Cancer risk estimates from radiation therapy for heterotopic ossication prophylaxis after total hip arthroplasty

Michalis Mazonakis and Theoharris Berris

Department of Medical Physics, Faculty of Medicine, University of Crete, P.O. Box 2208, 71003 Iraklion, Crete, Greece

Efrossyni Lyraraki
Department of Radiotherapy and Oncology, University Hospital of Iraklion, 71110 Iraklion, Crete, Greece

John Damilakis
Department of Medical Physics, Faculty of Medicine, University of Crete, P.O. Box 2208, 71003 Iraklion, Crete, Greece

(Received 25 April 2013; revised 13 August 2013; accepted for publication 15 August 2013; published 10 September 2013) Purpose: Heterotopic ossication (HO) is a frequent complication following total hip arthroplasty. This study was conducted to calculate the radiation dose to organs-at-risk and estimate the probability of cancer induction from radiotherapy for HO prophylaxis. Methods: Hip irradiation for HO with a 6 MV photon beam was simulated with the aid of a Monte Carlo model. A realistic humanoid phantom representing an average adult patient was implemented in Monte Carlo environment for dosimetric calculations. The average out-of-eld radiation dose to stomach, liver, lung, prostate, bladder, thyroid, breast, uterus, and ovary was calculated. The organequivalent-dose to colon, that was partly included within the treatment eld, was also determined. Organ dose calculations were carried out using three different eld sizes. The dependence of organ doses upon the block insertion into primary beam for shielding colon and prosthesis was investigated. The lifetime attributable risk for cancer development was estimated using organ, age, and genderspecic risk coefcients. Results: For a typical target dose of 7 Gy, organ doses varied from 1.0 to 741.1 mGy by the eld dimensions and organ location relative to the eld edge. Blocked eld irradiations resulted in a dose range of 1.4146.3 mGy. The most probable detriment from open eld treatment of male patients was colon cancer with a high risk of 564.3 105 to 837.4 105 depending upon the organ dose magnitude and the patients age. The corresponding colon cancer risk for female patients was (372.2541.0) 105 . The probability of bladder cancer development was more than 113.7 105 and 110.3 105 for males and females, respectively. The cancer risk range to other individual organs was reduced to (0.00368.5) 105 . Conclusions: The risk for cancer induction from radiation therapy for HO prophylaxis after total hip arthroplasty varies considerably by the treatment parameters, organ site in respect to treatment volume and patients gender and age. The presented risk estimates may be useful in the follow-up studies of irradiated patients. 2013 American Association of Physicists in Medicine. [http://dx.doi.org/10.1118/1.4820366] Key words: heterotopic ossication, total hip arthroplasty, radiation therapy, organ dose, cancer risk 1. INTRODUCTION The abnormal bone formation in soft tissues, also known as heterotopic ossication (HO), is a frequent complication following hip surgery. The incidence of HO after total hip arthroplasty may vary from 8% to 90% by patient inclusion criteria and individual risk factors.1 Patients with extensive HO after surgery may suffer from pain and a limited range of motion. HO may lead to functional problems and an impaired rehabilitation.2 The use of radiotherapy either alone2 or in combination with nonsteroidal anti-inammatory drugs3 has been considered as an effective treatment option for HO prophylaxis. However, the main concern related to the patients irradiation for a benign disease with megavoltage beams always remains the induction of side effects.
101702-1 Med. Phys. 40 (10), October 2013

Radiotherapy for HO in the hip region might result in temporary sperm abnormalities.4 There is no risk for permanent gonadal failure whereas the risk for hereditary disorders in future generations is low but nontrivial.5 The carcinogenic risk attributable to radiation therapy for the prevention of HO still remains questionable. Seegenchmidt et al.1 referred the paucity of these complications. The above data might be attributed to the low radiation doses of 68 Gy delivered to the target site and to the limited follow-up of these patients. The BEIR-VII report6 proposed a minimal latent period of 5 years for radiation-induced cancer, whereas Welte et al.7 found a median latency of 7.4 years for the appearance of second malignancies related to radiotherapy. These latent periods exceed the typical patient follow-up of less than 18 months as summarized in a meta-analysis of seven randomized trials related
2013 Am. Assoc. Phys. Med. 101702-1

0094-2405/2013/40(10)/101702/8/$30.00

101702-2

Mazonakis et al.: Cancer risk from radiotherapy for heterotopic ossication prophylaxis

101702-2

to HO after hip procedures.8 Radiotherapy-induced sarcomas have been recently observed 11 and 15 years after hip irradiation for HO prophylaxis.9, 10 According to the review of Balboni et al.,11 the probability of cancer development from prophylactic treatment of HO needs to be investigated in a large number of patients followed for a long interval after radiation exposure. The current study was conducted (a) to calculate the absorbed dose to radiosensitive organs from hip radiotherapy for the prevention of HO and (b) to estimate the projected lifetime attributable risk (LAR) for cancer induction using organ-, age, and sex-specic risk coefcients.

