeters water tank scanner is about 0.4 mm, expressed as k = 2.16 The een the main dosimeter used for measureaccuracy is more dif cult to specify, in part, because of the rapy source dose. Typically, these meauncertainty in the source-positioning device and also because been made in solid-water phantoms comof the uncertainty in the effective point of measurement for ics having radiological characteristics similar et al.17 provided characteristics that should be the detector. Considering only the effects of geometry i.e., the inverse-square relationship and ignoring signal variation ime a TLD measurement is made. A calibraacross the detector i.e., a pointlike detector , the dosimetric s to a known energy and dose is necessary to effects of a 0.04 cm positional uncertainty at distances of 1 ry. Two major sources of uncertainty are the cm and 5 cm are 8% and 1.6%, respectively. e used by different investigators and the inFor dose rate measurements of the same duration at these dependence kBq Q , which is per unit of activpositions, the reading at 5 cm is 25 times lower than the 1 cm erel . Depending on the temperatures and materials, the uncertainty can increase drasreading due to the inverse-square effect alone, not account% to 5%. The uncertainty is reduced when ing for medium attenuation. For measurements involving is used in the handling, reading, and irradialow-energy photon emitters, the relative signal at the greater . The other large source of uncertainty is the distance is considerably lower due to medium attenuation e TLD absorbed-dose sensitivity between the that is not compensated by increased scatter. Most often, the calibration and that of the brachytherapy detectors used for brachytherapy dosimetry measurements the uncertainty in the relation of the energy are not limited by counting statistics, but rather intrinsic he absorbed-dose sensitivity relative to that properties such as signal-to-noise ratio and detector reprolity used for calibration. Each reading regime ducibility. This often produces an uncertainty at 5 cm about ame to reduce the variation. The characteris-

and volume aver

III.B.1. Thermo TLDs have b ment of brachythe surements have prised of plast to water. Kron reported each t tion of the TLD perform dosimet annealing regim trinsic energy ity i.e., Becqu cooling for the tically, from 1 meticulous care tion conditions variation in th energy used for source. This is dependence of t in the beam qua should be the s

ten times larger than that at 1 cm. When compared to sourcet thermoluminescence are elaborated upon in to-detector positioning uncertainty, there is partial compen2009 AAPM Summer School text.18 If care sation between these two effects. The decreased signal with h of the regimes, an overall estimate of the distance can sometimes be overcome when using integrating ainty to measure absorbed dose would be dosimeters simply by leaving the dosimeters in place for a longer time. Radionuclides with short half-lives limit the improvement that can be achieved by increasing the exposure

tics that affec Chap. 24 of the is taken in eac expanded uncert 5.58% k = 2 .19

III.B.2. Radioc hromic lm duration. For 103Pd, with a 17-day half-life, the dose rate at 5 cm is only 0.4% of that at 1 cm in water. To obtain 1 Gy at Radiochromi c lm has become a common detector for 1 cm from a 1 U source requires about 6.9 days. At 5 cm brachytherapy m easurements. Advantages of EBT lm comfrom this same source, the maximum dose possible after pared to silver halide lm include the following: relative Medical Physics, Vol. 38, No. 2, February 2011 787 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unce rtainty recommendations 787 IV. MONTE CARLO U NCERTAINTY energy insensitivity, insensitivity to visible light, selfIN BRACHYTHERAPY DOSIMETRY developing characteristics, greater tissue equivalency, and dose-rate independence.2023 Different investigators have While MC method s may be used to characterize brachynoted up to 15% variation in the lm response throughout a therapy source dosi metry accurately, there are both obvious lm that was exposed to a uniform dose of radiation. Sources and hidden uncertai nties associated with the process that for these uncertainties have been pointed out by Bouchard et must be accounted for. For large numbers of histories where al.24 Looking at two orthogonal directions, the lm response Poisson statistics applies, the uncertainty in the estimated is more uniform in one direction. Applications of various results decrease by the inverse square root of the number of models of radiochromic lm in radiation dosimetry have particle histories. This uncertainty is referred to as the type A been discussed in detail in AAPM TG-55 Ref. 25 and more uncertainty for M C methods and should be kept to 0.1% recently by Soares et al.26 Radiochromic lm response is when feasible so as to be negligible in comparison to other independent of dose rates in the clinical range of 0.1 4 Gy/ computational uncer tainties. In many cases, it is unfeasible min. Dini et al.23 showed that the responses of both XR type to simulate addit ional histories due to processing power and T and type R lms were independent of the dose rate. The time constraints. W hile variance reduction techniques are results of their investigations showed a 5% variation for dose sometimes used to diminish type A uncertainties, careful

rates ranging from 0.16 to 7.55 Gy/min. These results were equired for radiation transport codes and in good agreement with the nding of Giles and atures and subroutines. The MC dosimetric Murphy,27who had shown that XR type lms are dose-rates presented in Table XII of the TG-43U1 independent within 5%. In an independent investigation, r separate components2 and has been subSaylor et al.21 showed 5% variation in optical densities of ed here into eight separate components all HD-810 lm for dose rates ranging from 0.02 to 200 Gy/ B . These roughly correspond chronologimin. However, many of the reports in the literature pertain to int particle transport with the MC simulaolder lms that are not useful for current brachytherapy meast be estimated by each dosimetry invessurements. The manufacturer discontinued production of peci c source and circumstance being EBT lm and now only provides the EBT2 model. The doly, example tables are not provided since simetric uncertainties of brachytherapy source measurements ependent on the energy of the source emismade with EBT2 are increasingly being investigated.28 30 gn, simulation goals, and MC code. This Before use, the dosimetry investigator should be aware of the s the simulation process and current statecharacteristics of the individual type of lm. rtainty analyses. It is important to clarify In general, the handling of the lm can be important so arrive at values for the dosimetric comthat exposure to ultraviolet light and other conditions are ies and aspire to minimize these uncertainminimized; again the uncertainty can be reduced if this care d sources may differ from their design; MC is taken. An estimate of the expanded uncertainty to measure d be performed with representations of the absorbed dose is 10%.30,31 Due to the increasing number of ivered product. What follows are descripdifferent radiochromic lms and their dependence on scanies that arise throughout the process of ning techniques, caution is recommended. In addition, it is to simulate dose-rate distributions in the important to realize that the scanner can have a signi cant herapy sources. Dosimetry investigators effect on the results of the lm.32 While investigations have er these analyses and introduce detailed been made for various scanners such as by Hupe and re brachytherapy dosimetry publications. Brunzendorf31 and by Alva et al.,33 there have been con icting results requiring further research. truction ion of brachytherapy dose-rate distributions l understanding of the source construction. In III.B.3. Diamonds, diodes, and MOSFETs erapy sources contain radionuclides that are Occasionally, measurements in a water phantom use diode

benchmarking is r their individual fe uncertainty analysi report listed fou stantially expand but one being type cally for nonadjo tion process and mu tigator for the s studied. Consequent the results are d sions, capsule desi subsection review of-the-art for unce the methods used to ponent uncertaint ties. Manufacture simulations shoul nal clinically del tions of uncertaint using MC methods vicinity of brachyt are urged to consid estimates in futu

IV.A. Source cons Characterizat starts with a ful general, brachyth sealed in a singl

e capsule. High dose-rate HDR sources or diamond detectors, but their dosimetric uncertainties can attached to a delivery cable used to position exceed 15% k = 1 for low-energy photon-emitting brachyurce at multiple locations within the patient. therapy sources.34 These uncertainties result from the large sources are similar to HDR, but the treatenergy dependence of its absorbed-dose sensitivity, nonlinn a protracted manner. LDR sources may be earity, directional dependence, temperature dependence, and vidual entities and do not utilize a delivery bias dependence, especially when used for low-energy hey may be contained within metal or plasbrachytherapy sources. Diode characteristics are given in the a surgical suture material as is the case for AAPM TG-62 report by Yorke et al.35 MOSFET dose reith the current TG-43 dosimetry formalism sponse is also energy and dose-rate dependent.36,37 While urces based on superposition of individual MOSFETs have been used for brachytherapy in vivo 30 cm diameter water phantom to provide dosimetry,38,39 they have not been used to date for direct itions for r 10 cm, all that is required is

have the capsule the individual so Pulsed dose-rate ment is applied i described as indi cable. However, t tic cylinders or stranded seeds. W for low-energy so sources within a full-scatter cond

characterization of the active radionuclide and the source.2 dosimetric parametrization of brachytherapy sources. Medical Physics, Vol. 38, No. 2, February 2011 788 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unc ertainty recommendations 788 ascribe MC dosi metric uncertainties to this component, the The dosimetry investigator should independently assess full range of m otion should be considered, along with possiavailable manufacturer data on source construction, estimate bilities for co n guring internal components if multiple items the uncertainties associated with each dimension, and estiare free to mov e and subtend different geometries upon setmate the distribution of results within the available range of tling within th e capsule. An example is provided. results. A theoretical example is provided for how to characterize the source geometry uncertainties for a hypothetical a For the example given in the source geometry uncerbrachytherapy source. tainty d escription, the Ir pellet could move 0.25 mm along th e capsule long axis and 0.035 mm in the a The capsule is a right cylinder made of pure 100% titanium = 4.51 0.05 g / cm3 , with inner and outer lateral direction within the capsule due to a combination of dimensional tolerances. diameters of 0.70 and 0.80 mm rectangular distribub In addit ion to the aforementioned shifts, the pellet

tion over a tolerance of 0.02 mm , respectively, overssibly rotate within the capsule. all length of 4.52 mm rectangular distribution over a tolerance of 0.05 mm , and end-weld thicknesses of e single internal component Ir pellet is well 0.15 mm rectangular distribution over a tolerance of d dosimetric uncertainties due to a dynamic 0.03 mm . ent would be small compared to other dosib The capsule is lled with room temperature Ar gas nty components. However, this would not be = 1.78 0.04 mg / cm3 and an Ir pellet. internal component containing a low-energy The Ir source pellet = 22.56 0.01 g / cm3 is a right c radionuclide were much smaller and nestled cylinder with a 0.66 mm diameter rectangular distriopaque marker where the radiation emissions bution over a tolerance of 0.01 mm and 4.10 mm ntially attenuated in comparison to an optiactive length L rectangular distribution over a tolerfor the internal components. 0.02 mm with a 192Ir loading of ance of that the dynamic internal components of 1011 atoms uniformly distributed 3.2 0.2 e the largest in uence on dose rate variations throughout the pellet. be considered for the source models, posist, and source orientation relevant to the cliniThis description presents k = 1 uncertainties associated . In general, the dosimetric uncertainty related only with capsule dimensions, internal components, and loponent movement increases as photon energy cation of radiation emission. A sophisticated MC dosimetric e not an important aspect for all sources, the analysis would simulate the in uence of varying each of tigator should assess the impact of this effect these components and estimate the resultant effect of these source being examined since some sources are

