A technique for total skin electron therapy (TSET) of an anesthetized pediatric patient

Abstract Purpose Total skin electron therapy (TSET) is a technique to treat cutaneous lymphomas. While TSET is rarely required in pediatric patients, it poses particular problems for the delivery. It was the aim of the present work to develop a method to deliver TSET to young children requiring anesthetics during treatment. Methods A customized cradle with a thin window base and Poly(methyl‐methacrylate) (PMMA) frame was built and the patient was treated in supine position. Two times six fields of 6 MeV electrons spaced by 60° gantry angles were used without electron applicator and a field size of 36 × 36 cm2. The two sets of six fields were matched at approximately 65% surface dose by rotating the patient around an axis 30 cm distance from beam central axis, effectively displacing the two sets of fields in sup/inf direction by 60 cm. Electron energy was degraded using a 12 mm PMMA block on the gantry. Focus to skin distance was maximized by displacing the patient in opposite direction of the beam resulting in a different couch position for each field. Results A 2‐yr‐old patient was treated in 12 fractions of 1.5 Gy over 2.4 weeks. Dose to skin was verified daily using thermoluminescence dosimetry and/or radiochromic film. The treatment parameters were adjusted slightly based on in vivo dosimetry resulting in a dose distribution for most of the treatment volume within ±20% of the prescribed dose. Six areas were boosted using conventional electron therapy. Conclusion TSET can be delivered to pediatric patients using a customized couch top on a conventional linear accelerator.

Irradiation of large parts or all of a patient's skin is technologically challenging and several techniques have been developed over the years. 3,4 However, modern techniques such as intensity modulated radiation therapy (IMRT) and image guidance had very little impact on the development of TSET techniques and the report of the American Association of Physicists in Medicine (AAPM) task group 30 of 1988 is still widely used as a guidance. 5 Most treatment techniques rely on extended source to skin distance (SSD) to cover larger areas of the patient. At a distance of about 2 m from isocentre, it is possible to cover the whole length of a patient standing up using two angled radiation beams (up and down matched at about 65% dose). To cover the whole circumference of the patient, the patient is rotated continuously 4,6 or in typically six increments ("Stanford" technique). 7 Electron energy used varies between 4 and 10 MeV in the literatureonce oblique incidence and distance are taken into account this results in the desired depth of usually between 5 and 10 mm receiving 80% of the radiation dose. It is also accepted that not all parts of the body will equally be irradiated and in vivo dosimetry measurements are commonly employed to assess which parts of the anatomy need boosting and by what dose. 8 Mycosis fungoides occurs mainly in adults above the age of 20.
As such reports on techniques used for TSET in children are rare and mostly confined to case reports as a recent review by S. Malgorzata shows. 9 The review identified seven reported cases in the literature based on several reports. [10][11][12][13][14] Techniques used included treatment on the floor and the design of a specialized frame holding the child in an upright position at a distance. 12 We are reporting a different technique that utilizes lateral and vertical couch movement to maximize the SSD for the beam in each direction by moving the couch position away from the radiation source. Combined with a custom-made thin tabletop this allows the patient to be treated in supine position for the whole of the treatment which was considered essential for safe anesthetics.

2.A | Patient and prescription
The patient was a 2-yr-old female to be treated with total skin electron irradiation. The patient was approximately 90 cm tall with aver- Areas of low dose were to be identified using in vivo dosimetry and boosted using conventional electron irradiation. Figure 1 shows the technique that was developed. The patient was positioned on a customized thin window Mylar top (thickness 0.3 mm) inserted in a Varian treatment couch. The couch top can be seen in Fig. 2. Six beam directions were chosen with gantry angles 60°apart as indicated in Fig. 1(b). In order to maximize source to skin distance (SSD) and therefore field size, the couch was moved in the direction opposite of the radiation beam. The maximal lateral motion of the couch was 20.8 cm which limited the displacement possible. In order to facilitate similar distances to patient central axis, couch height and lateral displacement were calculated using trigonometry taking the patient ant/post separation of 10 cm into consideration. Table 1 lists couch positions used relative to isocentre with the tan(30) × 20 cm = 11.54 rounded up to 12 cm.

2.B | Technique
After delivery of the first six fields to the upper part of the body, the patient was rotated and the lower half treated. Beam central axis location was verified by light field and the junction was marked on the patient's skin daily. The effective distance from the isocentre to the junction between the two sets of six beams was 30 cm, which was approximately at 65% dose in the electron beam profile of both beams assessed at 114 cm SSD and 1 mm depth under 30°oblique incidence reflecting "average" dose delivery conditions. The location of the junction was also informed by profiles acquired at perpendicular incidence and consideration of depth dose at different locations.

2.C | Commissioning and verification measurements
Once the technique had been agreed upon, measurements were performed using several measurement setups and dosimetric equipment: 1. Output and depth dose were assessed using a PPC05 plane parallel chamber (IBA dosimetry, Schwarzenbrueck) and a thin window advanced Markus chamber (Exradin A10, Standard Imaging, Middleton, WI) in a solid water slab phantom.