2. METHODS
2.A. Monte Carlo simulations

The MCNP radiation transport code has been previously employed to simulate the head of a medical linear accelerator (Philips/Elekta, SL75, The Netherlands) generating 6 MV x-rays.12 The model consisted of the main beam modifying parts such as the bremstrahlung target, primary and secondary collimators, attening lter and attening lter holder. All structural shielding components of the head were also simulated by the model. The simulation of the linear accelerator was performed in two steps. In the rst step, the spectrum of the simulated 6 MV photon beam was determined by placing F2 type tallies under the attening lter holder. These tallies were capable of registering the relative uence of photons through a surface. The required energy binning was added for the accurate registration of the photon spectrum. In the second step, the spectrum tallied below the attening holder was emitted from a point source and passed through secondary collimators. These simulated collimators were employed to shape the emitted beam into rectangular elds. In-eld and out-of-eld radiation doses calculated by the Monte Carlo model were found to be in a very good agreement with direct dose measurements performed in a water tank (RFA-300, Scanditronix Wellhofer, Upasla, Sweden) and in a Rando anthropomorphic phantom (Alderson Research Labs, Stanford, CA), respectively.12 An androgynous mathematical phantom was produced by a commercially available software package (Body Builder, White Rock Science, Los Alamos, New Mexico) and implemented into the Monte Carlo environment. The produced phantom was based on the descriptions obtained by Cristy and Eckerman report.13 The phantom may represent a typical adult patient with a height of 179 cm and a weight of 73.5 kg. The phantom has been modied to include the missing red bone marrow tissue that was not present in the original phantom structure. The distribution of the red bone marrow in the phantoms bony regions was dened based on previously reported data.13 The bone marrow composition was considered to be the same with that of the soft tissue. An extra modication was also made to represent the prostate gland in the humanoid phantom. The missing gland was simulated by a sphere of 1.8 cm radius residing in the pelvic region between the urinary bladder and rectum.
Medical Physics, Vol. 40, No. 10, October 2013

The linear accelerator model together with the modied humanoid phantom was employed to simulate radiotherapy in the hip region. Two parallel opposed anteroposterior (AP) and posteroanterior (PA) treatment elds (gantry angles: 0o and 180o ) were simulated by the Monte Carlo code. The collimator and couch angles were set to 0o . The aforementioned irradiation technique is currently applied for HO prophylaxis.3 The following eld sizes: 12 14, 13 16, and 14 18 cm2 were used for each beam orientation. These eld sizes represented the small, medium, and large dimensions that may be employed in clinical practice. A midplane cross section of the phantom with a superimposed AP beam is shown in Fig. 1. Additional simulations were made by modeling cerrobend blocks placed on a lucite tray to protect the region of prosthesis used in total hip arhroplasty and the part of the colon which is included in the primary irradiated area. Simulations with blocks were carried out for the typical medium eld size. The blocked area was equal to 39.9 cm2 . All treatment parameters were dened with the aid of a radiotherapist experienced in the management of benign diseases.
2.B. Organ dose-cancer risk calculations

