- Email: [email protected]
Evaluation of Field-in-Field Technique for Total Body Irradiation
International Journal of
Radiation Oncology biology
physics
www.redjournal.org
Physics Contribution
Evaluation of Field-in-Field Technique for Total Body Irradiation Cem Onal, M.D., Aydan Sonmez, M.Sc., Gungor Arslan, M.Sc., Serhat Sonmez, M.Sc., Esma Efe, M.Sc., and Ezgi Oymak, M.D. Department of Radiation Oncology, Baskent University Faculty of Medicine, Adana, Turkey Received Feb 15, 2011, and in revised form Oct 12, 2011. Accepted for publication Oct 23, 2011
Summary We demonstrate the feasibility of the FIF plan for TBI with a treatment planning system, and the results are verified with in vivo dosimetry using MOSFET (metal-oxide semiconductor field-effect transistor) detectors. The MOSFET readings and planning system doses in the body were similar (percentage difference range, 0.5% to 2.5%). Because the target volume for TBI is the whole body, we conclude that the FIF plan is a safe, easy, and time-sparing technique.
Purpose: To evaluate the clinical use of a field-in-field (FIF) technique for total body irradiation (TBI) using a treatment-planning system (TPS) and to verify TPS results with in vivo dose measurements using metal-oxide-semiconductor field-effect transistor (MOSFET) detectors. Methods and Materials: Clinical and dosimetric data of 10 patients treated with TBI were assessed. Certain radiation parameters were measured using homogenous and regular phantoms at an extended distance of 380 cm, and the results were compared with data from a conventional standard distance of 100 cm. Additionally, dosimetric validation of TPS doses was performed with a Rando phantom using manual calculations. A three-dimensional computed tomography plan was generated involving 18-MV photon beams with a TPS for both open-field and FIF techniques. The midline doses were measured at the head, neck, lung, umbilicus, and pelvis for both open-field and FIF techniques. Results: All patients received planned TBI using the FIF technique with 18-MV photon energies and 2 Gy b.i.d. on 3 consecutive days. The difference in tissue maximum ratios between the extended and conventional distances was <2%. The mean deviation of manual calculations compared with TPS data was þ1.6% (range, 0.1e2.4%). A homogenous dose distribution was obtained with 18-MV photon beams using the FIF technique. The mean lung dose for the FIF technique was 79.2% (9.2 Gy; range, 8.8e9.7 Gy) of the prescribed dose. The MOSFET readings and TPS doses in the body were similar (percentage difference range, 0.5% to 2.5%) and slightly higher in the shoulder and lung (percentage difference range, 4.0e5.5%). Conclusion: The FIF technique used for TBI provides homogenous dose distribution and is feasible, simple, and spares time compared with more-complex techniques. The TPS doses were similar to the midline doses obtained from MOSFET readings. Ó 2012 Elsevier Inc. Keywords: Total body irradiation, Field-in-field technique, Conformal radiotherapy planning, Metal-oxide-semiconductor field-effect transistor (MOSFET) detectors
Introduction Reprint requests to: Cem Onal, M.D., Baskent University Faculty of Medicine, Adana Research and Treatment Center, Department of Radiation Oncology, 01120, Yuregir, Adana, Turkey. Tel: (þ90) 322-3444444/1304; Fax: (þ90) 322-3444445; E-mail: [email protected] Conflict of interest: none. Int J Radiation Oncol Biol Phys, Vol. 83, No. 5, pp. 1641e1648, 2012 0360-3016/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.ijrobp.2011.10.045
Total body irradiation (TBI) is one of the essential treatment modalities used in certain hematologic, oncologic, and immunologic diseases and has been used in conjunction with chemotherapy as a conditioning regimen, principally for bone marrow transplantation (BMT) or peripheral blood stem cell transplantation
1642 Onal et al.
International Journal of Radiation Oncology Biology Physics
(1, 2). Because the target volume for these treatments is the entire bone marrow and body, there are some limitations for treatment planning and delivering homogenous radiation doses. A conventional treatment plan involves administering a prescribed dose to a point within the body in the pelvic region at the height of the umbilicus. With this plan, the general consensus is that the dose received by the rest of the body varies by 10% of this prescribed dose (3). Computed tomography (CT)-based treatment-planning systems (TPSs) provide detailed volumetric information in inhomogeneous tissues and enable more accurate dose distributions compared with point dose calculation methods. Because the aim in conformal radiotherapy is to achieve dose homogeneity without increasing toxicity, various methods have been proposed to optimize these objectives (4, 5). The incidence of TBI-induced complications is correlated with dose. Common problems include pulmonary complications, renal toxicities, cataracts, and reduced pituitary function (6). Acute TBI-induced interstitial pneumonitis (IP) is the most important and dose-limiting TBI complication (6, 7). To improve dose distribution, the field-in-field (FIF) technique has been used in the treatment of certain cancers, especially in breast cancer (8). Field-in-field is a manually based forward intensitymodulated radiotherapy (IMRT) plan for which the calculated dose is modified in certain dose distribution areas by creating multiple lower-weighted reduction fields based on the primary field. Although the planning system may seem to be a complicated approach to enhance homogeneity, daily patient setup is ultimately simplified and more consistent than with other techniques. Importantly, treatment time is significantly shorter. Currently few data exist with respect to using FIF techniques to treat various cancer types, and no studies have yet investigated the FIF technique for TBI. In this study we evaluated the clinical use of the TPS-generated FIF technique in patients treated with TBI. Additionally, in vivo dosimetric measurements were made using metal-oxide-semiconductor field-effect transistor (MOSFET) detectors to validate TPS accuracy.
maximum ratios (TMRs) were measured in a 40 cm 40 cm 30-cm solid water phantom. The TMRs were measured at a depth that ranged between 1 and 20 cm. The off-axis dose values were measured at a depth of 10 cm, with an interval of 5 cm through 90 cm, by shifting the position of the rectangular solid water phantom. The machine was set to a 100 monitor unit (MU)/min dose rate, and the dose was measured at the midplane at an extended sourceeaxis distance of 380 cm. A 1.5-cm thickness Perspex beam spoiler was positioned close to the phantom to increase the skin doses. The build-up region dose measurement was established using a Roos PTW plane parallel ionization chamber (PTW-Freiburg GmbH, Freiburg, Germany). To confirm TPS results, these data were compared with the midline doses for head, neck, shoulder, lung, and umbilicus regions that were calculated manually. For this purpose, an Alderson Rando phantom (Alderson Research Labs, Stanford, CA) was used. For manual calculations, midline points were defined on CT images of the phantom. The distance from defined midline points to the surface was measured to find TMR data for related areas. We first manually calculated MU according to the depth of the body at the umbilicus level. The doses to other body regions were calculated on the basis of the umbilicus dose by individual TMR for open fields without any compensators. Dose distribution was significantly affected by low tissue density, especially in the lungs. We corrected the results by multiplying the lung depth with the electron density coefficient for the lung. For manual calculations to the lungs, we used the generally accepted electron density coefficient of 0.3. We also obtained lateral and back scatter factors to compensate for lack of scatter, as defined by Sanchez-Nieto et al. (9). A three-dimensional dose distribution of the phantom was calculated by the TPS, which has a pencil beam convolution (PBC8.0.3-ec001) calculation algorithm and used a modified Batho inhomogeneity correction. The manual calculation TPS findings were then compared.
