29 Skin Cancer

SECOND EDITION

The GEC ESTRO Handbook of Brachytherapy

PART II: CLINICAL PRACTICE 29 Skin Cancer Jose Luis Guinot, Jose Pérez-Calatayud, Erik Van Limbergen

Editors Erik Van Limbergen Richard Pötter

Peter Hoskin Dimos Baltas

Skin Cancer

3

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

29

Skin Cancer Jose Luis Guinot, Jose Pérez-Calatayud, Erik Van Limbergen

1. Summary 2. Introduction

3 3 3 3 4 4 4 4

9. Treatment Planning

9

10. Dose, Dose Rate, Fractionation

12 12 13 16 17 18

3. Anatomical Topography

11. Monitoring

4. Pathology 5. Work Up

12. Results

13. Adverse Side Effects

6. Indications, Contra-indications 7. Tumour and Target Volume

14. Key messages 15. References

8. Technique

1. SUMMARY

Non-melanoma skin cancer can be treated with brachytherapy as an alternative to surgery, with some advantages in dosimetry compared with external beam radiation due to the rapid fall-off of the dose. LDR is no longer available, and HDR looks for repro- ducing the good results achieved during decades. A standard optimal schedule for HDR is not well defined, and different options have been published. In superficial tumours, contact brachytherapy is simple and effective, with flaps, personalized moulds or surface applicators. The larger the area to be irradiated, the lower the dose should be per fraction. In carcinomas deeper than 5mm an interstitial technique with plastic tubes or needles is mandatory.

2. INTRODUCTION

There is no consensus on the standard techniques and required doses for skin brachytherapy. A report of the American Brachytherapy Society (ABS) on aspects of dosimetry and clinical practice of skin brachytherapy describes a dosimetric summary and different available approaches [4].

Non-melanoma skin cancer is a disease of older age, more than half of new patients are more than 65 year old. About one half of patients with a skin carcinoma will develop a new primary lesion within 5 years of the initial diagnosis. Due to a longer life expectancy, the incidence rate is increasing by 3-6% per year, and it is becoming epidemic [1]. Basal cell and squamous cell carcinomas of the skin usually occur on sun-exposed sites, the face being one of the sites of predilection, accounting for 95 % of cases. Surgery in these sites (nose, ears, eyelids, lips)may bemutilating or require complex plastic reconstruction techniques under general anaesthesia. Surgery and radiotherapy appear to be themost effective treatments with surgery showing the lowest failure rates [2]. Only one randomized trial [3] of surgery versus radiotherapy in basal cell carcinoma of the face had primary outcome data at four years, showing significantlymore persistent tumours and recurrences in the radiotherapy group as compared to the surgery group but using a mixture of techniques and fractionations. Radiotherapy and brachytherapy have been used successfully for years, as an alternative or complement to surgery. Brachytherapy offers several advantages in dosimetry, is easy to apply, and the use of the linear accelerators for skin cancer can be decreased. Carefully tailored brachytherapy is a good alternative, if not the treatment of choice, for those lesions that cannot be safely removed by surgery with primary wound closure under local anaesthesia. All these factors contribute to encourage the use of HDR-BT for cutaneous carcinomas.

3. ANATOMICAL TOPOGRAPHY

As most skin cancers arise in the face, they are usually diagnosed at an early stage. However, deep extension to the orbit, ear or bone should always be ruled out. Deep extension along an embryonic pathway should always be suspected, especially in case of tumours involving the naso-labial fold, and adequate deep safety margins for resection and for brachytherapy must be taken [5].

4. PATHOLOGY

Most skin cancers are basal cell (65%) or squamous cell (30%) carcinomas. Brachytherapy is not indicated for skin melanomas. Around 90%of lesions are small, less than 20mmand their thickness is limited to a few mm [4].

Skin Cancer

4

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

5. WORK UP

implant patients. Moreover, it allows control of the active source length making the treatment volume conformal to the target volume and respecting the healthy surrounding tissue, with dose optimization. The short irradiation time allows for some extra shielding on critical structures such as the eyes. With regard to the more conventional orthovoltage photons and electrons, the dose distribution on the skin surface with contact HDR-BT shows a sharp gradient, similar to the conventional low- energy X-ray, in contrast to the electron dose distribution. This gradient allows a relatively higher dose to the first mm and a lower one to the deeper tissues. Moreover, the multicatheter applicator in moulds allows us to conform the isodose to irregular surfaces, which is not possible using orthovoltage photons or electrons. Electron beam therapy often requires cut-outs, in most cases the cut-out sizes are small and then specific dosimetry needs to be evaluated. Moreover, if collimation is done at a distance from the skin, then larger penumbra should be considered. Typically, additional collimation at skin with lead inserts is required to improve penumbra. Full-scalp irradiation with brachytherapy may have a lower risk of excessive brain irradiation than external beam irradiation because the intensity of surface-mould brachytherapy drops off rapidly away from the source [7]. A study on the technical considerations and dosimetry of HDR surface applicators shows that the availability of optimization techniques results in superior dose uniformity at the prescription depth [8].

All suspected lesions should have a biopsy to confirm the diagnosis. Local tumour extension should be carefully documented. Exact measurements (in mm) are required. Photographs may help to document local tumour extension. Suspected deep infiltration into the orbit, ear or other structures, should be studied byCTor other investigations. Agood collaboration with dermatologists is very helpful in order to define the real extension of the tumour, especially in basal cell carcinomas where a subdermal extension is not easily evaluated. Examination under direct vision with a dermatoscope can help to delineate better the margins. High-resolution ultrasound with frequencies higher than 18MHz (30MHz) is a good tool to document the widespread and depth of small lesions [6].