For each AP and PA simulated eld treatment, organ dose calculations were performed by positioning 26 different F6 tallies within the androgynous humanoid phantom. Each F6 tally was assigned to a geometrical cell representing either the whole organ-at-risk or a part of the organ. The F6 tally calculated the average collision kerma at each geometrical cell. The radiation dose was determined for all organs having a specic risk coefcient for the development of malignant diseases as proposed by the BEIR-VII report.6 The organs-at-risk were divided in two categories depending upon their site in respect to the treatment volume. The rst category consisted of the following organs: stomach, liver, lung, prostate, bladder, thyroid, female breast, uterus, and ovary. The above organs were excluded from the primary irradiated area and, therefore, they were exposed to low doses due to scattered radiation and head leakage. The average out-of-eld dose in mGy was calculated for each organ-at-risk. In accordance with the BEIR-VII report,6 a linear relationship between radiation dose and solid cancer risk was assumed even at very low doses. This linearity is valid for doses of approximately up to 2.5 Sv. The appropriate organspecic risk factors6 together with the calculated average organ doses were employed to estimate the risk of cancer induction for individual out-of-eld organs.14, 15 The patients gender and age were taken into account for cancer risk assessment. The second category referred to colon that partly included within the treatment volume. A portion of the colon received primary radiation. The radiation dose to this portion was the same with the target dose reaching a value of 7 Gy. For such high doses inside the beam ports, the use of the linearnonthreshold model is incompatible.16, 17 The average organ dose should not be used for cancer risk assessment from radiotherapy, when the organ is partly exposed to primary radiation and receives an inhomogeneous dose distribution.18 The

101702-3

Mazonakis et al.: Cancer risk from radiotherapy for heterotopic ossication prophylaxis

101702-3

F IG . 1. A midplane cross section view of the androgynous phantom used for Monte Carlo simulations. An anteroposterior radiation eld of medium size is presented with the dotted line. The following organs-at-risk and structures are marked with the corresponding numbers: sigmoid colon-1, descending colon-2, prostate-3, uterus-4, ovaries-5, small intestine-6, ascending colon-7, arm bones-8, pelvic bones-9, leg muscles-10, leg bones-11, target site-12, and skin-13.

organ equivalent dose (OED), as introduced by the model of Schneider et al.,18 was adopted in this study. The above model suggests that any dosimetric distribution within an organ of interest is equivalent and corresponds to the same OED when it results in the same radiation-induced cancer incidence.18, 19 The model has already been used to assess the second cancer risk for organs inside the primary radiation eld.2022 The OED to colon was determined using the following formula: OED = 1 N
N

Di exp(aDi ),
i =1

where N is the total number of dose calculation points, Di is the dose at a point i, and a is the colon-specic cell sterilization parameter.18 The calculated OED value was used to assess the risk for colon cancer induction. Moreover, the risk for the development of any other cancer was estimated. The dose to colon may be used for the residual category entitled other cancer.6 In our study concerning hip irradiation, the other cancer solely referred to organs/tissues located outside the treatment eld. Hence, the risk assessment of other cancer was approximated by the average radiation dose to colon parts (denoted as out-ofeld colon dose) that were entirely outside the treatment eld. The out-of-eld colon dose was combined with the proper risk factor to nd the probability for other cancer induction. The lifetime attributable risk (LAR) of site-specic solid cancer incidence associated with hip radiotherapy was assessed using the framework introduced by the BEIR-VII report.6 This report has proposed the appropriate organ-, gender-, and age- specic risk coefcients for the LAR
Medical Physics, Vol. 40, No. 10, October 2013

assessment. The projected LAR for any individual refers to the probability of developing a radiation-induced cancer at any time subsequent to the age at postoperative radiation therapy. For each organ-at-risk, the LAR of cancer induction was estimated for 50- and 60-year-old female and male patients. These patient ages were selected based on a meta-analysis8 including seven studies and 1143 patients who underwent prophylactic treatment for the prevention of HO after total hip arthroplasty. The mean age of the above patients ranged from 43 to 66 years.8 All organ doses and cancer risk estimates presented here correspond to a target dose of 7 Gy which is currently given during radiotherapy for HO prophylaxis.2 The radiation dose to the target from each eld was considered to be equal to 3.5 Gy. Organ dose calculations were limited by the uncertainty associated with Monte Carlo experiments. A sufcient number of source electrons was used during radiation therapy simulations in order to ensure statistical errors below 7%.12 The organ dose values corresponded to a typical adult patient undergoing hip irradiation. Possible deviations from these typical somatometric characteristics might result in unknown errors in the prediction of the organ dose. 3. RESULTS The mean radiation dose received by the out-of-eld organs and the OED to colon from radiotherapy for HO are summarized in Table I. For a 7-Gy target dose, the radiation dose range to organs excluded from the treatment volume was 1.0172.3, 1.1218.2, and 1.3302.9 mGy for irradiation with small, medium, and large eld sizes, respectively. The OED to colon varied from 600.3 to 741.1 mGy. The organ dose

