GEC-ESTRO ACROP skin brachytherapy recommendations
TL;DRDepth sets the technique: contact BT for skin tumours <5 mm deep, interstitial for ≥5 mm or curved surfaces; no standard dose schedule endorsed.
The dosimetric fine print gates safety: with the Valencia applicator prescribed at 3 mm, skin surface dose runs around 135%, and omitting the plastic cap raises skin dose 2.5× (Valencia) or 10× (Leipzig). Recording surface dose, not only prescription-depth dose, is what lets late toxicity be compared across schedules.
In a NMSC under 5 mm deep on a flat or small surface, this supports contact BT (mould, flap or shielded applicator) as the technique; it does not extend to lesions 5 mm or thicker or on curved facial surfaces, where interstitial implant is the stated option.
Lesion depth drives the technique: contact BT below 5 mm without anaesthesia, interstitial at 5 mm or more or on curved facial surfaces, with catheters at least 3 mm under the skin. Surface dose exceeds the prescription in every contact technique, so documenting it is the practical change for an RT program building a skin BT service.
10 details
GEC-ESTRO ACROP recommendation document (Radiother Oncol 2018) drawing on published series and an ABS report; no new outcome data and no formal grading of evidence stated in source.
Contact techniques prescribe to 3–5 mm (moulds/flaps) or 3–4 mm (shielded applicators); interstitial catheters sit 8–12 mm apart and at least 3 mm under the skin. Shielded applicators need at least 5 mm GTV→CTV plus 1–2 mm PTV, and a CT with surface markers is the standard for moulds and implants.
The anchor cited is the ABS report summarizing 19 published protocols, with local control in the 90–100% range. Head-to-head data vs surgery are limited to a nasal vestibule HDR comparison (no difference in locoregional control or survival, better function and cosmesis) and an electronic BT vs Mohs matched-pair cohort.
Schedule grid mixes LDR, HDR and electronic sources with different gradients (8% vs 12% per mm), so totals are not interchangeable. BED is set aside for surface work because overall treatment time varies, leaving no common dose currency.
The firm content is geometric and dosimetric; the dose content is a menu, not a ranking. Hypofractionation in frail pts is accepted at the cost of reduced cosmesis, which is the trade a reader is actually making when choosing once-weekly schedules.
| Technique | Indication | Prescription depth | Geometry |
|---|---|---|---|
| Surface moulds / flaps | Regular surfaces; flaps for larger volumes; custom moulds for nose, pinna, hand, fingers | 3–5 mm | Source-to-skin distance 5 mm |
| Shielded applicators (radionuclide or electronic) | Small lesions on regular surfaces | 3–4 mm | GTV→CTV at least 5 mm, plus 1–2 mm PTV |
| Interstitial | Thickness 5 mm or more, irregular surfaces | PTV equals CTV | Catheters 8–12 mm apart, at least 3 mm under skin |
| Histology | CTV margin |
|---|---|
| BCC, well defined | 5 mm |
| BCC, poorly defined, large, infiltrative, morphoeic or sclerotic | 7–10 mm |
| SCC <20 mm | 10–15 mm |
| SCC >20 mm | 15–20 mm |
| Merkel cell | 20–30 mm peripheral; deep 5–10 mm below visible extension |
| Post-operative | Whole surgical site; margins per histopathology |
| Applicator | Dose/fx | Fractions | Frequency | Total |
|---|---|---|---|---|
| Moulds / flaps | 3 Gy | 17–18 | 3×/wk | 51–54 Gy |
| Moulds / flaps | 4 Gy | 10–12 | 3×/wk | 40–48 Gy |
| Moulds / flaps | 5 Gy | 10–12 | 2×/wk | 50–60 Gy |
| Moulds / flaps | 5 Gy | 8 | Twice a day, daily | 40 Gy |
| Leipzig | 3 Gy | 17–18 | 3×/wk | 51–54 Gy |
| Leipzig | 5 Gy | 8 | 2×/wk | 40 Gy |
| Leipzig | 10 Gy | 3 | 1×/wk | 30 Gy |
| Valencia / Esteya / Xoft | 7 Gy | 6 | 2×/wk | 42 Gy |
| Valencia / Esteya / Xoft | 5 Gy | 8–10 | 2×/wk | 40–50 Gy |
| Modality | Mean total dose | Mean recurrence |
|---|---|---|
| External | 13.5 ± 3.3 Gy | 22.2 ± 16% |
| HDR BT | 13.7 ± 2.6 Gy | 10.5 ± 15% |
| LDR BT | 19.3 ± 1.2 Gy | 21.3 ± 2.1% |
- Optimal dose per fraction for skin surface brachytherapy
- Randomized comparison of skin brachytherapy vs surgery
- Long-term cosmesis of large-fraction schedules in frail pts
📚 Sources · 📄 1 paper
The longer read
This document reads best as a technique manual issued under a guideline banner. Its firm content is geometric and dosimetric: which applicator family fits which lesion depth and surface, how far catheters sit from the skin, how wide the CTV extends by histology, and what happens to skin dose when setup details slip. Its dose-fractionation content is deliberately soft. The authors say outright that no standard schedule can be recommended and that prescriptions rest more on experience than on evidence, then list the regimens in common use. A reader should treat that schedule grid as a catalogue of published practice, not as a ranked recommendation, and should not infer equivalence between rows simply because they share a table.
There is little to set this against in the trial sense. The authors concede that randomized trials, systematic reviews and meta-analyses are missing and that most series come from single institutions with small numbers. The nearest anchor is the ABS report summarizing 19 protocols, with local control between 90 and 100%. Those figures come from selected cohorts, often elderly, treated by experienced centres, and a less selected population with thicker or poorly defined lesions would plausibly sit lower. The comparisons with surgery are thin: a nasal vestibule HDR series showing no difference in locoregional control or survival but better function and cosmesis, and an electronic BT versus Mohs matched-pair cohort. Matching is not randomization; lesion size, site and histologic subtype can steer pts toward either arm. On eyelid tumours the document itself names surgery as the gold standard and positions BT for cases where functional structures cannot be preserved.
The most consequential methodological point sits in the dose-calculation section, not the conclusions. Every surface technique delivers more at the skin than at the prescription point, and the excess depends on bolus thickness, prescription depth and gradient. With the Valencia applicator prescribed at 3 mm, surface dose runs around 135%; gradients differ between devices (8% per mm for Esteya, 12% per mm for the others); and a missing plastic cap multiplies skin dose 2.5-fold or 10-fold. Two centres using the same nominal schedule can therefore deliver materially different skin doses, which confounds any cross-series comparison of cosmesis or late toxicity. That is why the request to record surface dose matters more than any single regimen: without it, published control and cosmesis rates cannot be mapped onto a dose-response.
What the document changes is selection and setup: contact BT without anaesthesia for tumours under 5 mm deep, interstitial implants at 5 mm or more or on curved facial surfaces, histology-based margins borrowed for SCC from NCCN, and same-day post-excision HDR for keloids. What it leaves untouched is whether BT should displace excision for a resectable lesion, which dose per fraction best balances control against long-term cosmesis, and how much cosmetic cost is acceptable when hypofractionating for frail pts. The guidance would be wrong in practice if large-fraction schedules such as 10 Gy weekly or 7 Gy twice weekly produced late skin changes that protracted schedules avoid; late skin effects take years to emerge, and nothing in the document establishes long-term cosmesis by schedule.