Methods and Materials
Treatment planning
Patient characteristics
A planning CT scan was conducted with patients in a supine position, with legs flexed at the knees and feet fixed in a commercially available knee support device. The whole body was scanned from head to feet at a slice thickness of 1 cm. During the CT scan, a radio-opaque marker was placed on the umbilicus to define the isocenter. The CT dataset was transferred to the Eclipse TPS (Varian Medical Systems, Palo Alto, CA). The target volume, including whole bones and bone marrow, were delineated with the TPS autosegmentation tool. Bilateral lungs, kidneys, and the liver were also contoured. Two sets of three-dimensional treatment plans were generated with Eclipse: one plan used a classic paralleleopposed lateral technique (either non-FIF or open-field technique), and the other set was an FIF technique. For open-field calculations, the dose was normalized to the prescription point, which was the midline at the position of the umbilicus. Field-in-field plans were then generated manually using subfields strategically placed in areas where the dose was considerably higher than the prescription dose. For the FIF plan, main fields and subfields were merged in one portal, including several multileaf collimator (MLC) segments for sequential irradiation. Two or three subfields were used for each portal. Lung doses were kept to <80% of total dose using MLC. For the non-FIF (open-field) technique, no subfields or
Clinical and dosimetric data of 10 patients treated with TBI between January 2007 and December 2009 were evaluated at Baskent University Department of Radiation Oncology in Adana, Turkey. Average patient age was 33 years (range, 17e65 years). Four patients (40%) were male, and 6 (60%) were female. Nine patients (90%) were irradiated for treatment of acute lymphoblastic leukemia, and 1 patient (10%) was irradiated for treatment of leukemic transformation of lymphoma.
Dosimetric measurements for TBI To determine whether our TPS could yield accurate TBI calculations, some radiation parameters were measured using homogenous and regular phantoms at an extended distance of 380 cm. These measurements were compared with results achieved from a conventional standard distance of 100 cm. Relative and absolute dose measurements were performed with a Scanditronix Wellho¨fer FC65-G (Wellho¨fer Dosimetrie GmbH, Schwarzenbruck, Germany) and a 0.6-cm3 ionization chamber connected to a Scanditronix Wellho¨fer electrometer. Off-axis factors and tissue
Volume 83 Number 5 2012
Field-in-field technique for TBI 1643
compensators were used to decrease doses at the head and legs. The midline doses for the two different techniques were measured at the head, neck, lung, umbilicus, and pelvis. Additionally, the FIF and non-FIF technique doses were compared between certain anatomic regions. Mean lung doses were measured, and dosee volume histograms of the lung were also created.
TBI delivery
entrance dose, Rent is the reading from the in vivo dosimeter, Fcal is the calibration factor under reference conditions, and CFi is the correction factor used to obtain the entrance dose when the irradiation conditions differ from reference conditions. To prevent uncertainty during detector readings due to lack of electronic equilibrium, Fcal was described under full scatter conditions. In accordance with International Atomic Energy Agency Protocol 398 and European Society for Therapeutic Radiology and Oncology (ESTRO) recommendations (13), the linear accelerator
A 2-Gy treatment dose was prescribed to the umbilicus and given in 6 fractions over 3 consecutive days. The treatment was given twice daily with a minimum of 6 h between fractions, an interval that was previously proven to be effective for conditioning BMT (10, 11). All patients were treated with a linear accelerator (Varian DHX 3323) with 18-MV energy using the FIF technique, with a treatment field of 40 40 cm, a gantry angle of 90 , and collimators at 45 in the first phase (Fig. 1). Subfields were used to reduce higher-dose regions (Fig. 2). The lungs were spared using MLC; lung sparing was verified at each setup by portal images, and appropriate corrections were made if needed. Because the sternum was shielded by MLC to spare the lungs, an additional boost field to the sternum was added to keep target volume doses at 10% of dose limits. Because dose homogeneity was achieved with subfields, no bolus or compensators were used during treatment. A beam spoiler was instead used to increase the skin doses.
Dose verification with MOSFET detectors Patient measurements were performed to investigate dose homogeneity and the influence of treatment technique in each patient and to compare the TPS-predicted data with the MOSFET detectors. To verify the dose across the mid-plane, the patient dose verification model TN-RD-70-W (mobile MOSFET system; Thompson and Nielsen Electronics, Ottawa, ON, Canada) was used. Radiation dose was determined Q from the MOSFET reading with the equation Dent ZRent Fcal CFi (12), where Dent is the i
Fig. 1. Initial open treatment field of 40 40 cm, with a gantry angle of 90 and collimators at a 45 angle.
Fig. 2. Subfields were used to reduce higher-dose regions, especially in (a) the head-and-neck regions and (b) the legs.
1644 Onal et al.
International Journal of Radiation Oncology Biology Physics
Results TBI dosimetric results The difference of the measured TMR between the extended distance and conventional distance was <2%. The radiation beam along the diagonal axis was almost flat and varied by 3.5%. The absolute dose measurement was very similar to the dose calculated via the inverse square law. We selected 100 MU/min from the machine options to reduce the dose rate based on the treatment distance. The dose rate was 8.27 cGy/min and was measured at the treatment distance for a large field. We used a 1.5-cm-thick beam spoiler to increase the skin doses up to 97.5%. Fig. 3. Doses of treatment-planning system (TPS) and metaloxide-semiconductor field-effect transistor (MOSFET) readings at various anatomic sites.
was initially calibrated with a polystyrene phantom at reference conditions of a 10 10-cm field size at the isocenter and a sourceesurface distance of 100 cm, because 1 cGy equals 1 MU at dmax. According to the ESTRO formulae (13), to eliminate electron contamination and to achieve a consistent correction factor (CF) during MOSFET calibration, a depth of 10 cm is preferred. Measurements are not sensitive to electron contamination provided that the reference depth is sufficiently large. The MOSFET detectors were put into the MOSFET calibration jig, which was then placed at a 10-cm depth, sandwiched between solid water phantom plates, and irradiated with the prescribed dose. During calibration the detectors do not need a build-up cap, because it is built into 10-cm depth of phantom, which is adequate for build-up conditions. Measurements are not sensitive to electron contamination provided that the reference depth is sufficiently large. Therefore, MU calculations were conducted without any alteration in physical conditions and using sourceesurface distance techniques at a 200-cGy dose and 10-cm depth. Because MOSFET readings and the conditions of known radiation doses were equivalent, CF was equal to 1, and Fcal was calculated for each detector by dividing the measured values from the detectors and prescribed dose. During the in vivo measurements at high energies, covering the detectors at the skin surface either with bolus material or a buildup cap is preferred (14). In treatment conditions, we measured the entrance dose at the desired levels with MOSFET detectors taped to the body surface. For achieving build-up conditions, a 1-cmthick, 3 3-cm-wide bolus was placed on the detector on the patient’s head, neck, shoulder, umbilicus, and pelvic regions. An additional 1.5-cm-thick beam spoiler was used to increase the skin dose and to remedy the lack of appropriate build-up material. Using the beam spoiler and the boluses led us to create a full build-up on the detector for TBI conditions. The detectors were read within 5 min of completing the irradiation. At points of interest, the ratio of the mid-separation dose to the entry dose sum was calculated. Estimation of the midline dose can take place only at the skin surface, and thus a relationship between the dose at the midline (Dm) and at the surface (DYS) was established from the TPS to evaluate the dose with the formula Dm ZDs ðTMRm Þ=ðfm Þ2 (15), where fm is the distance to the fs midline point, fs is the distance to surface point, and TMRm is the TMR for the midline depth.