101702-4

Mazonakis et al.: Cancer risk from radiotherapy for heterotopic ossication prophylaxis

101702-4

TABLE I. Organ doses from radiation therapy for heterotopic ossication prophylaxis with a target dose of 7 Gy given by equally weighted anteroposterior and posteroanterior elds. Organ dose (mGy) Small eld size: 12 14 cm2 Organ Colon Stomach Liver Lung Prostate Bladder Thyroid Breast Uterus Ovary Out-of-eld colon AP eld 307.2 8.1 3.1 0.5 73.3 71.3 0.6 1.5 38.9 49.3 3.9 PA eld 293.1 6.9 2.8 0.5 73.3 101.0 0.4 1.4 43.2 49.5 3.9 Total 600.3 15.0 5.9 1.0 146.6 172.3 1.0 2.9 82.1 98.8 7.8 Medium eld size: 13 16 cm2 AP eld 325.7 9.1 3.5 0.5 87.8 89.9 0.8 1.8 50.0 68.0 4.7 PA eld 323.8 8.1 3.3 0.6 88.0 128.3 0.5 1.7 55.6 68.2 4.8 Total 649.5 17.2 6.8 1.1 175.8 218.2 1.3 3.5 105.6 136.2 9.5 Large eld size: 14 18 cm2 AP eld 371.9 10.3 4.0 0.6 103.7 111.0 0.8 2.2 64.3 95.9 5.6 PA eld 369.2 9.4 3.7 0.7 103.8 191.9 0.6 1.8 72.7 95.9 5.8 Total 741.1 19.7 7.7 1.3 207.5 302.9 1.4 4.0 137.0 191.8 11.4

calculations from prophylactic radiotherapy with blocked elds of medium size are presented in Table II. The block insertion into the primary beam resulted in a dose range to out-of-eld organs of 1.4105.7 mGy. The OED to colon was equal to 146.3 mGy and, therefore, it was reduced by more than four times compared with the respective OED from open eld treatment. The LAR for incidence of malignant diseases associated with open eld irradiation of male patients is presented in Fig. 2. For specic organs, the cancer risk range from radiotherapy at the age of 50 years old was (1.0837.4) 105 depending upon the organ of interest and the eld size employed. The corresponding range at the age of 60 years old was reduced to (0.8696.6) 105 . The risk for thyroid cancer induction (not shown in Fig. 2) at both patient ages was much smaller and varied from 0.003 105 to 0.01 105 . The risk range for other cancer from radiation therapy at the age of 50 years was (10.916.0) 105 . The corresponding range at the advanced patient age was (7.611.2) 105 .

The LAR for cancer development in female patients is shown in Fig. 3. For 50-year-old females, the cancer risk estimates for individual organs varied from 0.5 105 to 541.0 105 by the organ of interest and the eld size employed. The risk for other cancer was (11.516.9) 105 . Radiotherapy of female patients at the advanced age of 60 years old resulted in a cancer risk range for specic organs of (0.4459.5) 105 . The probability for the development of other malignancies was (8.512.4) 105 . The incidence of thyroid cancer (not shown in Fig. 3) in females following radiation therapy was (0.010.06) 105 .

4. DISCUSSION Limited data have been published about the radiation dose and cancer risk from radiotherapy for HO prophylaxis after total hip arthroplasty. Patel et al.4 focused only on the radiation dose to testes inside and outside testicular shields. Jansen et al.23 provided dose calculations for several organs-at-risk. Their results related to the ovarian and bladder exposure were similar with those presented here. The rest of their results referred to different organs or organ parts than those examined in our study. The risk for carcinogenesis associated with radiation prophylaxis for HO has been previously assessed using the effective dose quantity as derived by a Monte Carlo model.23 This simplied model was based on the use of energy spectra that are readily available from the literature, and no information was given about the accuracy of the obtained dosimetric calculations. The International Commission on Radiological Protection (ICRP) has introduced the concept of the effective dose which is calculated using the appropriate organ-specic weighting factors.24 These factors are age- and sex-averaged.24 However, the patient population undergoing radiotherapy after total hip arthroplasty belongs to an old age group making the use of the weighting factors to be questionable. For exposures involving high radiation doses, the ICRP has advised against the use of the effective dose and