could po

Clearly, th constrained, an internal compon metric uncertai the case if the

photon-emitting behind a radiowould be substa mized geometry

It appears

sources can hav and thus should tions of intere cal application to internal com decreases. Whil dosimetry inves for the type of

uncertainties on the calculated dose distribution. Karaiskos et al.40 investigated the effect of varying the silver halide coat- fairly suscepti ble to this effect previously mentioned factor ing thickness i.e., 1 10 m for an 125I source; and g r of 2 where other sources exhibit less than a 0.1% dosimetric effect at the r eference position.43 Time-averaged internal were unchanged within 1%. Koona41 assessed variable 125I component posit ions should be used for reference data, and source capsule wall thickness i.e., 30 100 m and found the dosimetric uncertainties for all possible internal compoan in uence on ranging from +16% to 1%. For similar nent positions should be considered. endweld thicknesses, differences in ranged from 0.2% to 0.9%. However, the variation in the endweld thickness led to a signi cant impact on F r , for small polar angles. IV.C. Source emis sions Brachythera py sources generally contain radioactive maIV.B. Movable components terials and hav e capsules to prevent direct contact of the radioactive mat erials with patients. Exceptions include elecAs shown by Rivard, the internal components within the capsule may change position.42 The dimensions from source tronic brachyth erapy sources, which generate radiation without radionuclid es,44,45 and the 103Pd RadioCoil source.46 to source may vary also. At distances of a few millimeters Since nuclear d isintegration processes are well understood, from some sources, the dose rate can change more than a factor of 2 upon varying the capsule orientation.42 Since there is little uncertainty associated with knowledge of the radiation spect rum from the radioactive materials. A general most low-energy sources do not have their internal compouncertainty in dose rate per unit source strength at P r0 , 0 nents rigidly attached to the encapsulation, it is possible that of 0.1% for low -energy sources43 and 0.5% for high-energy the internal components may move about based on the sources47 may b e assumed. However, physical fabrication of source orientation. Especially for a low-energy photonbrachytherapy s ources often involves radiochemistry and emitting source containing radio-opaque markers for localother processes to purify the isotopic and elemental compoization, such dynamic aspects may be of clinical relevance sition of the r adioactive product. With radiocontaminants under certain circumstances. While this effect can be obhaving differen

t half-lives than the desired radionuclide, served experimentally when the source orientation is rotated bstantial uncertainty concerning the radionu180, this behavior is readily assessable using MC methods, d in the source. When simulated using MC but more challenging with experimental techniques where

there may be su clides containe

methods, the do simetry investigator is advised not to assume localization of the internal components may be unknown. To Medical Physics, Vol. 38, No. 2, February 2011 789 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unce rtainty recommendations 789 investigator aim s to simulate the geometry of a physical exa pure radioactive product and to include the contaminant periment. Here, the setup will often include a plastic medium radionuclides and daughter products in the carrier material if in place of liqu id water. Due to the variable nature in fabrithe presence of such contaminants has been veri ed masscating these pla stic media, the dosimetry investigator is adspectroscopy measurements and/or photon spectrometry vised to determi ne the composition and mass density indemeasurements . Further, electron dose contributions from pendently and as sign uncertainties to this assessment. sources generally considered as photon emitters should be considered.48 50 Furthermore, these uncertainties directly impact the resultant dosimetric uncer tainties, which should be assigned to the The National Nuclear Data Center NNDC at phantom composit ion. In contrast to phantom size, the MC Brookhaven National Laboratory is an internationally regarded reference for radionuclide radiation spectra.51 This dosimetric uncer tainties due to phantom composition generally increase wi th decreasing photon energy and increase database includes all of the commonly used radionuclides in with increasing radial distance. brachytherapy, often listing the precision of photon and elecSpeci cation o f a solid phantom material is important for tron energies to four signi cant digits and the emission indosimetric evalu ation of brachytherapy sources, particularly tensities to three signi cant digits and probabilities to parts for low-energy p hoton-emitting sources.16,55 Meigooni et per million. As a result of uncertainties in the source photon al.55 showed tha t a 0.4% difference in the calcium content of energies and the exaggerated precision of emission probabili-

hantom material may lead to 5% and 9% ties, the dosimetry investigator should consider the in uence for 125I and 103Pd sources, respectively. of an inaccurate spectral characterization on the resultant e in good agreement with the published data dose distribution. This latter feature would be most meaning56 who performed a robust material analysis of ful for considering relatively new radionuclides, sources with osition. In addition, Meigooni et al. novel means of generating radiation, and sources that contain t of the phantom composition on g r for radionuclides which emit both photons and electrons. 3Pd sources.55 Small differences in phantom IV.D. Phantom geometry to large differences in g r for low-energy Differences were more signi cant at larger Phantom size has a signi cant effect on brachytherapy source, and they concluded that one must dose distributions.5254 Although variations in radiation scatection factors based on correct chemical ter and attenuation are readily accounted for with modern cross-section data when extracting a conexternal-beam TPS, brachytherapy TPS generate dosimetry tric parameters for a brachytherapy source data based on brachytherapy dosimetry parameters and may TG-43U1 protocol.2 Dosimetric uncertainnot account for full-scatter conditions or appropriate scatter m uncertainties in phantom composition are conditions for the task at hand. Thus, the dosimetry investid as type B. gator should describe the phantom size used in the simulations and should estimate the in uence of scatter conditions over the positions in which dose was calculated. The current transport code brachytherapy dosimetry formalism,2 based on the AAPM use approximations and assumptions when TG-43 report,3 stipulates that MC calculations be performed logical interactions. For example, generation in a 15 cm radius liquid water phantom to provide at least 5.0 on emissions following characteristic x-ray cm of radiation backscatter for low-energy photon-emitting

the Solid Water p

differences in These results ar by Patel et al., the phantom comp showed the impac both 125I and 10 composition lead photon emitters. depths from the use updated corr composition and sensus of dosime by means of the ties arising fro typically classi e

IV.F. Radiation All MC codes simulating radio of multiple-phot

e simpli ed to the most probable photons, sources such as 125I and 103Pd at the farthest position from nore electron binding effects, and electron the source. By the current AAPM de nition, low-energy en reduced to a multigroup algorithm or igphoton-emitting sources are those which emit photons of enAlthough molecular form factors can be used ergy less than or equal to 50 keV.2 Under these circumhere is no signi cant dosimetric effect when stances for a 50 keV photon-emitting source, approximately dent-atom approximation for coherent scat5.0 and 7.5 cm of backscattering material are needed to ors.54 Speci c to the use of radiation transsimulate in nite scatter conditions within 3% and 1%, etermining brachytherapy dose-rate distriburespectively.53 Thus, the initially recommended 5.0 cm of a practical energy limit for simpli cation to a backscatter to simulate in nite scatter conditions within 1% sport technique at the exclusion of coupled applies only for photon-emitting sources with E 40 keV. transport, and high-energy photon-emitting ch as 192Ir and 137Cs may not be simulated IV.E. Phantom composition close to the source. Electron contributions Presently, the TG-43 dosimetry formalism does not acc uncertainty could be negligible given accucount for material heterogeneities and recommends liquid quations, empirically derived atomic form water as the reference media for speci cation of in vivo doser implementation of the code by the dosimrate distributions. Being a simple and readily available mar. However, dosimetric differences within 1 urce capsule between photon-only and terial, it is not challenging to simulate the composition lectron transport may exceed 15%.49,50,57 H2O and mass density = 0.998 g / cm3 at 22 C of liq-

production may b some MC codes ig transport is oft nored entirely. in some codes, t using an indepen tering form fact port codes for d tions, there is photon-only tran photon-electron radionuclides su accurately when to the dosimetri rate transport e factors, and prope etry investigato mm of a 192Ir so coupled photon-e

Estimates of k = 1 dosimetric uncertainties due to the physics uid water. However, care must be taken when the dosimetry Medical Physics, Vol. 38, No. 2, February 2011 790 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unce rtainty recommendations 790 inversely propor tional to the square root of the voxel volume implementation within MC radiation transport algorithms at

nced by voxel thickness along the radial r = 1 cm are 0.3% and 0.2% for low- and high-energy sources, respectively, and 0.7% and 0.3% at r = 5 cm.43,47 ation of brachytherapy dosimetry parameters F r , , and an r using MC methods inIV.G. Interaction and scoring cross sections tion of results over various tallied voxels, s based on solid angle, or taking ratios of With the computational geometry established, progression ates. Since all brachytherapy dosimetry paof radiation transport is governed by atomic and nuclear ios of dose rates, except for , it is often cross sections that dictate the type and frequency of radioto simply take ratios of the raw simulated logical interactions. These cross sections are organized into tic uncertainties in postsimulation processing libraries that are maintained by international agencies such nergy thresholds ,2 intentional volume avas the NNDC. Uncertainties in the cross sections within the y energy modi ers are employed. Further resource affect radiation emitted in the phantom. These cross uncertainties is needed. sections are typically calculated and compared to experimental cross sections, determined at discrete energies. Given the N THE TG-43 DOSIMETRY physics model used to characterize the element and radioTERS logical interaction, a tting function such as a log-log t is used by the radiation transport code to interpolate between is a quantitative assessment of dosimetric reported cross-section values. Since the interpolation t may he brachytherapy dosimetry parameters used not be robust for all element and energy possibilities, it is e calculation formalism. The reader is direcommended to use the recently derived cross-section li04 AAPM TG-43U1 report for de nitions of braries with high resolution in energy. Sensitivity of dosimy dosimetry parameters.2 The tables in the etric results on cross-section libraries was illustrated by Deresent best practice values for propagated unMarco et al.58 are not meant to be used for uncertainty budMC-based radiation transport codes utilize en / toward calculating dose rates and are separated from / as, for example, one could determine dose to muscle in water intrength stead of dose to water in water. Here, the / and en /

and are thus in ue direction. Deriv such as , g r , volves the summa weighting result simulated dose r rameters are rat straightforward results. Systema may arise when e eraging, or tall search on these

