Comparison of peripheral dose from image-guided radiation therapy (IGRT) using kV cone beam CT to intensity-modulated radiation therapy (IMRT)

Radiotherapy and Oncology 89 (2009) 304–310 www.thegreenjournal.com

Cone beam CT

Comparison of peripheral dose from image-guided radiation therapy (IGRT) using kV cone beam CT to intensity-modulated radiation therapy (IMRT) Julian R. Perks*, Jo ¨erg Lehmann, Allen M. Chen, Claus C. Yang, Robin L. Stern, James A. Purdy Department of Radiation Oncology, UC Davis Medical Center, CA, USA

Abstract Purpose: The growing use of IMRT with volumetric kilovoltage cone-beam computed tomography (kV-CBCT) for IGRT has increased concerns over the additional (typically unaccounted) radiation dose associated with the procedures. Published data quantify the in-field dose of IGRT and the peripheral dose from IMRT. This study adds to the data on dose outside the target area by measuring peripheral CBCT dose and comparing it with out-of-field IMRT dose. Materials and methods: Measurements of the CBCT peripheral dose were made in an anthropomorphic phantom with TLDs and were compared to peripheral dose measurements for prostate IMRT, determined with MOSFET detectors. Results: Doses above 1cGy (per scan) were found outside the CBCT imaged volume, with 0.2cGy at 25 cm from the central axis. IMRT peripheral dose was 1cGy at 20 cm and 0.4cGy at 25 cm (per fraction). Conclusions: An appreciable dose can be found beyond the edge of the IGRT field; of similar order of magnitude as peripheral dose from IMRT (mGy), and approximately half the dose delivered to the same point from the IMRT treatment (0.2cGy c.f. 0.4cGy 25 cm from the isocenter). This shows that peripheral dose, as well as the in-field dose from CBCT, needs to be taken into account when considering long term care of radiation oncology patients. c 2008 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 89 (2009) 304–310.



Keywords: Peripheral dose; IGRT; Cone beam CT

Image-guided radiation therapy (IGRT) is a rapidly growing field, especially with the introduction of on-board kilovoltage cone beam computed tomography (CBCT) systems to linear accelerators [1–3]. In the CBCT approach, the patient is, as well as possible, aligned to the room isocenter using skin tattoos made during CT simulation, then a CBCT image data set is obtained and a registration of this daily image to the CT used for planning is performed. Based on the movements of the images during registration, the patient is moved in a combination of table shifts and/or rotations to correct his/her position, and treatment is then delivered. Because IGRT is being implemented for therapies where daily imaging is common, e.g., for prostate cancer [4,5], it holds that the CBCT will also be taken daily [6]. This naturally leads to the question of imaging dose and then to the issue of peripheral dose. The dose delivered to the patient to create the CBCT is dependent on a number of factors relating to the quality desired in the image. In addition to factors such as patient size and anatomical site to be imaged (these are outside the operator’s control), the factors that can be altered for a CBCT include field size, kV, the



number of images used to reconstruct the image (frames), the mA per frame, and the time for each frame acquisition (ms). The dose within the imaged area for a CT used for radiation therapy patient alignment is well reported. Our group has reported exposure measurements for a CBCT system [7], and other users of CBCT systems have published their measurements of patient dose, with similar dose amounts being reported [8–13], in-field doses for a single kV CBCT on the order 1.5–3.0cGy to the isocenter are typical, but can be as high as 6cGy depending on the settings. However, details of the dose fall-off outside the irradiated (imaged) volume, in particular the periphery for a CBCT scan and how it compares with the peripheral dose of IMRT have not been reported to date. Understanding peripheral dose is important in radiation oncology, as any radiation distant to the disease being treated is undesirable and should therefore be minimized. There are a number of terms/definitions associated with out-offield and peripheral dose, but for this study, peripheral dose of IGRT is defined as a dose anywhere outside the imaged

0167-8140/$ - see front matter c 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2008.07.026