TABLE II. Organ doses from radiation therapy for heterotopic ossication prophylaxis with a target dose of 7 Gy given by equally weighted anteroposterior and posteroanterior blocked elds of size 13 16 cm2 . Organ dose (mGy) Organ Colon Stomach Liver Lung Prostate Bladder Thyroid Breast Uterus Ovary AP eld 73.1 8.1 2.7 0.7 52.6 48.0 1.2 2.9 30.8 35.2 PA eld 73.2 6.8 2.4 0.7 53.1 58.1 0.8 1.8 31.8 34.8 Total 146.3 14.9 5.1 1.4 105.7 104.4 2.0 4.7 62.6 70.0

Medical Physics, Vol. 40, No. 10, October 2013

101702-5

Mazonakis et al.: Cancer risk from radiotherapy for heterotopic ossication prophylaxis

101702-5

Stomach

Colon

Liver 12x14 cm2 Lung 13x16 cm2 14x18 cm2 Prostate

Bladder

Other cancer 0 1 2 3 4 5 5 20 35 50 65 80 80 240 400 560 720 880

LAR (x10 -5)

(a)

Stomach

Colon

Liver 12x14 cm2 Lung 13x16 cm2 14x18 cm2 Prostate

Bladder

Other cancer

5 5

20

35

50
-5

65

80 80 240 400 560 720 880

LAR (x10 )

(b)
F IG . 2. Cancer risk estimates for (a) 50- and (b) 60-year-old male patients undergoing open eld radiation therapy for heterotopic ossication prophylaxis.

recommended that the absorbed doses in organs and tissues have to be employed for risk evaluation.24 Based on the above data, previous studies pointed out that the adoption of the effective dose method and the subsequent assessment of a whole-body cancer risk should be avoided in radiotherapy.16, 25, 26 Trott and Kamprad26 reported that the effective dose may overestimate the real risk for cancer induction attributable to radiotherapy of benign diseases up to one order of magnitude. To overcome the difculties associated with the effective dose concept in this study, the probability for cancer development to individual organs was estimated
Medical Physics, Vol. 40, No. 10, October 2013

using the absorbed dose to each organ of interest and the appropriate organ-, sex-, and age- specic risk coefcients. Monte Carlo simulations showed that the organ dose magnitude is mainly affected by the organ position relative to the eld edge. The highest organ dose was observed for colon that was partly included within the treatment volume. As expected, the radiation dose to out-of-eld organs located above the diaphragm was lower than that calculated for organs in the abdominopelvic region. The urinary bladder was the outof-eld organ receiving the largest amounts of scattered radiation. The increase of eld size resulted in an elevation of the

101702-6

Mazonakis et al.: Cancer risk from radiotherapy for heterotopic ossication prophylaxis

101702-6

Stomach Colon Liver Lung Breast Uterus Ovary Bladder Other cancer 0 1 2 3 4 5 5 20 35 50 65 80 80 240 400 560 720 880

12x14 cm2 13x16 cm2 14x18 cm2

LAR (x10 -5)

(a)

Stomach Colon Liver Lung 12x14 cm2 Breast Uterus Ovary Bladder Other cancer 0 1 2 3 4 5 5 20 35 50
-5

13x16 cm2 14x18 cm2

65

80 80 240 400 560 720 880

LAR (x10 )

(b)
F IG . 3. Cancer risk estimates for (a) 50- and (b) 60-year-old female patients undergoing open eld radiation therapy for heterotopic ossication prophylaxis.

organ doses. For irradiation with the large eld size instead of the standard medium one, the organ dose increase was up to 1.4 times. This dose increase needs to be taken into account during the denition of the margins of the planning-tumorvolume. The risk for radiotherapy-induced malignant diseases for male and female patients was estimated in this study based on organ-specic data. The age progression led to a risk reduction due to the decreased risk coefcients at the advanced patient age.6 The appearance of bladder and colon cancer were the most probable carcinogenic effects after open eld
Medical Physics, Vol. 40, No. 10, October 2013

radiotherapy of females with risk ranges of (110.3224.1) 105 and (372.2541.0) 105 , respectively, depending upon the patient age and the eld dimensions employed. The corresponding ranges for male patients were (113.7230.2) 105 and (564.3837.4) 105 , respectively. The above probability for bladder cancer induction is much lower than the risk value of 1.3% associated with intensity-modulated radiation therapy (IMRT) for localized prostate cancer.27 The management of prostate carcinoma with IMRT may lead to a risk for the development of in-eld colorectal and out-of-eld colon malignant diseases of 0.6% and 1.0%, respectively.27