Treatment planning system doses The mean deviation between the manual calculations and the TPS data was þ1.6% (range, 0.1e2.4%). These results confirmed that the TPS data and manual calculations were very similar. We then determined the monitor calculations from the TPS data and treated the patients according to these TPS findings, which were verified with in vivo dosimetry. All patients received planned TBI using the FIF technique with 18-MV photon energy on schedule with no delays in treatment time, and no major complications were observed during treatment. Because they were treated in the supine position, the patients were comfortable while this technique we being performed. Setup was rapid and generally took <10 min. The entire treatment time needed per patient for a course of treatment, including planning time, was approximately 5 h. The changes in mean midline dose values of the treatment fields for 10 patients according to energy values are presented in Table 1. Except for the thoracic region surrounding the lungs and shoulder, which included the apical lung segment, doses to all measured points on the trunk and head were within 3.5% of the prescribed dose. The measurements from the shoulder and lung doses were lower than in other regions of the body. The mean patient lung volume was 2588 cm3 (range, 2156e3769 cm3). To decrease IP risk, lungs were shielded with MLC to keep the lung dose at <80% of the prescribed dose; the mean lung doses as a result were 79.2%. The mean lung doses of the FIF and non-FIF techniques were 9.5 Gy (range, 9.2e9.7 Gy) and 11.8 Gy (range, 11.4e12.1 Gy), respectively. The FIF technique enabled a more homogenous dose distribution compared with the non-FIF technique (Fig. 4). The largest dose differences were evident at the Table 1 Quantitative analysis of percentage dose for 10 patients in different anatomic sites from computed tomographyebased treatment plan calculations using the field-in-field technique Dose (%) Anatomic site
Mean SD
Head Neck Shoulder Lung Umbilicus Pelvis
100.2 101.4 82.0 78.8 100.5 96.7
2.4 3.3 4.9 1.5 1.0 1.3
Range 95.3e103.9 96.3e106.8 74.6e88.4 76.8e80.7 100.0e103.3 94.9e98.9
Volume 83 Number 5 2012
Field-in-field technique for TBI 1645
Discussion
Fig. 4. Mean percentage doses of field-in-field (FIF) and nonFIF techniques for different anatomic sites from computed tomographyebased treatment-planning calculations. head-and-neck region (Fig. 5). In these regions the doses were lower with the FIF technique compared with the non-FIF technique (Fig. 6).
MOSFET detector doses To assess the midline doses at certain anatomic regions with increased accuracy, the skin doses measured by MOSFET detectors were converted to midline doses, and these results were compared with the TPS doses. A comparison of midline dose determined from MOSFET and TPS is presented in Table 2. As shown in Fig. 3, many of the MOSFET readings were very similar to the TPS doses. However, the TPS doses at the shoulder and lung regions were considerably lower than the MOSFET doses.
In the present study we evaluated a simple and feasible method involving the FIF technique for patients being treated with TBI for conditioning before BMT. We demonstrated that using the FIF technique results in reasonable dose homogeneity and acceptable pulmonary doses. Additionally, the TPS results are supported by the MOSFET dosimetric readings. Importantly, the FIF technique is simpler and has a more rapid treatment time and easier setup compared with more-complex techniques. Total body irradiation has an important role in conditioning patients for BMT in that it both eradicates malignant cells and prevents rejection of donor hematopoietic cells via immunosuppression. The optimal TBI regimen is still being investigated, mainly because of toxicities. Gains in disease control are associated with an increase in regimen-related toxicities and nonrelapse mortality, resulting in no improvement in overall survival. We used modern combined treatment strategies for the patients in this study, and our aim was to increase survival rates without causing significant treatment-related morbidity and mortality. A major consideration in designing TBI treatment strategies is dose homogeneity throughout the body. The dose administered to the rest of the body other than lungs should be within 10% of this prescribed dose (3). Conventional methods for TBI include anteroposterioreposteroanterior and paralleleopposed lateral techniques. Achieving homogenous dose distribution with either of these techniques is challenging because of variations in patient tissue thickness and density throughout the body. Contour variations cause changes in both the primary and scatter dose at certain points within the patient. A detailed calculation is possible if precise patient-specific anatomic and density information is available from CT imaging slices. Many authors reported commercially available TPSs capable of calculating a patient’s
Fig. 5. Dose profile of a corresponding patient delivered from the treatment-planning system for the non-FIF technique. Higher doses were evident, particularly at the head and neck regions; dose reductions were seen at the shoulder and lung.
1646 Onal et al.
International Journal of Radiation Oncology Biology Physics
Fig. 6. Dose profile of a representative patient delivered from the treatment-planning system for the field-in-field (FIF) technique. Highdose regions were reduced at the head and neck area, and the dose distribution was more homogenous. contour at an extended distance (16). If these pixel-based data can be used in a treatment plan to directly determine the dose, an accuracy of 3% can be achieved with the dose calculation (17). When noncomputerized calculation results were compared with those of the TPS, the values obtained from our TPS (Varian Eclipse) could be safely used for TBI dose calculations, as was seen in our manual and TPS calculations that demonstrated a small deviation of þ1.6%. Several techniques, either simple or complex, were assessed to achieve dose homogeneity. The FIF technique, also known as forward IMRT planning, is a simple technique with demonstrated feasibility in treating certain cancers, especially breast cancer, gastrointestinal cancers, and tumors treated with craniospinal irradiation (8, 18). During the FIF plan, both open fields and subfields were used to reduce hot spots, and fewer MU were required to achieve a homogenous dose distribution compared with standard IMRT. Uniform dose delivery with these techniques ensures that critical organs are spared, particularly the lungs. To evaluate the feasibility of the FIF technique for TBI, dosimetric assessment is essential. We therefore used MOSFET detectors to verify the TPS findings. Recently, Bloemen-van Gurp et al. (4) demonstrated that MOSFETs were suitable for in vivo dosimetry purposes during TBI when appropriate conversion factors are used. In the present study, we used MOSFET detectors to measure in vivo doses and found that the measurements were very similar to TPS findings, with the exception of those at the lung and shoulder. These findings may be explained by two possibilities. First, we used 0.3 as the homogeneity correction factor value for the lungs. However, we know that the dose estimation of the entire lung from a single point calculation poses a significant risk of inaccuracy. Furthermore, the electron density of lung tissue varies according to patient age, level of the section, and volume of air inside the lung. A second explanation is that MLC blocks the lungs and beam spoilers. Electrons have limited range due to the spoiler, and the doses close to the spoiler are affected. The skin dose under
the MLC was increased, which may cause dose elevations in the MOSFET readings of the lung and shoulder region. The effectiveness of TBI is limited by the low tolerance of normal tissues, especially the eye lens, lungs, liver, and kidneys, which are particularly susceptible to injury from ionizing radiation. Damage to the lung is of particular concern in TBI as IP, the major dose-limiting TBI-related toxicity, which is highly correlated with mean lung dose (10, 19). In a retrospective review conducted by Sampath et al. (6), a conditioning regimen of 12-Gy TBI in 6 daily fractions induced IP at an approximate incidence of 11% in the absence of lung shielding; with lung shielding, the estimated incidence decreased to approximately 2.3%. Whether lung shielding should be used is still matter of debate, because lung sparing may also spare cancer cells and/or reduce the immunosuppression required for BMT. Although research thus far indicates that lung sparing may adversely affect the effectiveness of BMT, it is important to note that most tissues being shielded still receive approximately 60e80% of the target dose, which is Table 2 MOSFET
Prescribed dose measurements in TPS and
Anatomic site
TPS dose (cGy)
MOSFET dose (cGy)
Difference (%)
Head Neck Shoulder Lung Umbilicus Pelvis
200 203 164 158 201 193
199 202 172 169 204 188
0.5 0.5 þ4.7 þ6.5 þ1.5 2.5
Abbreviations: TPS Z treatment-planning system; MOSFET Z metal-oxide-semiconductor field-effect transistor.
