Stereotactic Radiosurgery · Technical Foundations
Radiobiology of High Dose per Fraction
Why a single ablative dose behaves differently from thirty small ones
Conventional radiotherapy is built on fractionation — spreading dose over many sessions to exploit the differential repair of tumor and normal tissue. Radiosurgery deliberately abandons that, delivering one or a few very large fractions. The radiobiology that justifies conventional fractionation does not cleanly extrapolate to these doses, and understanding where it breaks down — and what mechanisms take over — is what separates principled dose selection from recipe-following.
Orientation
The classical "five Rs" of fractionated radiotherapy — repair, reassortment, repopulation, reoxygenation, and intrinsic radiosensitivity — explain why dividing a course into many small fractions spares late-responding normal tissue while still controlling tumor. Radiosurgery does the opposite: it concentrates dose so that the steep physical gradient, not fractionation, protects the surrounding brain. At these doses, additional biology appears — vascular and stromal injury, and possibly immune effects — that the standard model was never built to capture. The practical upshot is that radiosurgical doses are anchored in clinical dose-response data (RTOG 90-05 and disease-specific series), not in naive linear-quadratic extrapolation.
The Linear-Quadratic Model and Its Limits
1.What the LQ model says
The linear-quadratic (LQ) model describes cell survival as a function of dose d through two terms: a linear component (αd, single-hit lethal damage) and a quadratic component (βd², accumulated sublethal damage). The α/β ratio — high (~10 Gy) for early-responding tissues and most tumors, low (~2–3 Gy) for late-responding normal tissues — captures why fractionation spares late-responding tissue: small fractions sit on the shallow part of the survival curve for low-α/β tissue. The model, and derived quantities like the biologically effective dose (BED), are the backbone of conventional radiotherapy planning.
2.Where it strains at radiosurgical doses
Most of the data underpinning the LQ model come from in-vitro survival curves at doses far below those used in radiosurgery. Extrapolated to 15–24 Gy single fractions, the LQ model's continuously bending curve tends to over-predict cell kill (and conversely can mis-estimate normal-tissue effect). Kirkpatrick and colleagues argued prominently that the LQ model is inappropriate for the high-dose-per-fraction regime, while Brown, Brenner and others countered that the LQ model remains reasonable up to roughly 15–18 Gy per fraction and that invoking new mechanisms is often unnecessary. The honest position is that this is genuinely contested: the LQ model is a useful approximation that becomes progressively less reliable as dose per fraction rises, and clinicians should treat BED comparisons across very different fractionations with caution rather than as exact equivalences.
| Tissue / endpoint | Approx. α/β | Implication |
|---|---|---|
| Most tumors, early-responding tissue | ~10 Gy | Relatively insensitive to fraction size; control tracks total dose |
| Late-responding normal CNS (brain) | ~2–3 Gy | Spared by small fractions; vulnerable to large single doses |
| Spinal cord (myelopathy) | ~2 Gy | Low tolerance drives strict cord Dmax limits in spine SRS |
| Optic apparatus | ~1.6–2 Gy | Steep dose-response near threshold → single-fraction ~8–10 Gy cap |
| AVM / vascular endothelium | n/a (vascular mechanism) | Above ~8–10 Gy/fx, ceramide-pathway endothelial injury dominates |
Mechanisms Beyond the Five Rs
3.Vascular and stromal injury
A defining feature of ablative single-fraction biology is endothelial and microvascular damage. Garcia-Barros and colleagues showed in 2003 that tumor response to high single doses depends substantially on endothelial cell apoptosis driven by the acid sphingomyelinase/ceramide pathway — a mechanism that becomes prominent only above roughly 8–10 Gy per fraction and is largely absent at conventional fraction sizes. The resulting vascular shutdown contributes to the durable obliteration seen in arteriovenous malformations and to tumor control beyond what classical clonogenic killing would predict. Whether this vascular mechanism is the dominant driver of tumor control or a contributor remains debated, but its existence is the clearest reason the conventional model is incomplete at these doses.