6. INDICATIONS, CONTRA-INDICATIONS

6.1 Indications • Basal cell as well as squamous cell carcinoma is radiosensitive, andmost of these lesions are detected at an early stage, when the chance for cure by radiation is high. Both external radiotherapy (low energy X-ray or electron beam) and brachytherapy can be used, but brachytherapy is preferred to X-rays or to electron beams when these are difficult to apply, for example on curved surfaces. • Exclusive brachytherapy: the main indications are epidermal skin cancers T1 - T2 N0 on the face for which curative surgery with adequate margins cannot be offered without mutilation or without the need for extensive reconstructive surgery under general anaesthesia. • As a boost in larger T2 - T3 or in N+ cases after external beam radiotherapy to the primary tumour and lymph nodes. • As a postoperative procedure when there are close or positive margins, or in some cases of nerve involvement. • Upper eyelid lesions for LDR or PDR BT, but not for HDR • Where the anatomical situation makes the source positioning needed to provide adequate covering of the target volume impossible: Pinna tumours involving both the concha and the external auditory canal, ear conduct or any other site, unless special devices can be delivered to get adequate dosimetry. 6.3 Comparison of HDR Brachytherapy with LDR and external beam therapy Contact BT with HDR has some relative advantages compared to LDR interstitial BT with fewer radioprotection issues; the treatment is administered on an outpatient basis, in fast sessions, and isolation is not required, a very important issue whenmanaging elder patients. The moulds, flaps and surface applicators are easy to place and patients do not report discomfort, unlike interstitial 6.2 Contraindications • Malignant melanoma of the skin. • Skin cancers invading bony structures.

7. TUMOUR AND TARGET VOLUME

The clinical target volume for well-delineated squamous cell or basal cell carcinomas is the palpable or visible tumour with a safety margin of 3-5 mm for skin cancers and 5 mm for lip cancers. For poorly defined lesions, such as morphea like basal cell carcinomas, a wider safety margin is taken (7-10 mm). The depth of the tumour can be evaluated by clinical assessment or by means of high frequency ultrasound.

8. TECHNIQUE

Two different techniques are useful for skin cancers: contact (plesiotherapy) and interstitial brachytherapy. Contact brachytherapy can be used with the three modalities (LDR, PDR or HDR) through flaps, moulds or contact surface applicators that are applied on the skin area to be treated; no anaesthesia required. Interstitial implants have been used in several ways with low dose rate (LDR) by means of hypodermic needles, silk wires, and inner nylon tubes with flexible 192-Ir wires. Nowadays, high dose rate (HDR) or pulsed brachytherapy (PDR) use afterloading rigid needles or plastic tubes.

Skin Cancer

5

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Fig. 31.1: Leipzig applicator TM of 10, 20 and 30 mm internal diameter. The HDR source is positioned vertically (first 3 applicators) or horizontally (last 3 applicators) in the tip of the applicator, having different dose distributions.

Fig. 31.2: Treatment of a superficial basal cell carcinoma of the left inner canthus in a 78 year old lady with the Valencia applicator. The CTV (GTV plus 3 mm margin) is delineated. The delineator cap of 20mm inner diameter is brought to the correct position and the outer diameter is delineated on the skin. Verification of the correct positioning by controlling the indentation of the inner diameter around the CTV. Applying the applicator within the delineated area and fixation with tape and net (not shown).

Skin Cancer

6

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Fig. 31.4: Hand-made flaps using 5mm bolus and plastic tubes taped to the surface (from Maroñas et al, 2011, with permission [51] )

Fig. 31.3: Esteya system

8.1 Contact Brachytherapy (Plesiotherapy) 8.1.1 Surface contact applicators The surface applicators are cup-shaped shielded applicators with a single dwell position. Their shielding is used to reduce the dose to normal tissue and focalizing the treatment volume. The use of these applicators is indicated on flat surfaces where there is good contact without air gaps and small target areas smaller than 3 cm. Several surface contact applicators are available for use with PDR or HDR afterloaders such as the Leipzig applicators™ (Fig. 1) [9, 10, 11, 12]. The Valencia applicators™ (Fig. 2) are an improved modification that gives a more homogeneous dose to the skin [13, 14]. Current Valencia applicators version covers 20 and 30mmfield diameter sizes. There is a new version in progress covering up 50 mm in diameter field size [15]. Electronic brachytherapy is an alternative to the radionuclide based surface applicators [16]. Because of their very low energy (50-69.9 kV) no shielded room is required. There are three systems currently available: Xoft Axxent™ [17, 18], Esteya™ [19] (Fig. 3), and IntrabeamTM.

8.1.2 Flaps Flaps are constructed with plastic tubes (diameter 1.6 - 2mm, 5 or 6 French) and strips of tape (Fig. 31.4). This allows the treatment of large flat or slightly curved skin areas such as the forehead, legs, and the back. The mats are simply taped or fixed to the skin. No anaesthesia is needed. Flaps of different size and inter source spacing can be constructed. These applicators can be used with PDR or HDR afterloading machines. They are not suitable for very curved areas, or areas where skinmay move during treatment (such as the peri-oral area, shoulder, groin....). Diverse flexible flaps with the plastic tubes placed at an exact distance from the surface (usually 5mm) are available: Elekta Freiburg Flap, the Varian Catheter Flap set, and the Harrison-Anderson-Mick (HAM) from Mick Radio- Nuclear Instruments (now E&Z Bebig, Berlin, Germany). They can be cleaned, re-sterilised and used several times. The Freiburg flap applicator™ is made of small spheres 10mm of diameter joined to get a flexible layer that can adapt to the curvature of round areas of the head (Fig. 5). Then the 5mm source-surface distance is guaranteed. Taking into account the shape of the applicator and the surrounding air between spheres, the deviations on a typical flap have been evaluated withMonte Carlo for a standard plan, resulting in a dose reduction of 5% as compared to TG43 in infinite water [20]. The Varian catheter flap has 5 mm between the catheters and

Skin Cancer

7

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Fig. 31.5: Freiburg flap on the skin of a SCC with neural involvement after surgery treated with HDR 4Gy three times a week x 11 fractions. Results at three year.

