Stereotactic Radiosurgery · Follow-Up

Surveillance & Response Assessment

What to measure, when enlargement is not failure, and when uncertainty requires action

Radiosurgery ends in one day; interpretation lasts for years. This page organizes post-treatment surveillance across malignant, benign, vascular, functional, endocrine, and spine indications, with a practical framework for separating expected evolution from progression and adverse radiation effect.

Timoteo Almeida, MD, PhD | Departments of Neurological Surgery and Radiation Oncology

Orientation

After SRS, enhancement is not the same thing as viable tumor, size is not the same thing as clinical success, and one MRI is rarely a verdict. The correct interpretation combines the treatment goal, pretreatment trajectory, dose distribution, timing, symptoms, steroid requirement, systemic disease, and serial imaging. Follow-up must be planned at treatment, not improvised when the first scan looks uncomfortable.

Part I

The Universal Follow-Up Frame

1.Follow the endpoint you treated

A metastasis is treated for local control and avoidance of intracranial morbidity; a vestibular schwannoma for growth arrest and cranial-nerve preservation; an AVM for angiographic obliteration; a functioning PitNET for biochemical remission; trigeminal neuralgia for pain control; and a spine metastasis for durable local control, pain relief, mechanical stability, and neurologic preservation. Imaging is only one component of response, and the relevant clinical endpoint changes by disease.

2.Compare like with like

Use the same thin-slice postcontrast MRI technique when possible, review the original planning images and dose distribution, and measure the same lesion in the same planes. Note steroid dose, neurologic status, seizure control, hearing or endocrine testing, and systemic therapy at each time point. For spine, compare MRI with CT when fracture, collapse, or hardware obscures the picture. A radiology report without the treatment plan and clinical timeline may describe change accurately while interpreting it incorrectly.

Five questions for every suspicious scan When did the change appear? Is it inside the high-dose region? Is the patient clinically worse? Is edema or steroid requirement changing? Is the trajectory convincing on more than one time point? If these answers do not cohere, do not force a binary diagnosis.

Part II

Surveillance by Indication

3.Representative schedules and endpoints

Intervals below are common starting patterns, not universal rules. Shorten surveillance for symptoms, aggressive biology, uncertain imaging, high intracranial failure risk, or an actionable salvage option; lengthen it only when disease biology, prior stability, and patient goals support doing so.

Post-SRS surveillance is disease-specific and continues beyond the first reassuring scan.
Indication	Typical early surveillance	What defines success	Common trap
Brain metastases	Brain MRI about every 2–3 months initially, least negotiable after SRS alone, where distant failure is the expected event and salvage depends on catching it early	Local control plus detection of new distant brain disease while salvage remains feasible	Calling treatment effect progression from one enlarging scan
Resection cavity	Early postoperative MRI for baseline, then close MRI surveillance	Cavity control and detection of nodular or leptomeningeal recurrence	Confusing evolving cavity geometry with marginal recurrence
Vestibular schwannoma	MRI at roughly 6–12 months, then annually before extending, with formal audiometry on the same schedule while hearing is serviceable	Long-term growth arrest, hearing and cranial-nerve preservation	Intervening on transient enlargement without serial confirmation
Meningioma	MRI at about 3, 6, and 12 months, then annually while stable	Growth arrest and, for skull-base tumors, cranial-nerve preservation	Reading early volume change in either direction as a settled answer
Pituitary/PitNET	Serial sellar MRI plus lifelong annual endocrine testing, since new deficits accrue for years after the tumor is radiographically stable	Tumor control and, for functioning tumors, biochemical remission	Stopping endocrine surveillance because the MRI is stable
AVM	MRI with MRA about every 6 months through the 2–3 year latency period; catheter angiography usually between 1 and 3 years once MRI suggests obliteration	Angiographic obliteration, then long-term surveillance for rare recurrence	Equating reduced flow or MRI improvement with proven obliteration
Trigeminal neuralgia / functional	Clinical review; imaging when symptoms or lesion evolution warrant it	BNI pain outcome, medication reduction, sensory and neurologic safety	Declaring failure before the expected latency has passed
Spine SBRT	Clinical review and MRI commonly every 2–3 months initially, then individualized	Local control, pain response, neurologic preservation, and mechanical stability	Missing fracture/instability while focusing only on tumor signal

Part III

Response, Progression, and Treatment Effect

4.Brain metastases: RANO-BM is a framework, not a substitute for judgment

RANO-BM integrates target-lesion size, nontarget disease, new lesions, steroid use, and clinical status. For trial measurement, a lesion generally must be at least 10 mm to be measurable, which means many SRS-treated metastases are below formal thresholds. In daily practice, volumetric change can be more sensitive, but reproducibility and segmentation method matter. New distant lesions indicate intracranial progression even when every treated lesion is controlled; that distinction is central to an SRS-alone strategy.

After SRS, a treated metastasis may enlarge, enhance more intensely, or develop edema without viable tumor growth. Conventional MRI cannot always separate progression from adverse radiation effect. Serial morphology, perfusion MRI, spectroscopy, diffusion, and amino-acid PET can shift probability; pathology remains the reference when the answer will change management and noninvasive studies remain equivocal. Amino-acid PET now has dedicated response criteria for brain metastases, PET RANO BM 1.0, intended chiefly for trials and equivocal cases rather than routine surveillance. Timing also carries diagnostic weight: transient post-SRS enlargement can appear as early as 6 weeks and persist beyond 15 months, and enlargement first appearing more than 12 to 15 months after radiosurgery more often reflects treatment effect, whereas true progression tends to enlarge earlier in the course. Timing shifts probability rather than settling the question.