J.R. Perks et al. / Radiotherapy and Oncology 89 (2009) 304–310

volume, with the imaged volume being that anatomy of the patient that receives primary kV X-rays and is then reconstructed in the CBCT image. An important concern regarding peripheral dose is a secondary malignancy. Followill et al. [14,15] early on in the IMRT era, expressed concern that the move from 2DRT/ 3DCRT to IMRT may result in an increased rate of secondary malignancies because of the significantly larger number of monitor units (MUs) required to deliver a comparable prescribed dose. Followill et al. reported the estimated percent probability of a fatal secondary cancer due to a 70.00 Gy course of radiation therapy as follows: for 6 MV X-rays: 0.3% for 2DRT un-wedged technique and 1.0% for IMRT. Furthermore, the probability of developing such a secondary malignancy has been calculated, both in general terms [16,17] (where transformation of cells lines in vitro increased linearly with dose from approximately 1 to approximately 4–5 Gy) and with the most modern approaches. Eric Hall [18] states ‘‘...because IMRT involves more fields, a bigger volume of normal tissue is exposed to lower radiation doses. Intensity-modulated radiation therapy may double the incidence of solid cancers in longterm survivors.’’ The topic of secondary malignancy is still ‘‘hot’’, as most recently computed tomography has been studied in these terms [19]. A number of studies exist on peripheral dose from linear accelerators [20–24] and the American Association of Physicists in Medicine published task group reports [25,26], where the dose from megavoltage beams is classified by the distance out of field, and at least one of which has been qualified by a reduction in the estimates of dose [27]. Additionally, an interesting study of organ doses associated with a modern fractionated radiation therapy for prostate cancer that includes imaging dose was produced by Harrison et al. [28]. For the linear accelerator MV beam, typical values of peripheral dose are quoted as a percentage of the central axis dose, so 0.5% of the central axis dose may be found 25 cm from the field edge (depending on field size, energy, etc.). Additionally, the dose distribution, and particularly the out-of-field dose, for the (again megavoltage) helical Tomotherapy (Tomotherapy Inc., WI, USA) has been determined [29]. Here, the peripheral dose dropped to 0.4% of the prescribed (central axis dose) dose at 20 cm. The concept of peripheral dose is strongly associated with intensity-modulated radiation therapy (IMRT) [23], due to the increased monitor units used. This leads to increased scatter and leakage. Thus, this study specifically compares peripheral dose for an IGRT image with that of typical IMRT. The aim here was to assess the amount of peripheral dose of a CBCT with that of a well-established norm (IMRT) in order to more fully assess the long-term impact of the new technique on patients where IGRT is utilized.

Materials and methods The IGRT and IMRT techniques used for the two measurements of peripheral dose are typical for a patient undergo-

305

ing IMRT: a seven-field step and shoot pattern is used with one CBCT to verify patient positioning before treatment, and no confirmatory CBCT scans being taken after patient alignment or treatment.