Fig. 31.6: Custom made contact BT with plastic tubes fixed on an individual mask for a squamous cell ca of the right hand in a 80 year old man. The CTV is first drawn on the skin and then on the mask. A radio-opaque marker is applied to indicate CTV position on the planning CT

Skin Cancer

8

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Fig. 31.7: Large flap wax-made for scalp contact brachytherapy with nine parallel plastic tubes. A second mask is added and tied, to keep the wax at place.

Fig. 31.9: Wax moulds for nose and curved areas. Isodose curves for a nasal mould with five plastic tubes.HDR 3Gy three times a week x 16 fractions. Results at two years.

Fig. 31.8: Custom-made polymethyl methacrylate surface moulds with six plastic tubes. (from Guix et al. 2000, with permission [49])

they are located in the middle axis of the flap with 5 mm from the skin. The HAM flap applicator has an array of catheters spaced 10 mm apart and at 5 mm from the skin. Thermoplastic masks aid to fix the position for the daily exact placement in contact with the skin (Fig. 6). Large areas such the scalp can be treated with different systems using thin wax layers, silicon or similar materials to fix the parallel plastic tubes to the head [21, 22, 23] (Fig. 7).

8.1.3 Customized Moulds Individual off prints can be taken from different anatomical areas to construct surface-moulds containing plastic tubes for, PDR or HDR afterloading: these moulds allow contact brachytherapy of several anatomical tumour sites, including irregular shaped areas such as the external ear or nose (Fig. 31.8 and 31.9).

Skin Cancer

9

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Fig. 31.10: Interstitial implant with three 5 French plastic tubes of an infiltrating >8mm thick squamous cell carcinoma.

Fig. 31.10: Interstitial implant with three 5 French plastic tubes of an infiltrating >8mm thick squamous cell carcinoma.

Fig. 31.12: Squamous carcinoma in the columella. Two plastic tubes HDR 4Gy x 12 twice a day. Results at six months

8.2 Interstitial techniques Local or regional anaesthesia is required. If possible, nerve block anaesthesia is performed to avoid swelling in the target area. This is possible with a mentalis nerve block for the medial two thirds of the lower lip, an infraorbital nerve block for the upper lip and cheek, or partial or total ring anaesthesia around the ear. If local infiltration anaesthesia of the skin is used all visible and palpable tumour should be marked with a pen before the skin is infiltrated with lidocaine 1% or other local anaesthetic. The positioning of the lines is drawn on the skin, taking into account the Paris System and Stepping Source Dosimetry System (SSDS) [24] rules to cover the target volume (for details see 9.A). Most skin cancers can be treated by a single plane implant, using parallel lines, spaced 10 to 16 mm apart (Fig. 31.10, 11, 12). Implants which are too superficial may result in late visible telangiectasia along the source positions. For curved planes, frequently occurring in skin cancers of the scalp, face and limbs the prescription isodose projects deeper at the concave than at the convex side. The arrangement of implanted or applied sources in those cases must take account of this shift in the isodose lines. Lesions thicker than the thickness of the reference prescription isodose of a single plane configuration, have to be treated by double plane implants (the second plane may be constructed "in air" above the tumour adding some bolus or tissue equivalent to cover the catheters). However, sometimes it is easier to shave the exophytic part of the tumour with electrocoagulation. With HDR, the optimization allows to correct the small differences in distance between the plastic tubes. Some catheter spacing and stabilization technique is useful to maintain geometry for multicatheter interstitial implants [25]. If rigid needles are used,

a pair of templates with perforated holes at exact distances, usually triangles or squares 1cm side, are useful to achieve very homogeneous implants. Most of the commercially available Treatment Planning Systems (TPS) are based on the TG-43 assumptions [26] and then the scatter default on this implants type is not considered. In some hospitals, this deviation is theoretically compensated with the use of bolus over the skin surface. In a recent work using Monte Carlo the conclusion is that no bolus is required for Ir-192 but few mm of bolus are required for HDR Co-60 [27].

9. TREATMENT PLANNING

9.1 2D planning In contact brachytherapy, treatment planning is often 2D and the prescription is to distance in depth depending on the contact applicator type. It is always required to report the depth of prescription dose, as well as the surface dose delivered at the epidermal surface. For surface contact applicators the prescription depth is 3-4 mm under the skin surface, no more than 5 mm. Dosimetry can use library pre-calculated curves (atlas) for surface applicators. We must keep inmind that the surface is receiving a higher dose than prescribed, with surface contact applicators (Leipzig™, Valencia™

Skin Cancer

10

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Fig. 31.13: a) Dosimetry of the Valencia applicator™, and b) comparison of isodoses with the Leipzig applicator™ (in red the prescription isodoses at 3 mm depth for both applicators).

Table 31.1: Geometric relations between active source length, intersource spacing, and treated length, thickness and width in single plane, double plane squares and triangles for a stepping source implant with parallel and equidistant source configurations [28].

Treated length / Radioactive length

Treated thickness / spacing

Lateral margin/ spacing

Safety margin/ spacing

Implant type

2 lines

0.8

0.5

0.37

_

n lines in one plane

0.8

0.6

0.33

_

n lines in «square»

0.8

1.55 to 1.60

_

0.27

n lines in triangle

0.8

1.3

_

0.20

and Xoft™), the gradient is 10% more dose per mm, and 7% with Esteya™ (Fig. 31.13). Withmoulds or masks, prescription doses may lie between 3-6 mm depth. In those cases, some points of the skin can receive even 150% of the prescribed dose due to the steep dose gradient. Because of this reason, it is advisable to keep plastic tubes at a distance at least of 2 -5mm from the skin, by using wax, bolus or masks, in order to get a more homogeneous distribution on the skin and a lower gradient of dose in depth (Fig. 31.14 and 15). If we need to treat a deeper target, an interstitial procedure is advisable. Unlike interstitial implants, where no variations can happen between fractions and then CTV is equal to PTV, contact brachytherapy requires several fractions and generates some uncertainties, which should be compensated with a small margin in the lateral direction, therefore, a PTV is required. Interstitial implants should be planned with forecast dosimetry, using the rules of the Paris System. Based on the dimensions of the CTV: thickness, width and length, the number of planes, the number of source lines and the spacing between them are determined. A CTV smaller than 12 mm can be treated with 1 plane. The treated thickness is 0.5 times the spacing for 2 lines, and 0.6 times for more than 2 lines. The treated width is n times the spacing plus 2 times the lateral margin (see table 1). The length is 0.8 times the Stepping Source Length (compare chapter 7).