V. UNCERTAINTY I FORMALISM PARAME What follows uncertainties in t in the TG-43 dos rected to the 20 the brachytherap current report p certainties and gets. V.A. Air-kerma s

values for water and muscle would be used, respectively. ty in NIST primary standard for Thus, the uncertainties k = 1 in both / and en / are of hoton-emitting sources concern and are about 1.2% and 1.0% for low- and highional primary standard of air-kerma strength energy sources, respectively.59,60 The in uence of the crossnergy 50 keV photon-emitting section uncertainties on the absorbed dose is a function of urces, containing the radionuclide 103Pd, distance from the source with larger distances subject to is realized using the NIST wide-angle free-air larger dosimetric uncertainties. For low-energy sources, the 3 The WAFAC is an automated, free-air dosimetric uncertainties at 0.5 and 5 cm are about 0.08% and er with a variable volume. As of October 0.76%, respectively; with high-energy sources, dosimetric sources of 41 different designs from 19 uncertainties are 0.01% and 0.12% for these same ve been calibrated using the WAFAC since distances.43,47 Further research on a modern assessment of ed uncertainty k = 2 in SK,NIST is given as cross-section uncertainties is needed. i2 + u j2 , 1 IV.H. Scoring algorithms and uncertainties l to the standard deviation of the mean of All the prior steps set the simulation framework in which ements type A and the quadrature sum of the calculations are performed. The dosimetry investigator nents of uncertainty is represented by u j .64 must select the scoring algorithm used to determine the doserate distributions. While some estimators are more approprie SK,NIST measurement, the responses of sevate than others,61 none will truly represent the desired output onization chambers of different designs are resultant from the dosimetry calculations. Typically, some . To understand the relationship between form of volume averaging or energy-weighted modi cation nse I and WAFAC-measured SK,NIST for will be used to determine the dose rate at a given location n brachytherapy sources, emergent photon within the calculation phantom. These uncertainties should ured with a high-purity germanium specbe 0.1% for all classes HDR/LDR and low/high energy lative response of calibration instruments has of brachytherapy sources. For path-length estimators used to depend on both emergent spectrum and owledge of source spectrum allows separadetermine collisional kerma, decreases in voxel thickness along the radial direction will diminish volume averaging mber response effects due to spectrum differwithin the voxel without signi cant in uence on the type A sed by variations in the spatial anisouncertainties.62 However, MC estimators based on energy

V.A.1. Uncertain LDR low-energy p The U.S. nat SK,NIST for low-e brachytherapy so 125 I, or 131Cs,

chamber WAFAC .6 ionization chamb 2010, over 1000 manufacturers ha 1999. The expand VWAFAC = 2 s where si is equa replicate measur all type B compo less than 0.8% Following th eral well-type i measured at NIST well-chamber respo low-energy photo spectra are meas trometer. The re been observed to anisotropy.64 Kn tion of well-cha ences from those cau tropy of emissio

ns due to self-absorption by internal source deposition within the voxel will have type A uncertainties components. Medical Physics, Vol. 38, No. 2, February 2011 791 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unce rtainty recommendations 791 atmosphere.69 Fo r routine calibrations, a spherical-Al cavity To verify that sources of a given design calibrated at chamber with sev eral 137Cs working standard sources is NIST are representative of the majority of those calibrated in used. The expand ed uncertainty k = 2 in SK,NIST for LDR the past, several additional tests have been implemented. The 137 Cs sources is 2%. As is the case with LDR 192Ir sources, distribution of radioactive material within a source is mapped using radiochromic lm contact exposures. The inwell-chamber respo nse is relatively insensitive to small air anisotropy of sources is studied by taking WAFAC and changes in sourc e construction. Additional characterization x-ray spectrometry measurements at discrete rotation angles measurements per formed on the sources following calibraabout the long axis and the axis perpendicular to the midtion include wel l-chamber response and radiochromic lm contact exposure measurements.70 point of the source long axis, respectively. The airanisotropy ratio, calculated from the results of angular x-ray At NPL, air-ke rma rate calibrations are performed for 192 measurements, has proven to be a useful parameter for exIr wires and pins using the secondary standard radionuplaining differences in well-chamber response observed for clide calibrator , which is traceable to the NPL air-kerma primary standard. T he expanded uncertainty k = 2 for an 192Ir different source models having the same emergent spectrum on their transverse plane.65 The rst primary standard device air-kerma rate mea surement is stated to be 1.5%.65 in Europe for calibration of low-energy photon sources was the large-volume extrapolation chamber built at the PTB V.A.3. SK uncert ainty for HDR high-energy sources where the procedures are, in principle, the same as at NIST.66 NIST traceab ility for the measurement of air-kerma For each seed type not necessarily for each individual seed strength for HDR 192Ir sources is based on the interpolation of same type , the spectral photon distribution to obtain the of air-kerma cal ibration coef cients of a secondary standard spectrum dependent correction factors for air attenuation, ionization chamb er.71 The weighted average-energy of these scattering, etc., is determined. Details are given in Ref. 66. sources is 397 k eV and thus an interpolated value between

Using a sensitive scintillation detector free-in-air at 1 m, points of 137Cs and 250 kVp x rays is used. both polar and azimuthal anisotropies are measured for each gorous methodologies for the ionization individual seed to be calibrated. The results of the anisotropy r-kerma calibration coef cient have been measurements are part of the calibration certi cate. The NPL with Eq. 2 from Eq. 1 of Ref. 72, also provides air-kerma rate calibrations of 125I sources using their secondary standard radionuclide calibrator, a well-type 1 1 1 Cs-137 + x ray , 2 ionization chamber for which the calibration coef cient is 2 NK NK traceable to the NIST primary air-kerma standard.64 agreement within 0.5%, falling within the Ir-192 is the ionization chamber y k = 2 . N S V.A.2. Uncertainty in NIST primary standard for LDR K ation coef cient for 192Ir or as designated high-energy photon-emitting sources The U.S. national primary standard of SK,NIST for LDR o techniques to measure SK using an ionizahigh-energy gamma-ray-emitting brachytherapy sources conibrated as above, the shadow shield method taining the radionuclide 192Ir is realized using a spherical stance technique. The seven-distance techgraphite-wall cavity chamber that is open to the e ned and the results for SK from all HDR atmosphere.67 Since arrays of approximately 50 sources were ufacturers have been found to agree to required to perform the cavity chamber measurement due to ir-kerma strength is thus given as low detector-sensitivity, the SK,NIST of individual sources is

the calibration However, more ri chamber 192Ir ai suggested,72,73 1

= Ir-192 N SK which results in

2.15% uncertaint

air-kerma calibr 137

Cs or x ray . There are tw

tion chamber cal and the seven-di nique has been r 192 Ir source man

within 0.5%.74 A Ir

-192 Md MS d + c

2 NS

determined by using a spherical-Al re-entrant chamber workK , 3 SK =

ing standard with a 226Ra source to verify the stability of the t re-entrant chamber over time. The expanded uncertainty k 2 is the air-kerma calibration coef cient for 192Ir, = 2 in SK,NIST for LDR 192Ir sources is 2%. Well-chamber response is not as sensitive to small changes in source cont measurement including the primary beam struction due to manufacturing variability for high-energy tance to the source center d, setup distance photon emitters in comparison to low-energy sources.68 Nevadiation time t. The value of SK is then ertheless, additional characterization measurements are perwell-type ionization chamber. r-kerma standards are established at LNHB, formed on the sources following calibration, including wellAn intercomparison of the University of chamber response, photon spectrometry, and radiochromic lm contact exposure measurements. The results of these ed Dosimetry Calibration Laboratory measurements are used to verify that no signi cant modi andard with the LNHB calibration agreement for a speci c HDR 192Ir source cations to the LDR low-energy source design have been ntercomparisons done between NPL and implemented by the manufacturer. d agreement to within 0.3% to 0.5%.77 Similar to 192Ir, the U.S. national primary standard of SK,NIST for LDR high-energy photon-emitting brachytherapy analysis is performed for all other HDR sources containing 137Cs is also realized using a spherical els and intercomparisons, the overall ex-

Ir-19

where N S

K M d is the direc scatter M s, dis error c, and irr transferred to a HDR 192Ir ai PTB, and NPL.75 Wisconsin Accredit ADCL calibration st standard showed within 0.3%.76 I LNHB demonstrate When uncertainty 192 Ir source mod

panded uncertain ty k = 2 for SK is 2.15%.73,74 LNHB graphite-wall cavity chamber that is open to the Medical Physics, Vol. 38, No. 2, February 2011 792 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry uncer tainty recommendations 792 TABLE I. Propagation of best practice uncertainties k = 1 un less stated otherwise associated with the transfer of air-kerma strength from NIST through the ADCL to the clinic for LDR low-energy brachytherapy sources.

Row Quantity units SK,NIST U SK,NIST / IADCL U/A 1 2

Relative propagated Measurement description uncertainty % 0.8 NIST WAFAC calibration

ADCL well ion chamber calibration 0.9 1.1 ADCL calibration of source from manufacturer 1.2 ADCL calibration of clinic well ion chamber Clinic measures source air-kerma strength 1.3

SK,ADCL U

SK,ADCL / ICLINIC U/A 4 5 SK,CLINIC U SK,CLINIC U

2.6 Expanded uncertainty k = 2 achieves a HDR 192Ir calibration uncertainty k = 2 of 1.3% ents as well as the ability to monitor possible modi cations for well-type ionization chambers.76 Given the assortment of n LDR low-energy seed construction. Data from NIST, the DCLs, and the source manufacturer for each seed model HDR high-energy sources and a variety of calibration methre plotted as a function of time such that the integrity of the ods used at the various primary standards laboratories, the easurement traceability chain is veri ed. This process proaforementioned calibration uncertainties are not necessarily ides assurance that any ADCL secondary standard has not indicative for other sources or other laboratories. hanged since the initial transfer within the uncertainty level, erving as a monitor for consistency. Based on the data colV.A.4. Transfer of NIST standard to the ADCLs ected by NIST and the ADCLs over many years, it appears The AAPM ADCLs are responsible for transferring a hat the accuracy achievable in a secondary standard is not traceable calibration coef cient to the clinics. Therefore, the he same for all source models. Variations in emergent specADCLs maintain secondary air-kerma strength standards usrum and spatial anisotropy of emissions in uence well ing well-type ionization chambers, which are directly tracehamber to WAFAC response ratios, and how well such able to NIST to a great precision and add about 0.1% to the ariations are minimized during source fabrication affects