4.Reoxygenation, repopulation, and the immune question
Single-fraction treatment forgoes reoxygenation between fractions, so a hypoxic subpopulation may be relatively protected — one rationale some cite for a few-fraction regimen in larger or hypoxic targets. With a single session there is no meaningful tumor repopulation during treatment, removing the accelerated-repopulation concern of protracted courses. High-dose radiation also produces immunomodulatory effects (release of tumor antigens, interferon signaling, and the much-discussed but clinically inconsistent abscopal effect); these are an active research area in combination with immunotherapy and are covered, for brain metastases, on the disease pages rather than asserted as established practice here.
From Biology to Dose Selection
5.How doses are actually chosen
Because the models are imperfect at these doses, radiosurgical prescriptions are anchored empirically. The single most cited safety anchor is RTOG 90-05, which established maximum tolerated single-fraction doses by tumor diameter — 24 Gy for ≤ 20 mm, 18 Gy for 21–30 mm, and 15 Gy for 31–40 mm — in previously irradiated patients, and demonstrated that larger targets carry steeply higher toxicity. Disease-specific marginal doses (for example, 12–13 Gy for vestibular schwannoma, or 16–25 Gy depending on AVM size and eloquence) come from outcome series, not from BED calculation. When a target is too large or too close to a critical structure to treat safely in one fraction, the response is hypofractionation (typically 3–5 fractions), which trades some of the single-fraction vascular effect for improved normal-tissue tolerance — a pragmatic, evidence-guided compromise rather than a model-derived certainty. Normal-tissue tolerances for these regimens are compiled in HyTEC and AAPM TG-101 and applied on the planning and constraints page.
Key points
- Conventional fractionation exploits the five Rs and the α/β difference; radiosurgery instead relies on the physical dose gradient for normal-tissue sparing.
- The LQ model over-predicts cell kill at ablative doses; its validity above ~15–18 Gy/fraction is genuinely contested (Kirkpatrick vs Brown/Brenner) — treat cross-fractionation BED comparisons cautiously.
- Above ~8–10 Gy/fraction, endothelial/microvascular injury via the ceramide pathway (Garcia-Barros) contributes to tumor and AVM response beyond classical clonogenic killing.
- Single fractions forgo reoxygenation (hypoxia caveat) and avoid intratreatment repopulation; immune effects are real but clinically inconsistent.
- Doses are anchored empirically: RTOG 90-05 (24/18/15 Gy by size) and disease-specific series, not naive model extrapolation; large/eloquent targets → hypofractionation.
References
- Shaw E, Scott C, Souhami L, et al. Single dose radiosurgical treatment of recurrent previously irradiated primary brain tumors and brain metastases: final report of RTOG protocol 90-05. Int J Radiat Oncol Biol Phys. 2000;47(2):291–298. PubMed
- Kirkpatrick JP, Meyer JJ, Marks LB. The linear-quadratic model is inappropriate to model high dose per fraction effects in radiosurgery. Semin Radiat Oncol. 2008;18(4):240–243. PubMed
- Brown JM, Carlson DJ, Brenner DJ. The tumor radiobiology of SRS and SBRT: are more than the 5 Rs involved? Int J Radiat Oncol Biol Phys. 2014;88(2):254–262. PubMed
- Garcia-Barros M, Paris F, Cordon-Cardo C, et al. Tumor response to radiotherapy regulated by endothelial cell apoptosis. Science. 2003;300(5622):1155–1159. PubMed
- Grimm J, et al. (HyTEC). High Dose per Fraction, Hypofractionated Treatment Effects in the Clinic (HyTEC): an overview. Int J Radiat Oncol Biol Phys. 2021. PubMed
Educational synthesis for neurosurgery and radiation-oncology trainees. The adequacy of the LQ model at ablative doses is an area of legitimate scientific disagreement, presented as such. Radiobiology references verified against PubMed during review.