101702-7

Mazonakis et al.: Cancer risk from radiotherapy for heterotopic ossication prophylaxis

101702-7

The risk for the induction of pelvic malignancies, including prostate, uterus and ovarian cancer from radiotherapy for HO prophylaxis was less than 68.5 105 . The risk for stomach, liver, lung, breast, and thyroid cancer was reduced to (0.0036.3) 105 . The probability for the appearance of any other malignant disease was also found to be low. The previously mentioned nontrivial risk for pelvic malignancies implies the need for long-term follow-up of patients undergoing treatment for HO prevention. The block introduction reduced the radiation dose to colon due to the interception of primary radiation. Monte Carlo simulations showed that the radiation dose to infradiaphragmatic out-of-eld organs decreased with blocked eld irradiations in comparison to that with open portals, whereas the dose to supradiaphragmatic organs increased. It is well known that the components of the out-of-eld dose are head leakage, internal scatter generated within the phantom and external scatter produced by the machine head and beam modiers. The major contributor of the out-of-eld dose close to the treatment volume is the internal scatter whereas the sum of the two other components overweighs in the far periphery.28, 29 Stovall et al.30 reported that internal scatter and head leakage are approximately equal at distances of about 30 cm from the eld edge, and beyond that point, head leakage becomes the dominant component of the peripheral dose. The use of blocks reduced treatment volume leading to a reduction of the phantom scatter which mainly affected abdominoplevic organs located in the periphery of the irradiated area. The additional scattered radiation from the tray and blocks increased the external scatter resulting in slightly elevated doses for breast, thyroid, and lung. The presented organ dose calculations and cancer risk assessments solely refer to hip radiotherapy with a 6 MV photon beam. This limitation should be attributed to the inability of the linear accelerator of our department to generate different photon beam energies. Radiation therapy with 6 MV x-rays has been widely employed for the prevention of HO.3, 4, 9, 31 Higher photon energies might also be used during prophylactic irradiation especially for thick patients.4 Nasr et al.32 have reported that postoperative irradiation with 18 MV x-rays is effective in preventing HO. Radiotherapy with an 18 MV photon beam may result in lower out-of-eld doses than those associated with treatment using 6 MV x-rays.30, 33, 34 Moreover, the current dosimetric data were derived by using a mathematical phantom simulating a normal sized adult patient. Further research is required to compare these data with Monte Carlo calculations obtained with the aid of voxelized patient models. The lifetime cancer risk assessments of this study might contain large uncertainties associated with the applied model6 and the risk coefcients proposed by this model. Regarding the low dose region, the BEIR-VII report6 assumed that the linear-no-threshold model should be extended to radiation doses of less than 0.1 Gy. However, a deviation from linearity might exist in the aforementioned dose range, as previously reported,16, 35 leading to different cancer risk values than those obtained by the linear dose-response curve. It should also be noticed that the formalism for OED calculation has been
Medical Physics, Vol. 40, No. 10, October 2013