m i A a m v c s l t t t c v

uncertainty budget. The AAPM Calibration Laboratory Ache magnitude of variability in well-chamber measurements creditation subcommittee monitors this traceability. ADCLs or sources of supposedly identical construction. establish their on-site secondary standard by measuring the response of a well chamber to a NIST-calibrated source. The ratio of air-kerma strength SK to I yields a calibration coefcient for a given source type. The ADCLs use their cali.A.5. Transfer of NIST standard from ADCLs to brated well chamber and manufacturer-supplied sources to he clinic calibrate well chambers for clinics. Calibrations of electromThe use of an ADCL-calibrated well-ionization chamber eters and instruments monitoring atmospheric conditions are s the usual manner for clinics to measure the strength of also necessary to complete the system. For low-energy heir brachytherapy sources. Therefore, the uncertainty in the sources, intercomparisons among ADCLs and pro ciency ell-chamber calibration coef cient for the speci c type of tests with NIST ensure that each ADCL is accurate in its ource used is the key component that creates the nal undissemination, and that the calibrations from different ADertainty in the air-kerma strength measured at the clinic. CLs are equivalent. Europe does not yet have the same scale Following the primary standard measurement of airof infrastructure for low-energy source calibrations as does erma strength SK,NIST , the response usually a measured the U.S. urrent I of a well-ionization chamber is determined. The For LDR low-energy photon-emitting brachytherapy K / I ratio yields a calibration coef cient for the wellsources, the NIST air-kerma strength standard for each new onization chamber for a given source type. Such calibration source model is initially transferred to all ADCLs that are oef cients enable well-ionization chambers to be employed accredited by the AAPM to perform brachytherapy source t clinics for calibration of source air-kerma strength. To calibrations by sending a batch of three WAFAC-calibrated odel the traceability of measurements performed on sources, in turn, to each ADCL. To ensure that the NISTrachytherapy sources from the primary standard measure-

t f

V t

i t w s c

k c S i c a m b

traceable standard at each ADCL remains consistent over ent of air-kerma strength at NIST to the transfer of this time with the initial baseline values, subsequent batches of tandard to the ADCLs and source manufacturers to a nal three sources of each model are calibrated by NIST and circulated among all ADCLs at least annually.78 A NISTeri cation of source strength at a clinic prior to their use in reatment, uncertainties have been assigned based on NIST traceable air-kerma strength standard for both high-energy gamma-ray-emitting brachytherapy sources i.e., 192Ir and easurement histories to SK,NIST and I as %uc,WAFAC 137 0.8% k = 1 and %uc,I = 0.5% k = 1 . These values are Cs has been available from all ADCLs for many years. ropagated through the measurement traceability chain in Supplementary measurements performed at NIST, including wo paths, the rst of which is shown in Table I. Although I, photon spectrometry, and anisotropy characterization, pro-

m s v t m = p t

t his model is applied to measurements of a single low-energy vide quality assurance QA checks for WAFAC measureMedical Physics, Vol. 38, No. 2, February 2011 793 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unce rtainty recommendations 793 TABLE II. Propagation of best practice uncertainties k = 1 u nless stated otherwise associated with the transfer of the air-kerma strength standard from NIST to the manufact urer for LDR low-energy brachytherapy sources. Row Relative propagated Measurement description uncertainty % 0.8 NIST WAFAC calibration

Quantity units SK,NIST U

1 2

SK,NIST / I M U/A

Manufacturer well ion chamber calibration 0.9 1.1 Manufacturer calibration of QA source 1.2 Manufacturer instrument calibration for assay Manufacturer assays production sources 1.3 1.4 or 2.4 Manufacturer places sources in 2% or 7% bins 2.8 or 4.8 Expanded uncertainty k = 2

S K, M U SK,M / I M U/A S K, M U SK,M bin U SK,M bin U

3 4 5

ti

on current is measured and multiplied by the calibration photon-emitting source, the same analysis may be applied to ef cient, yielding an air-kerma strength SK,M for the source high-energy photon-emitting sources by using the appropriate uc values. ow 5 . Finally, in row 6, the source is placed in a 2% wide In row 1 of Table I, the air-kerma strength SK,NIST of a n with other sources of air-kerma strength SK,M bin 1%. source is measured, which is then sent to an ADCL. The me manufacturers have larger bin sizes, up to 7% wide. response of an ADCL standard well-ionization chamber is erefore, a range is included in row 6 of Table II to account measured, yielding a current IADCL. A calibration coef cient the range in bin sizes. The source is then sent to a clinic for the chamber SK,NIST / IADCL is then calculated row 2 . The r patient treatment. According to this model, the propagaADCL receives a source from the manufacturer row 3 , and on of uncertainties from the various well-chamber measurethe air-kerma strength SK,ADCL is calculated based on the nts involved in the transfer of the source-strength standard standard well-chamber current measurement and the calibrathe manufacturer, including binning, results in a minimum panded uncertainty k = 2 in SK,M bin of 2.83%. To evalution coef cient for the chamber. To transfer the source calibration to the clinic, a well chamber from the clinic is sent to e the uncertainty due to binning, the binning process is an ADCL, where the calibration coef cient SK,ADCL / ICLINIC ted as an additive perturbation such that is determined row 4 . Finally, in row 5, the well-chamber S K, M bin = S K, M + B , ionization current is measured and multiplied by the calibration coef cient, yielding an air-kerma strength SK,CLINIC for e B is the bias associated with placing a seed of airthe clinical source. According to this model, the propagation rma strength SK,M in a bin of center value SK,M bin. The bin of uncertainties from the various well-chamber measuredth is modeled by a rectangular distribution, yielding a ments involved in the transfer of the source-strength standard mponent of uncertainty due to binning of 0.6% for a 2% to the clinic results in a minimum expanded uncertainty k de bin and 2.0% for a 7% wide bin. The minimum uncer= 2 in SK,CLINIC of 2.56%. This level of uncertainty assumes inty in SK,M bin k = 2 is therefore 2.81%, increasing to that the clinic is measuring a single seed with a high-quality 78% for the widest bin in this model row 6 in Table II . electrometer and other reference-quality measurement equipNow the question may be asked, How well should the ment. An alternate method of calibration, instead of the wellinical determination of source air-kerma strength chamber calibration, is for the clinic to purchase a source and SK,CLINIC based on an ionization current measurement in a send it to the ADCL for calibration. When this calibrated librated well chamber agree with the value SK,M bin prosource is sent to the clinic, it is used to calibrate the clinics d by the manufacturer? To answer this question, one well chamber. This procedure results in an additional uncerst rst establish a source acceptance criterion. One possi4

co r bi So Th for fo ti me to ex at trea

wher ke wi co wi ta 4.

cl

ca vide mu

tainty of 0.6%, resulting in a total uncertainty of 2.83% at lity is to require that the absolute value of the difference k = 2. tween the air-kerma strength stated by the manufacturer The second path of the measurement traceability chain is ,M bin and that determined by the clinic SK,CLINIC be less illustrated in Table II. Following measurement of air-kerma an the propagated uncertainty of that difference with an strength SK,NIST at NIST, a source is returned to the manupropriate coverage factor according to facturer. The response of a manufacturer s well-ionization 2 2 K,CLINIC SK,M VS + VS chamber is measured, yielding a current IM . A calibration bin K, M bin K,CLINIC coef cient for the chamber SK,NIST / IM is then calculated row 2 . For QA purposes, the air-kerma strength SK,M of a refer VS . 2

bi be SK th ap

K,WAFAC

5 Si ment no from me wi unce = 1 clin 0.5

nce VSK,WAFAC is common to both paths of the measureence source is calculated based on well-chamber current measurements and the chamber calibration coef cient row traceability chain, it is removed in quadrature so as 3 . This reference source is used to determine the calibration t to be added twice. Using the uncertainties determined coef cient SK,M / IM for a well-ionization chamber located on the model at the ends of the two paths of the measurethe source production line row 4 . To verify source strength nt traceability chain, SK,CLINIC must agree with SK,M bin to as part of the production process, the well-chamber ionizathin 3.4% assuming 2% bins in order for the source to be Medical Physics, Vol. 38, No. 2, February 2011 794 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry rtainty recommendations 794 TABLE III. Propagation of best practice uncertainties k unless stated otherwise associated with the transfer of air-kerma strength from NIST through the ADCL to the ic for LDR high-energy brachytherapy sources. Well-chamber measurement uncertainty is estimated to be %. Row Relative propagated Measurement description uncertainty % 1.0 NIST calibration

Quantity units SK,NIST U

SK,NIST / IADCL U/A 2 3 SK,ADCL U

1.1 ADCL well ion chamber calibration ADCL calibration of source from manufacturer 1.2

SK,ADCL / ICLINIC U/A 4 SK,CLINIC U SK,CLINIC U 5

1.3 ADCL calibration of clinic well ion chamber 1.4 Clinic measures source air-kerma strength 2.8 Expanded uncertainty k = 2

.B. Dose-rate constant acceptable for use by the clinic. This result is for a set of measurements made on a single source and does not include As is de ned as the ratio of dose rate at the reference uncertainties due to source-to-source variability. Thus, 3.4% D r0 , 0 / SK, the osition to the air-kerma strength, is the lower limit for the source acceptance criterion. Critencertainty is simply rion for acceptance of calibration is discussed in Ref. 79, where the lower-third of its Table II for 100% source assay is + %uS 2 . 2 %u = %uD r 0, 0 K directly comparable to Table II of the current report. In the case of high-energy sources, the procedure is simihile Sec. V A 5 discussed uSK,CLINIC, clinical users do not lar to that given above with some minor differences. For easure the reference dose rate and thus do not directly obLDR high-energy sources, there are long-lived sources, such ain %u . Instead, %u values are taken from the literature as 137Cs, and shorter-lived sources, such as 192Ir sources. f dosimetry investigators upon deriving . For instance, Table III is presented for the clinic measurement uncertainty hen the AAPM issues consensus datasets, and %u conwith an ADCL-calibrated well-ionization chamber and is cerensus values may be provided with %u values generally tainly appropriate for a short-lived source. Following the maller than the individual investigator %u value due to same model of uncertainty propagation as above assuming ncreased sampling of candidate datasets. For low- and high%uc,I = 0.5% for each well-chamber measurement , the mininergy photon-emitting brachytherapy sources, the measured mum expanded uncertainty k = 2 of clinical air-kerma k = 1 are approximately 2.9%; MCalues of %u strength measurements for LDR high-energy sources is 2.8% imulated values of %u k = 1 are approximately 2.1%. Table III . In the case of a long-lived source, the original 6

p u

W m t o w s s i e

v s

NIST-calibrated source may be used, in which case, rows 2 .C. Geometry function and 3 are not present. In this case, the uncertainty in the The geometry function is dependent on L or effective ADCL calibration of the clinic well chamber is 1.12% and ength , r, and . Since L is primarily used to minimize inthe uncertainty in the clinical measurement is 1.22%, with erpolation errors during treatment planning, it can take on the expanded uncertainty of 2.45% k = 2 . The HDR highlmost any value.62,80,81 However, realistic dose distributions energy sources have a NIST-traceable calibration through an re usually best-approximated through using realistic L valinterpolated calibration coef cient from two photon beams es. In practice, the geometry function is used by dosimetry as given in Ref. 71. Following the same model of uncertainty nvestigators to determine other parameters such as g r and propagation as above assuming 0.5% uncertainty on each r , . In both cases, the geometry function is used to rewell-chamber measurement , the minimum expanded uncertainty k = 2 of clinical SK measurements for HDR highove the effects of solid angle when evaluating measure-

l t a a u i F m

m ents or calculations of dose rate around a source. Conseenergy sources is 2.94% from Table IV. TABLE IV. Propagation of best practice uncertainties k = 1 u nless stated otherwise associated with the transfer of air-kerma strength from a traceable NIST coef cient from th e ADCL to the clinic for HDR high-energy brachytherapy sources. Row Relative propagated Measurement description uncertainty % 1.1 ADCL calibration