CBCT We determined the dose delivered to a patient receiving a CBCT by irradiating an anthropomorphic phantom loaded with thermoluminescent detectors (TLD 100; Harshaw/ Thermo Fisher Scientific Inc., MA, USA). The phantom (RPHAT) [30] allowed us to simulate both thin and thick patients because it is sliced in the coronal plane. In this study, the thin patient measured 23 cm from ventral to dorsal surfaces, and the thick patient measured 28.1 cm. TLD chips were loaded in the center of the phantom and on the surface along a line in the cranial–caudal direction. The phantom itself is 42 cm long and the CBCT scan has a length of 26 cm, so dose could readily be measured in both the central area of the scan (the pelvis in this case) and up into the abdomen. All measurements were performed on an Elekta Synergy S linear accelerator with the XVI CBCT system (Elekta, Crawley, UK). TLD chips were placed at 0 cm (central axis, two TLD chips) and then at 4, 8, 12, 15, 20, and 25 cm along the cranial–caudal axis, both on the surface and on the central (internal) plane, of the phantom. This arrangement made best use of the space available in the phantom to give a good sampling of both the central and the peripheral dose with a total of 48 readings (a number of points having multiple TLDs for increased accuracy). TLD chips were not placed laterally across the phantom as this, while interesting, would not have addressed the issue of peripheral dose. A thin layer (5 mm) of standard bolus material (Superflab – Mick Radio-Nuclear Instruments Inc., NY, USA) was used to cushion the internal TLDs. Fig. 1 shows the RPHAT phantom on the linear accelerator couch, with an inset showing how the TLDs are positioned in the opened phantom. Fig. 2 shows a lateral, cutaway view of the RPHAT phantom and the volume imaged by a 26 cm-long CBCT scan; note how the surface of the phantom (patient) is imaged when a thin patient is simulated and how the same size CBCT only images the internal anatomy of a larger patient. The CBCT used to derive the peripheral doses was a clinically validated protocol, ‘‘Prostate S20’’. We chose ‘‘Prostate S20’’ because it results in the highest dose setting available in the clinic (6cGy at isocenter) [7]. Because previous work by our group has shown that CBCT dose scales linearly with mA [7], the highest dose CBCT protocol was used to study peripheral dose, with the results being scalable for the lower dose (strictly lower mAs and/or fewer acquired frames) protocols that are in clinical use. The ‘‘Prostate S20’’ CBCT protocol uses the S20 setting of the Elekta Synergy, which means the small field of view with an effective cranial–caudal imaging field size of 26 cm. The protocol takes a 360° scan, has 120 kV accelerating potential, 40 mA and 40 ms per frame, with approximately 650 frames collected per scan. It should be noted that lower dose CBCT protocols are generally used clinically, if patient size allows, particularly

306

IGRT peripheral dose

Fig. 1. RPHAT phantom in place on the treatment couch and opened to show the positioning of the TLD chips (inset).

one referred to as ‘‘Prostate M10’’, which uses a larger lateral field of view and a shorter cranial–caudal imaging field size. The dose to the patient from this protocol is less than 3cGy to the isocenter. In order to deposit sufficient dose to the TLD chips to obtain reasonable readings, the scan was delivered ten times to the phantom and the readings were divided by ten. To fully characterize the CBCT, the kVp and the ms per frame were also measured and compared with the manufacturer’s stated values. To then convert the output of each TLD to dose, a number of factors were incorporated. First, the TLD set was calibrated against a 6 MV linear accelerator beam, with each TLD having an individual calibration factor. Then, to account for the change in response from 6 MV to 120 kV nominal energy, a factor of 1.3 was applied, reducing the reading of each TLD [31,32]. The TLD set is maintained to an accuracy of 1%; any TLD exhibiting a difference between calibration dose and readout greater than 1% is replaced. With a derived dose reading, one should consider the relative biological effectiveness (RBE) of the kV used. There is evidence that the RBE value of the X-rays present in a CBCT is not unity [33]. An RBE value of 1.13 for 100 kV photons has been published, as have higher RBEs for lower kV [34–36], but as the spectrum of the X-ray source is not known for the CBCT system these factors have not been applied here.

Standard IMRT To put the peripheral doses measured for a CBCT scan into context with another modern clinical radiation oncology technique, the out-of-field dose from a standard prostate IMRT was also measured. The IMRT technique used to irradiate the phantom was copied from a prostate cancer patient currently under treatment and consisted of 7 coplanar fields, using beam energy of 6 MV and a prescription of

2 Gy (set to the 95% isodose) with a maximum in-field dose of 217cGy. The IMRT field has a maximum length in the cranial–caudal direction of 8 cm, typical of prostate IMRT. The RPHAT phantom was used again, but MOSFET detectors (Thomson Nielsen, Ottawa, Canada) were used in place of TLD. MOSFET detectors were chosen as they have proved successful for peripheral dose measurements in the past [37] and allow a more rapid readout than TLD; MOSFET detectors was not used for the CBCT peripheral dose measurement as more points were measured than detectors available in the clinic. The MOSFET detectors used are in clinical operation for patient dose measurements and have been calibrated in a 6 MV beam with an accuracy and reproducibility of 3%. Similar to the CBCT peripheral dose measurements, the MOSFET detectors were placed internally in the phantom, but they were not placed on the surface of the phantom (as build-up dose is not considered here). Confirmatory measurements of the dose in the treated volume were also made. Because MOSFET detectors have a finite lifetime, dependent on the dose delivered to them, separate measurements inside and outside the field were made to avoid excessive damage to the detectors. The first measurement used the standard patient prescribed dose (2 Gy); here, two detectors were placed at isocenter and three were placed at 4, 8, and 12 cm from isocenter, respectively. This first IMRT measurement was used to confirm the accuracy of the MOSFET detectors in the treatment field and to provide an overlap between the readings on the edge of the field (4 cm from the central axis) with the measurements taken to determine the peripheral dose. The second measurement was performed with a greatly increased dose (20 Gy at isocenter) to increase the MOSFET reading at distances of 8, 12, 15, 20, and 25 cm from the