2.
A cylindrical PMMA phantom with density 1.16 g/cm 3 , length 20 cm and diameter 16 cm (shown in Fig. 2) was used for dose assessment around the circumference of the patient. This phantom was readily available as it is also used for assessment of Computed Tomography Dose Index in radiological procedures.
Thermoluminescence dosimetry (TLD) using LiF:Mg,Cu,P was employed for the measurements. 15,16 TLD chips were individually calibrated and read out using an automatic TLD reader (Harshaw 5500, Thermo Fisher Scientific, Waltham, MA, USA). Standards were exposed using the same nominal energy as the patient irradiation (6 MeV electrons, exposed at depth of maximum dose). view from the feet: The isocentre of the linac is shown for two of the six couch positions for two of the six electron fields in each of the patient's position. The couch shifts were chosen so the centre of the patient assumed to be 5 cm above the couch top is always at a distance of 124 cm to the radiation source. For the posterior beams, the couch rails were adjusted to minimize their impact on dose; (c) anterior view: the position of PMMA hand shielding which was applied for parts of the treatment to improve dose homogeneity is indicated. Eyes were shielded for the whole treatment; toenail shielding was used from the third fraction onwards while fingernail shielding was added after fraction 8. Except for the hands, all shielding was manufactured in house and made from lead.
Other aspects of the measurement method are described in more detail by Lonski et al. 17,18 3. The head of an anthropomorphic phantom (ATOM dosimetry verification phantom, Computerized Imaging Reference Systems, CIRS, Norfolk Virginia) was used for radiochromic film measurements to assess dose in contour areas. The adult head phantom has similar dimensions to the body of a 2-yr-old child (20 × 15 cm 2 cross section). Radiochromic film (EBT3, Ashland, Bridgewater, NJ, USA) was cut to fit into the phantom and exposed using approximately three times the dose to be delivered per fraction (150 MU per beam) to the patient to ensure relatively low doses could be assessed. Calibration films were exposed using the same nominal energy (6 MeV electrons at depth of maximum dose). The film was read using a flat bed scanner (EPSON 700) with 72 dpi resolution. The red channel with 16 bit depth was used for dosimetry. 19

2.D | In vivo dosimetry
In vivo dosimetry was performed for most fractions using TLD and radiochromic film measurements using the same methods as described above. Figure 3 illustrates the measurement locations used. In addition to the point dose measurements indicated in the figure, rows of TLDs and strips of radiochromic film were used to identify dose across the junction of the two sets of fields and in locations where the edge of a boost field needed to be identified.   T A B L E 1 Initial couch positions relative to isocentre (0/0/0) for the six beam directions used in the work. The beam numbers are identical to the ones shown in Fig. 1(b). Positions were adjusted based on in vivo dosimetry as indicated in Table 2. In the initial setup, the distance to the patient midline is identical (124 cm) for each beam.

Gantry
Couch position (cm, seen from feet)

3.C | In vivo dosimetry
For nearly all fractions in vivo dosimetry was performed. Both TLD and radiochromic film were evaluated overnight prior to the next treatment fraction by relying on standards irradiated very close to the actual treatment time.  Fig. 1(b) and used in patient treatment).
KRON ET AL.

| 113
Thermoluminescence dosimetry in vivo dosimetry measurements over the first 6 days of treatment were used to improve the dose distribution and adjustments made to couch vertical height and number of monitor units per field as can be seen in Table 2. Dose to the back of the patient was low with lowering the couch resulting in somewhat improved dose.
Eyes were shielded and partial shielding with PMMA was added to hands. All lead shielding was wrapped in thin plastic foil ("cling wrap") to avoid direct contact with the skin. Also based on in vivo dosimetry, toenail shielding was used from the third fraction onwards while finger nail shielding was added after fraction 8. All shielding was manufactured in house and made from lead.
In total, 34 measurement points were taken in areas that were not shielded or considered for boost. The average dose of these points was 1.45 ± 0.29 Gy (20%, 1SD).
Based on patient geometry and the in vivo dosimetry measurements, six areas were identified for boost treatment using conventional electron irradiation fields: • Scalp -6 Gy in 3fx, 5 × 7 cm oval, 1.5 cm bolus, 110 cm SSD

| DISCUSSION
The technique used here for pediatric TSET uses a similar approach to the Stanford technique for adults being based on two sets of six large field electron beams. 7 However, using six gantry positions 60°a part, it was possible to treat a pediatric patient in supine position.
By extending the SSD using a different couch position for each beam and employing two sets of fields 60 cm apart, it is possible to treat patients up to a body height of approximately 1 m. The field size chosen (36 × 36 cm 2 ) was based on our technique for adults and could be modified.
In vivo dosimetry was found to be an essential part of the treatment approach monitoring progress and identifying opportunities to improve the dose distribution, which would have been difficult to predict with a cylindrical phantom. This has been reported by several authors for adult TSET 8,[21][22][23] and in the present study more than 400 TLD chips have been read to ensure adequate dosimetry. The clinical situation did not allow for extensive commissioning exploring variations in patient size and shape. As such, some adjustment of the technique based on in vivo dosimetry was expected as Table 2 shows. In these modifications, we adopted a stepwise process to allow assessment of the dosimetric impact of changes made over the next fractions. Future treatments would benefit from this. centre of the patient to less than 3% of the incident dose. As many structures are closer to the skin in children than in adults, care was taken to optimize dose fall off as can be seen in Fig. 5

| CONCLUSION
Total skin electron therapy can be delivered to pediatric patients in supine position under general anesthetics using a conventional linear accelerator with a customized patient support. The treatment technique described here allows treatment of the whole skin of a young patient with acceptable dose accuracy but limited dose homogeneity.
A considerable amount of work is required to commission the technique and in vivo dosimetry can inform personalization as required.

ACKNOWLEDG MENTS
The authors would like to thank the mechanical workshop staff at Peter MacCallum Cancer Centre for manufacturing the thin window tabletop. Also, the financial support of the Gross foundation for physical dosimetry at Peter MacCallum Cancer Centre is acknowledged.

CONFLI CT OF INTEREST
None of the authors have any conflicts of interest to declare.

R E F E R E N C E S
T A B L E 2 TLD measurement results for four locations on the patient's mid body. Shown are doses in Gy per fraction and the changes in treatment made based on the measurements. To account for the changes made, dose for 6 and 12 fractions was estimated by assuming each fraction without a measurement would yield the same dose as the previous one (indicated in parenthesis).