Gamma Knife Planning and Dose: MRI, the Team and the Marginal Dose

Gamma Knife dose planning is a live session on the day of treatment in which a neurosurgeon, a radiation oncologist and a medical physicist build your dose directly from your own MRI and CT scans, setting a marginal dose to the edge of the target that is chosen condition by condition. Nothing is decided in advance in a vacuum: the plan is shaped around your anatomy while you wait, which is why it takes one to several hours¹.

When I had my acoustic neuroma treated, I understood the frame and the machine but not the hours in between, when three people I never really met were working out where every beam would land. The planning is the part that makes radiosurgery precise, and it is also the part that is least explained. This is what actually happens, and how the numbers are chosen. For the whole journey, start with Gamma Knife radiosurgery; for the physics underneath the plan, see how Gamma Knife works.

How the dose is planned from your scans

The plan is built directly on your own imaging: an MRI, usually combined with a CT, gives the exact position and shape of the target, and the beams are then arranged to match it. The imaging on the day takes about 30 minutes, and for an arteriovenous malformation an angiogram is added to map the abnormal vessels¹. Because the frame (or the mask) is in place during the scan, the images carry a fixed coordinate system, so a point on the picture corresponds to a precise point in your head.

From those images the team outlines the target and marks the structures to protect, such as the optic nerves or brainstem. The software then places convergence points, called isocentres, so the combined dose fills the target shape. The beams meet to an accuracy of under about 0.5 mm, which is the tolerance that lets a target sitting right against a nerve be treated at all². If you want the coordinate system explained through the hardware that creates it, see the Gamma Knife frame and frame-based versus frameless radiosurgery.

Who is in the room: the planning team

Three professionals plan the dose together: a neurosurgeon, a radiation oncologist and a medical physicist, and the plan is not delivered until all of them are satisfied. The neurosurgeon and the radiation oncologist define the target and the dose limits on your scans, deciding what must be hit and what must be spared; the medical physicist builds and verifies the plan that delivers it. They work while you wait, which is precisely why planning takes one to several hours rather than minutes¹.

I found it oddly reassuring, afterwards, to learn that no single person had signed off my dose alone. My contribution to that hour was to sit very still with a frame on and let them argue it out in a room I never saw. For the wider question of who decides that radiosurgery is right in the first place, see am I a candidate for Gamma Knife, and for what to raise with them beforehand, questions to ask before Gamma Knife.

The marginal dose and the isodose line

The marginal dose is the dose delivered to the outer edge of the target, prescribed to a chosen isodose line, most often the 50% line, so the edge receives about half of the peak dose at the centre. An isodose line simply joins the points that get the same percentage of the maximum dose. Prescribing to the 50% line gives a steep fall-off: the dose drops away sharply just outside the target, which is what protects the healthy brain around it³.

This is the single number people mean when they ask “how much radiation did you have”. It is measured in gray (Gy), and it is set condition by condition rather than by a one-size figure. The centre of the target receives more, but the marginal dose is what guarantees the whole target is covered while the surrounding tissue is not. The trade-off is always the same: enough dose to control the target, little enough beyond it to spare what matters. For how that dose then acts over time, see Gamma Knife results and follow-up.

Dose examples by condition

Marginal doses follow published ranges tuned to the aim of treatment, so a benign tumour is treated with a modest dose while an AVM or a trigeminal nerve root needs a much higher one. The team chooses within these ranges based on the size, position and type of your target. The common examples are:

• Acoustic neuroma (vestibular schwannoma): margin dose most commonly 12 to 13 Gy, giving 5-year tumour control of roughly 90 to 99%, with facial-nerve preservation of about 95 to 100% at 5 years⁴. See Gamma Knife for acoustic neuroma.
• Meningioma (benign, grade I): margin dose 12 to 16 Gy, with 5-year local control of roughly 85 to 100% (median about 94%). See Gamma Knife for meningioma.
• Brain metastases: the dose is strongly linked to control, with about 92% at 21 Gy or more, falling to about 72% at 16 to 20 Gy. See Gamma Knife for brain metastases.
• Arteriovenous malformation (AVM): margin dose commonly about 18 to 25 Gy, giving obliteration of roughly 65 to 90% over a latency of 2 to 3 years⁵. See Gamma Knife for AVM.
• Trigeminal neuralgia: about 70 to 90 Gy (commonly 80 Gy) to the nerve root, with initial pain relief of roughly 70 to 90%. See Gamma Knife for trigeminal neuralgia.
• Pituitary adenoma: margin dose 12 to 15 Gy for non-functioning tumours and higher, about 20 to 30 Gy, for hormone-secreting ones, with tumour control over 90%. See Gamma Knife for pituitary adenoma.

My own dose sat in the acoustic-neuroma range, and seeing where it fell within that band made the follow-up scans easier to read: the team had not pulled a number from nowhere, they had placed mine on a known curve.

How planning limits the risks

Careful planning is the main way the delayed risks of radiosurgery are kept low, because the dose to healthy tissue and to nearby nerves is a direct result of how the plan is built. The most important delayed effect is radiation necrosis, roughly 5 to 25% depending on the size of the target (particularly over 20 mm) and the dose¹. A tighter plan, with a steep fall-off outside the target, is what keeps that risk towards the lower end.

That is also why targets very close to the optic nerves or brainstem may be treated with the dose split over a few sessions rather than in one: fractionating gives healthy structures time to recover between doses. Planning is not a formality; it is where the balance between controlling the target and protecting you is actually struck. For the full risk picture, see Gamma Knife risks and side effects and radiation necrosis, what I learned.

The planning wait, from the inside

Planning is the long, quiet middle of the day, one to several hours during which you wait, awake and receiving no radiation, while the team builds and checks your plan. You feel nothing, because planning is a computer task done on your scans away from the machine¹. The treatment that follows is often quicker than the planning that precedes it.

For me this wait was the strangest hour of the whole day: frame on, nothing to do, no way to hurry the people working out where my beams would go. Knowing in advance that this pause is normal, and that it means care rather than delay, would have helped me enormously. If you want the day mapped out end to end, see the day of Gamma Knife, hour by hour and, for the wait between the later scans, radiosurgery and scanxiety.