derived by data from irradiated Hodgkins patients.18 The well-known genetic susceptibility to Hodgkins disease might affect the development of other malignant diseases and might probably lead to conservative risk assessments.21 5. CONCLUSIONS The current study provides analytical Monte Carlo calculations for the radiation dose to organs-at-risk attributable to hip radiotherapy for HO prophylaxis. Patient irradiation with a 6 MV photon beam may result in an increased risk for the induction of colon and bladder cancer. The risk for the development of any other tumor is low or even negligible. The current risk assessments may be of value for radiotherapists and clinicians in the follow-up studies of irradiated patients.
H. Seegenschmiedt, O. Micke, and R. Heyd, Heterotopic ossications: General survey for all sites, in Radiotherapy for Non-Malignant Disorders, edited by M. H. Seegenschmiedt, H. B. Makoski, K. R. Trott, and L. W. Brady (Springer, Berlin, Heidelberg, 2008), pp. 333355. 2 M. H. Seegenschmiedt, Radiotherapy of non-malignant diseases, in Principles and Practice of Radiation Oncology, 5th ed., edited by E. C. Halperin, C. A. Perez, and L. W. Brady (Lippincott Williams & Wilkins, Philadelphia, 2008), pp. 19331958. 3 E. E. Pakos, K. S. Stadas, P. G. Tsekeris, A. N. Politis, G. Mitsionis, and T. A. Xenakis, Combined radiotherapy and indomethacin for the prevention of heterotopic ossication after total hip arthroplasty, Strahlenther. Onkol. 185, 500505 (2009). 4 H. Patel, C. L. Silverman, L. Carrascosa, A. Malkani, and C. M. Yashar, Evaluation of scrotal and testicular radiation doses for heterotopic ossication prophylaxis, Am. J. Orthop. 37, E163E166 (2008). 5 G. Kokona, M. Mazonakis, H. Varveris, E. Liraraki, and J. Damilakis, Treatment of benign diseases with megavoltage X-ray beams: Is there a risk for gonadal damage? Clin. Oncol. 18, 658662 (2006). 6 BEIR, Health Risks from Exposure to Low Levels of Ionizing Radiation: BEIR VII, Phase 2 (National Academy of Science, Washington, DC, 2006). 7 B. Welte, P. Suhr, D. Bottke, D. Bartkowiak, W. Dorr, K. R. Trott, and T. Wiegel, Second malignancies in high-dose areas of previous tumor radiotherapy, Strahlenther. Onkol. 186, 174179 (2010). 8 E. E. Pakos and J. P. A. Ioannidis, Radiotherapy vs nonsteroidal antiinammatory drugs for the prevention of heterotopic ossication after major hip procedures: A meta-analysis of randomized trials, Int. J. Radiat. Oncol., Biol., Phys. 60, 888895 (2004). 9 M. K. Farris, V. K. Chowdhry, S. Lemke, M. Kilpatrick, and M. Lacombe, Osteosarcoma following single fraction radiation prophylaxis for heterotopic ossication, Radiat. Oncol. 7, 140 (2012). 10 W. F. Mourad, S. Packianathan, and R. A. Shourbaji, Radiation-induced sarcoma following radiation prophylaxis of heterotopic ossication, Pract. Radiat. Oncol. 2, 151154 (2012). 11 T. A. Balboni, R. Gobezie, and H. J. Mamon, Heterotopic ossication: Pathophysiology, clinical features, and the role of radiotherapy for prophylaxis, Int. J. Radiat. Oncol., Biol., Phys. 65, 12891299 (2006). 12 T. Berris, M. Mazonakis, J. Stratakis, A. Tzedakis, A. Fasoulaki, and J. Damilakis, Calculation of organ doses from breast cancer radiotherapy: A Monte Carlo study, J. Appl. Clin. Med. Phys. 14, 133146 (2013). 13 M. Cristy and K. F. Eckerman, Specic absorbed fractions of energy at various ages from internal photon sources, ORNL Report No. TM-8381/Vols. VI and VII (Oak Ridge National Laboratory, Oak Ridge, TN, 1987). 14 B. Bednarz, B. Athar, and X. G. Xu, A comparative study on the risk of second primary cancers in out-of-eld organs associated with radiotherapy of localized prostate carcinoma using Monte Carlo-based accelerator and patient models, Med. Phys. 37, 19871994 (2010). 15 E. M. Donovan, H. James, M. Bonora, J. R. Yarnold, and P. M. Evans, Second cancer incidence risk estimates using BEIR VII models for standard and complex external beam radiotherapy for early breast cancer, Med. Phys. 39, 58145824 (2012).
1 M.

101702-8
16 X.

Mazonakis et al.: Cancer risk from radiotherapy for heterotopic ossication prophylaxis
26 K.