When geometrical optimization is applied in interstitial brachytherapy with stepping source afterloaders, the Paris System rules can be modified to the Stepping Source Dosimetry System (SSDS) of van der Laarse [24]. The treated length improves to 85% of the Stepping Source Length and the prescription isodose may be changed is to the 90% isodose instead of the 85% of the Mean Central Dose. 9.2 3D-dose planning 3 D dose planning is not possible with surface contact applicators (not CT-compatible metallic applicators), but is feasible with plastic tubes applied in individual shaped contact moulds, wax, bolus or masks. With the Stepping Source technology the depth of treatment and individual adaptation on individual target contours become possible.The GTV and CTV can be delineated on CT scans (MRI) with the applicator in place. Using 3D treatment planning an individual 3D treatment plan can be generated. As at present there is no clinical evidence available for 3D prescription and reporting, the 3D treatment planning starts with the traditional prescription of depth and surface dose according to 2D clinical practice and then improves on D90 and D98 values for optimal target coverage. All values from 2D and 3D should be recorded and reported.

Skin Cancer

11

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Fig. 31.14: Example for a dose plan for a flap made with five parallel plastic tubes for contact brachytherapy (shown in Fig. 4). Prescription dose is to 3mm under the skin surface. The distance is calculat- ed by adding 5mm bolus plus 1mm radius, total 9mm from the axis. The 150% isodose curve does not arrive to the skin.

Fig. 31.15: Isodose curves for a nasal mould with five plastic tubes. High doses remain inside the mould, with a more homogeneous distribution on the skin surface.

Skin Cancer

12

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Fig. 31.16: Lower eyelid implant with 2 plastic tubes and dosimetry for interstitial HDR brachytherapy.

For 3D image based treatment planning with interstitial brachytherapy CT scan slices or MRI, with dummies inside the tubes or needles are used, with markers on the edges of the GTV or CTV. GTV and CTV are delineated. In the case of parallel tubes or needles the modified Paris System (SSDS van der Laarse) can be used and the prescription is usually to 90% isodose of the Mean Central Dose (Fig. 31.16). In 3D planning D90 and D98 values should be reported which are related to the GTV and the CTV as defined on 3D imaging. Optimized dosimetry will also adjust isodose curves to the CTV drawn on the scan, to ensure target coverage and avoid hot spots by keeping the DNR (Dose-non-uniformity-ratio) under 0.36 [29]. The DNR (which is the ratio of the volume that receives 150% of the prescribed dose over the volume that receives the prescribed dose = V150/ V100) is a simple and easy to interpret parameter to quantify the inhomogeneous dose distribution around interstitial sources. To achieve the highest degree of homogeneity it is advised to prescribe not only to an isodose related to the MCD in the central plane, but also to the mean longitudinal minimum doses between the sources throughout the whole treated volume. The lowest DNR is achieved by the combination of dose point optimization with geometrical optimization.

between fractions, dose: typically 8 times 4 Gy twice a day in 4-5 days. Depending on the volume to be treated and the organs at risk (eye, cartilage, bone) a lower or higher per fraction (with corresponding total dose) is used. Doses between 3 and 5 Gy can be chosen, in order to finish the treatment in a similar time to the LDR techniques: five days to avoid the gap of the weekend, or eight-ten days with lower doses per fraction. When using contact brachytherapy, the schedule is more similar to electrons, and 3-4 Gy per fraction three days per week in 4-5 weeks is effective. In large areas such as scalp, a 2-3 Gy per day fractionation is preferable. In small epithelial tumours, 5-7 Gy per fraction 2-3 days per week can be used. With skin surface applicators the chosen dose is 7 fractions of 6 Gy or 5- 6 fractions of 7 Gy, twice a week. In cases of very thin skin or with underlying cartilage, such as the nose, lower doses per fraction probably allow better cosmetic long-term results. No clear recommendations can be done due to the great variety of published schedules and the prescribed dose is based more in experience that in evidence. The depth of prescription must be always indicated, because the same prescribed dose can result in different doses to the skin surface. The total dose depends on the chosen dose per fraction, common regimens may include 51-54 Gy (17-8 fractions of 3Gy), 44-48 Gy (11-12 fractions of 4 Gy ), 40 Gy (8 fractions of 5Gy), 42 Gy (6 fractions of 7 Gy or 7 fractions of 6 Gy), 35 Gy (5 fractions of 7 Gy).

10. DOSE, DOSE RATE, FRACTIONATION

11. MONITORING

With LDR Ir-192 wires, the prescribed dose was 60 Gy at the 85% of the Mean Central Dose isodose. This assumed to be the minimum target dose (peripheral dose) covering the CTV, at dose rates between 45-70 cGy/h. Depending on the linear activity and the source spacing and length, the required dose used to be delivered in 4 to 6 days. Although doses up to 70 Gy were given in some large tumours, without unacceptable sequelae, the increase in cosmetic damage from a dose increase above 60 Gy is greater than the gain in local control. With PDR similar doses are recommended. With interstitial HDR brachytherapy, high dose per fraction is used, twice a day in interstitial implants, separated at least 6 hours

Moist desquamation develops 1 week after implantation in the mucosa and after ± 12 days in the cutaneous areas. The reaction is maximal at about three weeks, and heals progressively (depending on the area) in 5 to 8 weeks. No special care is required, except for daily cleaning and application of a topical antiseptic. Local application of silicone-coated wound dressing, seems to be a good alternative to treat larger zones of moist desquamation. Sun exposure must be avoided, and a sunscreen should be routinely used.