J.R. Perks et al. / Radiotherapy and Oncology 89 (2009) 304–310

307

Fig. 2. Schematic of the TLD chip positions in both the thick and thin versions of the RPHAT phantom, with the shaded area representing the CBCT field.

isocenter [38–41]. This increase in dose was achieved by simply increasing the MU used in each field of a regular prostate patient tenfold, with the record-and-verify system (Mosaiq, IMPAC Medical Systems, Inc., CA, USA) accounting for the relative increase in MU per segment whilst running in quality assurance mode. The MOSFET readings were then simply divided by ten to obtain the 2 Gy equivalent values.

Results The kVp of the CBCT was measured as 124.0 kV for a nominal (manufacturer’s specified) accelerating potential of 120 kV. The manufacturer’s specification of frame time (40 ms) was measured as 39.2 ms. For the CBCT scan, the dose across the image and the fall-off outside the imaged region is shown in Fig. 3. Here, the field edge is marked at 13 cm from the center of the imaged field of view as the nominal scan length is 26 cm. The largest in-field dose is 7.2cGy, and this is seen on the

surface of the thin patient. Outside the imaged volume, the dose falls but is still measurable, with averages (across the thin and thick patients) of 0.5cGy at 20 cm and 0.2cGy at 25 cm from the central axis (7 and 12 cm from the field edge, respectively). By comparison, the peripheral doses recorded for a routine IMRT prostate treatment are shown in Fig. 4. Here, the dose is displayed as a percentage of the central axis dose, because the main comparative out-of-field (leakage) requirement for the linear accelerator is 0.1% of the central axis dose at 50 cm from the central axis. A reasonable overlap of the results is seen for the IMRT measurements between the central readings at 2 Gy prescription and the peripheral readings at 20 Gy setting, as the readings at 8 cm from the central axis are within 1% for both the thin and thick phantoms. The readings at 12 cm from the central axis are also within 1% of each other for the thin phantom; no dose was detected at the 12 cm point in the thick phantom for the 2 Gy prescriptions due to a lack of resolution of the MOSFET detector.

308

IGRT peripheral dose

Fig. 3. Dose measurements for a 120 kV CBCT scan, showing dose both within the imaged area and in the peripheral region. The dotted line represents the edge of the CBCT image.

Fig. 4. Peripheral dose in the RPHAT phantom measured for a typical prostate IMRT. The dotted line represents the edge of the IMRT field.

The results generated for the IMRT peripheral dose are in accordance with the published data [21] and yield 1cGy at 20 cm and 0.4cGy at 25 cm from the central axis (16 and 21 cm from the field edge, respectively) for 2 Gy prescription. Hence, the peripheral doses from a 120-kV CBCT are about half the level found from a prostate IMRT for this particular arrangement, with no RBE corrections for the kV dose.