101702-8

G. Xu, B. Bednarz, and H. Paganetti, A review of dosimetry studies on external-beam radiation treatment with respect to second cancer induction, Phys. Med. Biol. 53, R193R241 (2008). 17 R. A. Halg, J. Besserer, and U. Schneider, Systematic measurements of whole-body dose distributions for various treatment machines and delivery techniques in radiation therapy, Med. Phys. 39, 76627676 (2012). 18 U. Schneider, D. Zwahlen, D. Ross, and B. Kaser-Hotz, Estimation of radiation-induced cancer from three-dimensional dose distributions: Concept of organ equivalent dose, Int. J. Radiat. Oncol., Biol., Phys. 61, 1510 1515 (2005). 19 U. Schneider and B. Kaser-Hotz, Radiation risk estimates after radiotherapy: Application of the organ equivalent dose concept to plateau doseresponse relationships, Radiat. Environ. Biophys. 44, 235239 (2005). 20 M. Yoon, S. H. Ahn, J. Kim, D. H. Shin, S. Y. Park, S. B. Lee, K. H. Shin, and K. H. Cho, Radiation-induced cancers from modern radiotherapy techniques: Intensity-modulated radiotherapy versus proton therapy, Int. J. Radiat. Oncol., Biol., Phys. 77, 14771485 (2010). 21 H. Paganetti, B. S. Athar, M. Moteabbed, J. A. Adams, U. Schneider, and T. I. Yock, Assessment of radiation-induced second cancer risks in proton therapy and IMRT for organs inside the primary radiation eld, Phys. Med. Biol. 57, 60476061 (2012). 22 C. B. Simone, K. Kramer, W. P. OMeara, J. E. Bekelman, A. Belard, J. McDonough, and J. OConnell, Predicted rates of secondary malignancies from proton versus photon radiation therapy for stage I seminoma, Int. J. Radiat. Oncol., Biol., Phys. 82, 242249 (2012). 23 J. T. M. Jansen, J. J. Broerse, J. Zoetelief, C. Klein, and M. H. Seegenschmiedt, Estimation of the carcinogenic risk of radiotherapy of benign diseases from shoulder to heel, Radiother. Oncol. 76, 270277 (2005). 24 International Commission on Radiological Protection, The 2007 recommendations of the International Commission on Radiological Protection. ICRP publication 103, Ann. ICRP 37, 1332 (2007). 25 R. M. Harrison, Second cancers following radiotherapy: A suggested common dosimetry framework for therapeutic and concomitant exposures, Br. J. Radiol. 77, 986990 (2004).

R. Trott and F. Kamprad, Estimation of cancer risks from radiotherapy of benign diseases, Strahlenther. Onkol. 182, 431436 (2006). 27 M. J. Zelefsky, D. M. Housman, X. Pei, Z. Alicikus, J. M. Magsanoc, L. T. Dauer, J. St. Germain, Y. Yamada, M. Kollmeier, B. Cox, and Z. Zhang, Incidence of secondary cancer development after high-dose intensity-modulated radiotherapy and image-guided brachytherapy for the treatment of localized prostate cancer, Int. J. Radiat. Oncol., Biol., Phys. 83, 953959 (2012). 28 M. Mazonakis, E. Lyraraki, C. Varveris, E. Samara, K. Zourari, and J. Damilakis, Conceptus dose from involved-eld radiotherapy for Hodgkins lymphoma on a linear accelerator equipped with MLCs, Strahlenther. Onkol. 185, 355363 (2009). 29 N. Chofor, D. Harder, K. C. Willborn, and B. Poppe, Internal scatter, the unavoidable major component of the peripheral dose in photon-beam radiotherapy, Phys. Med. Biol. 57, 17331743 (2012). 30 M. Stovall, C. R. Blackwell, J. Cundiff, D. H. Novack, J. R. Palta, L. K. Wagner, E. W. Webster, and R. J. Shalek, Fetal dose from radiotherapy with photon beams: Report of AAPM Radiation Therapy Committee Task Group No. 36, Med. Phys. 22, 6382 (1995). 31 E. E. Pakos, E. J. Pitouli, P. G. Tsekeris, V. Papathanasopoulou, K. Stalas, and T. H. Xenakis, Prevention of heterotopic ossication in high-risk patients with total hip arthroplasty: The experience of a combined therapeutic protocol, Int. Orthop. 30, 7983 (2006). 32 E. Nasr, R. Nehme, I. Ghanem, F. Azoury, D. N. Nasr, and F. Dagher, Role of radiotherapy in heterotopic ossication, Cancer Radiother. 13, 4246 (2009). 33 S. Mutic, J. Esthappan, and E. E. Klein, Peripheral dose distributions for a linear accelerator equipped with a secondary multileaf collimator and universal wedge, J. Appl. Clin. Med. Phys. 3, 302309 (2002). 34 M. Mazonakis, F. Zacharopoulou, H. Varveris, and J. Damilakis, Peripheral dose measurements for 6 and 18 MV photon beams on a linear accelerator with multileaf collimator, Med. Phys. 35, 43964403 (2008). 35 E. J. Hall, Intensity-modulated radiation therapy, protons, and the risk of second cancers, Int. J. Radiat. Oncol., Biol., Phys. 65, 17 (2006).

Medical Physics, Vol. 40, No. 10, October 2013