Skin Cancer

13

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Table 31.2: Local control, cosmesis and adverse side effects after Interstitial skin brachytherapy with LDR in basal cell and squamous cell skin carcinoma: various patient cohorts with heterogene- ous treatment schedules (retrospective evaluation)

Author [reference]

Tumour localization Number

Local control

Good cosmesis

Adverse side effects

Year

Mode

Follow up

97% if primary 94% if recurrence

13% 31%

Daly [34]

1984

Eyelid

165

192 Ir

5 y

Nasal vestibule

Baris [33]

1985

22

192 Ir

2 y

96.4%

0%

13% ulcers 4% necrosis

Mazeron [36]

1986

Ear/pinna

70

192 Ir

5 y

99% <4cm: 78% >4cm: 11%

95% if primary 88% if recurrence

Mazeron GEC [32]

93% 87%

1988

Nose

1676

RT/ 192 Ir

2 y min

2%

Crook [30]

1990

Nose

468

192 Ir

5 y

97.5%

94%

2%

97% if primary 94% if recurrence

Debois [31]

1994

Nose

370

137 Cs

2y

97%

0%

Maes [37]

2001

Face

173

192 Ir

45m

95%

89%

3.6%

Gambaro [35]

2001

Eyelid

50

192 Ir

7 y

96%

92%

92.5% if radical 88% postoperativ.

Rio [38]

2005

Face

97

192 Ir

Ducassou [39]

2011

Face

147

192 Ir

5 y

87.3%

12. RESULTS

12.2 Local control rates after HDR brachytherapy HDR Brachytherapy has substituted LDR nowadays. Better surgical techniques likeMohs’ surgery [40] have been developed, and other therapeutic options, cryocoagulation, photodynamic therapy or external skin treatment with imiquimod can be considered [41]. Therefore few studies are available that report the results of high-dose rate brachytherapy with custom-made surface moulds for skin tumours. The results appear to be very favourable with 5 year actuarial local control rates of 88-98% and without severe late complications (Table 31.3). Svoboda et al. [42] used moulds made with two 0.75 cm square silicone bases and 2 mm endobronchial catheters fixed between them. In lesions close to eyes, 4 mm lead shields were used for protection. The dose was prescribed to the surface of the applicator. They treated 87 cutaneous carcinomas (BCC and SCC) obtaining in each of thema complete tumour response and very good to excellent cosmetic result, with recurrence in four cases, all them with an initial diameter greater than 2cm and a depth greater than 3mm. Allan et al. [43] described complete response in 13 patients with cutaneous carcinomas of the pinna treated by means of moulds for HDR-BT. There were no cases of Grade 4 dermatitis, the cosmetic results were excellent and they report no relapse during the 18-month follow-up.

12.1 Local control after LDR brachytherapy With doses of 60 – 65 Gy, local control is excellent for T1 – T2 skin cancers. Five-year recurrences rates range from 1 to 5% for non-melanoma skin cancers of the nose [30, 31, 32] and nasal vestibule [33], the eyelids [34, 35], ear and the pinna [36] or different areas of the face [37]. However, if the patient is treated for a recurrent tumour after previous surgical resection, the recurrence rate is higher, ranging from 6 to 13%. Most of these studies are retrospective mono-institutional cohort studies, only one is a multicentre study (Table 31.2). The more recent series, with interstitial 192-Ir LDR Brachytherapy, confirm good results: Rio et al. [38] treated 97 patients with periorificial facial skin carcinomas, and the local control rate was 92.5% for those cases treated with radical intention and 88% for those with postoperative intention. AndDucassou et al. [39] treated 147 facial carcinomas with local-regional-relapse-free survival of 96.6% at 2 years and 87.3% at 5 years, local control was better after primary treatment than after recurrence, and for basal cell carcinoma than for squamous cell carcinoma.

Skin Cancer

14

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Table 31.3: Local control after HDR skin brachytherapy in in basal cell and squamous cell skin carcinoma: various patient cohorts with heterogeneous treatment schedules (retrospective evalua- tion)

Author [reference]

Local control

Year

Number

System Fractions Prescription Total dose Frequency

Single/ weekly/ twice a day/ Daily Twice a day

4-22 Gy 9 Gy x 3 at surface

Silicone moulds

Svoboda [42]

1995

130

1-15

12-50 Gy

100%

Allan [43]

1998

13

moulds

8

5.5 Gy x 8

44Gy

100%

Kohler-Brock [44]

1999

520

Leipzig app

5-10 Gy

30 Gy

1-2/week

92%

33-36 + Boost 42-44

1.8 Gy at 5mm

60-65 Gy 75-80 Gy

99% radical 87% adjuvant

Guix [49]

2000

136

moulds

Daily

4 Gy at 5

Montero [50]

2009

11

moulds

11-12

100%

mm 44-48 Gy

Wax/ bolus molds

3-4 Gy at 3 mm 48-57 Gy 3 times per week

Maroñas [51]

2011

51

12-18

89% at 5y.

6,5 Gy at 5-10mm 39Gy 3 G-4y at 3-4mm 36 Gy

90% 50%

Azad [53]

2011

20 eyelid Interstitial

6

6 days

Gauden [45]

2013

236

Leipzig app

12

Daily

98%

Electronic BT

Twice a week Twice a week

100% at 10m.

Bathnagar [54]

2013

171

8

5 Gy

40 Gy

Tormo [47]

6-7 Gy at 3-4mm 42 Gy 5 Gy at 3-4mm 40-50Gy

2014

45

Valencia app

6-7

97.7% at 4y.

2-3 times per week

96% at 1 year

Delishaj [48]

2015

57

Valencia app

8-10

Leipzig app 101 Molds 33

95% 88% at 5 y.