Volume 83 Number 5 2012 within accepted limitations. Additionally, chemotherapy administered during the conditioning regimen adds to the therapeutic effect in these shielded areas. To reduce lung doses, we used lung shielding with MLC. However, there might be concerns with using this technique because MLC might spare the sternum and vertebrae in addition to sparing the lungs. To avoid this unwanted dose reduction in the sternum and vertebrae, we designed a boost field for these regions. Performing calculations in a number of different planes using appropriate CT scans enables the design of MLC-based plans to compensate for contour irregularities. Many centers prefer bilateral fields in which rice bags or a tissue-equivalent bolus are positioned around the head, neck, and extremities to reduce the dose at these areas during irradiation and create a more homogeneous dose distribution. The technique described here needs no compensating materials; instead, subfields using MLC are used for dose reduction in these regions. The procedure described for simple IMRT plans is both easy and practical and does not require much more time for planning and delivery. Importantly, the highdose regions are reduced significantly. To improve the sparing of organs at risk while achieving conformal dose delivery to bones and bone marrow, conformal techniques have been assessed, especially total marrow irradiation (TMI), an IMRT technique. Feasibility studies of TMI using the helical tomotherapy system (5, 20) and linear accelerators (21, 22) have been reported. Although improved target dose coverage has been achieved with these techniques, the major difficulty was the treatment time and setup difficulties (21). With longer treatment times, patient movement could negatively impact dose distribution. For this reason, time-sparing techniques with appropriate dose distributions are likely more feasible. Another concern with respect to more sophisticated techniques for TMI is the possibility of missing circulating clonogenic cells located outside the bone marrow. Although induction chemotherapy used as a part of the conditioning regimen provides systemic treatment of circulating cells, clonogens may still remain in circulation outside the target volume (5, 23). Wheldon et al. (23) noted that the isoeffective dose should correlate with the log of the cell density, resulting in a reduction in the dose required to sterilize circulating cells. With TMI, circulating clonogens remaining after induction chemotherapy is a clinical problem that still needs to be solved. The simplified method described in this study, which covers the entire body, is therefore currently more feasible than TMI to eradicate both bone marrow and circulating clonogenic cells.
Conclusion Total body irradiation is an important part of patient conditioning before BMT. Because prolonged survival is achieved with combined-modality treatment, quality of life should also be taken into consideration. For this purpose, the aim is to treat patients effectively without inducing chronic toxicities in the surrounding organs, and especially in the lungs. Because the treatment field encompasses the whole body, accomplishing dose homogeneity is difficult and results in an elevated risk of IP. We thus evaluated the clinical use of the FIF technique delivered from a CT-based TPS in patients receiving TBI, and results were supported by in vivo dosimetry using MOSFET detectors. We conclude that the FIF technique is a safe, feasible, and time-sparing technique for TBI.
Field-in-field technique for TBI 1647
References 1. Appelbaum FR. The influence of total dose, fractionation, dose rate, and distribution of total body irradiation on bone marrow transplantation. Semin Oncol 1993;20:3e10. quiz 11. 2. Gilson D, Taylor RE. Total body irradiation. Report on a meeting organized by the BIR Oncology Committee, held at The Royal Institute of British Architects, London, 28 November 1996. Br J Radiol 1997; 70:1201e1203. 3. Van Dyk J, Galvin JM, Glasgow GP, et al. AAPM report no. 17. The physical aspects of total and half body irradiation. Report of Task Group 29, Radiation Therapy Committee, American Association of Physicists in Medicine. New York: AAPM; 1986. 4. Bloemen-van Gurp EJ, Mijnheer BJ, Verschueren TA, et al. Total body irradiation, toward optimal individual delivery: Dose evaluation with metal oxide field effect transistors, thermoluminescence detectors, and a treatment planning system. Int J Radiat Oncol Biol Phys 2007;69: 1297e1304. 5. Schultheiss TE, Wong J, Liu A, et al. Image-guided total marrow and total lymphatic irradiation using helical tomotherapy. Int J Radiat Oncol Biol Phys 2007;67:1259e1267. 6. Sampath S, Schultheiss TE, Wong J. Dose response and factors related to interstitial pneumonitis after bone marrow transplant. Int J Radiat Oncol Biol Phys 2005;63:876e884. 7. Planskoy B, Tapper PD, Bedford AM, et al. Physical aspects of totalbody irradiation at the Middlesex Hospital (UCL group of hospitals), London 1988e1993: II. In vivo planning and dosimetry. Phys Med Biol 1996;41:2327e2343. 8. Prabhakar R, Haresh KP, Kumar M, et al. Field-in-field technique for upper abdominal malignancies in clinical radiotherapy. J Cancer Res Ther 2009;5:20e23. 9. Sanchez-Nieto B, Sanchez-Doblado F, Terron JA, et al. Lateral scatter correction algorithm for percentage depth dose in a large-field photon beam. Med Dosim 1997;22:121e125. 10. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with acute myeloid leukemia in first remission: A randomized trial of two irradiation regimens. Blood 1990;76:1867e1871. 11. Harden SV, Routsis DS, Geater AR, et al. Total body irradiation using a modified standing technique: A single institution 7 year experience. Br J Radiol 2001;74:1041e1047. 12. Carrasco P, Jornet N, Duch MA, et al. Comparison of dose calculation algorithms in phantoms with lung equivalent heterogeneities under conditions of lateral electronic disequilibrium. Med Phys 2004;31: 2899e2911. 13. Huyskens D, Bogaerts R, Verstraete J, et al. Practical guidelines for the implementation of in vivo dosimetry with diodes in external radiotherapy with photon beams (entrance dose). ESTRO physics for clinical radiotherapy booklet 5. 1st ed. Brussels: European Society for Therapeutic Radiology and Oncology; 2001. 14. Jornet N, Carrasco P, Jurado D, et al. Comparison study of MOSFET detectors and diodes for entrance in vivo dosimetry in 18 MV x-ray beams. Med Phys 2004;31:2534e2542. 15. Kirby TH, Hanson WF, Cates DA. Verification of total body photon irradiation dosimetry techniques. Med Phys 1988;15: 364e369. 16. Hui SK, Das RK, Thomadsen B, et al. CT-based analysis of dose homogeneity in total body irradiation using lateral beam. J Appl Clin Med Phys 2004;5:71e79. 17. Sanchez-Nieto B, Sanchez-Doblado F, Terron JA. A CT-aided PC-based physical treatment planning of TBI: A method for dose calculation. Radiother Oncol 1997;42:77e85. 18. de la Torre N, Figueroa CT, Martinez K, et al. A comparative study of surface dose and dose distribution for intact breast following irradiation with field-in-field technique vs. the use of conventional wedges. Med Dosim 2004;29:109e114.
1648 Onal et al.
International Journal of Radiation Oncology Biology Physics
19. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with chronic myeloid leukemia in the chronic phase: A randomized trial of two irradiation regimens. Blood 1991;77: 1660e1665. 20. Hui SK, Kapatoes J, Fowler J, et al. Feasibility study of helical tomotherapy for total body or total marrow irradiation. Med Phys 2005;32:3214e3224.
21. Wilkie JR, Tiryaki H, Smith BD, et al. Feasibility study for linac-based intensity modulated total marrow irradiation. Med Phys 2008;35:5609e5618. 22. Aydogan B, Mundt AJ, Roeske JC. Linac-based intensity modulated total marrow irradiation (IM-TMI). Technol Cancer Res Treat 2006;5:513e519. 23. Wheldon TE. The radiobiological basis of total body irradiation. Br J Radiol 1997;70:1204e1207.
