Activation of hip prostheses in high energy radiotherapy and resultant dose to nearby tissue

Abstract High energy radiotherapy can produce contaminant neutrons through the photonuclear effect. Patients receiving external beam radiation therapy to the pelvis may have high‐density hip prostheses. Metallic materials such as those in hip prostheses, often have high cross‐sections for neutron interaction. In this study, Thackray (UK) prosthetic hips have been irradiated by 18 MV radiotherapy beams to evaluate the additional dose to patients from the activation products. Hips were irradiated in‐ and out‐of field at various distances from the beam isocenter to assess activation caused in‐field by photo‐activation, and neutron activation which occurs both in and out‐of‐field. NaI(Tl) scintillator detectors were used to measure the subsequent gamma‐ray emissions and their half‐lives. High sensitivity Mg, Cu, P doped LiF thermoluminescence dosimeter chips (TLD‐100H) were used to measure the subsequent dose at the surface of a prosthesis over the 12 h following an in‐field irradiation of 10,000 MU to a hip prosthesis located at the beam isocenter in a water phantom. 53Fe, 56Mn, and 52V were identified within the hip following irradiation by radiotherapy beams. The dose measured at the surface of a prosthesis following irradiation in a water phantom was 0.20 mGy over 12 h. The dose at the surface of prostheses irradiated to 200 MU was below the limit of detection (0.05 mGy) of the TLD100H. Prosthetic hips are activated by incident photons and neutrons in high energy radiotherapy, however, the dose resulting from activation is very small.

Furthermore, patient implants, pacemakers, or prostheses composed of non-biological materials may invite significantly different interactions with incident photon and neutron radiation. Modern radiotherapy typically involves highly conformal photon beams directed toward a target volume. Contaminant neutrons may be scattered by collimators, but are not efficiently absorbed by them. Compared to the photon field, neutrons do not exist as a collimated beam, but as a diffuse fluence incident on the entire patient. 5 It is common practice to avoid directly irradiating metallic prostheses in treatment planning as they are known to perturb the radiation field and reduce the dose to the tissue downstream of the prosthesis. 6 Treatment planning systems do not accurately account for beams traversing metallic implants and estimate the reduction in dose poorly. 7 However, beams are commonly allowed to pass nearby or even through prostheses on exit from the patient.
Prostheses may be activated by photons and/or neutrons in high energy radiotherapy. High energy photons can activate prostheses via the photonuclear effect; causing the emission of neutrons from within the patient. Neutrons produced in the linac can also activate the prostheses, even when planned treatment beams are not directly incident upon them, due to the diffuse nature of the neutron contamination arising from the high frequency of neutron scattering events. These neutrons have mean energies in the vicinity of 0.14 MeV in tissue 8 and, as such, have high cross-sections for capture with the nuclei of hip prostheses. Neutron capture often induces radioactivity in the target nuclei.
In addition to the direct interaction of neutrons with patient tissue, neutrons pose another potential exposure pathway to patients via the production of unstable product nuclei following neutron capture. These radioactive nuclei may emit secondary radiations with half-lives that may be much longer than the beam irradiation times.
We have previously reported on the isotopes, half-lives, and doses resulting from activity produced in components of linacs operated at high energies. 9 In the present work, we extend this to considering the neutron activation of metal objects inside the patient. The decay of activation products may involve radiation emissions that will deposit energy in nearby tissue.
We have investigated the activation of metallic hip prostheses by 18 MV radiotherapy photon beams which are commonly used for these treatments. Prostheses were irradiated in water phantoms in a number of geometries, both in-and out-of direct photon beams.
Gamma-ray spectroscopy was used to identify the isotopes and halflives produced within metallic hip prostheses. The resultant doses at the surface of a prosthesis were measured with high sensitivity TLD100H dosimeters, to address the question of whether these secondary doses might be significant.

| ME TH OD
Prosthetic hips (Thackray, UK) were irradiated in and out-of-field by 18 MV photon beams in a water phantom in different geometries.
The isotopes induced and their half-lives were identified through gamma ray spectroscopy. The dose at the surface of an irradiated prosthesis was also measured using high sensitivity thermoluminescence dosimetry (TLD-100H).

2.A | Irradiation
The four representative irradiation geometries of interest are shown in Fig. 1. The beam entering through the prosthesis (Fig. 1a) should ideally be avoided clinically because the prosthesis directly shadows the target. However, this geometry is sometimes necessary to avoid adjacent organs at risk. It is also of interest as it provides a "worstcase" scenario for activation. The other three geometries are more commonly encountered clinically. The beam exiting through the prosthesis, Fig. 1(c), is identical to the "entry" beam (a) except the beam has been significantly attenuated before entering the prosthesis. Beams (b) and (d) pass laterally to the prosthesis, which is exposed only to out-of-field photons and neutrons scattered from the linac head. The water phantom configurations used to simulate these geometries are shown in Fig. 2

2.B | Gamma-ray spectroscopy
Acquisition of the gamma ray energy spectra was initiated within 1 minute of termination of the beam with a 3 9 3″ NaI(Tl) scintillation detector (Saint-Gobain, France) and an Ortec DigiBASE-E Ethernet multichannel analyser PMT base (Ametek, USA). The spectra were each acquired for 15 min to allow sufficient counts for peak identification.

2.C | Thermoluminescence dosimetry
To induce sufficient activity to produce a measurable dose, one prosthesis was irradiated at isocenter in water by 10,000 MU from

| RESULTS
The gamma-ray spectra acquired following irradiation show the characteristic gamma-ray peaks of 53 Fe, 56 Mn, and 52 V (Fig. 4). The shape of the spectrum at the low energy end is due to Compton scattering and over lapping back scatter peaks. The total count rates measured during spectroscopy of in-field irradiated prostheses were between 6 and 7.5 times higher than those measured from detailed calculations relating activity from one exposure to another cannot be made without accurate energy spectrum data. The activity of those prostheses irradiated in-field was much higher than those irradiated out of the primary photon field. The surface dose rate of a prosthesis irradiated out-of-field would be expected to be reduced by a similar factor, and it would be below the limit of detection of 0.05 mGy.

| CONCLUSION S
Contaminant neutrons in high energy radiotherapy can induce radioactivity in metallic prostheses, even when prostheses are outside the primary field. 53 Fe was found only in prostheses directly irradiated by 18 MV radiotherapy beams. 56 Mn and 52 V were present in prostheses directly irradiated and those exposed between~1 and 10 cm from the edge of the field. 53 Fe was observed only from in-field irradiation and is therefore attributed to the photonuclear effect. Since 56 Mn and 52 V are produced regardless of whether the prosthetic hip was in-or out-of the photon field, they are attributable to neutron activation. The dose measured at the surface of a prosthesis irradiated to 10,000 MU was 0.20 mGy over 12 h (five half-lives of the longest lived isotope produced in the prosthesis).
The surface dose rates for fewer MU and out-of-field irradiations were below the limit of detection for the TLDs. This is very low dose when compared to prescribed radiotherapy doses and even the out-of-field photon dose a patient receives from scatter and leakage radiation.

CONFLI CT OF INTEREST
None.

Photon energy (MeV)
Fe-54 F I G . 6. The cross-section for neutron production by photons with 54 Fe nuclei as a function of photon energy. These data are from the JENDL 4.0 nuclear data library. 16