Arenas [46]

2015

134

15-18

3Gy at 5mm 45-54Gy 3 times per week

Kohler-Brock et al. [44] treated 520 skin lesions with Leipzig applicators™, reaching a total dose of 30-40 Gy, 5-10 Gy/fraction, 1 or 2 fractions/wk with complete response in 91% of the cases, partial in 6%, and a relapse rate of 8%, without late complications. Another study with Leipzig applicators™ used lower doses per fraction, 3 Gy daily x 12 fractions to deliver 36 Gy prescribed to 3-4mm, in 236 lesions with median follow-up of 66 months and achieved 98% of local control with cosmesis good or excellent in 88% [45]. In 101 cases with 3Gy three times per week x 15-18 fractions, achieved local control in 95% at 5 years [46] (Fig. 13.17). With Valencia™ applicators, the first results treating 45 cases of BCC, 6-7 Gy twice a week to achieve 42Gy at 3-4mmobtain a local control of 97.7% with a mean follow-up of 47 months [47] (Fig. 13.18). With 5Gy 2-3 times per week to 40-50 Gy 96% showed a complete response after 12 months [48]. Guix et al. [49] treated 136 cutaneous carcinomas with contact HDR-BT reaching a total dose of 60-65 Gy, 1.8 Gy/session, and for lesions greater than 4cm, a boost of 18 Gy was added.The actuarial local control rate at 5 years was 98%; 99% for the treatments with radical intention (n=73), and 87% for the adjuvant treatments (n=63). Montero et al. [50] used custom-made moulds in 11 cases of non- melanoma skin carcinoma, with 11-12 fractions of 4 Gy over 4

weeks to achieve 44-48 Gy, with no relapses at 15 months Maroñas et al. [51] treated 51 facial skin carcinomas, with a size between 0.5 and 4 cm and a maximum thickness of 3 mm, making customized moulds, with a dose per fraction of 3-4 Gy three times a week. The relapse rate was 9.8% with median follow-up of 45 months. No failure was recorded on flat surfaces. To treat nose carcinomas, a wax mould with plastic tubes was made, only one of 20 cases (5%) on the nasal bridge or alae had recurrences, but 4 out of 15 cases (27%) on the nasal tip, because the applicator did not fully cover the margins to be radiated (Fig. 13.19). Unusual locations as eyelid carcinomas, have been treated with HDRwith successful results, achieving local control in 17 cases at a mean follow-up of 24 months, with 10 fractions of 4 Gy prescribed at 3mmwith no visual complication reported [52]. Azad et al. [53] treated 20 carcinomas of the eyelid with six fractions of 6.5 Gy in six days with HDR brachytherapy, and with a median follow up of 40 months, the 5-year disease free survival rate was 90%, 57% and 50 % for squamous, sebaceous cell and basal cell carcinoma respectively. Initial experience with electronic brachytherapy appears to show acceptable acute safety and favourable cosmetic outcomes. Treatment of 171 skin carcinomas delivering eight fractions of 5 Gy twice weekly showed no recurrence with a median follow-up of 10 months [19]. In 40 patients treated with six fractions of 7 Gy or

Skin Cancer

15

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Fig. 31.17: Small and superficial basal cell carcinoma in a 84 year old lady, treated with contact BT by Leipzig applicator 5 times 7Gy at 3mm depth, delivered in 2.5 weeks. Results at 3 years.

Fig. 31.18: Results with Valencia applicator™, six fractions of 7 Gy twice a week.

Fig. 31.19: Results of a basal cell carcinoma of the nose treated with HDR contact brachytherapy with wax moulds, 3 Gy three times a week, 18 sessions, total dose 54 Gy.

Skin Cancer

16

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

Fig. 31.20: Basal cell carcinoma of the left inner canthus in a 76-year old man with results 3 years after hypodermic needle implant with LDR. The deeply infiltrating tumour has disappeared, leaving a re-epithelized skin depression surrounded by post RT acromia in the irradiated area.

and additional excisions were needed in 42%. These very good results are thus obtained with complex surgical procedures which are in fact not feasible in the majority of typically older patients who present with non-melanoma skin tumours (in this study the patients who were not fit for general anaesthesia were excluded). The cosmetic outcome was good in 87% of surgical treatments versus 69% in the radiotherapy group (5% necrosis). However, the brachytherapy dose was very high (66 - 70 Gy), whichmay explain the worse cosmetic results in the radiotherapy group. 13.2 Adverse side effects after HDR brachytherapy Moulds with different techniques using 3-4 Gy three times a week produced acute G1-2 dermatitis in 78%, solved in a few days with topical treatment, 22% developed severe acute toxicity G4, which was healed in 3 weeks time using local cures on an outpatient basis. No cases of later complications were reported with good or very good late cosmetic results. Moulds treating low doses per fraction, 1.8 Gy five days a week, produced acute G4 complication rate in 10% (ulceration) and late good cosmetic results in 98%. In some cases, a slight hypopigmentation of the skin (G1) or small telangiectasia’s (G2) in the radiated area, mainly in cases of large tumours are described, with patchy pigmentation in a reduced number of cases. Cosmetic results are good or very good in all cases. With Leipzig applicators, no late complications are reported. In one study, grade 1 acute skin toxicity was detected in 168 treated lesions (71%) and grade 2 in 81 (34%), late skin hypopigmentation changes were observed in 13 cases (5.5%) and cosmesis was good or excellent in 208 cases (88%). With Valencia applicators the highest skin toxicity was grade 1 RTOG/EORTC, having resolved with topical treatment at 4 weeks in all but one case which required 2 months and there were no grade 2 or higher late adverse events. Electronic brachytherapy produces similar acute toxicity but long term side effects are not yet available.

6.1 Gy, local control at one year was 95 and 90% respectively [54]. Other tumours can benefit fromHDR brachytherapy, as cutaneous metastasis of Merkel cell carcinoma [55], Kaposi sarcoma lesions [56], or cutaneous T-cell carcinoma (mycosis fungoides) [57].