Discussion Image-guided radiation therapy utilizing CBCT is a relatively new technique that is being rapidly implemented into clinical use for radiotherapy patient alignment. A number of studies are emerging to quantify both the technique itself [1,3,7] and the dose that is associated with a particular scan [8,9,12,26]. CBCT complements the use of IMRT because it provides a highly accurate setup for the highly conformal dose distributions associated with IMRT. One of the concerns with IMRT, though, is the increased low dose to normal tissues (peripheral dose) compared to 3D conformal techniques. Studies of peripheral dose from IMRT are pub-

lished, with examples for pediatric and prostate patients [9,23], as are studies on how the latest technology needs to be fully assessed for its impact on peripheral dose [42]. Because CBCT is known to deliver dose to the patient, it is a logical step to also determine the peripheral dose for this technique and to put it into context by comparing it with the peripheral dose for IMRT. Followill et al. [14,15], early on in the IMRT era, expressed concern that the move from 2DRT/3DCRT to IMRT may result in an increased rate of secondary malignancies because of the significantly larger number of monitor units (MUs) required to deliver a comparable prescribed dose. Typically, the increase in the number of MUs is by a factor of 2 to 3 for conventional MLC IMRT (but can be as large as 8 to 10 for some other forms of IMRT, e.g., serial tomotherapy) [43]. This results in an increase in the dose outside the boundary of the primary collimator due to the increased amount of leakage and scattered radiation [14,15,44]. Followill et al. reported the estimated percent probability of a fatal secondary cancer due to a 70.00 Gy course of radiation therapy as follows: (a) for 6MV X-rays: 0.3% for 2DRT un-wedged technique; 1.0% for IMRT: 2.7% for tomotherapy; and (b) for 18 MV X-rays: 1.8% for 2DRT un-wedged technique; and 5.1% for IMRT: 14.9% for tomotherapy [14,15]. This present study has shown that a measurable and appreciable dose exists outside the field of view when a CBCT is taken for patient alignment during radiotherapy. The peripheral dose measured for the CBCT has the same order of magnitude as that of 6 MV IMRT, since both deliver mGy per fraction. The CBCT peripheral dose found in this study is approximately half the amount associated with a prostate IMRT at a distant point outside the field; 0.2cGy (CBCT) and 0.4cGy (IMRT) were measured 25 cm from the central axis. Additionally, at a point 10 cm outside the IMRT field the doses from the peripheral IMRT field and the CBCT would be approximately equal (see Fig. 4). Hence, by considering a typical patient receiving daily IMRT and one CBCT scan per day for IGRT for a fractionated course of radiation therapy, the total peripheral dose in our clinic would be of the order of 0.2–0.5 Gy. However, the fact that the peripheral dose (from either IMRT or CBCT) scales linearly with technique should be borne in mind. IMRT as a technique introduces more peripheral dose than a 3D conformal technique as more monitor units are required; this is justified by the increase in conformity of the dose distribution to the target shape. By analogy, the peripheral dose from CBCT will scale with the technique, in proportion to the in-field dose, and by the number of times the CBCT is used. A CBCT image can be produced with a wide variety of parameters (mA, voltage, degrees of rotation, frame rate, etc.), and our clinic, for example, has 8 CBCT protocols validated for clinical use that deliver 0.5cGy for the lowest dose and up to 6cGy for the highest (this one was used in the peripheral dose measurements), with a 3cGy in-field scan being the most commonly used protocol on a daily basis. The clinical relevance of the IGRT peripheral dose measured here has to put into context against the intent of the therapy and the individual patient. With only one CBCT scan per day and a prescription of the order of 70 Gy for radical therapy the peripheral dose accumulated may be less