Quantity units SK,NIST U

SK,NIST / IADCL U/A 2 3 SK,ADCL U SK,ADCL / ICLINIC U/A 4 SK,CLINIC U 5

1.2 ADCL well ion chamber calibration ADCL calibration of source from manufacturer 1.3 1.4 ADCL calibration of clinic well ion chamber

1.5 Clinic measures source air-kerma strength Expanded uncertainty k = 2 SK,CLINIC U 2.9 Medical Physics, Vol. 38, No. 2, February 2011 795 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unc

ertainty recommendations

795

ts.2 For low- and high-energy sources, meaquently, the geometry function appears in both the numerator 1 uncertainties are approximately 2.4% and the denominator of the expressions used to determine these parameters. A proper uncertainty analysis will recogtively; MC-simulated values of %uF r, nize the arti cial decoupling of the TG-43 brachytherapy domately 1.1% and 0.6%, respectively. These simetry parameters, and that the geometry function cancels e weighted over all polar angles and are subout once dose-rate values are obtained in the TPS as long as r near the source long axis where dynamic it is used consistently in the other parameters such as g r nts may cause large dose variations. and F r , . Variability in dose measurements resulting from the associated variability in source positioning contributes to py function dosimetric uncertainties, not geometry function uncertainties. anisotropy function is the average of the Thus, the practical implementation of the geometry function the source at a given r divided by the dose means there is no associated uncertainty. That is, %uG r, sverse plane at the same r, it is a relative = 0. While sources of a given model have L variations, these ke g r and F r , . Because of the volume variations manifest themselves with physical dose rates and more complicated to express the dosimetric other parameters because a single consistent L is used for a given radius since the 4 sr averaging may given source model.81 n of the capsule. However, its expression is for the 2D anisotropy function, V.D. Radial dose function 2 %u + % u D r, 2 . 9

candidate datase sured %uF r, k = and 1.3%, respec k = 1 are approxi uncertainties ar stantially large internal compone V.F. 1D anisotro Since the 1D dose rate around rate on the tran function just li averaging, it is uncertainty at a require exclusio similar to that

%u

an r The radial dose function uncertainty is the square root of D r, d 0 the sum of the squares of the relative dose-rate uncertainties In practice, %u an r is less than %uF r, due to diminishment at the reference position and point of interest on the transof positioning u ncertainties due to volume/angular averagverse plane. In Sec. V C, it was shown that the geometry ing. As for g r

and F r , , uncertainties increase for large r function uncertainty was zero. Thus, dose rate and for small r based on dosimies close to the source. Estimates are based on 2 2 %ug r = %uD r + % u D r, . 0, 0 0 n of F r , uncertainty Sec. V E . For lowsources, measured %u an r k = 1 uncertainIn general, the uncertainty increases for large r more for mately 1.5% and 1.1%, respectively; MClow-energy sources where attenuation is greater and for

diminishment of etric uncertaint 7 the determinatio and high-energy ties are approxi

simulated values of %u an r k = 1 are approximately 0.6% small r based on dosimetric uncertainties close to the source . Estimates of this type B uncertainty are based on the and 0.4%, respec tively. experience gained through the derivation of a large number of AAPM consensus datasets from candidate datasets.2 For V.G. TPS uncerta inties summary 0.5 cm r 5 cm, low- and high-energy photon-emitting The uncertai nty in TPS-calculated dose will be based on brachytherapy source measured values of %ug r k = 1 are the combination of uncertainties of NIST-traceable SK and approximately 2% and 1%, respectively; MC-simulated valthe dose rates d etermined by the dosimetry investigator. ues of %ug r k = 1 are approximately 1% and 0.5%, respecHowever, there a re additional uncertainties introduced by the tively. These dose uncertainties increase for r 0.5 cm due TPS. Commissioning to the in uence of dynamic internal components and for of the brachytherapy source for dose calr 5 cm due to cross-section uncertainties in the phantom culations requir es the physicist or other responsible person to material. install source c haracterization data into the TPS computer. Since primary ca lculations for patient treatment are almost V.E. 2D anisotropy function never performed today using manual methods, other than for a check, the unc ertainty associated with manual calculations The 2D anisotropy function uncertainty is the square root will not be disc ussed. Therefore, the question becomes, what of the sum of the squares of the relative dose-rate and geomadditional uncer tainty is associated with the installation of etry function uncertainties. It was shown that the geometry source character ization data, and the use of those data in the function uncertainty was zero in Sec. V C. Thus,

e dose distributions?

TPS, to calculat 2 2 . 0 8 and spacing of t terpolation on t verse proportion certainty is likel ts to experimenta are entered, the The t approach an When dosimet

ry parameters are entered, the frequency % u F r, = % u D r, + % u D r,

he data are key since the TPS performs inIn general, the uncertainty increases with increasing r and he entered data. Unless spacing varies in inwhen approaches the long axis of the source due to diminto the contribution of a parameter, the unished dose rates. As approaches 90, %uF r, approaches y to be different at different distances. When zero. The numerator and denominator of F r , share the l- or MC-derived dosimetry parameters uncertainty relates to the quality of the t. same r, and uncertainties due to cross section or medium d model used will affect the uncertainty. corrections are minimized. Estimates of this type B uncerdose calculation uncertainty depends on the tainty are based on the experience gained through the deri-

Further, the TPS

implementation o f the algorithm, the calculation matrix spacvation of a large number of AAPM consensus datasets from Medical Physics, Vol. 38, No. 2, February 2011 796 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unce rtainty recommendations 796 TABLE V. Propagation of best practice uncertainties k = 1 un less stated otherwise in dose at 1 cm on the transverse plane associated with source-strength measurement s at the clinic, brachytherapy dose measurements or simulation estimates, and treatment planning system datas et interpolation for low-energy low-E and highenergy high-E brachytherapy sources as relating to values pr esented in Fig. 1. Relative propagated uncertainty % Row 1 2 3 4 5 Uncertainty component high-E SK measurements from row 5 of Tables I an 1.5 Measured dose 3.0 Monte Carlo dose estimate 1.6 TPS interpolation uncertainties 2.6 Total dose calculation uncertaint 3.4 Expanded uncertainty k = 2 6.8 e

low-E d IV 1.3 3.6 1.7 3.8 y 4.4 8.7

xpanded uncertainty should be given along with the meaing, and the veracity of the output mechanisms. Conseured value of the quantity using a coverage factor of 2 quently, it is impossible to determine explicitly the uncerk = 2 . Moreover, the current report has adopted the symbol tainty introduced by model tting and interpolation. Based on the experience gained through the derivation of a large to indicate expanded uncertainty to avoid confusion with number of AAPM consensus datasets from candidate he symbol U, which is commonly used by the medical physdatasets,2 %uTPS values k = 1, type B of 3.8% and 2.6% are cs community to indicate SK units. In addition, all comporecommended for low- and high-energy sources, respecents of uncertainty, identi ed as type A or type B, should be tively, unless speci c data indicate otherwise. These values ulated along with the calculated value of the combined are slightly higher than the 2% k = 1 value in the 2004 tandard uncertainty. The statistical methods used to obtain he various components of uc should be described in detail, TG-43U1 report which pertained to individual dosimetry pand a level of con dence interpretation of the results may be rameters. ncluded, if appropriate. Propagating the uncertainties from all Sec. V and Table V to obtain the dose at 1 brachytherapy source transverse plane, the I.B. Clinical medical physicists for low- and high-energy sources are %VD = components see cm on the k = 2 uncertainties 8.7% and %VD

V t i n tab s t a i

I.B.1. SK and TPS data entry = 6.8%, respectively. Note that these uncertainty estimates To minimize uncertainties, clinical medical physicists are generalized for the broad variety of available sources in hould use the consensus brachytherapy dosimetry data. The each source photon energy classi cation and are restricted to se of nonconsensus data would lead to a mistake see Sec. single-source dose distributions in a standardized liquid waI rather than an increase in uncertainties. The primary aster spherical phantom. ects under control by the clinical medical physicist are meaurements of SK and TPS data entry. For the rst aspect, the VI. RECOMMENDATIONS linical medical physicist should follow the 2008 AAPM rachytherapy source calibration recommendations.79 For Uncertainty analyses should include all dosimetric properties of clinical brachytherapy sources and follow a com-

s u I p s c b T

PS data entry, the physicist should carefully consider the mon set of guidelines and principles, analogous to TG-43 ecommendations of Sec. V G and avoid inadvertently inparameters for brachytherapy sources. We recommend folreasing the uncertainties by, for example, deviating from the lowing the principles described in Secs. I and II of the curumerical or spatial resolution of the AAPM-recommended onsensus dataset.2 Here, the 2% tolerances associated with rent report. This will provide more accurate and meaningful determination of dose in treatment plans and facilitate comataset interpolation may increase with a coarser dataset. Anparison between multiple investigators. The goal is to quanther example of a local uncertainty exceeding the best practify overall uncertainty in the delivered dose and maintain it ice values in the current report would be the use of a novel at the lowest possible level. ource with a calibration certi cate indicating higher SK unertainties than presented in Sec. V A. VI.A. General uncertainty I.B.2. Treatment planning system developments Uncertainty analyses should be performed using a universal methodology. The recommended methodology i.e., It is important for the clinical medical physicist to keep an GUM was described in detail in Sec. II of the current report ye toward the future regarding efforts to improve the curand is fully documented in NIST Technical Note 1297.10 ent TG-43 dose calculation formalism. These improvements AAPM/GEC-ESTRO recommends that when reporting unight include development of dose calculation algorithms to certainties of physical quantities relevant to brachytherapy ccount for intersource attenuation, phantom scatter, and ma-

r c n c d o t s c V

e r m a

t erial heterogeneities.7 Currently, there is an infrastructure in e.g., air-kerma strength, absorbed dose, and dose rate , the Medical Physics, Vol. 38, No. 2, February 2011 797 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry uncer tainty recommendations 797 place for dosimetry investigators, source manufacturers, TPS i Positional u ncertainty: When evaluating measurement manufacturers, clinical medical physicists, and professional position unc ertainty, both source and detector posisocieties to promote consistent usage of a standardized tional uncer tainty should be evaluated. In addition to dataset i.e., TG-43 dosimetry parameters for a singlesource-to-de tector distance uncertainty, angular uncersource model. As dose calculation algorithms become more tainty and i ts effect on the measured quantity should be sophisticated, these standardized datasets will no longer be addressed. T olerances for speci c source positioning directly used for derivation of patient dose.82 Consequently, jigs and pha ntom construction should be included in the clinical medical physicist must note the changes in dose the uncertai nty analysis. Moreover, due to the nature of calculation uncertainty as TPS manufacturers migrate toward the radiatio n emitted from brachytherapy sources, the more sophisticated algorithms. This topic is under investigamagnitude of the uncertainty often depends on the dis-