13. ADVERSE SIDE EFFECTS

13.1 Adverse side effects after LDR brachytherapy In general, an excellent or a good cosmetic result is obtained in 78 to 92%of patients after LDR brachytherapy (Fig. 17). Complication rates range between 0 and 13% and are similar to those of external beams. Complications are dose and dose rate dependent. For small tumours (T1-T2) a dose of 60Gy should therefore be recommended. Higher doses result in only a small increment in local control, but a significant rise in complications. For larger tumours, local control rates remain good, but complication rates are higher (5.5% skin necrosis or healing ulcer for lesions larger than 2 cm) and cosmetic outcome is less favourable, because of skin destroyed by the tumour and retraction of the healing scar which occurs afterwards. The risk of complications is highest for tumours on the pinna (4 - 18.5% persistent ulcers) because of poor blood supply. It has been suggested that for basal or squamous cell carcinomas on the pinna, a dose of 55 Gy may be sufficient, since these tumours show very good local control (100%) after a dose of 60 - 66 Gy but a high risk of complications (permanent ulcers 18.5%) [37]. There is one published randomised study compares surgery with radiotherapy (LDR brachytherapy or external radiotherapy) for basal cell carcinomas of the face [3]. Surgery appeared to be better than radiotherapy, both in terms of local control rate and cosmetic outcome. The 4-year actuarial local control was 99.3% in the surgical group compared to 93.5% in the radiotherapy group. However, in 30% of the surgical cases, general anaesthesia was used in 46%, flap reconstructions were needed. In 91% of the treatments, systematic frozen section examination was performed

Skin Cancer

17

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

14. KEY MESSAGES

• Carefully tailored brachytherapy is a good alternative, if not the treatment of choice for those lesions that cannot be safely re- moved by surgery. • Local control rates with interstitial LDR using doses of 60 - 65 Gy are excellent for T1 - T2 skin cancers. HDR has become the standard treatment, with the aim of delivering the equivalent dose.

• HDR and PDR can be used in interstitial implants with a B.I.D schedule, usually 3-4 Gy per fraction.

• In skin tumours with a depth of <5mm, contact brachytherapy through flaps, moulds or contact surface applicators without anaesthesia are effective. • In small size tumours, Leipzig, Valencia applicators or electronic devices yield excellent results with high doses per fraction twice a week. • In extensive flat lesions, customized flaps or moulds with taped plastic tubes at a distance from skin of at least 3-5mm to avoid overdose on the skin, can be used at 3-5Gy per fraction once daily or every other day. • No standard schedule can be recommended and total doses are based in experience. The dose on the skin surface should be recorded to correlate the outcome with late side effects.

Skin Cancer

18

THE GEC ESTROHANDBOOKOF BRACHYTHERAPY | Part II Clinical Practice Version 1 - 30/04/2017

15. REFERENCES

22. Semrau S, Kunz M, Baggesen K, et al. Successful treatment of field cancerization of the scalp by surface mould brachytherapy. Br J Dermatol 2008; 159:753-5. 23. Sachse MM, Wagner G, Aydin H, et al. Treatment of extensive lentigo maligna melanoma of the scalp by brachytherapy moulage technique. Br J Dermatol 2011; 164:219-21. 24. van der Laarse R. The stepping source dosimetry system as an extension of the Paris system. Chapter 34. In “Brachytherapy from radium to Optimization”. Editors: Mould RF, Battermann JJ, Martinez AA, Speiser BL. 1994. Published by Nucletron. 25. Demanes DJ, Friedman JM, Park SJ, et al. Brachytherapy catheter spacing and stabilization technique. Brachytherapy . 2012; 11:392-7. 26. Rivard MJ, Coursey BM, DeWerd LA, et al. Update of AAPM Task Group No. 43 Report: A revised AAPM protocol for brachytherapy dose calculations. Med . Phys 2004; 31:663-674. 27. Granero D, Perez-Calatayud J, Vijande J, Ballester F, Rivard MJ. Limitations of the TG-43 formalism for skin high-dose-rate brachytherapy dose calculations. Med Phys 2014; 41(2):021703. 28. Dutreix A, Wambersie A, Marinello G: Dosimétrie en curiethérapie. Masson, Paris, 1982 29. Major T., Polgar C., Somogyi R, Nemeth G. Evaluation of the dose uniformity for double plane high dose rate interstitial breast implants with the use of dose reference points and dose non-uniformity ratio. Radiotherapy and Oncology 2000; 54:213-220. 30. Crook JM, Mazeron JJ, Marinello G, et al. Interstitial Iridium 192 for cutaneous carcinoma of the external nose. Int J Radiat Oncol Biol Phys 1990; 18: 243-48. 31. Debois JM. Cesium-137 brachytherapy for epithelioma of the skin of the nose: experience with 370 patients. JBR-BTR 1994; 77: 1-4. 32. Mazeron JJ, Chassagne D, Crook J, et al. Radiation therapy of carcinomas of the skin of nose and nasal vestibule: A report of 1676 cases by the Groupe Europeen de Curiethérapie. Radiother Oncol 1988; 13: 165-73 33. Baris G, Visser AG, van Andel JG. The treatment of squamous cell carcinoma of the nasal vestibule with interstitial iridium implantation. Radiother Oncol 1985; 4: 121-25. 34. Daly NJ, de Lafontan B, Combes PF. Results of the treatment of 165 lid carcinomas by iridium wire implant. Int J Radiat Oncol Biol Phys 1984; 10: 455-59. 35. Gambaro G, Negri E, Prino, et al. Interstitial brachytherapy with Iridium-192 for carcinomas of the eyelid and inner canthus. Radiother Oncol 2001; 60 (suppl.1): 16. 36. Mazeron JJ, Ghalie R, Zeller J, et al. Radiation therapy for carcinoma of the pinna using iridium 192 wires: a series of 70 patients. Int J Radiat Oncol Biol Phys 1986; 12: 1757-63. 37. Maes A, Van Limbergen. E. LDR – Brachytherapy (BT) for non-melanoma skin cancer of the face: Local control rate, functional and cosmetic outcome in 173 patients. Radiother Oncol 2001; 60 (suppl 1): 16. 38. Rio E, Bardet E, Ferron C, et al. Interstitial brachytherapy of periorificial skin carcinomas of the face: A restrospective study of 97 cases. Int Radiat Oncol Biol Phys 2005;63:753-757. 39. Ducassou A, David I, Filleron T, et al. Retrospective analysis of local control and cosmetic outcome of 147 periorificial carcinomas of the face treated with low-dose rate interstitial brachytherapy. Int J Radiat Oncol Biol Phys 2011; 81:726-31. 40. Mosterd K, Krekels GA, Nieman FH, et al. Surgical excision versus Mohs’ micrographic surgery for primary and recurrent basal-cell carcinoma of the face: A prospective randomised controlled trial with 5-years’ follow-up. Lancet Oncol 2008;9:1149-1156. 41. Lansbury L, Leonardi-Bee J, Perkins W, et al. Interventions for basal cell carcinoma of the skin. Cochrane Database Syst Rev 2010;4: CD007869. 42. Svoboda VH, Kovarik J, Morris F. High dose-rate microselectron molds in the treatment of skin tumors. Int J Radiat Oncol Biol Phys . 1995 Feb 15;31(4):967- 72.