Radiation Oncology biology
physics
www.redjournal.org
Physics Contribution
Evaluation of Field-in-Field Technique for Total Body Irradiation Cem Onal, M.D., Aydan Sonmez, M.Sc., Gungor Arslan, M.Sc., Serhat Sonmez, M.Sc., Esma Efe, M.Sc., and Ezgi Oymak, M.D. Department of Radiation Oncology, Baskent University Faculty of Medicine, Adana, Turkey Received Feb 15, 2011, and in revised form Oct 12, 2011. Accepted for publication Oct 23, 2011
Summary We demonstrate the feasibility of the FIF plan for TBI with a treatment planning system, and the results are verified with in vivo dosimetry using MOSFET (metal-oxide semiconductor field-effect transistor) detectors. The MOSFET readings and planning system doses in the body were similar (percentage difference range, 0.5% to 2.5%). Because the target volume for TBI is the whole body, we conclude that the FIF plan is a safe, easy, and time-sparing technique.
Purpose: To evaluate the clinical use of a field-in-field (FIF) technique for total body irradiation (TBI) using a treatment-planning system (TPS) and to verify TPS results with in vivo dose measurements using metal-oxide-semiconductor field-effect transistor (MOSFET) detectors. Methods and Materials: Clinical and dosimetric data of 10 patients treated with TBI were assessed. Certain radiation parameters were measured using homogenous and regular phantoms at an extended distance of 380 cm, and the results were compared with data from a conventional standard distance of 100 cm. Additionally, dosimetric validation of TPS doses was performed with a Rando phantom using manual calculations. A three-dimensional computed tomography plan was generated involving 18-MV photon beams with a TPS for both open-field and FIF techniques. The midline doses were measured at the head, neck, lung, umbilicus, and pelvis for both open-field and FIF techniques. Results: All patients received planned TBI using the FIF technique with 18-MV photon energies and 2 Gy b.i.d. on 3 consecutive days. The difference in tissue maximum ratios between the extended and conventional distances was <2%. The mean deviation of manual calculations compared with TPS data was þ1.6% (range, 0.1e2.4%). A homogenous dose distribution was obtained with 18-MV photon beams using the FIF technique. The mean lung dose for the FIF technique was 79.2% (9.2 Gy; range, 8.8e9.7 Gy) of the prescribed dose. The MOSFET readings and TPS doses in the body were similar (percentage difference range, 0.5% to 2.5%) and slightly higher in the shoulder and lung (percentage difference range, 4.0e5.5%). Conclusion: The FIF technique used for TBI provides homogenous dose distribution and is feasible, simple, and spares time compared with more-complex techniques. The TPS doses were similar to the midline doses obtained from MOSFET readings. Ó 2012 Elsevier Inc. Keywords: Total body irradiation, Field-in-field technique, Conformal radiotherapy planning, Metal-oxide-semiconductor field-effect transistor (MOSFET) detectors
Introduction Reprint requests to: Cem Onal, M.D., Baskent University Faculty of Medicine, Adana Research and Treatment Center, Department of Radiation Oncology, 01120, Yuregir, Adana, Turkey. Tel: (þ90) 322-3444444/1304; Fax: (þ90) 322-3444445; E-mail: [email protected] Conflict of interest: none. Int J Radiation Oncol Biol Phys, Vol. 83, No. 5, pp. 1641e1648, 2012 0360-3016/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.ijrobp.2011.10.045
Total body irradiation (TBI) is one of the essential treatment modalities used in certain hematologic, oncologic, and immunologic diseases and has been used in conjunction with chemotherapy as a conditioning regimen, principally for bone marrow transplantation (BMT) or peripheral blood stem cell transplantation
1642 Onal et al.
International Journal of Radiation Oncology Biology Physics
(1, 2). Because the target volume for these treatments is the entire bone marrow and body, there are some limitations for treatment planning and delivering homogenous radiation doses. A conventional treatment plan involves administering a prescribed dose to a point within the body in the pelvic region at the height of the umbilicus. With this plan, the general consensus is that the dose received by the rest of the body varies by 10% of this prescribed dose (3). Computed tomography (CT)-based treatment-planning systems (TPSs) provide detailed volumetric information in inhomogeneous tissues and enable more accurate dose distributions compared with point dose calculation methods. Because the aim in conformal radiotherapy is to achieve dose homogeneity without increasing toxicity, various methods have been proposed to optimize these objectives (4, 5). The incidence of TBI-induced complications is correlated with dose. Common problems include pulmonary complications, renal toxicities, cataracts, and reduced pituitary function (6). Acute TBI-induced interstitial pneumonitis (IP) is the most important and dose-limiting TBI complication (6, 7). To improve dose distribution, the field-in-field (FIF) technique has been used in the treatment of certain cancers, especially in breast cancer (8). Field-in-field is a manually based forward intensitymodulated radiotherapy (IMRT) plan for which the calculated dose is modified in certain dose distribution areas by creating multiple lower-weighted reduction fields based on the primary field. Although the planning system may seem to be a complicated approach to enhance homogeneity, daily patient setup is ultimately simplified and more consistent than with other techniques. Importantly, treatment time is significantly shorter. Currently few data exist with respect to using FIF techniques to treat various cancer types, and no studies have yet investigated the FIF technique for TBI. In this study we evaluated the clinical use of the TPS-generated FIF technique in patients treated with TBI. Additionally, in vivo dosimetric measurements were made using metal-oxide-semiconductor field-effect transistor (MOSFET) detectors to validate TPS accuracy.
maximum ratios (TMRs) were measured in a 40 cm 40 cm 30-cm solid water phantom. The TMRs were measured at a depth that ranged between 1 and 20 cm. The off-axis dose values were measured at a depth of 10 cm, with an interval of 5 cm through 90 cm, by shifting the position of the rectangular solid water phantom. The machine was set to a 100 monitor unit (MU)/min dose rate, and the dose was measured at the midplane at an extended sourceeaxis distance of 380 cm. A 1.5-cm thickness Perspex beam spoiler was positioned close to the phantom to increase the skin doses. The build-up region dose measurement was established using a Roos PTW plane parallel ionization chamber (PTW-Freiburg GmbH, Freiburg, Germany). To confirm TPS results, these data were compared with the midline doses for head, neck, shoulder, lung, and umbilicus regions that were calculated manually. For this purpose, an Alderson Rando phantom (Alderson Research Labs, Stanford, CA) was used. For manual calculations, midline points were defined on CT images of the phantom. The distance from defined midline points to the surface was measured to find TMR data for related areas. We first manually calculated MU according to the depth of the body at the umbilicus level. The doses to other body regions were calculated on the basis of the umbilicus dose by individual TMR for open fields without any compensators. Dose distribution was significantly affected by low tissue density, especially in the lungs. We corrected the results by multiplying the lung depth with the electron density coefficient for the lung. For manual calculations to the lungs, we used the generally accepted electron density coefficient of 0.3. We also obtained lateral and back scatter factors to compensate for lack of scatter, as defined by Sanchez-Nieto et al. (9). A three-dimensional dose distribution of the phantom was calculated by the TPS, which has a pencil beam convolution (PBC8.0.3-ec001) calculation algorithm and used a modified Batho inhomogeneity correction. The manual calculation TPS findings were then compared.