tion by AAPM TG-186. he source, as described in Sec. III A 2. ld be made to address this behavior. ment: Brachytherapy source dosimetry VI.B.3. Clinical dosimetric uncertainties ns usually involve the quanti cation of While lower uncertainties are clearly better, what maxie source. When performing such measuremum uncertainty should be clinically acceptable? Like the nvestigator must account for speci c detecjoint ABS/ACMP/ACRO report,83 the AAPM and GECristics for the energy being measured and ESTRO also recommend actions be taken to reduce the unn overall uncertainty. The lowest possible certainty in dose delivery for a particular patient implant that is achievable will come from choosing such as applicator repositioning, written directive adjusttrument for the experimental investigation. ment, or procedure termination. However, the AAPM and osimeters should be chosen with care. The GEC-ESTRO recognize that at this time the clinical medical ertainty should re ect the authors underphysicist is unlikely to be able to accurately determine the the various available dosimeters. For exdosimetric uncertainties in multiple sources because no spevestigation using TLDs should specify the ci c recommendations have been published. Clinical practice recommendations on the uncertainty of the dose deviation gime used as this can result in an increase have not been previously provided. Table V summarizes dotainty from 1% to 5%, depending on the and the cooling rate procedure.19 In addisimetric uncertainty contributions that lead to an overall expanded uncertainty of less than 10% k = 2 for conventional ainties arise from the differences in TLD photon-emitting brachytherapy sources. Yet there may be to differing photon energy of the calibrasources in which these dosimetric uncertainties are larger, e.g., 1.25 MeV and low-energy brachysuch as when using investigational sources that lack a robust ces e.g., 0.03 MeV . This energy depensource-strength calibration traceable to a primary standards divided into intrinsic energy dependence laboratory, or for sources whose calibration carries uncertainng detector reading to detector dose and

tance from t Efforts shou ii Dose measure investigatio dose from th ments, the i tor characte their role i uncertainty the best ins Therefore, d reported unc standing of ample, an in annealing re in the uncer temperature tion, uncert response due tion source therapy sour dence may be kBq Q relati

ties larger than those in row 1 of Table V due to design e energy dependence f Q relating dose variations and subsequent energy differences.84 These cirr to dose to medium in the absence of the cumstances and other factors may result in increased dosimWhen measuring the absorbed dose for etric uncertainties as recognized previously by Nag et al.83 hoton-emitting brachytherapy sources When these uncertainties add to those for sources of Table V ting with a 60Co beam, the kBq Q uncerand exceed 20% k = 2 , then the AAPM and GEC-ESTRO can be signi cantly less than 5%.2,18 recommend that brachytherapy implants be performed with medium: The AAPM TG-43 brachycautionpreferably under Institutional Review Board IRB metry protocol speci es a methodology to oversight with prior disclosure to the patient about the une absorbed dose to water for a brachycertain aspects of the procedure. ce. The dif culties involved with measureiquid medium often result in experiments VI.C. Dosimetry investigators d out in a solid medium that is designed to cally equivalent to liquid water. However, When performing physical measurements, investigators materials on the market today have been are encouraged to identify as many sources of uncertainty as be water equivalent at a particular energy possible. Several potential sources of uncertainty in physical ly megavoltage photon energies. These measurements performed on brachytherapy sources exist. y or may not be equivalent to water at Many of these have been presented in Sec. III. Other sources energies or for other types of radiation. of uncertainty may exist and, therefore, it is up to the indis should address the impact that measurevidual investigators to determine other potential uncertainties will have on the results as it pertains to and evaluate them appropriately. However, the speci c areas e to water. In addition, measurement of uncertainty presented in the current report should be adshould be speci ed in the investigators dressed in articles providing dosimetry parameters for

absorbed-dos to a detecto detector .18 low-energy p when calibra tainty k = 1 iii Measurement therapy dosi determine th therapy sour ments in a l being carrie be radiologi many of the designed to range, usual materials ma lower photon Investigator ment medium absorbed dos phantom size

brachytherapy sources and should include: publications . Medical Physics, Vol. 38, No. 2, February 2011 798 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry unce rtainty recommendations 798 standardized da tasets and development of TPS algorithm As with physical measurements, MC simulations also benchmarking pr ocedures toward minimizing type B dose contain uncertainties in their results. As such, MC investigacalculation unc ertainties. This can be accomplished through tors should have a thorough understanding of the MC procontinuing adop tion of the consensus dataset approach for cess and its associated uncertainties. Speci c areas to be adsingle-source d ose calculations in standardized geometries dressed are as follows: and through pro viding the information required to dosimetrii Type A uncertainties: MC methods are stochastic in cally character ize the clinical applicators and patient internature. By using probability distributions, appropriate faces which wil l be incorporated in these new TPS platforms. starting conditions, and suitable pseudorandom numbers, a problem may be simulated to produce a result consistent with a physical system. In general, converVII. SUMMARY AN D COMPARISON TO EXISTING gence of MC-based radiation transport simulations WRITTEN STANDAR DS obey Poisson statistics and, as such, have an associated Throughout the current report, the AAPM and GECstatistical uncertainty that decreases as the inverse ESTRO have re ned clinical expectations of brachytherapy square root of the number of samples in this case the dosimetric unce rtainty. Uncertainties are involved in all asnumber of particle histories . Thus, the investigator pects of the do simetry process. Every aspect of the process should provide simulations with a suf cient number of results in a gr eater uncertainty in the estimation of patient histories to provide an acceptable level of statistical dose. The AAPM TG-40 and TG-56 reports attempted to uncertainty 0.1% so these may be considered negprovide QA proc edures to reduce dosimetric uncertainty.1,70 ligible in comparison to other less constrainable uncerThe end result for consideration is the uncertainties involved tainties. in patient trea

tments. The rst aspect of these uncertainties ii Type B uncertainties: In addition to the type A unceransfer of the NIST calibration standard from tainties that arise naturally from a MC simulation, any clinics well chamber for the determination model of a physical system will include type B uncerrce strength. When the clinical medical tainties. This type of uncertainty will consist of uncerres this, a typical uncertainty k = 2 is about tainties in source dimensions, internal component locaIf each source is not measured, the corretion s , volume averaging, and material composition, ainty is increased through use of the manufor example. A thorough investigation to determine as based on batch averaging. If the physicist remany of the type B uncertainties as possible and their the manufacturer s value, then unknown effects on the dosimetric quantities should be perasurement uncertainties are passed along to formed in the course of completing a MC study of a ent , along with possible administrative errors brachytherapy source. Examples of determining the urer sending sources from the order placed type B uncertainties for a brachytherapy source have itution. Generally, the manufacturer sourcebeen given throughout Secs. III and IV. ainty is larger than if measured by the clinical st using an instrument with a calibration coe to a primary standards laboratory.79 The VI.D. Source and TPS manufacturers f dosimetric uncertainty involves treatment nsic to this process is derivation and utilization Brachytherapy source manufacturers should implement ters. If these parameters are based on tight tolerances on their manufacturing processes since the data, their uncertainties should have been clinical results are dependent on consistent source fabricaAAPM report. If data from multiple dosimtion. The largest potential dosimetric variation is from dy-

involves the tr the ADCL to the of measured sou physicist measu 3% Sec. V A 5 . sponding uncert facturer value lies solely on manufacturer me the clinic pati by the manufact by another inst strength uncert medical physici ef cient traceabl second aspect o planning. Intri of TG-43 parame AAPM consensus provided in the etry investigat

ors are entered into the TPS, the resultant donamic internal components Sec. IV B . Thus, the design ainty of the calculated dose is greater. Further, should constrain motion of these components. The source n the treatment planning process are not as design/version in regular clinical use should be the same on the patient treatment as is the initial deterdesign/version measured and simulated by the dosimetry inreference dose-rate distribution. When all vestigator and measured by the dosimetry laboratories. ties are combined, the k = 2 uncertainty of Moreover, detailed information on the source components low- and high-energy photon-emitting brachyincluding dimensions, tolerances, and material compositions used in treatment planning are approxishould be openly provided. If the manufacturer decides to %, respectively. Uncertainty in dose delivchange source design/version, the manufacturer must recogical implantation will add to these nize that this is equivalent to construction of a new source, nd surely be larger upon clinical implementawhich is subject to the processes described by DeWerd et al.,78 which include regular comparisons with dosimetry tly, it is paramount that the clinical medical gnizant of these uncertainties and endeavor to laboratories. Furthermore, manufacturers are advised to or the aspects within their responsibilities. minimize and keep constant any radiocontaminants per Sec. hytherapy source dosimetry investigators IV C. to minimize dosimetric uncertainties in their As mentioned in Sec. VI C, it is recommended that TPS

simetric uncert uncertainties i great an effect mination of the these uncertain dose rates for therapy sources mately 9% and 7 ery due to phys uncertainties a tion. Consequen physicist be co minimize them f Similarly, brac should continue

reference data. manufacturers continue to strive for clinical utilization of Medical Physics, Vol. 38, No. 2, February 2011 799 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry u ncertainty recommendations 799 1 4 R. C. Tailor, G. S. Ibbott, and N. Tolani, Thermoluminescence dosimetry The AAPM TG-56 report recommends brachytherapy measurements of brachytherapy sources in liquid water, Med. Phys. 35, dose delivery accuracy within 5%10% with source calibra-

4063 4069 2008 . tion accuracy within 3%.70 This latter tolerance was updated 5 R. Tailor, G. Ibbott, S. Lampe, W. Bivens-Warren, and N. Tolani, Dosiby Butler et al.79 to 6% for individual sources. While the metric characterization of a 131Cs brachytherapy source by thermoluminescence dosimetry in liquid water, Med. Phys. 35, 58615868 2008 . scope of the current report is limited to evaluation of pre6 S. W. Peterson and B. Thomadsen, Measurements of the dosimetric contreatment brachytherapy dosimetry uncertainties, it appears stants for a new 103Pd brachytherapy source, Brachytherapy 1, 110119 that the TG-56 10% criterion for accuracy of brachytherapy 2002 . 7 dose delivery cannot be adhered to within a 95% con dence . Kron, L. DeWerd, P. Mobit, J. Muniz, A. Pradhan, M. Toivonen, and M. Waligorski, A checklist for reporting of thermoluminescence dosimlevel. To our knowledge, there are no other existing societal etry TLD measurements, Phys. Med. Biol. 44, L15L19 1999 . standards on uncertainty for brachytherapy source calibration 8 L. A. DeWerd, L. J. Bartol, and S. D. Davis, Thermoluminescence doand dose delivery, and additional research in this area is simetry, in Clinical Dosimetry for Radiotherapy: AAPM Summer School, needed. A joint effort of GEC-ESTRO and AAPM brachyedited by D. W. O. Rogers and J. E. Cygler Medical Physics Madison, WI, 2009 , pp. 815 840. therapy physicists/physicians will explore more details of the 9 J. A. Raf , S. D. Davis, C. G. Hammer, J. A. Micka, K. A. Kunugi, J. E. clinical aspects of the total uncertainty budget for brachyMusgrove, J. W. Winston, Jr., T. J. Ricci-Ott, and L. A. DeWerd, Detertherapy treatment delivery. mination of exit skin dose for 192Ir intracavitary accelerated partial breast irradiation with thermoluminescent dosimeters, Med. Phys. 37, 2693 2702 2010 . 0 P. T. Muench, A. S. Meigooni, R. Nath, and W. L. McLaughlin, Photon ACKNOWLEDGMENTS energy dependence of the sensitivity of radiochromic lm compared with The authors extend their appreciation to the AAPM, GECsilver halide and LIF TLDs used for brachytherapy dosimetry, Med.