J.R. Perks et al. / Radiotherapy and Oncology 89 (2009) 304–310

than 1% of prescription. This may be considered readily acceptable in a number of circumstances but some specific cases should be noted, even beyond the concern of a secondary malignancy. Very low doses of radiation are associated with certain complications [45] including azoospermia (short-term 0.1 Gy, prolonged >2.5 Gy) [46], loss of ovarian function and oocyte death (<2 Gy) [47] and hypothyroidism/thyroid nodules (. . .about 28% of all solid nodules, 37% of malignant tumors, 31% of benign nodules, and 25% of cysts are associated with radiation exposure at a mean and median thyroid radiation dose of 0.449 and 0.087 Sv, respectively) [48,49]. So there is certainly a case to be made for a more judicious use of CBCT where a specific treatment site may be close to a critical area and in the pediatric/adolescent population [50]. Aside from reducing the mAs and the number of acquired frames of a given CBCT scan another method for reducing both the in-field and subsequently the peripheral dose from CBCT that is being investigated in our clinic is to image with a less than daily frequency. Since CBCT was introduced into clinical practice, daily imaging has been the norm at our center. There are two reasons for this, firstly the Elekta platform has proved to be incredibly stable and reliable so that CBCT scans can be easily incorporated to image the majority of patients daily, and also the ability to image and set the patient to isocenter based on those images with millimeter precision has proved to be very well received by the clinical staff. By maintaining a database of the daily shifts applied as a result of CBCT image matching, 100 patients are currently being analyzed with regard to reducing the frequency of imaging. Finally, the reader must bear in mind that this particular study is based on one particular combination of linear accelerator and CBCT system, namely the Elekta Synergy platform. Therefore, further investigation of peripheral dose from other imaging systems is warranted. The peripheral dose associated with IMRT has been proved to be consistent across different linear accelerators (compare this study with the work of Stern [21]) and this is understandable as the shielding and acceptance criteria for out-of-field dose from megavoltage beams are similar across manufacturers. However, this standardization does not exist for on board imaging systems and a recommendation would be for the user of IGRT to measure and fully understand the doses associated with the techniques and protocols they employ.

Conclusion It is clear that advances in imaging, treatment planning and treatment delivery are providing radiation oncologists the ability to conform dose closely to the target (tumor) volume while minimizing the dose to organs at risk. However, this transition from 2DRT to 3DCRT to IMRT to IGRT has resulted in clear changes to the dose distribution that previous clinical experience and second malignancy studies are based on. Particularly in the case of IMRT/IGRT, there is a larger volume of normal tissue that is irradiated to low radiation doses. Also, IMRT requires a significantly larger number of MUs to deliver a comparable prescribed dose, which results in an increase in the whole body dose as a result of leakage

309

and scattered radiation. Now, with the growing use of IGRT procedures, even more additional dose is being delivered to the patient undergoing radiation therapy. Thus, there is some potential that this era of conformal therapy may actually result in an increased rate of secondary malignancies, and that the treating physician at least acknowledges this issue in prescribing treatment regimens. This study is intended to remind the reader that peripheral dose contributed by CBCT can be on the same order of magnitude (and for multiple daily imaging can exceed) as the IMRT peripheral dose. Hence, we emphasize that, with the introduction of new radiation therapy techniques including imaging, all dosimetric consequences should be borne in mind, as all doses above the prescription require justification. Finally, in our opinion, the potential benefits of IGRT (with respect to improved precision, targeting etc.) outweigh any potential negatives (with respect to peripheral dose) although further follow-up is necessary. * Corresponding author. Julian R.Perks, Department of Radiation Oncology, UC Davis Medical Center, Suite G140, 4501 X Street, Sacramento, CA 95817, USA. E-mail address: julian.perks@ucdmc. ucdavis.edu Received 10 June 2008; received in revised form 11 July 2008; accepted 20 July 2008; Available online 31 August 2008

References [1] Letourneau D, Wong JW, Oldham M, et al. Cone-beam-CT guided radiation therapy: technical implementation. Radiother Oncol 2005;75:279–86. [2] Oldham M, Letourneau D, Watt L, et al. Cone-beam-CT guided radiation therapy: a model for on-line application. Radiother Oncol 2005;75:271–8. [3] Dawson LA, Jaffray DA. Advances in image-guided radiation therapy. J Clin Oncol 2007;25:938–46. [4] Herman MG, Abrams RA, Mayer RR. Clinical use of on-line portal imaging for daily patient treatment verification. Int J Radiat Oncol Biol Phys 1994;28:1017–23. [5] Mubata CD, Bidmead AM, Ellingham LM, et al. Portal imaging protocol for radical dose-escalated radiotherapy treatment of prostate cancer. Int J Radiat Oncol Biol Phys 1998;40:221–31. [6] Moseley DJ, White EA, Wiltshire KL, et al. Comparison of localization performance with implanted fiducial markers and cone-beam computed tomography for on-line image-guided radiotherapy of the prostate. Int J Radiat Oncol Biol Phys 2007;67:942–53. [7] Lehmann J, Perks J, Semon S, et al. Commissioning experience with cone-beam computed tomography for image-guided radiation therapy. J Appl Clin Med Phys 2007;8:2354. [8] Amer A, Marchant T, Sykes J, et al. Imaging doses from the Elekta synergy X-ray cone beam CT system. Br J Radiol 2007;80:476–82. [9] Wen N, Guan H, Hammoud R, et al. Dose delivered from Varian’s CBCT to patients receiving IMRT for prostate cancer. Phys Med Biol 2007;52:2267–76. [10] Ding GX, Duggan DM, Coffrey CW. Accurate patient dosimetry of cone-beam CT in radiation therapy. Med Phys 2008;35:1135–44. [11] Islam MK, Purdie TG, Norrlinger BD, et al. Patient dose from kilovoltage cone beam computed tomography imaging in radiation therapy. Med Phys 2006;33:1573–82. [12] Kan MW, Leung LH, Wong W, et al. Radiation dose from cone beam computed tomography for image-guided radiation therapy. Int J Radiat Oncol Biol Phys 2008;70:272–9.