Methods and Materials
Treatment planning
Patient characteristics
A planning CT scan was conducted with patients in a supine position, with legs flexed at the knees and feet fixed in a commercially available knee support device. The whole body was scanned from head to feet at a slice thickness of 1 cm. During the CT scan, a radio-opaque marker was placed on the umbilicus to define the isocenter. The CT dataset was transferred to the Eclipse TPS (Varian Medical Systems, Palo Alto, CA). The target volume, including whole bones and bone marrow, were delineated with the TPS autosegmentation tool. Bilateral lungs, kidneys, and the liver were also contoured. Two sets of three-dimensional treatment plans were generated with Eclipse: one plan used a classic paralleleopposed lateral technique (either non-FIF or open-field technique), and the other set was an FIF technique. For open-field calculations, the dose was normalized to the prescription point, which was the midline at the position of the umbilicus. Field-in-field plans were then generated manually using subfields strategically placed in areas where the dose was considerably higher than the prescription dose. For the FIF plan, main fields and subfields were merged in one portal, including several multileaf collimator (MLC) segments for sequential irradiation. Two or three subfields were used for each portal. Lung doses were kept to <80% of total dose using MLC. For the non-FIF (open-field) technique, no subfields or
Clinical and dosimetric data of 10 patients treated with TBI between January 2007 and December 2009 were evaluated at Baskent University Department of Radiation Oncology in Adana, Turkey. Average patient age was 33 years (range, 17e65 years). Four patients (40%) were male, and 6 (60%) were female. Nine patients (90%) were irradiated for treatment of acute lymphoblastic leukemia, and 1 patient (10%) was irradiated for treatment of leukemic transformation of lymphoma.
Dosimetric measurements for TBI To determine whether our TPS could yield accurate TBI calculations, some radiation parameters were measured using homogenous and regular phantoms at an extended distance of 380 cm. These measurements were compared with results achieved from a conventional standard distance of 100 cm. Relative and absolute dose measurements were performed with a Scanditronix Wellho¨fer FC65-G (Wellho¨fer Dosimetrie GmbH, Schwarzenbruck, Germany) and a 0.6-cm3 ionization chamber connected to a Scanditronix Wellho¨fer electrometer. Off-axis factors and tissue
Volume 83 Number 5 2012
Field-in-field technique for TBI 1643
compensators were used to decrease doses at the head and legs. The midline doses for the two different techniques were measured at the head, neck, lung, umbilicus, and pelvis. Additionally, the FIF and non-FIF technique doses were compared between certain anatomic regions. Mean lung doses were measured, and dosee volume histograms of the lung were also created.
TBI delivery
entrance dose, Rent is the reading from the in vivo dosimeter, Fcal is the calibration factor under reference conditions, and CFi is the correction factor used to obtain the entrance dose when the irradiation conditions differ from reference conditions. To prevent uncertainty during detector readings due to lack of electronic equilibrium, Fcal was described under full scatter conditions. In accordance with International Atomic Energy Agency Protocol 398 and European Society for Therapeutic Radiology and Oncology (ESTRO) recommendations (13), the linear accelerator
A 2-Gy treatment dose was prescribed to the umbilicus and given in 6 fractions over 3 consecutive days. The treatment was given twice daily with a minimum of 6 h between fractions, an interval that was previously proven to be effective for conditioning BMT (10, 11). All patients were treated with a linear accelerator (Varian DHX 3323) with 18-MV energy using the FIF technique, with a treatment field of 40 40 cm, a gantry angle of 90 , and collimators at 45 in the first phase (Fig. 1). Subfields were used to reduce higher-dose regions (Fig. 2). The lungs were spared using MLC; lung sparing was verified at each setup by portal images, and appropriate corrections were made if needed. Because the sternum was shielded by MLC to spare the lungs, an additional boost field to the sternum was added to keep target volume doses at 10% of dose limits. Because dose homogeneity was achieved with subfields, no bolus or compensators were used during treatment. A beam spoiler was instead used to increase the skin doses.
Dose verification with MOSFET detectors Patient measurements were performed to investigate dose homogeneity and the influence of treatment technique in each patient and to compare the TPS-predicted data with the MOSFET detectors. To verify the dose across the mid-plane, the patient dose verification model TN-RD-70-W (mobile MOSFET system; Thompson and Nielsen Electronics, Ottawa, ON, Canada) was used. Radiation dose was determined Q from the MOSFET reading with the equation Dent ZRent Fcal CFi (12), where Dent is the i
Fig. 1. Initial open treatment field of 40 40 cm, with a gantry angle of 90 and collimators at a 45 angle.
Fig. 2. Subfields were used to reduce higher-dose regions, especially in (a) the head-and-neck regions and (b) the legs.
1644 Onal et al.
International Journal of Radiation Oncology Biology Physics
Results TBI dosimetric results The difference of the measured TMR between the extended distance and conventional distance was <2%. The radiation beam along the diagonal axis was almost flat and varied by 3.5%. The absolute dose measurement was very similar to the dose calculated via the inverse square law. We selected 100 MU/min from the machine options to reduce the dose rate based on the treatment distance. The dose rate was 8.27 cGy/min and was measured at the treatment distance for a large field. We used a 1.5-cm-thick beam spoiler to increase the skin doses up to 97.5%. Fig. 3. Doses of treatment-planning system (TPS) and metaloxide-semiconductor field-effect transistor (MOSFET) readings at various anatomic sites.
was initially calibrated with a polystyrene phantom at reference conditions of a 10 10-cm field size at the isocenter and a sourceesurface distance of 100 cm, because 1 cGy equals 1 MU at dmax. According to the ESTRO formulae (13), to eliminate electron contamination and to achieve a consistent correction factor (CF) during MOSFET calibration, a depth of 10 cm is preferred. Measurements are not sensitive to electron contamination provided that the reference depth is sufficiently large. The MOSFET detectors were put into the MOSFET calibration jig, which was then placed at a 10-cm depth, sandwiched between solid water phantom plates, and irradiated with the prescribed dose. During calibration the detectors do not need a build-up cap, because it is built into 10-cm depth of phantom, which is adequate for build-up conditions. Measurements are not sensitive to electron contamination provided that the reference depth is sufficiently large. Therefore, MU calculations were conducted without any alteration in physical conditions and using sourceesurface distance techniques at a 200-cGy dose and 10-cm depth. Because MOSFET readings and the conditions of known radiation doses were equivalent, CF was equal to 1, and Fcal was calculated for each detector by dividing the measured values from the detectors and prescribed dose. During the in vivo measurements at high energies, covering the detectors at the skin surface either with bolus material or a buildup cap is preferred (14). In treatment conditions, we measured the entrance dose at the desired levels with MOSFET detectors taped to the body surface. For achieving build-up conditions, a 1-cmthick, 3 3-cm-wide bolus was placed on the detector on the patient’s head, neck, shoulder, umbilicus, and pelvic regions. An additional 1.5-cm-thick beam spoiler was used to increase the skin dose and to remedy the lack of appropriate build-up material. Using the beam spoiler and the boluses led us to create a full build-up on the detector for TBI conditions. The detectors were read within 5 min of completing the irradiation. At points of interest, the ratio of the mid-separation dose to the entry dose sum was calculated. Estimation of the midline dose can take place only at the skin surface, and thus a relationship between the dose at the midline (Dm) and at the surface (DYS) was established from the TPS to evaluate the dose with the formula Dm ZDs ðTMRm Þ=ðfm Þ2 (15), where fm is the distance to the fs midline point, fs is the distance to surface point, and TMRm is the TMR for the midline depth.