1 T

Phys. 18, 769775 1991 . ESTRO, and Medical Physics reviewers who helped to im1 M. C. Saylor, T. T. Tamargo, W. L. McLaughlin, H. M. Khan, D. F. prove this report while considering the practical aspects for Lewis, and R. D. Schenfele, A thin lm recording medium for use in clinical implementation. food irradiation, Radiat. Phys. Chem. 31, 529536 1988 . 2 S.-T. Chiu-Tsao, A. de la Zerda, J. Lin, and J. H. Kim, High-sensitivity a GafChromic lm dosimetry for 125I seed, Med. Phys. 21, 651 657 Electronic mail: mrivard@tuftsmedicalcenter.org 1 G. J. Kutcher, L. Coia, M. Gillin, W. F. Hanson, S. Leibel, R. J. Morton, 1994 . 3 J. R. Palta, J. A. Purdy, L. E. Reinstein, G. K. Svensson, M. Weller, and S. A. Dini, R. A. Koona, J. R. Ashburn, and A. S. Meigooni, Dosimetric

evaluation of GAFCHROMIC XR type T and XR type R lms, J. Appl. L. Wing eld, Comprehensive QA for radiation oncology: Report of AAPM Radiation Therapy Committee Task Group No. 40, Med. Phys. Clin. Med. Phys. 6, 114 134 2005 . 4 21, 581 618 1994 . . Bouchard, F. Lacroix, G. Beaudoin, J.-F. Carrier, and I. Kawrakow, 2 M. J. Rivard, B. M. Coursey, L. A. DeWerd, W. F. Hanson, M. S. Huq, G. On the characterization and uncertainty analysis of radiochromic lm S. Ibbott, M. G. Mitch, R. Nath, and J. F. Williamson, Update of AAPM osimetry, Med. Phys. 36, 19311946 2009 . Task Group No. 43 Report: A revised AAPM protocol for brachytherapy A. Niroomand-Rad, C. R. Blackwell, B. M. Coursey, K. P. Gall, J. M. dose calculations, Med. Phys. 31, 633 674 2004 . vin, W. L. McLaughlin, A. S. Meigooni, R. Nath, J. E. Rodgers, and 3 R. Nath, L. L. Anderson, G. Luxton, K. A. Weaver, J. F. Williamson, and C. G. Soares, Radiochromic lm dosimetry. Recommendations of A. S. Meigooni, Dosimetry of interstitial brachytherapy sources: RecomAPM Radiation Therapy Committee Task Group 55, Med. Phys. 25, mendations of the AAPM Radiation Therapy Committee Task Group No. 20932115 1998 . 6 43, Med. Phys. 22, 209234 1995 . G. Soares, S. Trichter, and S. D. Davis, Radiochromic lm, in Clini4 M. J. Rivard, W. M. Butler, L. A. DeWerd, M. S. Huq, G. S. Ibbott, A. S. cal Dosimetry for Radiotherapy: AAPM Summer School, edited by D. W. Meigooni, C. S. Melhus, M. G. Mitch, R. Nath, and J. F. Williamson, 5

2 H

d 2

Gal

2 C.

O. Rogers and J. E. Cygler Medical Physics, Madison, WI, 2009 , pp. Supplement to the 2004 update of the AAPM Task Group No. 43 Re59 813. port, Med. Phys. 34, 21872205 2007 . R. Giles and P. H. Murphy, Measuring skin dose with radiochromic 5 J. D. Honsa and D. A. McIntyre, ISO 17025: Practical bene ts of impleimetry lm in the cardiac catheterization laboratory, Health Phys. 82, menting a quality system, J. AOAC Int. 86, 1038 1044 2003 . 880 2002 . 6 8 J. A. C. Sterne and G. D. Smith, Sifting the evidenceWhats wrong ndsay, A. Rink, M. Ruschin, and D. Jaffray, Investigation of energy with signi cance tests?, BMJ 322, 226 231 2001 . dence of EBT and EBT-2 Gafchromic lm, Med. Phys. 37, 571 7 M. J. Rivard, J. L. M. Venselaar, and L. Beaulieu, The evolution of 76 2010 . 9 brachytherapy treatment planning, Med. Phys. 36, 2136 2153 2009 . Arjomandy, R. Tailor, N. Sahoo, M. Gillin, K. Prado, and M. Vicic, 8 P. Giacomo, News from the BIPM, Metrologia 17, 6974 1981 . dependence and dose response of Gafchromic EBT2 lm over a 9 Evaluation of measurement dataGuide to the expression of uncertainty ide range of photon, electron, and proton beam energies, Med. Phys. in measurement, International Organization for Standardization ISO , 37, 19421947 2010 . 0 Joint Committee for Guides in Metrology JCGM 100, 2008 , corrected S. Devic, S. Aldelaijan, H. Mohammed, N. Tomic, L.-H. Liang, F. Deversion 2010, http://www.bipm.org/utils/common/documents/jcgm/ Blois, and J. Seuntjens, Absorption spectra time evolution of EBT-2 JCGM_100_2008_E.pdf last accessed December 5, 2010 . model GAFCHROMICTM lm, Med. Phys. 37, 22072214 2010 . 10 1 B. N. Taylor and C. E. Kuyatt, Guidelines for evaluating and expressing . Hupe and J. Brunzendorf, A novel method of radiochromic lm dothe uncertainty of NIST measurement results, NIST Technical Note 1297 simetry using a color scanner, Med. Phys. 33, 408594 2006 . 2 7

7 2 E. dos 875 2 P. Li depen 5 2 B. Energy w

3 O

3 C

U.S. Government Printing Of ce, Washington, DC, 1994 , http:// . Richter, J. Paweike, L. Karsch, and J. Woithe, Energy dependence of physics.nist.gov/Pubs/guidelines/contents.html last accessed December EBT-1 radiochromic lm response for photon 10 kVp15 MVp and 5, 2010 . electron beams 6-18 MeV readout by a atbed scanner, Med. Phys. 36, 11 M. J. Rivard, C. S. Melhus, and B. L. Kirk, Brachytherapy dosimetry 506 5514 2009 . 3

5 3

parameters calculated for a new 103Pd source, Med. Phys. 31, 2466 2470 Alva, H. Mercado-Uribe, M. Rodrguez-Villafuerte, and M. E. Bran2004 . dan, The use of a re ective scanner to study radiochromic lm re12 M. J. Rivard, B. L. Kirk, and L. C. Leal, Impact of radionuclide physical ponse, Phys. Med. Biol. 47, 29252933 2002 . 4

H.

s 3 J b Sch

distribution on brachytherapy dosimetry parameters, Nucl. Sci. Eng. . F. Williamson and M. J. Rivard, Quantitative dosimetry methods for 149, 101106 2005 . rachytherapy, in Brachytherapy Physics: Joint AAPM/ABS Summer 13 M. G. Mitch and S. M. Seltzer, Model-speci c uncertainties in air-kerma ool, 2nd ed., edited by B. R. Thomadsen, M. J. Rivard, and W. M. strength measurements of low-energy photon-emitting brachytherapy Butler Medical Physics, Madison, WI, 2005 , Monograph 31, pp. 233 sources, Med. Phys. 34, 2337 2007 . 94. Medical Physics, Vol. 38, No. 2, February 2011 800 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry un certainty recommendations 800 35 57 E. Yorke et al., Diode in vivo dosimetry for patients receiving external R. E. P. Taylor and D. W. O. Rogers, EGSnrc Monte Carlo calculated dosimetry parameters for 192Ir and 169Yb brachytherapy sources, Med. beam radiation therapy: Report of Task Group 62 of the Radiation Therapy Committee, AAPM Report No. 87 Medical Physics Publishing, Phys. 35, 4933 4944 2008 . 58 36 Madison, WI, 2005 . J. J. DeMarco, R. E. Wallace, and K. Boedeker, An analysis of MCNP

V. O. Zilio, O. P. Joneja, Y. Popowski, A. Rosenfeld, and R. Chawla, cross-sections and tally methods for low-energy photon emitters, Phys. Absolute depth-dose-rate measurements for an 192Ir HDR brachytherapy Med. Biol. 47, 13211332 2002 . 59 D. E. Cullen, J. H. Hubbell, and L. Kissel, EPDL97: The Evaluated source in water using MOSFET detectors, Med. Phys. 33, 15321539 Photon Data Library, 97 Version, Lawrence Livermore National Labo2006 . r

R. Ramani, S. Russell, and P. OBrien, Clinical dosimetry using MOSatory Report No. UCRL-50400, Vol. 6, Rev. 5 19 September 1997 . 60 S. M. Seltzer, Calculation of photon mass energy-transfer and mass FETS, Int. J. Radiat. Oncol., Biol., Phys. 37, 959964 1997 .

37

38

J. E. Cygler, A. Saoudi, G. Perry, C. Morash, and E. Choan, Feasibility energy-absorption coef cients, Radiat. Res. 136, 147170 1993 .

61 J. F. Williamson, Monte Carlo evaluation of kerma at a point for photon study of using MOSFET detectors for in vivo dosimetry during permatransport problems, Med. Phys. 14, 567576 1987 . nent low-dose-rate prostate implants, Radiother. Oncol. 80, 296 301 62 M. J. Rivard, C. S. Melhus, D. Granero, J. Perez-Calatayud, and F. Bal2006 .