1. Rogers HW, et al, Incidence estimate of non-melanoma skin cancer in the United States, 2006. Arch Dermatol , 2010; 146 (3): 283-287. 2. Bath-Hextall FJ1, Perkins W, Bong J, Williams HC. Interventions for basal cell carcinoma of the skin. Cochrane Database Syst Rev . 2007; (1):CD003412. 3. Avril M-F, Auperin A, Gerbaulet A, et al. Basal cell carcinoma of the face: surgery or radiotherapy? Results of a randomized study. Br J Cancer 1997; 76: 100-6. 4. Ouhib Z, Kasper M, Perez-Calatayud J et al. Aspects of dosimetry and clinical practice of skin brachytherapy: The American Brachytherapy Society working group report. Brachytherapy 2015; 14(6): 840-58. 5. Mendenhall WM, Million RR, Mancuso AA, et al. Carcinoma of the skin. In: Million RR, Cassisi NJ, eds. Management of Head and Neck Cancer. A Multidisciplinary Approach. Philadelphia: J.B.Lippincott Company 1994; 643- 91 6. Ballester R, Pons O, Llavador M et al. Depth determination of skin cancers treated with superficial brachytherapy: ultrasound vs histopathology. J Contemp Brachyther 2014; 6(4): 417-423. 7. Alam M, Nanda S, Mittal BB, Kim NA, Yoo S. The use of brachytherapy in the treatment of non-melanoma skin cancer: a review. J Am Acad Dermatol. 2011; 65(2):377-88. 8. Sabbas AM, Kulidzhanov FG, Presser J, et al. HDR Brachytherapy with surface applicators: Technical considerations and dosimetry. Technol Cancer Res Treat 2004;3:259-267. 9. Evans MDC, Yassa M, Podgorsak EB, Roman TN, Schreiner LJ, Souhami L. Surface applicators for high dose rate brachytherapy in aids-related Kaposı sarcoma. Int J Radiat Oncol Biol Phys 1997;39:769-74. 10. Hwang IM, Lin SY, Lin LC, Chuang KS, Ding HJ. Alternative effective modality of Leipzig applicator with an electron beam for the treatment of superficial malignancies. Nuc Inst Meth A 2003;508:460-6. 11. Niu H, His WC, Chu JCH, Kirk MC. Dosimetric characteristics of the Leipzig surface applicators used in the high dose rate brachytherapy. Med Phys 2004;31:3372-7 12. Perez-Calatayud J, Granero D, Ballester F, Puchades V, Casal E, Soriano A, et al. A dosimetric study of the Leipzig applicators. Int J Rad Oncol Biol Phys 2005; 62:579-84. 13. Granero D, Perez-Calatayud J, Jimeno J, et al. Design and evaluation of a HDR skin applicator with flattening filter. Med Phys 2008; 35:495-503. 14. Granero D, Perez-Calatayud J, Ballester F, Ouhib Z. Radiation leakage study for the Valencia applicators. Phys Med . 2013; 29:60-4. 15. Candela-Juan C, Niatsetski Y, van der Laarse R,et al. Design and characterization of a new high-dose-rate brachytherapy Valencia applicator for larger skin lesions. Medical Physics 2016; 43(4):1639-1648. 16. Fulkerson, R.K., Dosimetric characterization of surface applicators for use with high dose rate Ir-192 and electronic brachytherapy sources, 2012, University of Wisconsin-Madison. p. 165. 17. Rong Y, Welsh JS. Surface applicator calibration and commissioning of an electronic brachytherapy system for non-melanoma skin cancer and treatment. Med Phys 2010; 37: 5509-5517. 18. Bhatnagar A. Nonmelanoma skin cancer treated with electronic brachytherapy: Results at 1 year. Brachytherapy 2013; 12: 134-140. 19. Garcia-Martinez T, Chan JP, Perez-Calatayud J, Ballester F. Dosimetric characteristics of a new unit for electronic skin brachytherapy. J Contemp Brachyther 2014; 6(1) 1-9. 20. Vijande J, Ballester F, Ouhib Z, Granero D, Pujades-Claumarchirant MC, Perez-Calatayud J. Dosimetry comparison between TG-43 and Monte Carlo calculations using the Freiburg flap for skin high-dose-rate brachytherapy. Brachytherapy 2012; 11:528-35. 21. Ozyar E, Gurdalli S. Mold brachytherapy can be an optional technique for total scalp irradiation. Int Radiat Oncol Biol Phys . 2002; 54:1286.