5.Benign tumors: temporary enlargement can be part of control

Vestibular schwannomas and some meningiomas may enlarge transiently after SRS, sometimes with central loss of enhancement or cystic change. Transient expansion of a vestibular schwannoma is typically seen about 6 to 18 months after SRS and may persist for months to years before regressing, commonly by 2 to 5 years. Meningioma pseudoprogression is reported in roughly 9 to 22% of treated tumors, peaks around 6 to 8 months, and usually regresses by 12 to 24 months. The converse also occurs: transient shrinkage, or pseudoresponse, has been reported in about 35% of meningiomas that ultimately progress, so early direction of change is weak evidence either way. Stable symptoms and a subsequent plateau or regression support treatment effect; sustained growth across serial studies, progressive cranial neuropathy, or growth outside the treated geometry raises concern for failure. One larger scan in an otherwise well patient is usually a reason to shorten the interval, not an automatic reason for salvage. The ISRS vestibular schwannoma guideline favors observation for asymptomatic enlargement occurring within 3 years of radiation.

6.Spine: control, fracture, and epidural disease are separate axes

SPINO response assessment emphasizes serial MRI and clinical outcomes, but osseous response after SBRT can be complex. Marrow signal change, sclerosis, and transient enlargement may not equal progression. Evaluate the epidural component, vertebral-body integrity, alignment, new compression fracture, pain phenotype, and neurologic examination separately. A controlled tumor can still produce a clinically important fracture, and a painful segment can fail mechanically without radiographic tumor progression.

Do not biopsy or operate on an image alone Escalate when the result will change care: progressive symptoms, serial growth, increasing steroid dependence, threatening mass effect, mechanical instability, or a realistic salvage option. When the patient is stable and the diagnosis remains uncertain, a short-interval study is often the most informative test.

Part IV

When the Scan Changes Management

7.Match the next step to the uncertainty

Likely treatment effect, asymptomatic: observe with a shorter interval; avoid reflexive steroids.
Symptomatic edema: treat symptoms and evaluate for adverse radiation effect; use the lowest effective steroid exposure.
Equivocal viable tumor versus injury: review plan and serial imaging; add perfusion, spectroscopy, or amino-acid PET when available and likely to change care.
Progressive mass effect or diagnostic impasse: consider resection or biopsy when pathology will alter treatment and surgery is appropriate.
Confirmed focal recurrence: select among surgery, repeat SRS/SRT, LITT, systemic therapy, or observation according to anatomy, prior dose, disease biology, and goals.
New distant brain disease: reassess total intracranial burden, velocity of new lesions, systemic control, prognosis, and feasibility of further SRS versus WBRT.
Spine progression or fracture: re-run NOMS/SINS/ESCC rather than treating the scan as a purely radiation-oncology problem.

Key points

Plan surveillance at treatment and tie it to the disease-specific endpoint.
Read serial imaging with the original plan, dose distribution, symptoms, steroid trajectory, and systemic therapy.
RANO-BM separates local target response from new distant brain disease and includes clinical status and steroid use.
Transient enlargement after benign-tumor SRS is common enough that one scan should rarely trigger salvage by itself.
AVM cure requires an obliteration endpoint, usually DSA confirmation; reduced flow is not the same as cure.
After spine SBRT, tumor control, pain, neurologic function, and mechanical stability must be assessed separately.
Short-interval follow-up is an active diagnostic strategy when the patient is stable and the answer is uncertain.

See also The diagnostic workup and treatment of symptomatic adverse radiation effect are developed on the Adverse Radiation Effects page.

References

Lin NU, Lee EQ, Aoyama H, et al. Response assessment criteria for brain metastases: proposal from the RANO group. Lancet Oncol. 2015;16(6):e270–e278. PubMed
Albert NL, et al. RANO criteria for response assessment of brain metastases based on amino-acid PET imaging. Nat Med. 2025;31(5):1424–1430. PubMed
Thibault I, Chang EL, Sheehan J, et al. Response assessment after stereotactic body radiotherapy for spinal metastasis: a report from the SPINO group. Lancet Oncol. 2015;16(16):e595–e603. PubMed
Fabi F, et al. Practical guidelines for spinal stereotactic body radiation therapy posttreatment follow-up: empirical consensus from Quebec oncology centers. Pract Radiat Oncol. 2026. Online ahead of print. PubMed
Tsao MN, Sahgal A, Xu W, et al. Stereotactic radiosurgery for vestibular schwannoma: International Stereotactic Radiosurgery Society practice guideline. J Radiosurg SBRT. 2017;5(1):5–24. PMC
Chen CJ, et al. Hemorrhage and recurrence of obliterated brain arteriovenous malformations treated with stereotactic radiosurgery. Stroke. 2022;53(8):e363–e368. PubMed

Educational follow-up framework for neurosurgery and radiation-oncology trainees; intervals are representative and should be individualized to indication, biology, symptoms, treatment options, and institutional protocol.