39

E. J. Bloemen-van Gurp, L. H. P. Murrer, B. K. C. Haanstra, F. C. J. M. lester, An approach to using conventional brachytherapy software for van Gils, A. L. A. J. Dekker, B. J. Mijnheer, and P. Lambin, In vivo clinical treatment planning of complex, Monte Carlo-based brachytherapy dosimetry using a linear MOSFET-array dosimeter to determine the uredose distributions, Med. Phys. 36, 1968 1975 2009 . 63 thra dose in 125I permanent prostate implants, Int. J. Radiat. Oncol., S. M. Seltzer, P. J. Lamperti, R. Loevinger, M. G. Mitch, J. T. Weaver, Biol., Phys. 73, 314 321 2009 . and B. M. Coursey, New national air-kerma-strength standards of I-125 40 P. Karaiskos, P. Papagiannis, L. Sakelliou, G. Anagnostopoulos, and D. and Pd-103 brachytherapy seeds, J. Res. Natl. Inst. Stand. Technol. 108, Baltas, Monte Carlo dosimetry of the selectSeed 125I interstitial brachy337358 2003 . therapy seed, Med. Phys. 28, 17531760 2001 . . G. Mitch and C. G. Soares, Primary standards for brachytherapy 41 R. A. Koona, Design and simulation of a brachytherapy source for the sources, in Clinical Dosimetry for Radiotherapy: AAPM Summer School, treatment of prostate cancer using Monte Carlo, M.S. thesis, University edited by D. W. O. Rogers and J. E. Cygler Medical Physics, Madison, of Kentucky, 2005. WI, 2009 , pp. 549565. 42 65 M. J. Rivard, Monte Carlo calculations of AAPM Task Group Report C. G. Soares, G. Douysset, and M. G. Mitch, Primary standards and No. 43 dosimetry parameters for the MED3631-A/M 125I source, Med. dosimetry protocols for brachytherapy sources, Metrologia 46, S80S98 Phys. 28, 629 637 2001 . 2009 . 66 43 64 M

M. J. Rivard, Brachytherapy dosimetry parameters calculated for a 131Cs H.-J. Selbach, H.-M. Kramer, and W. S. Culberson, Realization of reference air-kerma rate for low-energy photon sources, Metrologia 45, source, Med. Phys. 34, 754 762 2007 .

44

422 428 2008 . J. Beatty, P. J. Biggs, K. Gall, P. Okunieff, F. S. Pardo, K. J. Harte, M. J. Dalterio, and A. P. Sliski, A new miniature x-ray source for interstitial T. P. Loftus, Standardization of iridium-192 gamma-ray sources in terms radiosurgery: Dosimetry, Med. Phys. 23, 53 62 1996 . f exposure, J. Res. Natl. Bur. Stand. 85, 1925 1980 . 68 45 Z. Li, R. K. Das, L. A. DeWerd, G. S. Ibbott, A. S. Meigooni, J. PerezM. J. Rivard, S. D. Davis, L. A. DeWerd, T. W. Rusch, and S. Axelrod, Calatayud, M. J. Rivard, R. S. Sloboda, and J. F. Williamson, Dosimetric Calculated and measured brachytherapy dosimetry parameters in water prerequisites for routine clinical use of photon emitting brachytherapy for the Xoft Axxent x-ray source: An electronic brachytherapy source, sources with average energy higher than 50 keV, Med. Phys. 34, 37 40 Med. Phys. 33, 4020 4032 2006 . 2007 . A. S. Meigooni, H. Zhang, J. R. Clark, V. Rachabatthula, and R. A. Koona, Dosimetric characteristics of the new RadioCoil 103Pd wire . P. Loftus, Standardization of cesium-137 gamma-ray sources in terms of exposure units Roentgens , J. Res. Natl. Bur. Stand., Sect. A 74A, line source for use in permanent brachytherapy implants, Med. Phys. 31, 1 6 1970 . 30953105 2004 . 70 47 R. Nath, L. L. Anderson, J. A. Meli, A. J. Olch, J. A. Stitt, and J. F. M. J. Rivard, D. Granero, J. Perez-Calatayud, and F. Ballester, In uence Williamson, Code of practice for brachytherapy physics: Report of the of photon energy spectra from brachytherapy sources on Monte Carlo AAPM Radiation Therapy Committee Task Group No. 56, Med. Phys. simulations of kerma and dose rates in water and air, Med. Phys. 37, 24, 15571598 1997 . 869 876 2010 . 71 48 S. J. Goetsch, F. H. Attix, D. W. Pearson, and B. R. Thomadsen, CaliD. Baltas, P. Karaiskos, P. Papagiannis, L. Sakelliou, E. Lf er, and N. bration of 192Ir high-dose-rate afterloading systems, Med. Phys. 18, 462 69 T 67

46

Zamboglou, Beta versus gamma dosimetry close to Ir-192 brachytherapy sources, Med. Phys. 28, 18751882 2001 . 67 1991 . 72 49 E. Mainegra-Hing and D. W. O. Rogers, On the accuracy of techniques R. Wang and X. A. Li, Dose characterization in the near-source region for obtaining the calibration coef cient NK of 192Ir HDR brachytherapy for two high dose rate brachytherapy sources, Med. Phys. 29, 1678 1686 2002 . sources, Med. Phys. 33, 33403347 2006 . 50 73 F. Ballester, D. Granero, J. Perez-Calatayud, C. S. Melhus, and M. J. E. van Dijk, I.-K. K. Kolkman-Deurloo, and P. M. G. Damen, Determination of the reference air kerma rate for 192Ir brachytherapy sources and Rivard, Evaluation of high-energy brachytherapy source electronic disequilibrium and dose from emitted electrons, Med. Phys. 36, 4250 4256 he related uncertainty, Med. Phys. 31, 2826 2833 2004 . 74 51 2009 . K. E. Stump, L. A. DeWerd, J. A. Micka, and D. R. Anderson, Calibration of new high dose rate 192Ir sources, Med. Phys. 29, 14831488 NUDAT 2.5, National Nuclear Data Center, Brookhaven National Laboratory, Upton, NY, USA, http://www.nndc.bnl.gov/nudat2/index.jsp last 2002 . 75 52 accessed December 5, 2010 . A. M. Bidmead, T. Sander, S. M. Locks, C. D. Lee, E. G. A. Aird, R. F.

J. Prez-Calatayud, D. Granero, and F. Ballester, Phantom size in utbrown, and A. Flynn, The IPEM code of practice for determination of the reference air kerma rate for HDR 192Ir brachytherapy sources based brachytherapy source dosimetric studies, Med. Phys. 31, 20752081 2004 . on the NPL air kerma standard, Phys. Med. Biol. 55, 31453159 2010 .

C. S. Melhus and M. J. Rivard, Approaches to calculating AAPM TG-43 G. Douysset, J. Gouriou, F. Delaunay, L. DeWerd, K. Stump, and J. brachytherapy dosimetry parameters for 137Cs, 125I, 192Ir, 103Pd, and Micka, Comparison of dosimetric standards of USA and France for HDR 169 Yb sources, Med. Phys. 33, 17291737 2006 . rachytherapy, Phys. Med. Biol. 50, 19611978 2005 . 54 77 R. E. P. Taylor and D. W. O. Rogers, An EGSnrc Monte Carlo-calculated G. Douysset, T. Sander, J. Gouriou, and R. Nutbrown, Comparison of air kerma standards of LNE-LNHB and NPL for 192Ir HDR brachytherapy

53 76

55

database of TG-43 parameters, Med. Phys. 35, 4228 4241 2008 .

A. S. Meigooni, S. B. Awan, N. S. Thompson, and S. A. Dini, Updated sources: EUROMET Project No 814, Phys. Med. Biol. 53, N85N97 Solid Water to water conversion factors for 125I and 103Pd brachy2008 . therapy sources, Med. Phys. 33, 3988 3992 2006 . . A. DeWerd, M. S. Huq, I. J. Das, G. S. Ibbott, W. F. Hanson, T. W. 56 N. S. Patel, S.-T. Chiu-Tsao, J. F. Williamson, P. Fan, T. Duckworth, D. Slowey, J. F. Williamson, and B. M. Coursey, Procedures for establishShasha, and L. B. Harrison, Thermoluminescent dosimetry of the Syming and maintaining consistent air-kerma strength standards for lowmetra 125I model I25.S06 interstitial brachytherapy seed, Med. Phys. nergy, photon-emitting brachytherapy sources: Recommendations of the 28, 17611769 2001 . Calibration Laboratory Accreditation Subcommittee of the American AsMedical Physics, Vol. 38, No. 2, February 2011 801 DeWerd et al.: AAPM TG-138 and GEC-ESTRO brachytherapy dosimetry u ncertainty recommendations 801 2002 . sociation of Physicists in Medicine, Med. Phys. 31, 675 681 2004 . 79 78 L

W. M. Butler, W. S. Bice, Jr., L. A. DeWerd, J. M. Hevezi, M. S. Huq, G. M. J. Rivard, L. Beaulieu, and F. Mourtada, Necessary enhancements to S. Ibbott, J. R. Palta, M. J. Rivard, J. P. Seuntjens, and B. R. Thomadsen, commissioning techniques of brachytherapy treatment planning systems Third-party brachytherapy source calibrations and physicist responsibilihat use model-based dose calculation algorithms, Med. Phys. 37, 2645 ties: Report of the AAPM Low Energy Brachytherapy Source Calibration 2658 2010 . 3 Working Group, Med. Phys. 35, 38603865 2008 . Nag, R. Dobelbower, G. Glasgow, G. Gustafson, N. Syed, B. Thomad80 sen, and J. F. Williamson, Inter-society standards for the performance of J. A. Meli, Lets abandon geometry factors other than that of a point brachytherapy: A joint report from ABS, ACMP and ACRO, Crit. Rev. source in brachytherapy dosimetry, Med. Phys. 29, 19171918 2002 . 81 Oncol. Hematol. 48, 117 2003 . M. J. Rivard, B. M. Coursey, L. A. DeWerd, W. F. Hanson, M. S. Huq, G. 4 Ibbott, R. Nath, and J. F. Williamson, Comment on Lets abandon gellester, D. Granero, J. Perez-Calatayud, J. L. M. Venselaar, and M. J. Rivard, Study of encapsulated 170Tm sources for their potential use in ometry factors other than that of a point source in brachytherapy dosimetry Med Phys. 29, 19171918 2002 , Med. Phys. 29, 19191920 herapy, Med. Phys. 37, 16291637 2010 . Medical Physics, Vol. 38, No. 2, February 2011

8 S.

8 F. Ba

brachyt