310

IGRT peripheral dose

[13] Walter C, Boda-Heggemann J, Wertz H, et al. Phantom and in vivo measurements of dose exposure by image-guided radiotherapy (IGRT): MV portal images vs. KV portal images vs. cone-beam CT. Radiother Oncol 2007;85:418–23. [14] Followill D, Geis P, Boyer A. Estimates of whole-body dose equivalent produced by beam intensity modulated conformal therapy. Int J Radiat Oncol Biol Phys 1997;38:667–72. [15] Followill D, Geis P, Boyer A. Errata: estimates of whole-body dose equivalent produced by beam intensity modulated conformal therapy. Int J Radiat Oncol Biol Phys 1997;39:783. [16] Suit H, Goldberg S, Niemierko A, et al. Secondary carcinogenesis in patients treated with radiation: a review of data on radiation-induced cancers in human, non-human primate, canine and rodent subjects. Radiat Res 2007;167:12–42. [17] Hall EJ. Radiation the two-edged sword: cancer risks at high and low doses. Cancer J 2000;6:343–50. [18] Hall EJ. Intensity-modulated radiation therapy, protons, and the risk of second cancers. Int J Radiat Oncol Biol Phys 2006;65:1–7. [19] Brenner DJ, Hall EJ. Computed tomography – an increasing source of radiation exposure. N Engl J Med 2007;357:2277–84. [20] McParland BJ, Fair HI. A method of calculating peripheral dose distributions of photon beams below 10 MV. Med Phys 1992;19:283–93. [21] Stern RL. Peripheral dose from a linear accelerator equipped with multileaf collimation. Med Phys 1999;26:559–63. [22] Mutic S, Esthappan J, Klein EE. Peripheral dose distributions for a linear accelerator equipped with a secondary multileaf collimator and universal wedge. J Appl Clin Med Phys 2002;3:302–9. [23] Klein EE, Maserang B, Wood R, et al. Peripheral doses from pediatric IMRT. Med Phys 2006;33:2525–31. [24] Mansur DB, Klein EE, Maserang BP. Measured peripheral dose in pediatric radiation therapy: a comparison of intensity-modulated and conformal techniques. Radiother Oncol 2007;82:179–84. [25] Stovall M, Blackwell CR, Cundiff J, et al. Fetal dose from radiotherapy with photon beams: report of AAPM Radiation Therapy Committee Task Group No. 36. Med Phys 1995;22:63–82. [26] Murphy MJ, Balter J, Balter S, et al. The management of imaging dose during image-guided radiotherapy: report of the AAPM Task Group 75. Med Phys 2007;34:4041–63. [27] Mutic S, Klein EE. A reduction in the AAPM TG-36 reported peripheral dose distributions with tertiary multileaf collimation American Association of Physicists in Medicine Task Group 36. Int J Radiat Oncol Biol Phys 1999;44:947–53. [28] Harrison RM, Wilkinson M, Shemilt A, et al. Organ doses from prostate radiotherapy and associated concomitant exposures. Br J Radiol 2006;79:487–96. [29] Ramsey C, Seibert R, Mahan SL, et al. Out-of-field dosimetry measurements for a helical tomotherapy system. J Appl Clin Med Phys 2006;7:1–11. [30] Lehmann J, Stern RL, Levy J, et al. Radiation phantom with humanoid shape and adjustable thickness (RPHAT). Phys Med Biol 2004;49:N125–9.

