“Siebstrahlung” – An Old Idea with New Beams: Spatially Fractionated Radiotherapy with Proton Minibeams

More than a century ago, early radiotherapists discovered that irradiating through metal grids or sieves (“Siebstrahlung”) could reduce skin damage. The idea largely faded once megavoltage beams offered intrinsic skin sparing. In recent decades it has re-emerged in new forms of spatially fractionated radiotherapy, such as GRID, microbeam and minibeam radiotherapy (MBRT). In proton minibeam radiotherapy (pMBRT), narrow proton beamlets spaced a few millimetres apart produce sharply modulated dose patterns that have shown striking normal-tissue sparing in preclinical studies. Delivering such fields with a clinical proton system, and measuring them reliably, is a physics challenge in itself: steep sub-millimetre dose gradients push collimator design, detector resolution and Monte Carlo modelling to their limits.

Quantifying the biological effect is equally difficult, because conventional dose metrics lose their meaning in highly heterogeneous fields. To allow fair comparisons, we introduced an outcome-based framework that defines the therapeutic window directly through measurable endpoints such as tumour control and normal-tissue complication probabilities.

At the Danish Centre for Particle Therapy (DCPT), we have built dedicated minibeam collimators for our clinical proton system and performed in-vivo studies. Normal tissue tolerated almost twice the tumour dose that caused severe injury under conventional proton irradiation, without measurable acute toxicity, while tumour control was maintained. This places proton spatial fractionation among the most pronounced normal-tissue sparing effects reported in radiobiology.

Within the PRESTO consortium of the Euratom PIANOFORTE partnership, we will systematically quantify the spatial-fractionation effect with protons, identify optimal collimator configurations, and build radiobiological models on the experimental data, with the aim of translating pMBRT into clinical practice. Looking further ahead, DCPT has proposed variable collimator systems that change the beamlet spacing dynamically, for example using a moiré-based dual-collimator design. Recent preliminary results will also be presented.