Treatment planning system doses The mean deviation between the manual calculations and the TPS data was þ1.6% (range, 0.1e2.4%). These results confirmed that the TPS data and manual calculations were very similar. We then determined the monitor calculations from the TPS data and treated the patients according to these TPS findings, which were verified with in vivo dosimetry. All patients received planned TBI using the FIF technique with 18-MV photon energy on schedule with no delays in treatment time, and no major complications were observed during treatment. Because they were treated in the supine position, the patients were comfortable while this technique we being performed. Setup was rapid and generally took <10 min. The entire treatment time needed per patient for a course of treatment, including planning time, was approximately 5 h. The changes in mean midline dose values of the treatment fields for 10 patients according to energy values are presented in Table 1. Except for the thoracic region surrounding the lungs and shoulder, which included the apical lung segment, doses to all measured points on the trunk and head were within 3.5% of the prescribed dose. The measurements from the shoulder and lung doses were lower than in other regions of the body. The mean patient lung volume was 2588 cm3 (range, 2156e3769 cm3). To decrease IP risk, lungs were shielded with MLC to keep the lung dose at <80% of the prescribed dose; the mean lung doses as a result were 79.2%. The mean lung doses of the FIF and non-FIF techniques were 9.5 Gy (range, 9.2e9.7 Gy) and 11.8 Gy (range, 11.4e12.1 Gy), respectively. The FIF technique enabled a more homogenous dose distribution compared with the non-FIF technique (Fig. 4). The largest dose differences were evident at the Table 1 Quantitative analysis of percentage dose for 10 patients in different anatomic sites from computed tomographyebased treatment plan calculations using the field-in-field technique Dose (%) Anatomic site
Mean SD
Head Neck Shoulder Lung Umbilicus Pelvis
100.2 101.4 82.0 78.8 100.5 96.7
2.4 3.3 4.9 1.5 1.0 1.3
Range 95.3e103.9 96.3e106.8 74.6e88.4 76.8e80.7 100.0e103.3 94.9e98.9
Volume 83 Number 5 2012
Field-in-field technique for TBI 1645
Discussion
Fig. 4. Mean percentage doses of field-in-field (FIF) and nonFIF techniques for different anatomic sites from computed tomographyebased treatment-planning calculations. head-and-neck region (Fig. 5). In these regions the doses were lower with the FIF technique compared with the non-FIF technique (Fig. 6).
MOSFET detector doses To assess the midline doses at certain anatomic regions with increased accuracy, the skin doses measured by MOSFET detectors were converted to midline doses, and these results were compared with the TPS doses. A comparison of midline dose determined from MOSFET and TPS is presented in Table 2. As shown in Fig. 3, many of the MOSFET readings were very similar to the TPS doses. However, the TPS doses at the shoulder and lung regions were considerably lower than the MOSFET doses.
In the present study we evaluated a simple and feasible method involving the FIF technique for patients being treated with TBI for conditioning before BMT. We demonstrated that using the FIF technique results in reasonable dose homogeneity and acceptable pulmonary doses. Additionally, the TPS results are supported by the MOSFET dosimetric readings. Importantly, the FIF technique is simpler and has a more rapid treatment time and easier setup compared with more-complex techniques. Total body irradiation has an important role in conditioning patients for BMT in that it both eradicates malignant cells and prevents rejection of donor hematopoietic cells via immunosuppression. The optimal TBI regimen is still being investigated, mainly because of toxicities. Gains in disease control are associated with an increase in regimen-related toxicities and nonrelapse mortality, resulting in no improvement in overall survival. We used modern combined treatment strategies for the patients in this study, and our aim was to increase survival rates without causing significant treatment-related morbidity and mortality. A major consideration in designing TBI treatment strategies is dose homogeneity throughout the body. The dose administered to the rest of the body other than lungs should be within 10% of this prescribed dose (3). Conventional methods for TBI include anteroposterioreposteroanterior and paralleleopposed lateral techniques. Achieving homogenous dose distribution with either of these techniques is challenging because of variations in patient tissue thickness and density throughout the body. Contour variations cause changes in both the primary and scatter dose at certain points within the patient. A detailed calculation is possible if precise patient-specific anatomic and density information is available from CT imaging slices. Many authors reported commercially available TPSs capable of calculating a patient’s
Fig. 5. Dose profile of a corresponding patient delivered from the treatment-planning system for the non-FIF technique. Higher doses were evident, particularly at the head and neck regions; dose reductions were seen at the shoulder and lung.
1646 Onal et al.
International Journal of Radiation Oncology Biology Physics
Fig. 6. Dose profile of a representative patient delivered from the treatment-planning system for the field-in-field (FIF) technique. Highdose regions were reduced at the head and neck area, and the dose distribution was more homogenous. contour at an extended distance (16). If these pixel-based data can be used in a treatment plan to directly determine the dose, an accuracy of 3% can be achieved with the dose calculation (17). When noncomputerized calculation results were compared with those of the TPS, the values obtained from our TPS (Varian Eclipse) could be safely used for TBI dose calculations, as was seen in our manual and TPS calculations that demonstrated a small deviation of þ1.6%. Several techniques, either simple or complex, were assessed to achieve dose homogeneity. The FIF technique, also known as forward IMRT planning, is a simple technique with demonstrated feasibility in treating certain cancers, especially breast cancer, gastrointestinal cancers, and tumors treated with craniospinal irradiation (8, 18). During the FIF plan, both open fields and subfields were used to reduce hot spots, and fewer MU were required to achieve a homogenous dose distribution compared with standard IMRT. Uniform dose delivery with these techniques ensures that critical organs are spared, particularly the lungs. To evaluate the feasibility of the FIF technique for TBI, dosimetric assessment is essential. We therefore used MOSFET detectors to verify the TPS findings. Recently, Bloemen-van Gurp et al. (4) demonstrated that MOSFETs were suitable for in vivo dosimetry purposes during TBI when appropriate conversion factors are used. In the present study, we used MOSFET detectors to measure in vivo doses and found that the measurements were very similar to TPS findings, with the exception of those at the lung and shoulder. These findings may be explained by two possibilities. First, we used 0.3 as the homogeneity correction factor value for the lungs. However, we know that the dose estimation of the entire lung from a single point calculation poses a significant risk of inaccuracy. Furthermore, the electron density of lung tissue varies according to patient age, level of the section, and volume of air inside the lung. A second explanation is that MLC blocks the lungs and beam spoilers. Electrons have limited range due to the spoiler, and the doses close to the spoiler are affected. The skin dose under
the MLC was increased, which may cause dose elevations in the MOSFET readings of the lung and shoulder region. The effectiveness of TBI is limited by the low tolerance of normal tissues, especially the eye lens, lungs, liver, and kidneys, which are particularly susceptible to injury from ionizing radiation. Damage to the lung is of particular concern in TBI as IP, the major dose-limiting TBI-related toxicity, which is highly correlated with mean lung dose (10, 19). In a retrospective review conducted by Sampath et al. (6), a conditioning regimen of 12-Gy TBI in 6 daily fractions induced IP at an approximate incidence of 11% in the absence of lung shielding; with lung shielding, the estimated incidence decreased to approximately 2.3%. Whether lung shielding should be used is still matter of debate, because lung sparing may also spare cancer cells and/or reduce the immunosuppression required for BMT. Although research thus far indicates that lung sparing may adversely affect the effectiveness of BMT, it is important to note that most tissues being shielded still receive approximately 60e80% of the target dose, which is Table 2 MOSFET
Prescribed dose measurements in TPS and
Anatomic site
TPS dose (cGy)
MOSFET dose (cGy)
Difference (%)
Head Neck Shoulder Lung Umbilicus Pelvis
200 203 164 158 201 193
199 202 172 169 204 188
0.5 0.5 þ4.7 þ6.5 þ1.5 2.5
Abbreviations: TPS Z treatment-planning system; MOSFET Z metal-oxide-semiconductor field-effect transistor.