[31] Muhogora WE, Ngoye WN, Lema US, et al. Energy response of LiF:Mg, Ti dosemeters to ISO 4037 and typical diagnostic X-ray beams in Tanzania. J Radiol Prot 2002;22:175–84. [32] Zoetelief J, Jansen JT. Calculated energy response correction factors for LiF thermoluminescent dosemeters employed in the seventh EULEP dosimetry intercomparison. Phys Med Biol 1997;42:1491–504. [33] Winzel E, van der Merwe EJ, Groenewald W, et al. The relative biological effectiveness of 100 kV X-rays determined by the V-79 cell colony assay. S Afr Med J 1987;71:693–5. [34] Goggelmann W, Jacobsen C, Panzer W, et al. Re-evaluation of the RBE of 29 kV X-rays (mammography X-rays) relative to 220 kV X-rays using neoplastic transformation of human CGL1hybrid cells. Radiat Environ Biophys 2003;42:175–82. [35] Lehnert A, Lessmann E, Pawelke J, et al. RBE of 25 kV X-rays for the survival and induction of micronuclei in the human mammary epithelial cell line MCF-12A. Radiat Environ Biophys 2006;45:253–60. [36] Heyes GJ, Mill AJ, Charles MW. Enhanced biological effectiveness of low energy X-rays and implications for the UK breast screening programme. Br J Radiol 2006;79:195–200. [37] Butson MJ, Cheung T, Yu PK. Peripheral dose measurement with a MOSFET detector. Appl Radiat Isot 2005;62:631–4. [38] Butson MJ, Rozenfeld A, Mathur JN, et al. A new radiotherapy surface dose detector: the MOSFET. Med Phys 1996;23:655–8. [39] Ramani R, Russell S, O’Brien P. Clinical dosimetry using MOSFETs. Int J Radiat Oncol Biol Phys 1997;37:959–64. [40] Bower MW, Hintenlang DE. The characterization of a commercial MOSFET dosimeter system for use in diagnostic X-ray. Health Phys 1998;75:197–204. [41] Scalchi P, Francescon P. Calibration of a mosfet detection system for 6-MV in vivo dosimetry. Int J Radiat Oncol Biol Phys 1998;40:987–93. [42] Yang C, Guo F, Liu T, et al. Leakage radiation for Elekta synergy-SLINAC machine. Med Phys 2007;34:2466. [43] IMRTCWG: NCI IMRT Collaborative Working Group. Intensity modulated radiation therapy: current status and issues of interest. Int J Radiat Oncol Biol Phys 2001;51:880–914. [44] Williams P, Hounsell R. X-ray leakage considerations for IMRT (Correspondence). Br J Radiol 2001;74:98–102. [45] Hall E. Radiobiology for the radiologist. 5th ed. Philadelphia: Lippincott Williams & Wilkins; 2000. [46] Howell S, Shalet S. Gonadal damage from chemotherapy and radiotherapy. Endocrinol Metab Clin North Am 1998;27:927–43. [47] Wallace WH, Thomson AB, Kelsey TW. The radiosensitivity of the human oocyte. Hum Reprod 2003;18:117–21. [48] Boice Jr JD. Thyroid disease 60 years after Hiroshima and 20 years after Chernobyl. JAMA 2006;295:1060–2. [49] Imaizumi M, Usa T, Tominaga T, et al. Radiation dose– response relationships for thyroid nodules and autoimmune thyroid diseases in Hiroshima and Nagasaki atomic bomb survivors 55–58 years after radiation exposure. JAMA 2006;295:1011–22. [50] Robison LL, Green DM, Hudson M, et al. Long-term outcomes of adult survivors of childhood cancer. Cancer 2005;104:2557–64.