Too small to fail.
Top Line: Delivering high-dose single-fraction radiation to subcentimeter targets that can move on a swivel can be tricky.
The Study: Enter these international consensus best practices for radiosurgery for brain metastases less ≤1 cm. First up, imaging. A MRI acquired 10-15 minutes after contrast with max slice thickness of 1.5 mm should be fused to a CT simulation with slices equal to or thinner than that. And none of this should happen no more than 7 days prior to treatment—this part seems a bit ambitious, but we can all agree it should be as close to treatment as feasible. Also ideal is regular MRI quality assurance (QA) to minimize geometric distortion. On to contouring. To no surprise, there was an even split on whether or not any margin should be used. CTV-naysayers point to the partial volume effect, “where an individual voxel appears bright due to the presence of enhancing tumour in only part of its volume…which can in-effect add an unintended margin to the target.” PTV-naysayers point to the fact that a 1-mm expansion can double the volume of a subcentimeter target which can considerably increase risk of radionecrosis when treating multiple targets in close proximity. In the end, everyone can agree it should depend on individual machine QA reports on geometric and dosimetric accuracy. Finally, the tech. Interestingly, there was considerable disagreement on whether 5-mm or 2.5-mm multi-leaf collimators are better: “The only definitive conclusion we can derive [from the literature] is that an MLC width of no more than 5 mm is a prerequisite for SRS treatments.” The authors also highlight that, while using a higher energy (10 MV v 6 MV) flattening-filter-free setting drastically reduces treatment times, it also increases normal brain dose due to a wider penumbra, which is of particular concern when treating multiple small field sizes. Last, if you’re going to use a single-iso technique for multiple targets, a 6-degree-of-freedom couch with on-board imaging of patients at different couch angles is highly recommended.
TBL: We did our best to hit the high points, but these guidelines are worth a read E2E. | Grishchuk, Pract Radiat Oncol 2022