Volume 83 Number 5 2012 within accepted limitations. Additionally, chemotherapy administered during the conditioning regimen adds to the therapeutic effect in these shielded areas. To reduce lung doses, we used lung shielding with MLC. However, there might be concerns with using this technique because MLC might spare the sternum and vertebrae in addition to sparing the lungs. To avoid this unwanted dose reduction in the sternum and vertebrae, we designed a boost field for these regions. Performing calculations in a number of different planes using appropriate CT scans enables the design of MLC-based plans to compensate for contour irregularities. Many centers prefer bilateral fields in which rice bags or a tissue-equivalent bolus are positioned around the head, neck, and extremities to reduce the dose at these areas during irradiation and create a more homogeneous dose distribution. The technique described here needs no compensating materials; instead, subfields using MLC are used for dose reduction in these regions. The procedure described for simple IMRT plans is both easy and practical and does not require much more time for planning and delivery. Importantly, the highdose regions are reduced significantly. To improve the sparing of organs at risk while achieving conformal dose delivery to bones and bone marrow, conformal techniques have been assessed, especially total marrow irradiation (TMI), an IMRT technique. Feasibility studies of TMI using the helical tomotherapy system (5, 20) and linear accelerators (21, 22) have been reported. Although improved target dose coverage has been achieved with these techniques, the major difficulty was the treatment time and setup difficulties (21). With longer treatment times, patient movement could negatively impact dose distribution. For this reason, time-sparing techniques with appropriate dose distributions are likely more feasible. Another concern with respect to more sophisticated techniques for TMI is the possibility of missing circulating clonogenic cells located outside the bone marrow. Although induction chemotherapy used as a part of the conditioning regimen provides systemic treatment of circulating cells, clonogens may still remain in circulation outside the target volume (5, 23). Wheldon et al. (23) noted that the isoeffective dose should correlate with the log of the cell density, resulting in a reduction in the dose required to sterilize circulating cells. With TMI, circulating clonogens remaining after induction chemotherapy is a clinical problem that still needs to be solved. The simplified method described in this study, which covers the entire body, is therefore currently more feasible than TMI to eradicate both bone marrow and circulating clonogenic cells.
Conclusion Total body irradiation is an important part of patient conditioning before BMT. Because prolonged survival is achieved with combined-modality treatment, quality of life should also be taken into consideration. For this purpose, the aim is to treat patients effectively without inducing chronic toxicities in the surrounding organs, and especially in the lungs. Because the treatment field encompasses the whole body, accomplishing dose homogeneity is difficult and results in an elevated risk of IP. We thus evaluated the clinical use of the FIF technique delivered from a CT-based TPS in patients receiving TBI, and results were supported by in vivo dosimetry using MOSFET detectors. We conclude that the FIF technique is a safe, feasible, and time-sparing technique for TBI.
Field-in-field technique for TBI 1647
References 1. Appelbaum FR. The influence of total dose, fractionation, dose rate, and distribution of total body irradiation on bone marrow transplantation. Semin Oncol 1993;20:3e10. quiz 11. 2. Gilson D, Taylor RE. Total body irradiation. Report on a meeting organized by the BIR Oncology Committee, held at The Royal Institute of British Architects, London, 28 November 1996. Br J Radiol 1997; 70:1201e1203. 3. Van Dyk J, Galvin JM, Glasgow GP, et al. AAPM report no. 17. The physical aspects of total and half body irradiation. Report of Task Group 29, Radiation Therapy Committee, American Association of Physicists in Medicine. New York: AAPM; 1986. 4. Bloemen-van Gurp EJ, Mijnheer BJ, Verschueren TA, et al. Total body irradiation, toward optimal individual delivery: Dose evaluation with metal oxide field effect transistors, thermoluminescence detectors, and a treatment planning system. Int J Radiat Oncol Biol Phys 2007;69: 1297e1304. 5. Schultheiss TE, Wong J, Liu A, et al. Image-guided total marrow and total lymphatic irradiation using helical tomotherapy. Int J Radiat Oncol Biol Phys 2007;67:1259e1267. 6. Sampath S, Schultheiss TE, Wong J. Dose response and factors related to interstitial pneumonitis after bone marrow transplant. Int J Radiat Oncol Biol Phys 2005;63:876e884. 7. Planskoy B, Tapper PD, Bedford AM, et al. Physical aspects of totalbody irradiation at the Middlesex Hospital (UCL group of hospitals), London 1988e1993: II. In vivo planning and dosimetry. Phys Med Biol 1996;41:2327e2343. 8. Prabhakar R, Haresh KP, Kumar M, et al. Field-in-field technique for upper abdominal malignancies in clinical radiotherapy. J Cancer Res Ther 2009;5:20e23. 9. Sanchez-Nieto B, Sanchez-Doblado F, Terron JA, et al. Lateral scatter correction algorithm for percentage depth dose in a large-field photon beam. Med Dosim 1997;22:121e125. 10. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with acute myeloid leukemia in first remission: A randomized trial of two irradiation regimens. Blood 1990;76:1867e1871. 11. Harden SV, Routsis DS, Geater AR, et al. Total body irradiation using a modified standing technique: A single institution 7 year experience. Br J Radiol 2001;74:1041e1047. 12. Carrasco P, Jornet N, Duch MA, et al. Comparison of dose calculation algorithms in phantoms with lung equivalent heterogeneities under conditions of lateral electronic disequilibrium. Med Phys 2004;31: 2899e2911. 13. Huyskens D, Bogaerts R, Verstraete J, et al. Practical guidelines for the implementation of in vivo dosimetry with diodes in external radiotherapy with photon beams (entrance dose). ESTRO physics for clinical radiotherapy booklet 5. 1st ed. Brussels: European Society for Therapeutic Radiology and Oncology; 2001. 14. Jornet N, Carrasco P, Jurado D, et al. Comparison study of MOSFET detectors and diodes for entrance in vivo dosimetry in 18 MV x-ray beams. Med Phys 2004;31:2534e2542. 15. Kirby TH, Hanson WF, Cates DA. Verification of total body photon irradiation dosimetry techniques. Med Phys 1988;15: 364e369. 16. Hui SK, Das RK, Thomadsen B, et al. CT-based analysis of dose homogeneity in total body irradiation using lateral beam. J Appl Clin Med Phys 2004;5:71e79. 17. Sanchez-Nieto B, Sanchez-Doblado F, Terron JA. A CT-aided PC-based physical treatment planning of TBI: A method for dose calculation. Radiother Oncol 1997;42:77e85. 18. de la Torre N, Figueroa CT, Martinez K, et al. A comparative study of surface dose and dose distribution for intact breast following irradiation with field-in-field technique vs. the use of conventional wedges. Med Dosim 2004;29:109e114.
1648 Onal et al.
International Journal of Radiation Oncology Biology Physics
19. Clift RA, Buckner CD, Appelbaum FR, et al. Allogeneic marrow transplantation in patients with chronic myeloid leukemia in the chronic phase: A randomized trial of two irradiation regimens. Blood 1991;77: 1660e1665. 20. Hui SK, Kapatoes J, Fowler J, et al. Feasibility study of helical tomotherapy for total body or total marrow irradiation. Med Phys 2005;32:3214e3224.
21. Wilkie JR, Tiryaki H, Smith BD, et al. Feasibility study for linac-based intensity modulated total marrow irradiation. Med Phys 2008;35:5609e5618. 22. Aydogan B, Mundt AJ, Roeske JC. Linac-based intensity modulated total marrow irradiation (IM-TMI). Technol Cancer Res Treat 2006;5:513e519. 23. Wheldon TE. The radiobiological basis of total body irradiation. Br J Radiol 1997;70:1204e1207.